Re the comment about developing a large number of CCGT plant, I believe many of the existing open cycle plant will be quickly converted to CCGT once the full details of the ETS are know. Plant such as Braemar in Qld, Uranquinty in NSW were constructed with this in mind.
Even some of the smaller and/or older plants have been designed with the potential for conversion in mind (e.g. Quarantine), though the efficiency of some, post-conversion, might be too low to warrant the investment (compared to the option of newer, bigger, higher-efficiency CCGT).
Your pictures above are very colourful and good for people who know nothing, but those of us in the industry know your analysis is extremely simplified and shallow (eg a)’with holding capacity’ is worng, as their available capacity remained the same – should say output instead of capacity, b) no mention of constrained off output from Tumut due to high import flows into Victoria constraining 65/66 lines etc).
I commend you on taking electricity to the people because it will only become more ‘in their bface’ once CPRS commences. Keep it up, but to maintain industry credibility, please make sure it is correct and not sensationalised (which I know gets the media grabs).
1) You are correct in that we do try to simplify what’s a very complex industry, in order to make it more broadly understandable. At times this does mean that we can omit certain details – whilst we try to ensure this is not the case (especially if the details are core to what we are discussing), we do apologise for cases (such as this) when it does occur.
2) I believe you are referring to what we wrote specifically about what happened on 17th March 2008 (from p19 of this report).We tried to confirm this on email with you, but did not receive a response. In this instance, we recognise that it is possible that the output of the Snowy capacity – moreso Tumut, on the north side of the 65/66 lines could have been restricted as a result of constraints on the lines between these stations – however this was not checked, as we were focusing moreso on SA at the time. (For anyone else interested in learning more about this issue, there’s a mountain of reports and submissions on the issue on the AEMC’s Website, in the past couple of years, leading to their decision to abolish the Snowy region and allocate Tumut into NSW and Murray into Victoria for reasons related to Apol’s point – i.e. that constraints within the Snowy region could impact so dramatically on market outcomes and yet not be very understandable, except to those intimately aware of their detail).
3) We also appreciate your comment about the term, "withholding of capacity".We appreciate that that term is incorrect (because, as you say, the capacity is still offered to the market, just at a higher price). It is for this reason that:
(a) we tried to make this clear in the report (such as on p11 of this report where we state that "about 70% of their generation appears to have been bid above $9900")
(b) we are very careful to use the different term "economic withholding of capacity" to clarify. Based on our experience, this term is one of a few (others include "generator strategic bidding" and "generator market gaming") that have become widely used in the NEM, and more broadly internationally, over the past several years. Though, we infer from your comment, the term is perhaps not as broadly used as we had first thought.
Once again, we do appreciate your comments, as they help us to understand how others view the explanations and comments we provide. We look forward to continuing to enhance the quality of the service we provide, into the future.
Also, we’ve been told that:
(a) There is a discussion on nem-wide demands in section 3.3.6 of the 2008 SOO.
(b) It provides similar results to those quoted in our article and notes a significant difference in the level of diversity between summer and winter.
(c) More infomation on diversity is provided in chapter 13 of the SOO.
“National Electricity Market Management Company spokesman Paul Bird could not say exactly how many homes were affected.
“We instructed the transmission companies in Victoria to reduce the load because of that unusual event,” he said.
Energy and Resources Minister Peter Batchelor said electricity supply was “very tight” but there was “just sufficient power” to get by.
“Subject to no major plant failures and other catastrophic interruptions, we think we’ll get through today, the tightest of all days,” Mr Batchelor said.”
This is a conflict of information with what you have reported – does this mean that the media is covering up the fact that load was shed before basslink tripped due to the lack of reserve?”
We’re still pondering this one, to try to work it out. What we can see from the NEM-Watch snapshot included above is that:
1) The Market Notice about Basslink trip was published at 14:14 (i.e. first) and the LOR load shedding in SA at 14:26 – both of these times are NEM times, we believe.
2) The Market Notice itself refers to load shedding starting at 13:50 (i.e. half hour beforehand). We thought that this may have been a local time in SA, but that does not make sense, either, as SA is ahead of NEM time by 30 minutes in summer.
3) We can also see (snapshot not included) that the load started dropping in SA at 14:00 and that this continued to 15:00 (NEM time).
4) In total there was a reduction of 200MW or so – so it is possible that some of this was not the INVOLUNTARY curtailment, and might have just been a coincident outage in ETSA’s distribution system, which was under strain all day.
Thanks for your excellent analysis of NEM spot market behaviour. It would be even more worthwhile if you could complement this with information and discussion about related derivative trading.
We realise that the wholesale market is only 1/2 of the puzzle, and are working on ways in which we can shed more light on the operation of financial market trading (both OTC and ASX), but are not at a position where we can say much more yet.
The two photos in this article were both taken on the same camera and accurately reflect the differences in switched on lighting in Brisbane at the time.
The reason that they use different exposure times is that the fifteen second exposure shots I took during earth hour were over-saturated by the reflected light from the cloud cover.
The only alteration performed on the photos after they were taken was some slight angle correction due to my camera not being perfectly level when the photos were taken.
I was personally quite surprised at the level of lighting reduction during earth hour. Even after earth hour had finished there were less lights on in the city than there were the night before.
Disclaimer: I am personally skeptical of the benefits of earth hour and did not participate in the event, except to take these photos.
Thanks Brenna,
Unfortunately only get time to do this kind of thing in between our “”real work”” of software development. Hence, posts will continue to tend to be sporadic!
Cheers
Paul
PS – a couple of our friends at NEMMCO have pointed out a couple of things following from this article:
1) Technically, this is the 4th time the CPT has been reached (as both SA and VIC reached the CPT as a result of the heatwave this summer – as we noted here)
2) NEMMCO has provided this briefing paper on the Operation of Administered Pricing in the NEM – amongst other things, it clarifies how the pricing in other regions can be affected, due to the operation of the interconnectors. For instance, in the case above, prices in VIC could be affected because of flows over Basslink, meaning VIC also experiences a defacto cap, in some scenarios.
If NEMMCO has applied the APC and at the end of that business day, for Trading Interval 0400 hrs, the Cumulative Price is;
< CPT, then NEMMCO will not continue the application of the APC
> CPT, then NEMMCO will continue with the application of the APC
If at the Trading Interval 0400 hrs the Cumulative Price is less than the Cumulative Price Threshold, it is of NEMMCO’s opinion that determination of the price exceeding the Cumulative Price Threshold during the next trading day cannot be accurately predicted due to market volatility and as such will not be utilising NER 3.14.2 (c) (3). However, NEMMCO will exercise this option under the appropriate circumstances.
Thanks for the feedback, but I can’t let you get away without at least one comment. You mentioned reviewing the accuracy of the load forecasts. I suggest you have a look at Appendix C.
As a PS to my post above, Andrea has requested we make a donation, instead, to her favourite charity, the Permaculture Research Institute, which we were very pleased to do.
Andrea explains:
Maybe not the standard choice but these guys teach people how to develop sustainable living systems, considering the impacts of all their choices to design the best system for each location.
In that we’re firmly focused on making the world a better place, in our own small way (by making the electricity market more understandable – hence facilitating greater efficiencies, higher long-term sustainability and lower prices) we’re more than happy to help Andrea in this way.
I think Joe or anyone else who makes NEM price comparisons when spruiking benefits or costs for consumers is very brave to do so. The only reliable data source on retail electricity prices for all States in Australia is the CPI survey.
The electricity price indices included therein:
1. are based on incomplete coverage (being capital city only); and
2. do not by definition allow for price level comparisons between different states.
With respect to Joe Tripodi’s contribution at the conference, I would like to add that I was particularly pleased to hear his focus being on ensuring that (through the structure of the sale process so designed) the amount of competition in the market would not be diminished, and may be strengthened.
This was noted on two points in particular:
1) The need to ensure at least one new entrant in the market, and reserving the right to IPO Integral-Eraring if this is not achieved in trade sale.
2) Splitting Delta and MacGen trader rights into 4 parcels. As Joe noted, the co-insurance arrangements he spoke about would be one key to making this plausible (though there were no details provided of the specifics).
Philosophically, as a company we believe that the competitive market is (on balance) a better chance of delivering sustainable benefits to energy users – hence we see this as a positive step.
Might not be an “ideal” way to extract the NSW Government from it’s current conflicts of interest, but stands a good chance of being workable, in principle (though depends on the substance of where responsibilities lie under the gentrader model).
Will be very curious to see:
1) How many foreign bidders actually emerge after the global road-show earlier in the year.
I was speaking to a client earlier today about this event. They indicated that VOLL was hit by a combination of very high temperatures in Sydney combined with the trip of a unit in NSW (I do not remember which one, though from the chart in NEM-Review I have to assume it was one of the Eraring units).
I also got the impression that as this is still September, there is a number of units that are still out for maintainance that would have helped with the load had this occurred in December or January.
I would just like to clarify one of the points I made in this article.
Why does interconnection increase economic efficiency?
In general terms, the more interconnected a network and the larger a market, the more efficient it becomes. Connecting Mount Isa to the rest of the NEM will allow Mount Isa to purchase cheaper power during off peak periods, and will provide strong price signals for demand management during peak times. From the point of view of the generator at Mount Isa, exposure to the NEM makes them able to operate more efficiently as well, by allowing them to take advantage of high price periods.
A quick PS – I missed this Market Notice that AEMO issued this morning, which explains the tightness of supply/demand.
1) Taking one transmission line out of service reduced the transfer capacity into SA by half, hence exacerbating the shortfall.
2) Note the alert level was actually LOR2 (i.e. more significant).
See this notice in the list shown in NEM-Watch above (I just did not pick it up earlier!)
From : AEMO
To : NEMITWEB1
Creation Date : 12/11/2009 07:03:49
——————————————————————-
Notice ID : 28695
Notice Type ID : RESERVE NOTICE
Notice Type Description : LRC/LOR1/LOR2/LOR3
Issue Date : 12/11/2009
External Reference : Actual LOR2 South Australia Region – Thursday 12 November 2009
——————————————————————-
Reason :
AEMO ELECTRICITY MARKET NOTICE
Actual LOR2 South Australia Region – Thursday 12 November 2009
Refer Market Notice ID 28684
Due to the planned outage of the Heywood Moorabool No 1 500kV line in the Victoria Region, a credible contingent event may result in a separation of the South Australia region from the remainder of the NEM.
Load may be shed automatically as a result of low frequency in the South Australia region.
There are sufficient capacity reserves in the South Australia region but the technical characteristics of the capacity reserve generators may delay response to the contingency.
LOR2 declared from 0655 hrs until further notice.
Alan Honecker
Power System Operations Manager
——————————————————————-
END OF REPORT
——————————————————————-
In your description of the 10 year look ahead (ESOO) you have
“Because this document is published only every 10 years, these forecasts are provided in the form..”
You have missed a bit – it is published every year, looking ahead for 10 years … – or something similar.
Might be worth changing to avoid confusion.
Paul
There are several problems with load forecasting and predispatch for the 1 in 10 year event or even more extreme event.
a) AEMO do not appear to adjust the predispatch forecasts for humidity and wind speed. In contrast the Victorian gas industry does. (So AEMO has access to the skills set from VENCorp to make the approariate adjustment to the process.).
Over the years the Reliability Panel in their Annual Market Performance Review present information relating to variations in predispatch forecast and the actual dispatch in Figures 5 to 10 in their draft Report 2009. They appear to have convieniently exclude errors associated with peak loads. Several of the earlier AP Performance reports indicated significant predispatch errors. Each year they claim to be reducing the number and size of errors.
b) There is limited information relating to size and type of airconditioning for less than 160 MWh consumers across the NEM. My local Harvey Norman sold $ 180,000 worth of air-conditioners on the 6th of Feb 2009, the next summer they will add to peak demand along with the new consumer added to the system. Energy Australia is one of the few distribution businesses who penalise consumers for moving from one to three phase.
One can also interpret the data provided by ETSA from their Demand Mangement trial that 75% of the air-conditioners in SA need to be replaced so that they can effectly load control the business!
The analysis above (and what I have added today for the other 4 regions, and linked above as well) just relates to the MT PASA and ST PASA forecasts.
As I noted in my “part2” post, there are a number of things about these forecasts that I don’t fully understand.
The issues you raise with respect to pre-dispatch are additional to these (and relate to an even shorter time-horizon forecast), and I have not spent any time thinking through these.
Certainly I agree with you that one of the largest uncertainties facing everyone in the NEM is the fact that there is no central reference that indicates how much new demand capacity (particularly AC loads) has been added in each distribution substation area from one year to the next. As far as I am aware, no-one really knows the full picture!
1) I have heard some staggering numbers about similar sized sales of AC units from other people as well.
2) Assuming 2.5W/$ (rough guess, perhaps conservative) this would equate to another 450kW of peak demand sold in just one day at one of 25? Harvey Norman stores across Melbourne. That does not count all the other electrical appliance stores as well. The mind boggles when thinking about what this might mean for what the peak demand might be this summer in VIC, should the same events occur.
3) Hence it is certainly a recipe for being unpleasantly surprised, should extreme weather occur.
Finding ‘categoric’ data that quantifies Earth Hour results is difficult, and the variables on any given weekend often mask events as narrow as this. However, taking a look at the last 6 consecutive Saturday demands in each NEM sector does give me a little more clarity as to ‘apparent’ effects.
I wont post the graphs and spreadsheet here, but suffice to say that NSW overall appeared to have a subtle 5%-10% nibble, while Victoria and SA shared a larger, longer 10%-15% bite.
Compared to similar views of previous events it seems that Earth Hour is losing its Mojo in Sydney and waning elsewhere in Oz.
Maybe we’ve already achieved so much lighting efficiency by changing over to CFL’s that we barely register a change from turning off lights anymore.
Enough of my highly scientific analysis – over to the experts.
With Australia’s expanding population it is very hard to imagine that this reduction in emissions would turn into a trend. For this to happen the inevitable increase in demand for electricity would need to be more than offset by improvement in efficiency (unlikely) or by improvement in emission intensity. To my knowledge we haven’t closed any coal plants so presumable the drop in emissions has come from reduced capacity factors. As demand rises capacity factors for coal plants will increase again. New gas plants will reduce the emission intensity further but I would bet that emissions will return above 2008 levels before long.
You’ve given us the numbers but do you know the story behind the numbers? It was a mild day with little indication that such spikes would eventuate so I would be really interested in the factors contributing to this seemingly inexplicable outcome.
well done Paul – an excellent synopsis. You have obviously been paying attention at Roman’s conferences (which I think are very good also). I look forward to reading the next few chapters
cheers Jim
Paul,
A very good start.
One of the big problems with electricity transmission and distribution regulation is the Propose-Respond model that the AER has deveoped and had enshrined in the NEL (I think) whereby the DB’s can put forward whatever they want by way of Capex and Opex and it is up to the regulator to try and determine whether it is valid or not.
Unlike non-monopoly businesses these businesses do not have to demonstrate and capital restraint.
What they seem to do is underspend on capex for the first few years of a 5 year regulatory period, then overspend in the latter years.
They can keep the underspend as an “efficiency gain” and then they get to reset their asset base and recover the over-spend as a P 0 adjustment when the new regulatory period commences.
I believe this is called regulatory gaming.
There has to be a better way.
If users challenge an AER decision then we have to foot a huge legal bill whereas if the DBs do the same they have nothing to lose as any legal costs just get funded by an increase in their Opex which we pay for anyway and on balance they more often than not get a better outcome as the Appeals tribunal has no idea about reality it would seem.
The system is well and truly skewed in favour of the DBs.
I am still trying to get my head around some of the aspects of the concerns you have expressed (plus some others) about the revenue reset process for DNSPs and TNSPs.
It seems to me that, compared to 5 years ago (when it seemed like it was not much of an issue at all) it is certainly top of mind at present.
PS – today the AEMO has published this communication to explain yesterday’s interesting events:
From: AEMO Communications Sent: Friday, 2 July 2010 11:56 AM To: Subject: AEMO Communication No. 379 – NEM – Manifestly Incorrect Input and Over Constrained Dispatch – Thursday 1 July 2010
NEM – Manifestly Incorrect Input and Over Constrained Dispatch – Thursday 1 July 2010
AEMO has determined that dispatch intervals ending 1150, 1155, 1200 and 1205 hrs on Thursday 1 July 2010 contained a manifestly incorrect input. Under clause 3.8.24(a) of the National Electricity Rules, a scheduling error has occurred in those intervals. AEMO’s preliminary view is that the scheduling error is contained to those four intervals and no other intervals were affected by the manifestly incorrect input. Participants affected by a scheduling error may be entitled to compensation under clause 3.16.2 of the NER.
Spot prices were revised in those intervals through the application of clause 3.9.2B of the Rules, which also requires AEMO to publish a report on the price revisions as soon as reasonably practical after the event. This report is currently being prepared.
AEMO also advised through a market notice that dispatch intervals ending 1200 and 1205 were affected by over constrained dispatch and as such the prices are subject to review. A subsequent market notice confirmed the price revisions applied under the manifestly incorrect input process would stand.
For further information, please contact the AEMO Information Centre on 1300 361 011.
Brian Nelson
Senior Manager Electricity Market Operations and Performance
Australian Energy Market Operator Ltd
I will look forward to seeing the final report – and (if time permits) will post it here as well.
In case you were curious, these notes are normally copyright & confidential AEMO, so I ran this past Brian and he said it was ok.
The price hike in 2007 was due to two powerstations reducing output due to a government restriction on taking cooling water. I wonder if privately owned powerstations would have been so quick to cut production in lieue of finding alterate cooling methods? Consequently, it seems that the government owned power stations were the big winners during the drought. Perhaps while the Queensland government retains business interests in the power industry, they should recuse themselves from legislating where such legislation could be seen as favouring their business interests.
It’s like you know what I’m thinking Paul?
Sitting at my desk trawling through 100+ page reports at Melbourne Business School on the final stage of my MBA researching wind and embedded generation and up pops your graphs on the changing generation mix! A quick cut & paste, a bit of text and referencing (of course) and I’ve only got 39,500 words to go. Sweet. Thanks mate. Cheers Marc
I’m speculating whether a privately owned power station would not have found alternate methods for cooling, if it meant that in doing so, they would not have to shut down due to drought inflicted water restrictions.
Hey Paul
Great work on posting this article so promptly!!!
I’m studying power engineering at university, and I regularly visit your site. I really enjoy reading your reports on events within the NEM. Shame it wasn’t so hot down here in Vic, otherwise I also think that the NEM-Wide demand would have been clearly broken. Do you think that with most schools up and running now that the first term has started, the demand is likely to be higher due to extra HVAC load?
Also, do you know if the maximum supply available in S.A is limited by generation / transfer of power into S.A, or limits within the S.A transmission/Dist. system?
Furthermore, I would like to know whether you do a student edition of your NEM-WATCH software, for personal use.
Thanks
– Kieran
A lot of residential load was disconnected very early on Wednesday night (5:30pm where I live) when there wasn’t much wind yet. This would seem to suggest that the Ergon system really is quite fragile. I will need follow this up with Ergon once the dust has settled. With the rediculous hikes in prices they have been charging in recent years, I expect a corresponding improvement in service.
In section 2 you state that the difference in peakiness is primarily a result of weather patterns in the regions. I’d add (as you state in the final bit on Tasmania) it is also quite strongly influenced by the proportion of load being consumed by our major industrials. And maybe rather than Tassie having cooler summers we can say ‘less extreme weather changes’ 🙂 Hey, I’m looking for a job for Tourism Tasmania! And being a bit pedantic – it is not the cooler weather but the lack of air conditioners that impacts on consumption – the cooler weather makes it less important to buy a/c in Tassie but the lower average income in Tasmania also impacts on the number of people with a/c. Just some patriotic thoughts!
Tim
With respect to electricity, Tasmania is a fascinating place. Four big loads account for about 60% of the average demand shown in the table above, all with some flexibility to provide some form of DSR to varying degrees. Predominantly (zero SRMC) hydro with relatively large storage plus a single interconnector that is very large in relation to the demand in the region.
In terms of a tourism ambassador, sign me up as well. Lost count of how many times I have been there, and have enjoyed every minute.
Note that the NEM demand you use (TotalDemand in MMS) is actually a 5-minute forecast rather than an actual, and excludes demand served by significant non-scheduled wind farms in SA. If you include those, and use the half-hourly average demand then you will find that the NEM-wide peak summer demand was 35,568 MW which occurred for HH ended 13:30 on 1 Feb 2011.
The demand data we show in NEM-Watch (and hence use in the competition) is the dispatch target, as you note, for a couple of reasons:
1) The time stamp of the demand target aligns with the time stamp of dispatch pricing and the decision was made back when we first built NEM-Watch (in 2000) to ensure these were aligned.
2) In these early days, there were fewer other measures of demand available, as well.
The other difference you note is between 5-minute data and 30-minute data.
I know that the different measures of demand can confuse NEM observers, so I’ve tried to explain this a bit here.
We’ve made a note to look into what we can do to also include the metered numbers, as well, in some way.
Have you included non-scheduled generation, a classification which grew substantially during those years when the energy growth appears to have slowed?
Two comments/questions on these data:
1. Do you keep track of degree days for winter – say with 20C base temp perhaps? Or at least average ambient (not daily min/max as most commonly reported out of BOM) Degree days give a bit of insight into seasonal variation in heating load (for winter)
2. Is it just possible that the pink batts rebate is also having an impact at the margin? There was a lot of insulation put in over a short period before this program sank. I’d guess the impact would be to mitigate both aggregate heating load, and possibly peak heating load, if only a little.
1) Yes, we do have a record of weather data (including temperatures) – however we don’t currently have the time to do anything with them (such as comparing winters to try to understand why demand is declining). It seemed as if it was roughly a “normal” winter – though we have not analysed the data to back up that sense.
2) I’m sure that the insulation would be having some effect at reducing winter demand (and also summer) – it’s one of many contributing factors we have identified and will post about when we have time. The more challenging question is how much effect they have had – or even how we could determine this!
Just a couple of ideas you may want to add to your list (if you dont have them). Some are a bit left field but you may find something interesting in the data. The correlation between the demand data and:
– ASX Market Cap
– The AUD
– Severe weather events
– Variances to weather forecasts
– Demand Response products available
– Demand based retail contracts signed
– Co- Tri-Generation systems installed
– Number of Business Consumers
– Inverse Correlation to median business electricity prices.
Additional (albeit minor) reasons reducing electricity demand (apologies for the Queensland bias):
– Increased Demand Side Management e.g. Energex’s Cool Change Trial – A 13 per cent reduction in peak energy demand was achieved on the hottest day of the 2009/2010 summer (www.energex.com.au/__data/assets/pdf_file/0015/32352/Cool-Change-Cool-Change-Newsletter-November-2010-with-form-number.pdf)
– Household types and composition. Increasing densification of capital city populations is changing the historic relationship between population growth and electricity demand growth. IPART’s survey of residential energy demand illustrates the difference between house and unit electricity use by number of residents, as well as the impact of adults (>15 years old) versus children.
– Industrial strucure and growth. Low growth in energy intensive industries (e.g. manufacturing) and high growth in service industries (typically low energy intensity) has reduced business electricity demand growth.
– Resdiential high-draw devices. There has been sizeable growth in high-draw devices (primarily air conditioners) over the past 15 years, but the market is now close to saturation point in some categories (what is stopping the 20% of high-income households that don’t have an air conditioner from purchasing one?), which coupled with increased efficiency of high-draw devices, would be expected to reduce growth in electricity demand from these devices (closer to the growth in new dwellings).
Thanks for the thoughts – and apologies for the delay in responding.
I am curious to delve further into the detail, in order to understand the significance of these factors, and others people have noted. When time permits, I’ll be doing this with the database we’ve accumulated of sub-regional demand histories across the NEM – which should allow the benefit of (to some extent at least) looking at the effect in different sectors of the economy, and with respect to different weather patterns, etc….
If we are referring to forecast maximum seasonal demands, it could an overly optimistic forward projection influence mainly by outlandish economic forecast, the need for capx investment into aging network, influx of migrants, the weather, etc, etc.
Notice that post GFC period, the economy did not shrink but grow at a slower pace, which means that the behaviour on energy utilisation has changed. This is inline with what Paul has mentioned on energy efficiency measures and retail price of electricity. However not forgetting that in recently 1-3 years, the weather is cooler not any warmer (especially in Queensland for where i am from).
The numbers I have looked at, to date, have been NEM-wide. To some extent this means that regional patterns (e.g. cooler summers in recent years in QLD) have their effects masked.
There is another rather prosiac reason for a weak energy demand in 2010/11: the natural disasters over the summer, most particularly the Brisbane River flood which supressed Qld demand for over a week. It would be possible to estimate this effect.
On “embedded generation” a few points. Of course, what NEM commentators colloquially describe as “demand” is a misnomer, it is actually a measure of generation, and then only a subset of that-that which is easily measurable. And the “demand” excludes some generators, but includes losses in generating, transmitting and distributing electricity, which all change in turn with the pattern of dispatch!
The easily measured generation is all “scheduled” and “semi-scheduled” generation. This is most generators with connection point capacity>30MW and some larger windfarms built prior to 2010. AEMO publishes the large non-scheduled wind generation separately, so that “demand” can be recovered, but as you correctly point out there are still many other non-scheduled generators for which AEMO has no data, such as PV’s and co-generators below 30MW.
On the terminology, the rules define “embedded generation” as any distribution connected generation (which includes the 400MW Tallawarra!)-so I think what you mean is “non-scheduled”. We have lots of embedded scheduled generation, such as Tallawarra.
The ESI is full of commodities that are difficult to measure in actuality!
As noted to Evan (above) I am keen to delve more into the history of zonal data we have in order to at least be able to identify the extent to which the trend observed on a NEM-wide basis (with the approximation of real demand) is similar across the 26 zones for which we have data NEM-wide.
With so many data sets to analyse, it should be able to provide some insight into the extent to which various factors are at work – e.g. residential -vs- industrial -vs- commercial, north -vs- south, etc…
In terms of estimating demand for the future, I would also like to have time to complete this type of retrospective on forecasts. Perhaps AEMO might include this in the data it releases in 2012?
i don’t have numbers for you but have a book called “Brown Power – A jubilee history of the State Electricity Commission of Victoria” It is a self published work from 1969 and gets back into the halcyon days of brown coal expansion that lead to Loy Yang etc. There are many interesting aspects to this, not least the intent and associated policy and practice of promoting load growth. Linking this with the nature of large thermal plant there was a natural desire and deliberate process of growing what we currently call ‘base load’. Of course you know what baseload is, but it is an artifact of both technology and socio-politcal intent. Baseload is an outcome of historical priorities and triumph of one story (coal) over all competing others. Hence the title of the book.
Long story short, I’m contemplating revisiting this book over the next few weeks. If so, what would your review questions be in light of your current historical investigations?
Jim Castles
Unfortunately I can’t really comment in detail, not having read the original edition to understand its context.
Obviously there has been no new brown coal development since the VicPool market started – though my understanding is that utilisation rates increased, and have remained high. I’m not sure how this fits into the context of the older book.
thanks for providing a bit of fun and entertainment over the Christmas break, and you’re right: I really do need that barbeque!
But you are wrong in saying that I only submitted 9 guesses. In accordance with your Terms and Competitions, I provided a guess for every single day until 2 January 2012. But your analysis only includes guesses until 31 Dec 2011.
You wouldn’t want to deny us of those valuable last two guesses on 1 and 2 January now, would you?
Hahaha, gotta love people who work in the energy compliance area and read the fine print…
An excellent chance to test the theory of http://en.wikipedia.org/wiki/The_Wisdom_of_Crowds
Maybe we should take the average of those with multiple entries and then the average of each individual’s pick and see if the “crowd” is indeed wise?
Tim
It is an intriguing book – it’s one of the ones in our office library that has been read and reviewed by others in the company (and I appreciated the Wikipedia reference).
In updating the results table to correct my omissions, I also included another chart at the bottom designed to draw out, as much as possible, the Wisdom of the Crowds in this instance.
Gee, who would have thought that we’d have such low demand this summer!?! I can see my bbq ambitions going down the drain. But maybe we’ll still get a bit of a summer for the remainder of Feb, and temperatures will hopefully go up a bit…
Some wind energy advocates have argued that it is wind that is driving down spot prices – they like to refer to the “merit order” effect. Paul would you like to comment on this?
When I have more time, I might post a bit more about this.
In the meantime, the simple answer is that everything that’s having an effect of either:
(a) reducing demand, or
(b) netting off any generation that bids above $0/MWh (i.e. thermal generation and other) would be having some effect on lowering prices across the NEM.
My suspicion (without having looked at the numbers in detail) is that the volume of generation from wind is still too small to be having the most significant impact in this respect.
I suspect that any effect of wind might still be overshadowed by such greater impacts as:
1) Lower demand, which has been offset by a range of factors, including those linked to in the article above; and
2) The availability, for a couple years more, of cheap “ramp up gas” from coal seam gas fields in QLD, providing a cost incentive for open-cycle plant to run more heavily loaded than might “normally” be the case.
3) It’s possible (though I would need to check the data) that the higher production volumes from hydro, noted in this article, would be having more impact (because, through significant storage, hydro output can be targeted at times when prices might otherwise be high).
It’s my understanding that part of the enduring challenge with wind is that there is still a degree to which diversity patterns are not so great as to overcome the situation where localised wind patterns are independent of peak demand times, or peak price times. Interestingly, there’s some charts that surprised me for the UK here, showing less diversity than I had envisaged in that part of the world.
That said – through the remainder of the decade, as the increased MRET target continues to deliver more wind farms, the effect of wind on reducing peak prices will increase (all other things being equal).
However, if this was the sole objective of wind farm development, there’d be plenty of people who’d point out that there would be much cheaper ways of achieving the same goal – it would seem to represent a Pyrrhic victory, if done for just that reason.
Hi paul
we are studying the impact of the renewable wind and solar on WECC Western energy coordinating council and PJM e.g. the whole USA and N america. We see a big increas in CC generation and cycling to have 20-50% penetration of wind in the USA/ N america.
stay in touch
Your comment is not clear – are you saying that you’ve seen an increase in cycling of thermal plant attributable to increased penetration of wind & solar?
Thanks Paul for your response. It would be good if someone with your electricity market knowledge could write something on the effects of wind on electricity prices in Australia. A number of renewable energy advocates are pushing the line that wind is driving down wholesale prices. I would like to see an analysis from a less biased source.
The comparison table illustrating the changes does not include the NT. As a resident of the NT I would be interested in seeing the before & after affect of the CT
Pat, Paul,
NT is a fully regulated system so Power and Water Corporation would have to apply to the NT utilities commission for an increase to retailer prices based on the increases to their cost due to the carbon price. Their website is http://www.utilicom.nt.gov.au/ where you may find some more info.
Yes it was effective from 1st July. We were surprised to see it, too – there was probably an update from the AEMO or the Reliability Council, but we missed it…
despite your accurate comment at the beginning “it’s really too early to make any significant observations”, it is still interesting to watch what happens. But, as you say, too early yet to read too much in to it. Will watch though, and hope others do too.
Agreed, Rupert
Just because it’s early days does not make it uninteresting. Indeed, perhaps the opposite is true, which is fine – so long as we don’t run ahead of reality with conclusions.
Paul
Nice work Paul … Interesting as the wind increases it’s voilitility will drive more cycling and higher costs and more emissions. Remember wind is not free. We calculate wind integration costs in terms of extra cycling costs. This can be done for the carbon tax effect too.
Steve Llefton. Steve.lefton@intertekmcom.
Thanks Steve
Have not had time to look, yet, at which of the thermal plant are cycling more because of the intermittency of wind.
Had not thought of the increased emissions (from the cycling thermal plant) that would result from running at sub-optimal heat rate, but it’s not likely to be of the same order as emissions saved from wind production (as it’s not the brown coal plants that are cycling).
Concerns about cost, though, have been raised by others – but more with respect to capex for other build required to provide capacity when the wind is still (whereas you’re talking about increased O&M because of more creep and fatigue degradation etc, I think?).
Paul
Paul
Thanks again for another insight into the changes. With respect to the downturn in generation, we Engineers often overlook the economic perspective of cost elasticity of demand. The first time I was reminded of this was back in the 1980’s when NSW introduced the first inverted tariff. Fifty or so years of increasing public electricity use (thanks to price decreases) suddenly stopped!
That could clearly be part of what is happening now and, if cost (Carbon plus regulated costs) is a larger component than we expect, the downturn could accelerate. Not a win for renewables necessarily, but a win for energy efficiency and conservation.
Thanks Peter,
As you imply, reductions in demand due to higher costs could be significant – and could well be more significant than the effect of embedded generation of any type.
Will look into this further, at some stage.
Paul
Thanks Adam
I will put this on the “things to do” list and return to it when I can.
I suspect that reducing to individual states (or other areas) will highlight a greater individual variation of output from various wind farms.
Paul
Wattclarity – exactly as the name suggests
Would also suggests that rising power costs are a combination of infrastructure building, peaking demand leading to higher cost generation options etc etc. Carbon Tax may come into play with respect to cost increases but the data will reveal that in time.
It is nice to know when people appreciate how any of our products do help to “make the energy market more understandable”.
In relation to energy costs, you might be interested in a series of posts I started 2 years ago about the concerns of energy users, which included some indication of the drivers of higher costs that were apparent even at that time.
You’re not alone in pointing a finger at the rising cost of TUOS and DUOS charges, resulting from increased Capex allowance in the most recent AER regulatory findings. There are a number of reasons for this, some of which have been discussed by others elsewhere.
In addition, other factors are also at play, including the following:
1) The costs imposed this quarter by the introduction of the carbon tax (which we’ve started to look at recently); and
2) The rising cost of the various renewable schemes and subsidies (both federally in terms of the LREC and SRES schemes, and state-based in terms of attractive feed-in tarrifs).
3) Possibly there is still a hang-over from the period in 2007 when the drought drove up spot prices and retailers locked in hedges at these inflated prices which are still washing through into the retail space. We’ve not had time to investigate this.
Just a question or two:
is there anything akin to shorting in this market?
if so, who is taking a short position and are any particular generators being targeted?
are there any accurate/reliable figures showing the total energy generated from all sources, as opposed to this figure for demand (in the NEM context) that you have used here?
Hi Steve,
Being a cash settled futures market, electricity traders sell futures just as easily or as often as they buy futures. There’s no physical supply obligation attached to an Australian electricity futures contract, so it’s not like buyers or sellers of electricity futures can “corner physical supply” via the futures market. There’s no limit to the size that the cash settled derivatives (futures and OTC) can grow to as a multiple of the electricity consumed in the physical pool market. Ultimately, any futures positions held open as at contract expiry (i.e. at the end of the relevant quarter) are cash settled against the AEMO pool price average across the underlying quarter. It’s still the marginal generator(s) that set the pool price every half hour during that period, regardless of how many futures contracts have traded or at what price those futures contracts traded. Hope that helps.
Dean.
If anyone has any queries about the d-cyphaTrade ASX Electricity Futures & Options market please feel free to contact us on toll free 1800 330 101 or email info@d-cypha.com.au.
You attempted to reach d-cyphatrade.com.au, but instead you actually reached a server identifying itself as asxenergy.com.au. This may be caused by a misconfiguration on the server or by something more serious.
An attacker on your network could be trying to get you to visit a fake (and potentially harmful) version of d-cyphatrade.com.au.
You should not proceed, especially if you have never seen this warning before for this site.
You say:
“With reference to these daily production figures we see that average daily output as high as 1540MW has been seen from wind farms in 2011, which would have represented a daily capacity factor of the order of 70% – clearly a windy day.”
It would provide balance, and actually more important to point out, that there have been times when all wind farms in the NEM have produced no power. In fact, they have drawn more power than the generate.
In May 2010, there was near zero generation over a period of 4 days, under 10% capacity factor for 10 days, and 65 5-minute periods where the NEM wind farms were drawing more power than they generated. It is the minimum that is critical output that is critical, especially at times of peak demand.
The intermittent nature of wind generation means issues for the NEM at both ends of aggregate output:
1) As you note, some times the wind barely blows (even across multiple stations and regions) which means that, from a capacity point of view, only a small percentage of this capacity can be treated as firm at peak demand times.
2) At the other end of the scale, it’s often windy at night which has caused negative price events (mainly in SA) which has imposed costs on all generators – even the uncontracted wind farms.
These reinforce that wind power is primarily an energy-play, and not a capacity-play.
There has been heated debate since the report in the “Weekend Australian” (6/9) that wind energy is avoiding very little CO2 emissions and doing so at a cost of $1,484/t CO2 avoided. I would like to know how we, in Australia, can say how much emissions reductions can be attributed to wind generation?
Australia has no actual measurements of CO2 emissions from its power stations. We have crude estimates of emissions based on long term average fuel consumption. AEMO quotes the same emissions intensity for all units in the one power station no matter at what state of efficiency the unit is operating at the time and no matter whether the plant is ramping up or down. So clearly, the emissions intensity figures are very crude estimates and do not allow for changing intensity as the output and efficiency of the units changes.
We don’t even measure the fuel used in most power stations, let alone at the frequency needed to estimate the emissions avoided by wind generation.
We need at least the emissions data quality and frequency that Ireland has http://www.eirgrid.com/operations/systemperformancedata/co2intensity/ or, better still, we’d need what the US EPA requires for emissions measurement and reporting. But we’d need the information at preferably 5-minute intervals, or at worst 15-minute intervals, for every unit to allow us to determine how much CO2 emissions are avoided by wind generation.
CO2 savings per MWh of wind generation (relative to savings at zero wind penetration) are:
Wind Penetration CO2 Savings
0% 100%
10% 90%
20% 77%
30% 62%
40% 47%
At 50% wind proportion of total generation, 1 MWh of wind would be avoiding about 32% of the emissions from 1 MWh of fossil fuel generation, all else equal.
In fact it’s probably worse than that. I understand the Irish emissions intensity calculations do not include the extra emissions from fossil fuel units, for example: starting up and shutting down; in spinning reserve mode (i.e. not generating), powering up and powering down to back up for the changing wind power, etc.
Ireland’s figures apparently do not include all the extra emissions attributable to backing up for wind. But Australia has nowhere near the quality of emissions intensity data that Ireland has. So how can we make any reliable statement about how much emissions are being avoided by wind generation? Can we state what is the cost per tonne CO2 avoided by wind generation?
I did not read the article you mention, but understand that there are a number of views being promoted on aspects of:
1) Whether the emissions reduction from wind is large, or not; and
2) Whether the same emissions reduction is expensive, or not.
In the analysis we publish (and tools we provide) we strive to remain technology agnostic – and just to present facts, such that people can make up their own minds in terms of judgements such as large/small and expensive/inexpensive.
In the 5+ month period analysed above, for instance, Metered Wind Production in SA totalled 1,591GWh, whilst total SA demand over the same period totalled 5,795GWh.
This means that (based on these numbers) wind supplied 27.5% of the total demand over the period – and, as the chart above shows, oscillated to a significant degree over that period between cyclic high and low points. Both factors reinforce the fact that wind power is an energy-play moreso than a capacity-play. Both are important in the context of the broader energy supply industry.
Determining what emissions reduction resulted from this production is not a precise science as one needs to hypothesise about “what might have otherwise been”. At an extreme we might look at the emissions factor of Alinta’s Northern Power Station (whose output has been curtailed due to a combination of factors, including declining demand and increased wind production) and calculate an upper bound of emissions reduction as follows:
0.95 tonnes/MWh sent-out x 1,591GWh = approx 1.5Mt abated.
Is that not much, or a lot? There will be a range of views on this.
In terms of the cost of this abatement, this is analysis that we’ll have to leave for another time.
Thank you for your quick response and for your upper bound calculation of the emissions avoided by wind generation in the SA grid.
You say:
In the analysis we publish (and tools we provide) we strive to remain technology agnostic – and just to present facts, such that people can make up their own minds in terms of judgements such as large/small and expensive/inexpensive.
However, you’ve estimated an upper bound but not a lower bound or best estimate. I suggest it is really important to present this information objectively and with proper balance. The costs and benefits of wind energy is an important issue. Australia is committing to and spending billions of dollars on wind energy, apparently without objective analysis of the costs and benefits. So I’d urge that the estimates for CO2 emissions avoided and the CO2 avoidance cost be calculated and presented objectively.
Your upper bound estimate assumes 100% of the wind power displaces coal power. Many studies show this is not the case. The paper I linked to in my previous comment is one example. There are many others from Europe and North America that reach similar conclusions. Australia does not have any emissions measurements, so we are guessing. We don’t even have accurate emissions estimates, let alone at the 5 to 15 minute time spacing that is needed to do the equivalent of the study of the emissions avoided by wind generation in the EirGrid.
I’d urge you to balance your comment by also presenting the lower bound and the best estimate.
I would suggest the emissions avoided should be compared against three baselines:
1. the average emissions intensity of the SA grid, excluding wind power
2. the average emissions intensity of the fossil fuel generation that would have been the case now if the government had not mandated and subsidised wind power – if government policy had not mandated and subsidised wind power, the SA grid would have more CCGT and less OCGT, and the OCGT would be running more efficiently (lower emissions intensity) than they are, so the average emissions intensity of the fossil fuel generators in the SA grid would be lower than it is now.
3. Baseline #1 with the emissions embodied in the wind farms included.
Below I show an example of how I suggest the calculation could be done for the first case above. However, please note that numbers used in my example below are fictitious. I’d urge the analysis be redone using the best available numbers.
Assumptions and calculations:
CO2 emissions avoided:
SA grid average emissions, excluding wind = 0.7 t/MWh
Emissions avoided by wind generation = 65% x 0.7 t/MWh = 0.46 t/MWh
CO2 Avoidance cost:
Additional cost of electricity from wind generation (this cost should include all the hidden subsidies and hidden cost transfers to the fossil fuel generators) = $60/MWh
CO2 abatement cost with wind generation in SA = $60/MWh / 0.46 t/MWh = $132/t.
I’d urge that the best available numbers be used to calculate a lower bound and best estimate.
I appreciate what you’re seeking – however this is not something that WattClarity will provide, for three main reasons.
1) First and foremost, an answer to a question “what would emissions have otherwise been?” is, in truth, unknowable –
As I noted in my first reply, one would need to determine the scenario that would otherwise have unfolded – which obviously is not possible, in reality.
2) Of course it would be possible, through rigorous analysis, to progress down the asymptote to deliver an answer more refined than my crude upper bound estimate (above):
To do so would require dedicated knowledge, and a considerable period of time – and consultants would (quite rightly) charge a price for this piece of analysis.
This detailed analysis is not the role for WattClarity – a free service delivered as a value-add for existing clients.
3) More fundamentally, however, this is not an area of debate WattClarity will engage in.
It appears to us that much of the “debate” about these vexed questions of both:
(a) techniques for carbon reduction; and
(b) the costs (incl externalities) & benefits (incl externalities) thereof
has descended into a shouting match where people of a number of people selectively quote or misconstrue facts that support their own philosophical point of view.
It appears to us that much of the commentary that exists (supporting various points of persuasion) contains mistakes or simplifications that the authors may be:
Reason 1) Unaware of (e.g. because of limited knowledge);
Reason 2) Are aware of, and acknowledge as simplifications, but not to such an extent that some readers still don’t take out of context; or
Reason 3) That the author might be aware of, but still deliberately misrepresents to support their own point of view.
For instance, the “joewheatley” post on the Irish system (you point to above) contains some conclusions that fit into this category. Whether this is because of Reason 1 or Reason 3 is not of interest to us.
There may be other places you can find the assistance you’re looking for.
Regards
Paul
PS you asked about time-based emissions data, which is available from the AEMO for the Australian NEM – though I suspect it has been generated with the use of the same gross simplifying assumptions that are (to my knowledge) also the case for time-varying unit efficiency calculations right across the world.
Thanks for publishing this interesting analysis. Quick question – what definition / source of system demand are you using for plots 1 & 2 above? With nearly 400 MW of non-scheduled wind capacity in SA, obviously wind production and scheduled demand are going to show an exaggerated negative correlation.
What would be a great addition to the graph would be a line showing the operational capacity (the theoretical maximum that the bars are falling short of). At the moment the temptation is to guess it as the envelope of the peaks; but that ain’t necessarily so.
Also plotting the generation data as bars makes it hard to see how much low-generation there is. Could the same data be plotted as just points?
What about the impact of vertical integration of the majors? Could this be contributing due to the natural internal hedges they have by owning generation assets or rights to generated power due to privatisation?
Uncertainty about the medium-term landscape that will apply for carbon pricing might also be an influencing factor – along with the 3 other factors mentioned above.
At the end of the day, it’s a market – hence we won’t know for sure the motivations of each and every participant who trades (or doesn’t, as the case may be) all of the time. All we can do is think through possible contributory factors.
Paul,
Interesting times indeed. There were years not so long ago where a little price excursion such as $217/MWh would not have merited a post here.
Interestingly, this price would not even be able to trigger the $300 strike price of the ASX d-cyphatrade futures cap. Perhaps we will see a revival of $100 strike caps as a standard hedge instrument now that this is looking like a more sedate norm for the market.
On the other hand, the underlying structure of the market has not changed and all that is required is a few more plant retirements to bring the supply-demand balance back to something that can again generate price excursions in the $1000’s on a regular basis. I know we had one in VIC last week so they do still happen.
Interestingly, it was only 5-10 minutes after posting this that the price did spike (about $1600/MWh in the 16:35 and 16:40 dispatch intervals).
It’s overly simplistic, but one might say that “all it took” was for the demand to rise a little bit more (to 8490MW – so about 45MW higher than it had been prior to that, much smaller than the plant retirement you proposed).
Overall there is still plenty of capacity in the NEM, and while there has been a step change after 1 Jul 2012, spot prices are still historically low. I accept high price events are function of a properly working spot market however I wonder whether there were greater forces at play here.
Yes, there were a number of other factors at play (the 855-871 constraint being one of them). We have other tools that help our clients to understand more about these factors, and the interrelationships between them.
Occasionally we find the time to post some more detailed commentary (such as we did on 2nd July) about these more detailed factors.
Thanks Paul. In essence these things come down to generators taking advantage of their market power in favourable circumstances, which I would also do if I was in their shoes!
It might surprise some people – but, on at least some occasions, generators don’t need to “take advantage of their market power in favourable circumstances” for the price to spike.
Spot on with the ENSO (el nina etc) summers. Great analysis and very interesting!
Interestingly 2010/11 was a very strong La Niña but a year later it was relatively weak but seemed to have similar effects – just that one peak demand period in NSW. Also noteworthy that ENSO has a much less of an effect on VIC and SA than NSW and QLD which bucks the trend again in 2011/12
Paul, thank you for the analysis. Are you able to talk about the impact of the 855-871 constraint? I’d love to watch you identify it as a culprit in the stubbornly high prices.
Paul
Thanks for the analysis. You note that the current prices are not unprecedented and were seen when the drought was on in 2007 and generators withdrew capacity because they had limited cooling water. Why would prices reflect drought conditions when there is no drought?
I saw with interest the Courier Mail report http://www.couriermail.com.au/news/queensland/energy-firms-playing-games/story-e6freoof-1226553917675 which makes more sense to me and reinforces my view that generators have market power when there is congestion in the transmission system and withdraw capacity to increase prices.
This is why we introduced a rule change to stop this happening
Regards
David
Hi David,
The courier mail article was a very one sided view to stir up public opinion. The QLD generators have been losing money for most of the past 16 years – that “lost money” ends up subsidising the private retailers with generation capacity rather than the QLD government. So all Queenslanders are worse off than the states who have sold their assets. The oversupply of generation has been suppressing prices along with a carbon price that makes some generating units uneconomical. The idea of the carbon price was to make it expensive for coal generation so that they would eventually shut down and be replaced with generation with less carbon – ie gas. You would think with all the profits from the Carbon Tax that most of it would be put to creating cleaner energy funded by the federal government – like a national Solar or Thermal program.
Likewise – thanks for the comment, which is really three in one:
1) You ask ”Why would prices reflect drought conditions when there is no drought?”.
My apologies – I should have been clearer in making the comparison:
(a) In 2007, during the drought, we had 2 units at Tarong instructed to turn off (for a long period of time) by the Government, plus capacity at Swanbank as well.
(b) This year we have a similar situation, but for different reasons. As Paul Syvret acknowledged in this earlier article in the Courier Mail, the Tarong station has been running at a very marginal 60% capacity factor for quite some time – for something originally built to run at 90% or more. Hence, the owners of the station made the logical decision to close two of the units (for a long period of time)– as have several other generators around the country in the past 12 months.
As such, we have a similar outcome, though for different reasons. It’s not totally a surprise, then, to see the same outcomes in terms of prices.
2) You note that Paul’s subsequent article ”makes more sense to me and reinforces my view that generators have market power when there is congestion in the transmission system and withdraw capacity to increase prices.”.
Of course, transmission constraints can have a significant impact on pricing in the market:
The 855-871 constraint has been particularly problematic in Queensland this summer and, to a lesser extent, last summer – though I understand capital work (of the “non-gold-plating” variety?) is being done to resolve.
However transmission constraints having an impact on price can also be the case in any other market (witness the price of bananas in Australia after cyclone Larry in 2006 when foreign imports were constrained from entering the market).
With respect to your related point, there’s something I struggle to understand – so perhaps you can help me with this:-
If we look at the Iron Ore Spot Price and see it dropped to under $90/tonne in the September quarter 2012, wouldn’t we be able to use the same logic to say that:
(a) Rio and BHP were happy to sell iron ore at those prices in the September quarter,
(b) therefore they should be happy to be paid those prices currently (and not the current spot price, which is perhaps $40/tonne above its low point)?
I can’t imagine many Australians supporting that view – though I can imagine the Chinese (and other customers) trying to make the same argument (and being labelled “socialists” and “anti-market” by our media as a result).
In another different industry, I’d certainly like to be able to book a harbour view room at one of the best hotels in Sydney for the New Year’s Eve fireworks at the price I might normally be able to get on the discount travel sites – though I don’t like my chances.
What makes the energy market any different? Other than that (by virtue of the physics of supply) the price must change much more frequently, it’s been determined that we have a market, so we should expect market outcomes.
That means that, at times when the supply/demand balance is tight (across the NEM, or more locally with transmission constraints), prices will escalate. The benefits of this approach are that:
1) Most of the time, spot electricity is sold into the market at under its long-run cost of production – which benefits all energy users
2) At times when prices escalate, that is an excellent opportunity for Demand Side Response – which has benefits to the market more broadly, as well as being beneficial to the provider.
In the charts above, one of the striking things has been that, over the 14 years shown, there has not been an escalation in long-run revenue, or any indexation (save for the step-change in July 2012 for the Carbon Tax).
Some sensible arguments Paul. You might like to highlight that it was not just the drought affecting AGL hydro, Snowy hydro and QLD generators but also bidding behaviour in Q2 2007. Interestingly, NSW which was less affected by the drought than other regions usually had the highest spot price in most dispatch periods. The ETEF (Electricity Tariff Equalisation Fund) was still in place and the surplus was large (incentivising generators to increase spot prices). Also, the price sat around $90/MWh for most of that quarter until the last week, when it was continuously around $200-$300/MWh. Noticeably, on July 1, prices returned to historic normal levels. Given trading bonuses are based upon financial years and there wasn’t a drought breaking storm on July 1 maybe there were other factors at play…
Paul
Thanks for the further clarifications. Addressin your and other comenst in order
1. I accept that generators are having a hard time and I personally beleive that the energy only market has a flaw and that a capacity market would be more equitable and stable. I also agree that if there is a reduction in demand, the logical solution is that some generators will shut down, reducing available generation and that this should increase prices offered by generators.
What I don’t accept is that some generators can use their market power (a la AGL TIPS in SA during 2008 2009 and 2010). It is inequitable that some generators can have this market power but others don’t. Using network constraints to drive up power prices is not how a competitive market should operate as it is inefficient and creates significant other problems. Basically generators shpould bid their underlying costs becasue this results in the most efficient dispatch
2. I guess the difference between the iron ore example and electricity is that if i don;t like the price from RTA I can buy it from someone else and I know the price before I buy. In the case of electricity, I have used the electricity before I know the price and I don’t have the option of another source. I accept I can pass the risk to a retailer but the fundamantal issue still underlies the cost to my retailer and I pay it to accept that risk.
You comment that we have a market so we should accept market outcomes. I accept that but what I don’t accept is that some generators use market power to reduce the competitive pressures that ensures an efficient outcome. When bananas were high priced, many consumers didn’t buy bananas. When electricity is high priced I have no option but to pay the price. In today’s environment electricity is an essential commodity – bananas and iron ore are not – without elecricity the whole country stops.
To use market power to drive an inefficient price outcome is not what a market is about. Eminent economists Newbery, Tirole, Joskow all agree that using makret power is not efficient and have advocated various ways of preventing it
3. Whilst the MEU response to the draft decision not to make the rule change is quite definite, the response from the AER who have objected to the use of market power in the electricity markets is equally damning of the AEMC position.
Hit 45.8 hottest since 1939 with many other NSW records also falling. But heat across NSW was a bit below par with northern parts cooler. So the state’s overall temp fell well short of record levels!
This is a great tip particularly to those fresh to the blogosphere.
Simple but very precise info… Many thanks for sharing this one.
A must read article!
I’ll be happy to answer this one for you, one-on-one, Peter!
Will call you in the next day or two – or call me 07 3368 4064 if a particular time suits.
Regards
Paul
Thanks for this note, as well – yes, it would probably have had some effect.
SP Ausnet notes here that:
1) they expect the reduction in peak demand to be approximately 3-4% of their (distribution area) maximum demand – hence perhaps as much as 200MW.
2) Given that this DSR would have needed to be in place between 2pm and 6pm NEM time, it’s one plausible explanation for the dip in demand seen occurring around 2pm on the chart above (i.e. before the price spike).
However, we can also see that the demand climbed rapidly back up to the same level, preceding the spike. It seems likely that this was non-related demand (e.g. distributed air-conditioning load) and that the 400MW of DSR seen immediately after this price spike was more directly in response to price exposure (either directly at the energy user, or indirectly at a retailer – calling optional DSR).
Thanks for the analysis. Be aware that SPAusNet’s Distribution area is only one part of Victoria, probably about a fifth of Vic’s distribution connected industrial load. Their submission to the Power of Choice Review claims their CPP scheme achieves 70-80MW load reduction.
Thanks for the added points, Ben
At a more holistic level, I think we’re seeing the results of how Demand Side Response (in a variety of forms) is proving more widely accepted as a means of managing risk (physical and financial) than was the case a decade ago, or so.
Paul
Re the IRPM I followed the link to the Nem Watch glossary. Unfortunately it no longer provides a formally definition of the IRPM nor a formula. Would you let me know where I can find one?
IRPM translates the measure of RPM (traditionally used in a longer-term planning context) to real time:
IRPM = (Total Available Generation – Total Demand)/(Total Demand)
Within an Economic Island we substitute:
IRPM ei = (Total Available Generation – Net Demand)/(Net Demand)
where Net Demand = Demand in regions within the Economic Island + Net Exports, which are supplied by the local generators.
Shouldn’t the drop in brown coal production be attributed to the carbon price? Yallourn was back in action not that long after the carbon price came into play. Sure, the initial drop was due to Yallourn but it’s been 9 months already. If the drop in brown coal was only due to Yallourn then brown coal should have bounced back by now.
Hi Tom
Thanks for the question – and apologies for this belated response.
I’ve noted a number of people stating “the NEM is a complex place” the past couple weeks – the challenge in your question is that it’s not just one factor that is delivering change (and varying degrees of change dependent on participant, technology, commercial and other considerations).
Regards
Paul
‘Official’ is such a meaningless word, Macarthur has been dispatched for energy since late last year. Congratulations to the boys and girls at AGL, Vestas and Leighton on their hard work!
Thanks Peter,
With reference to NEM-Review it looks like it’s already produced about 371GWh over the period – so already contributing to reducing emissions.
Cheers
Paul
How much emissions has it abated? How much will it abate per year? How do you know?
Will it be more effective at abating emissions than wind power in Ireland (EirGrid)*? How do you know?
What will be the abatement cost ($/t CO2)?
* For those who don’t know, wind power in Ireland is about 53% effective at reducing emissions (17% wind energy proportion of total energy).
Joe Wheatley (2012) (in press) “Quantifying CO2 savings from wind power: Ireland” http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf
Hi Paul. I really appreciate your input into the electricity supply debate.
I have been reading the gushing reports about Macarthur wind farm and was wondering if the wind might come to the party. Probably a bit sad that it didn’t show up for the paperartzi.
A debate I am having on various blogs is the issue of dispatchability as the renewable advocates claim that baseload is a myth (read Mark Diesendorf) and even so, wind power can supply baseload (contradiction accepted).
So your comment about wind farms being a type of generation built for energy production, not capacity supply is what I’m trying to explain to these rather over zealous advocates who don’t seem to understand the importance of capacity if we are going to have a reliable network. It continues to be a tough sell for those that don’t want to really understand how electricity networks actual work.
The industry was already a complex beast pre-market, then the market came along and the complexity (and data load) increased. Add in a social drive to increase renewables and lower emissions and the complexity has ratcheted further.
Hi Martin,
I agree that capacity reserves are vital to a reliable network, however peak capacity is currently not a barrier to increased renewable installation. It will be a barrier in several decades perhaps.
Currently the main barrier is cost of RE vs. incumbent fossil fuels. As the price drops more and more RE will be installed at least to the point where oversupply will become the next hurdle. This is where the price will be unprofitable for significant periods of time for coincident RE generators (e.g. with 20 GW of PV noon will be the cheapest part of the day).
RE will not be able to mount this hurdle without energy storage or more flexible consumers. If these requirements are met then back-up capacity becomes an issue. It’s pretty easy to solve too – build OCGTs.
May I remind you of the current situation in SA? 25+% Wind power and no new peaking capacity added. Maybe when they get to 35% and the remaining coal plants shut down there will be a need to build some more peakers.
On your point about baseload; any technology can supply baseload, it’s just that no one cares to. If all the baseloaders were gone we could still run a peaker at night.
Paul,
With regards to your last points on Northern: We all get that wind is intermittent, and if average wholesale prices have fallen and the consumer is receiving that cost reduction then any market volatility is purely a problem for the retailers. Maybe instead of focusing on Northern which is mostly hobbled by the carbon tax you could be asking why several important peakers and intermediate plant weren’t made available for generation over the recent weeks? Summer was pretty weak and no-one had to push their assets hard. I count over 1 GW of unavailable gas plant. The entire point of some of these assets is to provide capacity yet they did not deliver. Maybe there is something in holding on to that gas…
You also brought up externalities for wind (of which neither example exists); why not talk about the health costs of coal generation? 😉
Thanks for these questions, Tom
This is the second lot from you I have not yet responded to. When I do find some time to look more thoroughly, I’ll be sure to post more on WattClarity.
Regards
Paul
It’s a pity that the offpeak load isn’t staggered. I guess ETSA have no incentive to do so.
It only takes 4 hours to charge a hot water system from cold, and 2 from warm, so there’s plenty of play in the timing.
Possibilities like these intrigue us, Tom – and we’ll look further into them when we have the time. At present we don’t know what the barriers might be to prevent this happening.
Regards
Paul
Note re Morwell Power Station – they confirmed last year that they would no longer operate to generate electricity into the grid – they are only operating one turbine (enough to supply on-site demands from briquette plant). The understanding was this is/was due to higher costs associated with carbon tax, but also combined with poor efficiencies, lower demands etc.
They received Govt funding to keep briquette plant operating for 2 years – industry noise is that they will close June 30 next year (2014) when funding runs out.
1) re Energy Brix power station: In June 2012 Martin Ferguson gave HRL – the ower of the Energy Brix plant – $50 million for a “a two-year restructuring package to allow the Latrobe Valley’s Energy Brix Australia Corporation to maintain briquette production while regional businesses that rely on briquettes transition to a cleaner fuel source”.
2) It is worth mentioning that a central reason why the carbon price has had little effect on the dirtiest coal fired power stations is that their owners have already been paid $1 billion in cash payments in 2011-12 from the $5.5 billion ‘Energy Security Fund’ and more subsidies to keep them going are on offer.
Amidst the fuss that switch to gas is an essential transition, have always been doubtful about this on suspicion that power stations extract brown coal for free, no $$$ per tonne royalty, resource rent depletion tax etc etc.
This being so, why would they switch, unless taxpayers make gas available for free? Any chance you can confirm or deny this suspicion?
“It’s a pity that the off-peak load isn’t staggered.” – –
I agree – how much discount would I win for a 2am start ? – I only heat the top half of the tank to 50*c for a hot morning shower- the thermostat range is 50 to75* – low pressure solar Beasley –
Perhaps ETSA’s “J” meters should have a built-in 3hr random number generator [1/180] – we live in a digital world now – – cheers.
Thomas
Thanks for this – a useful addition to the market intelligence (or is that an oxymoron?)
One aspect you did not seem to investigate is that the demand in NSW has been falling for the past few years and this would have had a significant impact on prices.
The fewer incidences of exercise of market power would also be related to the fact there have been fewer demand spikes due to the milder weather conditions in the past couple of years
Regards
David Headberry
Market intelligence is a big beast, we aim to provide some insight.
The demand will have a factor. Paul has already written a couple of blogs here on the falling demand (linked here).
A quick recap shows that the average metered demand decreased by 410 MW over the period (from 1 March 2009 – 28 Feb 2011 compared to 1 March 2011 – 28 Feb 2013) and maximum demand fell by 828 MW. This reduction in demand will have an effect on the opportunities for price spikes to occur and we agree that the NEM is a complex place.
There are always many reasons for prices being different. The Analytical Report that we will send out with our QLD market review is going to be over 100 pages and only covers four months!
The reason I am highlighting the change in ownership is that surplus capacity is not a lot higher than have seen in the past but the lack of price spikes is new. Since 1999, NSW has never had fewer than 53 prices above $300/MWh in any two year period. A lot of things have happened since then but the timing of the reduction in spikes points to the asset sale.
In summery: Agree market intelligence is a noble but difficult goal, and there are many temporary factors such as demand which plays into actual outcomes in the short term.
I wonder how much of this afternoon peak is caused by a fall off of rooftop solar in winter afternoons. Even in far north NSW, you don’t get much solar after 2 pm in winter. The fact that it occurs even on a Sunday might suggest it has more to do with domestic demand rather than commercial.
Hi Martin
Not sure I understand this question – the evening peak in demand has existed for many years. Perhaps the input of PV during daylight hours (prior to the peak) has increased the rate at which net demand (to be met by the wholesale market) ramps up to the peak has increased in recent years – is this what you were meaning?
Paul
Martin, unless the sun goes down at 3PM PV is still going to have a reasonable output. Like Paul said PV isn’t going to impact evening demand. There is only ~500 MW of PV installed in NSW so I don’t see it drastically effecting price. Check out the examples of the German multi-GW PV capacity cutting the middle out of the daily demand …and price. http://reneweconomy.com.au/2012/why-generators-are-terrified-of-solar-44279
Yes, sort of. PV in summer would reduce grid demand during the day exaggerating the peak in the summer evenings as the sun sinks increasing grid demand. I suppose a test of this theory would be whether the evening demand has remand flat over the last 5 years but the evening “peaks” (relative to midday) are becoming more prominant as winter approaches. Something like that 🙂
Guys,
I think it was the Germans who used to monitor short term changes in sewerage flows at night and related that to TV audiences and commercial breaks – do you want to co-ordinate with the appropriate authorities in NSW and Qld to investigate the correlation?
Good article – having the concept of surplus capacity helps provide greater context to high prices. A couple of questions:
1. Did wind generation reflect wind quality or were turbines disengaged?
2. Why is the Murraylink Interconnector down?
Hi Michael,
1. The low levels of wind generation reflected wind speed. It is possible that some individual turbines (around 2MW) may be out of service but no entire wind farms were offline. It is more difficult to see this graphically as wind availabilty always matches the output unlike fully scheduled generation where it is clear to see reductions in availability due to maintenance, system constraints etc in comparison to registered capacity.
2. The outage on the Murraylink interconnector was a preplanned outage in Victoria preventing the flow into SA. At this point I dont have more detail on whether maintenance or upgrades were taking place, but it could be obtained by contacting Electranet SA or SPAusnet in Vic.
Does this help answer your question?
Regards,
John Bartlett
PS – it’s been highlighted to me that the outage at Yallourn comes coincident with Industrial Relations action onsite, and with a fire that is under investigation at present.
I am more inclined to relate these demand increases to the domestic controlled load by the distributor. As 21:30 & 22:00 are common T33 switch ON times.
Not sure on the declines before the rises, but there is a decreasing trend in demand as the evening progresses most nights but it not linear. But back to sport and the NEM.. If you are looking for a sporting event that impacts the NEM then the winner in my mind is the Melbourne Cup the race that stops the nation. Demand decreases in QLD as manufacturing shuts down, office PCs go to standby etc… which is not offset by the thousands of TVs being switched on during the day.
Yes the audio does help with understanding. Even with the sound turned right up, it is still much quieter than most audio feeds and sometimes is difficult to hear. Can you boost the sound from your end?
Being a first attempt at adding audio, we used the tools we had at hand.
Based on the belief that this kind of thing does add value to clients (by helping them more easily understand the complex interactions at work in the NEM) we are having a look at some technology that will give better sound quality.
Pacific Hydro may have been referring to an analysis that was done earlier that I am responsible for, that utilised an erroneous demand figure (ie, non-scheduled generation had been subtracted from the denominator and thus, the demand figure was lower than it should have been in the formulas).
I’m working with the publisher to get it corrected, and I’ll let you know once it’s all fixed up. Apologies to you and Pacific Hydro for the error!
I agree with your thought process, but the other factor, which you may have considered is cost. I think that a far larger part of the population are being more careful about energy costs. I believe the cost of energy to moms and dads is now becoming a significant part of the household budget and therefore things are being done to keep the costs down.
We, for example, have dropped our ducted heating system by 1 degree to 19 deg C and on weekends be actually switch it off and just wear more clothes, where we previously left it on for the weekends.
I am also meeting more and more people who are looking for alternatives. Took a cab to the airport last month and the cabbie told me that using homemade vertical axis wind turbines using 44 gallon drums and old car batteries, he had so far converted his entire shed to an off grid system and was starting to expand into his house.
Increased awareness of energy usage, and costs (and their increasing share of budgets), both at work and at home were one of the factors noted here a couple years ago.
Your own example is not dissimilar to many, I suspect.
Given that the June-July 2013 profiles are lower for the whole day – not just when PV is generating – your conclusion is badly overstated. PV has no effect on the breakfast and evening meal peaks or overnight values, and they are all lower too, by the same margins.
There is some sign that the minimum is shifting to earlier in the day – which can perhaps be attributed to PV generation peaking at around noon, or just prior to noon for Brisbane (solar noon is at close to 12:00 for Rockhampton, a little earlier for Brisbane). But the effect is pretty marginal, and could also be attributed to the increasing use of air conditioners in the afternoon (it’s most likely a combination of both).
Thanks for the comments, Phil
In the post I linked to a range of factors that would be contributing to various degrees, and noted that the 4am demand was also lowest in July – hence, there’s more than just PV contributing.
Not sure how I have overstated a conclusion?
Paul
Paul, I noted your other factors, which are correct in the context of reducing demand generally, as we see over the family of curves. [Other readers, I can reccomend reading that list].
It just seemed to me that your article was inferring that the demand profile had reduced at midday, and that that was probably due to widespread PV installations (“Undoubtedly, the rush of people … getting their 5kW solar PV systems installed … has had some effect”). I did not feel that the midday reduction was significantly different to the reduction over the rest of the day, so to attribute it mainly to PV was overstatement of the PV effect.
From your response it seems that I had misunderstood your intended emphasis, and that you meant that PV contributed to the reduction, rather than being the primary cause of it.
I certainly agree PV is a factor, but it seems pretty minor from the evidence of these curves.
Hi Phil
Thanks for the added details. Part of the challenge in terms of interpreting PV (for us, and for everyone) is that it is behind the meter, so not as visible at the level of the NEM.
Cheers
Paul
Thanks for this.
A common response by advocates of renewables to the criticism of “unreliables” is to propose wider diversity of the renewables. As an example, when SA wind may be becalmed, wind farms in Vic and NSW can contribute instead. Apart from the question of power transfer over long distances, and the question of whether the customers in Vic and NSW might also want the power, the assumption that Vic and NSW won’t be becalmed at the same time seems somewhat wishful – as your analysis suggests. While it’s not that common, a large high pressure system centred somewhere over the SA/Vic/NSW common border would seem to generate the right conditions for that.
I would be interested to see a graph similar to your second graph, but using half hour (or shorter) demand data (MW), rather than energy (MWh). This will give a better indication of the time-matched energy production in both areas.
As I see it, the main problem with renewables is that the energy they produce isn’t time-matched to the energy demand of consumers. In many residential houses without air conditioning, a reasonably sized PV installation will produce more kWh’s than the house uses; but you’ll get nothing – zero kW – from the PV during the evening meal.
When (not if) we get some practical large capacity energy storage systems, renewables will really come into their own.
I will add these suggestion to our growing list of potential future analysis – though keep in mind you’d be able to do this, yourself, with NEM-Review.
You don’t have to go to the expense or trouble of signing up to NEM-Review if you just go to the link below where its all done for you based on 5 minute SCADA data from AEMO, with historical data as far back as 2009
Phil,
You might also like to have a look at a paper I have been fortunate to have had published
recently. Properly, anonymously, peer-reviewed, by real engineers in charge of grid-control.
It looks in detail at wind farm performance for the full calendar year 2010.
The reference is: Miskelly P 2012, “Wind Farms in eastern Australia – Recent Lessons”,
Energy & Environment Vol 23 No 8 December 2012. It’s available at: http://multi-science.metapress.com/content/f1734hj8j458n4j7/?p=3367b7f146d24db99c4ffb0a5361ab47&pi=4 .
As contributors Leith and Peter say above, you can do it yourself. I provide the methodology
with all links to relevant AEMO data in the paper.
Also, there is an earlier paper by two of my colleagues:
Miskelly A & Quirk T 2010, ” Wind Farming in South East Australia”, Energy & Environment Vol 21, Vol 20 Number 8 – Vol 21 Number 1 / December 2009 – January 2010. Available at: http://multi-science.metapress.com/content/j2l218v143733536/?p=5b117ed08c1e43a7a3ae567df515f5f2&pi=13 .
Note that http://windfarmperformance.info/ was designed and is maintained by Andrew Miskelly,
co-author of this second paper.
Hi Paul,
I absolutely agree with your findings here, simple probability suggests that there must be occasions where total wind generation falls to zero. However no one is suggesting that we run a country with a single energy source. It wouldn’t be a great idea to run solely on coal/gas/hydro/nukes either.
Like Phil pointed out we won’t be able to have a very high proportion of our energy coming from renewable without storage. In the meantime we can aim for a lower proportion safe in the knowledge that we have an enormous volume of fossil capacity ready to ramp up when required.
Tom Quirk and Andrew Miskelly published a largely ignored paper in Energy & Environment Vol. 20, No. 8, 2009/Vol. 21, No. 1, 2010 based on 5 minute data.
Someone might be able to help me, perhaps I have a misconception. The renewable power percentage is currently about half the 2020 target value, which should mean that about 10% of our electricity should currently be covered by RECs. The data in this blog suggest less than 5% on average coming from wind.
Is the RET under delivering?
Hi James,
Most of our renewable energy comes from hydro. The RET has been more or less on track to the fixed GWh target so far. However the large retailers (at least Origin and EA) are refusing to buy into new renewable projects which will make the target hard to meet. They are hoping they can inflate the cost of the scheme by reducing supply of RECs vs demand and then ask the government to reduce the target to protect consumers. Oh and their profits. Don’t forget the profits.
The chart demonstrates the volatility of wind power. Most reading this web site would be aware of some of the consequence.
The purpose of the RET and the many other policies intended to give advantage to wind and solar generation is to reduce Australia’s CO2 emissions. But seldom do people consider the cost of these policies in terms of “what is the CO2 abatement cost?”.
A rough estimate is that the CO2 abatement cost is about $120/tonne CO2 – based on the following assumptions:
– cost difference between coal generation and wind generation = $60/MWh (perhaps higher)
– Average CO2 intensity of NEM coal plants = 1 tonne/MWh (perhaps higher)
– Wind generation effectiveness at reducing emissions = 50% (has not be determined for Australia). This means that 1 MWh of wind generation avoids just 50% of the emissions from 1 MWh of NEM generation. This is based on the Joe Wheatley (2012) study [1] which demonstrated that the wind generation in EirGrid in 2011 was just 53% effective at abating CO2 emissions. EirGrid and SA’s grid share many important similarities, including: high wind energy penetration, similar generation mix, small interconnectors and small imports and exports.
The volatility shown in the WattClarity chart is a cause – not the only one – of the reduced effectiveness of wind power (and solar power) for CO2 abatement. However, it is seldom discussed and as far as I know there has been no real attempt to determine the CO2 abatement effectiveness for Australia. Given the cost of the CO2 abatement polices it seems amazing that we have no well justified estimate of the abatement cost.
Hi Peter,
The chart shows volatility in wind generation over a significant period of time. You would be hard pressed to find many instances where short term volatility in wind power caused upwards market volatility. However it is a lot more common for large fossil plant to trip offline during elevated prices and cause prices to hit $12k+. Even so our current generation base is such that the loss of 500 MW is no big deal in term of system security. My point here is that it will require a hell of a lot more wind installations before we have to get concerned about system security.
Your CO2 abatement cost is also pretty inaccurate. The Irish grid has nothing to do with ours. The effectiveness of emissions reductions would be much closer to 1 given the generators that are usually limited to make way for wind; namely black coal, brown coal, and lower efficiency gas (Torrens). Also the actual cost borne is $35 for a REC minus a reduction in the pool price. I reckon the cost of abatement would be more like $40/ton.
Tom,
What is the market mechanism by which brown coal generation is displaced by wind? I thought brown coal generation bids into the market bid stack at a lower price than gas generation. Wind will displace the most expensive alternative in the bid stack, which is most likely open cycle gas turbines.
The marginal carbon abatement due to variable wind generation does not appear to be measurable in Australia, hence conflicting claims from pro and anti wind lobbyists abound on the various blogs. Why don’t we all admit that nobody knows, reframe the problem in this context, and hold the wind industry accountable for providing evidence of displaced generation (gas or coal) due to wind generation.
Another misconception surrounds the ratio of wind energy to total supply. Retailers are accounting for about 10% via surrendered RECS, however only about 4% is being generated. This results in the cost of one MWhr of renewable energy being somewhat higher than the cost of one REC. The market would still appear to be awash with phony RECs created to support solar credits for phony rooftop solar energy.
Hi James,
It seems a bit unfair to tar wind with the solar multiplier brush doesn’t it? 🙂
Roughly speaking brown coal is displaced by wind when gas units are off or at their minimum capacity which is common during mediocre demand and/or high wind. If brown coal generators bid strategically above their marginal cost they may also be displaced at any other time. The interconnectors make it a bit tricky to pin down exactly what’s happening as one generator can be displacing another one 3 states over.
If we re-ran AEMO’s dispatch engine, with and without contribution from wind it would paint an accurate picture of the actual abatement.
For a more rough idea just look at the fact that gas and wind generation are rising, and coal generation is falling. Of course demand is a major driver but if wind was mostly displacing gas it would be taking a huge chunk out of gas generation.
Geez Paul – thought I would watch the video for fun…totally forgetting I had entered the competition, and low and behold my name was drawn out in the “nerdiest” way possible. Long live the Nerds!
By the way – I fly out to Paris in 10 days for a holiday (seriously) and was about to purchase a camera….thanks Paul, you saved me a head ache or two and quite a few dollars there.
Cheers Craig J.
This is interesting analysis. Thank you. What would also be interesting to see is the minimum aggregate average power output (MW) from wind generaton in a 5 min trading interval for the year and some sort of frequency analysis, such as aggregate wind generation was X MW or above 90% of the time and y MW or above 50% of the time for the year.
Hi, So what you are essentially saying is that the variability of the weather patterns over the last 4 years that the increasing number of wind farms in the NEM are operating in has not changed much.
Hi Muzz
It seems that weather patterns in the expanding area over which wind farms have been installed, thus far have not been sufficiently asynchronous to reduce the small but significant percentage of time when the entire portfolio of wind assets have produced a negligible amount.
Does this help?
Paul
Of course, the vast majority of the energy is coming still coming from SA and Vic, which have relatively strongly-coupled weather patterns, and most of the wind generation in NSW is concentrated in one area close to the Vic border.
What might be more useful to say here is that wind power is generated by “useful” weather patterns, and as long as wind farms are installed based on economic gain, the most useful wind patterns will continue to be exploited and improved reliability resulting from the geographical diversity of generation will not be realised.
The story becomes interesting when network limitations prevent wind farms from being installed in areas with the best weather patterns – we are approaching that situation now, and it is likely that as long as network augmentations designed specifically to “unlock” wind resources do not go ahead, that we will observe an increase in geographical diversity of wind generation (because the economic gain to the wind farm owner will be reduced under conditions of network congestion).
I’m sure the economists have a technical name for the behaviour you observe – wind farm developers maximising the benefits to themselves (in terms of locating where they will get maximum MWh generation, and so also RECs) but not having an incentive (yet?) to consider where diversity would be promoted.
With new planning restrictions in Victoria we’ll be interested to observe performance patterns as new wind farms sites further afield are opened up.
Interesting analysis. An interesting next step would be to estimate what would be the average cost of electricity for the 365 days.
Since the purpose of the RET, and for installing wind power, is to reduce CO2 emissions from electricity generation, the next step would be to calculate how much CO2 is actually being avoided. To do this, we need to know the effectiveness of wind power at cutting CO2 emissions. Many studies using empirical evidence show it is much less that is normally claimed based on modelling studies. The study by Joe Wheatley [1] suggest wind was only 53% effective at cutting CO2 emissions from the fossil fuel generators in Ireland in 2011; i.e. 1 MWh of wind generation reduces the emissions from the 1 MWh of displaced generation by just 53%.
Having calculated the average cost of electricity and the CO2 emissions from the NEM with and without wind generation it will be an easy step to calculate the CO2 abatement cost (in $ per tonne CO2 emissions abated). Expect the CO2 abatement cost to be around a hundred of dollars per tonne CO2 avoided.
1) Regarding the cost of electricity under the hypothetical case, each analyst will have their own point of view – with a wide diversity almost guaranteed. Suffice to say that this would be impossible to forecast what the actual price would be under such a scenario, as to do so would require predicting human behaviour in terms of trading decisions. It seems likely that the wind farm generators, were they to provide a more significant stake of energy in the NEM, would not continue to follow their “price taking” behaviour at present.
2) Regarding the MRET, my understanding of the original genesis of the scheme was that it was partly to do with carbon reduction, but also partly to do with support for a nascent renewable energy industry in Australia (for reasons broader than just carbon).
3) Regarding avoided carbon emissions, frequent readers here (and elsewhere) will see a tiring pattern in the fact that you frequently add this in, citing a single reference from a different market that aligns with your own particular perspectives. It’s not something we will be focusing on through WattClarity (the focus of which is energy supply, whatever the source).
Not sure if you have analysed the AEMO 100% RE study. The 2030 scenario 2 has a demand of 290 TWh/yr = 800,000 MWh per day. Of this wind provides 111 TWh/yr or 38% – around your maximum. Maximum demand in the network is 46,000 MW.
Total capacity installed in 2030 (all RE) is 98,000 MW of which wind is 30,000 MW (a generous capacity factor of 42% – must include off-shore wind as well). They admit that they would need >200% more capacity than maximum demand.
The 21GW figure above for idle capacity would probably be larger than that if we looked on a 5-minute by 5-minute basis, rather than the day-by-day basis of the above – hence maybe it would be somewhere near the “equal to peak demand” figure you note.
Great analysis Paul. Much more realistic than the usual overly optimistic/pessimistic reports I read.
With regards to your assumptions: I think it likely that geographic site diversity will be become more important than capacity factor over time. If your wind farm has the same output profile as everyone else you can expect to get lower spot prices and vice versa. Wind turbines designed for low wind conditions will be enable the move away from the southern coast. We haven’t even touched the wind resources that aren’t world class. Anyway it is likely that there will be a smoother profile than what you have extrapolated but it would not remove the need for back up supply.
I did some research on lithium batteries a couple of years back, and since then it seems that project costs have dropped by 50%. If the battery technologies keep moving at that rate it won’t be too long before they stop being horribly expensive and start playing some role in the energy market.
The Germans and Danes have put some effort into turning an oversupply of wind power into hydrogen. This may also become viable in the future with an international gas price on the east coast and occasional electricity gluts.
According to the ESAA’s latest Annual report, there is well over 45GW of non-wind generation capacity installed in the NEM. In addition, there are periods overnight on almost any evening, when over 21GW of electricity generation is “on stand-by”. Therefore, it is not clear what the significance is of your conclusion that 21GW of non-wind generation capacity will be requrired “on standby” should wind energy achieve a 37% market share (more than twice the current LRET target).
If you are making the point that no additional “back-up” generation is required to achieve much higher levels of wind generation then I would certainly agree with that. AEMO’s latest SOO states that no new thermal generation (“back-up” or otherwise) will be required in conjunction with the current LRET target, so AEMO would apparently agree with such a conclusion as well.
My main motive in have a look at this case was just in seeing how much wind would theoretically fit in the NEM under the assumptions listed (i.e. not related to current MRET target level).
In terms of the energy (and hence capacity) gap that would arise between windy and non-windy days, it would follow that this would also be a hypothetical case.
It’s probable that the challenges to be addressed are both:
1) Technical, in terms of balancing supply and demand over various time-frames (above is just day-by-day transitions)
2) Commercial, in terms of some form of changed compensation structure to provide the residual plant some incentive to remain operational for lower energy volumes (hence revenues in an energy only market), whilst still remaining available for much the same level of capacity reserve requirements. One option might be just to continue with the current structure, but understand that there’d probably be a lot more volatility present.
There is currently plenty of peaking capacity that rarely gets used. In the AEMO 5-minute production data for stations on the network, many of the Queensland gas power stations rarely get used. I have never seen power output listed from Snuggery (near Mt Gambier), nor Bell Bay gas. The hydro stations seem to be well below their potential for balancing the peakiness of wind. Tumut 3, for example, has 1500 MW of generation capacity and about 200 MW of pumped storage. If we were approaching the limits of wind into the network, it would be operating at close to capacity at times of low wind and high demand, while under the converse conditions its pumped storage would be operating at full capacity, and the 200 MW of pump from Lake Jyndabye to Eucumbene, After looking at many days of the AEMO data, I have never seen this occur. With wind currently supplying such a small proportion of the Vic-NSW demand, there is as yet little need for pumped storage to balance out the vagaries of wind.
The gas backup capacity will become much more expensive once the LNG plants at Gladstone start competing on the Eastern Australian gas market. Additional wind capacity will mean we won’t need to use as much gas in power stations. The price relativity between wind and gas will be less, but we still have the capacity built at a time when gas was cheaper.
My personal preference is that as wind is now a relatively mature technology, our public sector by incentives should be moving to geothermal, which is scheduled capacity. For example can we decrease the cost of power produced by this technology by the horizontal drilling now used by the fracking industry ? Instead the erratic wind power is making the price of renewable energy certificates too cheap, and hindering the development of geothermal.
I’m not sure I fully understand your analysis. Am I correct in saying that you’ve scaled up wind production until it achieves 100% of daily energy on a single day, and using that as the maximum capacity/energy that can be installed?
If so, I don’t think that’s a particularly meaningful number – it’s quite feasible you’d install more wind, and spill some energy on that day. After all, we currently install much more capacity than the peak demand day.
Ultimately, the penetration of wind comes down to two factors: economic and technical. There’s no technical reason why you couldn’t install much more wind than that, and curtail it on the highest wind/demand ratio days. Indeed, some of ROAM Consulting’s integrated resource planning modelling (slide 15, although the full detail isn’t on this slide) shows that a high penetration of partially curtailed wind can be least cost, particularly under moderate carbon prices and high gas prices. (E.g., if the LCOE of gas was $120/MWh and of wind was $100/MWh, then wind could be curtailed by 15% and still be least cost.) Longer term, this seems the sort of world we’re approaching. If costs are higher however, then your argument is probably more accurate, but you still need to consider economics.
On the technical side, this could potentially act as a limiting factor on penetration of renewables; AEMO suggested that the power system is operable with at least 15% of energy in each period coming from synchronous generation. More sophisticated studies are needed to understand exactly how much inertia, spinning reserve, etc., is required online.
This is something we’re thinking lots about, and would be very happy to discuss further.
An outcome I see is that, when we get over our infatuation with solar and wind, peak and intermediate gas generation will fit perfectly with baseload nuclear power. It is easy to envisage a future with 75% of electricity generated by nuclear (as in France), 6% hydro, minor renewables and the remainder by gas.
This would be the least cost way to generate electricity with low GHG emissions, as the CSIRO eFuture scenario analyser shows: http://efuture.csiro.au/#aboutsite
CSIRO reference BREE who state a LCOE of $100/MWh without considering disposal or decommisioning costs. They also assume it can be built in 6 years rather than the 12 that is the norm overseas. Pretty unlikely for a first of a kind build. They also assume a 60 year life span which certainly reduces the LCOE. What if it only makes 50 years? or 45?
Nuclear is great if you have cheap labour or no local resources. Both wind and coal-CCS are cheaper and more socially palatable.
It seems like it man be time to level the playing field for generators by removing the RET and other distortions favouring renewable energy. It makes no sense to encourage residential solar PV given it costs $600/tonne CO2 abated and wind costs over $100/tonne CO2 abated.
We need new distortions that favour nuclear! Or why doesn’t the taxpayer just fork out $180 billion for 30 GW of nuclear reactors. Are you with me Peter?
With the politicking about the repeal process underway, it’s probable that the same sort of pattern will occur (where circumstances allow) up to 30 June 2014 – whereby hydro generators will run hard in the period to 30 June in order to secure the benefit of the carbon-inclusive spot price before it disappears.
Thanks for another excellent example of the real costs of political interference (i.e. with carbon taxes, etc) in energy markets. What a blunder (net cost of ETS would be $1,345 billion to 2050 according to Treasury figures)! Will we ever learn?
Mortlake probably has the cheapest gas of any gas-fired power station on the NEM. Its owner Origin owns the gas pipeline from the Port Campbell gas plant, so it doesn’t need to pay third-party pipeline owners. While it doesn’t own the Iona gas plant at Port Campbell, it does have a majority share in two offshore gas wells. So it’s strongly vertically integrated, and thus not subject to different financial pressures to other operators. Compare that with Tamar Valley, which requires the use of a premium 3-rd party asset to transport gas under Bass Strait, then another premium third-party asset to convey the power back to the mainland. Furthermore, the spot price of gas in Victoria is usually lower than in Adelaide or Brisbane. So Origin have done their sums well in choosing the location.
In recent weeks, Mortlake has been producing power right through the night, which is unusual for a single cycle gas turbine station. Several NSW generation units at Liddell and Eraring have been out of action, and at night net power import into NSW has been over 1000 MW. So is the cost of coal + carbon tax off-lining some coal-fired generation ?
Thanks for your clear and precise comments – perhaps there’s a role for you in posting on WattClarity 🙂
The shoulder seasons are normally maintenance periods – but with the overhang of surplus capacity, the incentive to work around the clock to bring units back is not what it once was.
Paul, it is refreshing to see someone such as your self question the myth that because renewables effectively bid in at zero that it always results in lower prices. If generators that are dispatchable are relying on pool and contract revenues rather than REC payments, then they will endeavour to “game” prices to maximise total revenues. I think it’s also important to note the effect of less gas contracting by peaking OCGT’s that will mean higher SRMC’s, shorter run times and consequently higher prices.
For those particularly interested in where Electric Vehicles might be headed, these reports written by Don McPhail as part of his E.S. Cornwall scholarship-sponsored world travels will be of some additional value.
There are multiple proponents of “EV’s as energy storage”. While I concede the possibility, I’m not convinced of the actuality.
* The EV owner wants to drive home with a charged car. Peak lopping of commercial demand from EV V2grid could leave him with barely sufficient charge to get home. Yes this could be managed, – my point is that it needs to be managed, and that the “free lunch” is a pretty meagre feed.
* Charge cycling of conventional batteries decreases battery life. While this is apparently less of a problem for Li-Ion and some other new technologies, the EV owner still needs to be convinced of this
On the other hand, EV’s at parking stations throughout the day might well be a convenient sink for renewables generation. With some smarts in the charging systems, varying charge rates could significantly reduce the need for alternative back-up generation for renewables. For example, sudden reductions in PV in a large regional city as clouds passed over could be offset by suspension of EV charging, rather than requiring a step-up of gas turbine generation.
I agree Phil, until batteries are really cheap no-one would be crazy enough to cycle them for a measly dollar or two. And if offpeak flattens right out with all the new dispatchable demand there might be even less financial incentive to so on a day-to-day basis.
It was a good presentation from Owen, should this scenario eventuate I cannot wait for the relevant network and retail tariff for the new peak period between 11pm and 2am.
Why is the SA price close to $200/MWh, with a much higher spike into the thousands, when the Vic interconnector is running at only half capacity ? Surely the interconnector would curtail the price spikes unless it is at capacity.
It is also surprising that despite the forecasts of hot weather, there was absolutely no output from the Dry Creek gas peaking plant. Most of the peak was supplied from Torrens Island. Surely on days like this the Dry Creek plant would be made ready for production, and be placing bids below the $200/MWh spot price.
Hi Thomas,
Can you please tell me in simple terms why a private sector retailer had to be placed in the supply chain.
If a responsible government was going to keep prices to a minimum, would it not make sense to not have yet another layer of skimming in the supply chain.
I believe that in some instances the government has had to pay these retailers some compensation for lost revenue or something like that.
I simply can’t understand how anyone could take such a simple issue and complicate it.
My electricity bill has gone through the roof in recent years.
Your final comment is a bit timid on reasonable trend line analysis and extrapolation though Ric; to paraphrase “In the final analysis, it may not matter that much to the solar industry whether $40 is obtained through the Clearing House or slightly less than $40 is obtained through the market a little earlier.”
You are wrong I predict, and it will be a very important point when the possibility arises.
Consider this scenario; currently agents trading STC’s on behalf of liable entities still have a very slim business case that they can put to their customers, that they can continue to provide them a service and exist on the slim differential between current STC price and Registry floor price [$40], which the liable entities know they will have to pay once the clearing house comes into play.
So up to about $39.50 some traders may be inclined to still trade, but anything over $39.50 their ability to continue effectively and profitably evaporates. They will not volunteer to continue to act as a buyer for liable entities for no return. So the buyer will start to purchase from the clearing house direct without the services of an agent.
The agents will exit the business and move in another direction entirely. Once the clearing house comes into play it is likely to continue as the sole agent for STC sales unchallenged. Trend lines are very important tools and they tell us a lot about future likelihoods for PV installations in Australia.
Pardon my ignorance…but is this high, near record demand happening in spite of the high solar pv penetration in QLD? I had thought the solar influence would have considerably flattened the peak load curve which I presume is largely caused by air-conditioning.
How demand grows (or declines) is a result of the interaction of many factors – some supporting growth, some promoting decline.
1) A couple years ago we posted about these factors that were coming together to deliver the decline in (average and peak) demand seen in the past few years;
2) More recently, Hugh Saddler has put together this detailed analysis that highlights the magnitude of the effect these factors are each having on demand suppression.
Interestingly, the effect of solar PV is currently, according to Hugh’s numbers, five times less significant than the most significant factors (i.e. 7% contribution compared with 37% to a decrease from Business as Usual). I’d stress that this is “currently”, as it may be that the effect of solar might rapidly increase, if the take-up rate continues as it has in the past couple years – though, as Ric noted here, there has been a cooling in the take-up rate despite decreasing installed costs.
So, Neil, in answer to your question – scheduled demand (which AEMO has to meet in dispatching the market) would have been higher without the PV that’s operational now, but the amount of PV that’s currently installed is too small to prevent such peaks ever* occurring.
* as to whether or not that would be a good thing – economically or environmentally – each reader will have their own personal point of view
thanks Paul. Yes it does. I had obviously thought that solar pv was having a more significant impact. I will read Hugh Saddler’s report and others you’ve referred to.
Have been watching this with interest today Paul.
Can you offer any insight into why Tas has been trading at -$79/MWh all day? Struggling to understand the bid strategy that would cause this outcome…
The various emergency departments in southern Australia would probably also wish that the heatwave, at least, would be over.
I see the predispatch forecast is for over 33,000MW today – which would make it, I think, the highest the demand has been since this hot week in February 2011, nearly 3 years ago.
Maybe we should allow another round of entries, for a consolation, consolation prize?!
It will be even more interesting this time next year when the QLD LNG plants start operating and there is no fuel available for baseload CCGT power stations and a further 1900MW of generation in the NEM will not be running
While I loathe the heat, watching the NEM on these days is lots of fun. While South Oz is hot, you cannot underestimate how the wind generation has changed the SA market, compared to the similar temperatures 5-6 years ago.
It is interesting that these high network demands are occurring despite over 2 GW of solar PV capacity across Australia, and that market penetration is particularly high in Adelaide. A South Australian assessment concluded that PV had a capacity factor of 40% at the time of peak annual demand. As solar panel technology improves, we should expect this to increase as panels are developed that are less affected by high temperatures. Panels that perform well under high temperature conditions should be a much more cost-effective solution to heatwave demand that additional poles, wires and generators.
Another interesting feature is that MurrayLink (connecting South Australia to Mildura) has had minimal capacity to import electricity into SA, and has in some of your previous posts has been exporting from SA. Under normal conditions MurrayLink can import 200 MW into SA. However there appear to be constraints on the distribution system to the east of Mildura. There is one 220V line from Geelong, which serves Ballarat and Horsham before reaching Mildura. Another starts in NE Vic and serves Shepparton and Bendigo before Mildura. An third line runs from Wagga Wagga to Broken Hill with a cross-border connection to Mildura. The first 2 of these lines are no doubt close to capacity serving other cities along the route, particularly constraints on the Geelong-Ballarat sector. Extra capacity from Geelong to Ballarat could free up more capacity further up the line.
Apologies, Interested Observer – must be too much going on at present!
Error #1 – meant to type Torrens Island, which actually does have eight units. Have noted above.
Error #2 – one of our other readers has pointed out that the peak demand on 29th january 2009 was actually 10,496MW as we showed back then in this post. Will update image above also.
The arrow to Anglesea as having a transmission constraint is surely in error. It has its own 220 kV line to join the grid at Geelong. Surely the arrow is meant to go to Morton’s Lane. This wind farm shares the same 66 kV line to Terang, where it joins the 220 kV network. However today there would be plenty of local demand from the 66 kV line, with air temperatures of 42 C in the Western District. I work in the area, and our poor air-conditioner at work was blowing out air at 41 C !
It is interesting to note that Morton’s Lane Wind Farm consistently has a higher capacity factor than either Oaklands Hill or MacArthur, even though it shares essentially the same wind resource. Morton’s Lane is not located on a noticeable prominence – perhaps it’s simply better design and operation.
PS. Morton’s Lane shares the same 66 kV line as Oaklands Hill, and is subject to constraints, whereas the MacArthur Wind Farm feeds the 500 kV line to Portland, and does not have transmission constraints.
Thanks for all the comments the past couple days (and the compliment).
I have made a couple errors today, but the bit about Anglesea being subject to constraint is now (oops, not) one of them. The “+” symbol on the display for that station (and the “-” for the AGL Hydro plant) means that they are subject to a bound constraint – at this time it was the one AEMO calls “V>>SML_NIL_1”.
AEMO provides the following as the reason for the constraint “Avoid overloading Ballarat to Moorabool No.1 220 kV line for loss of Ballarat to Elaine to Moorabool 220 kV line and Mt Mercer Wind Farm” – the user can drill into the constraint in ez2view to get this detail.
The “+” symbol for Anglesea means that the constraint is (slightly, because of a low positive factor) trying to wind the output of the station down.
A user with a steady usual load doesn’t even need to participate in the NEM pool in order to benefit from negawatts. It is a two-step process: (1) They write their PPA with their retailer at pool pass-through prices (plus network and other charges). (2) They enter a hedge with a retailer (can be same or different retailer) or other financial intermediary for their usual consumption profile. They will collect the high spot price from the hedge counterparty when they reduce their consumption below the hedge quantity. And if they don’t reduce, they are still hedged against the pool price. That is, you don’t need a Rule amendment, and you don’t need to play the pool, in order to get value for your negawatts.
Thanks for the added insights – some of our clients do a variety of hedging (all sorts of permutations, we understand) to back up their physical ability to curtail.
This also varies by season, and with their order book, etc…
Hi
I’m enjoying your analyses – thank you. Given that Qld has more than double the PV capacity of SA, and around the same rate of penetration, I’d be keen to see your view some time as to the impact that PV has on summer demand in Qld, especially at critical peak times. You may or may not have seen this – http://reneweconomy.com.au/2014/solar-saved-southern-states-from-new-and-costly-demand-peaks-92609
Regards
Dean
Paul
Thanks for the article. I have looked at the SA demand and price curves for the hot days last week and they do show the same “jagged” peak you refer to in the article.
I know that some users that take spot risk have a staggered load shedding program with less essential demand being load shed at much lower spot prices than MPC, with trigger prices for least needed demand occurring below $1000, with other demand being shed at increasing price triggers – essential services stay on regardless. The challenge is to get the “essential; demand as low as possible – and some have found that this essential load can be less than 10% of normal demand.
Others install their own generation, still load shed and sell back into the grid and not only avoid the high rice events but contribute to the capital cost of their generation plant through the revenue earned from the market.
So there are number of ways to skin this cat!
Regards
David
Yes, we also know of people doing this – and there are, no doubt, others as well – some of whom probably keep their cards close to their chest as a piece of competitive advantage in their sector.
>”5) As others have noted, we see that the wind farms continued to produce with the expected intermittent pattern. In particular we see an inverse correlation between wind farm output and spot price, which is entirely expected (after all, as others have pointed out the wind farms help to lower spot prices when the wind blows – so the inverse is to be expected).”
Is this an impartial statement? I would have said the most obvious conclusion regarding wind in this chart is that when demand is high wind power drops.
We could draw the following conclusions about wind generation from this chart: Wind does not show up to work when most need and capacity credit should probably be lower than AEMO says.
It is likely wind is causing a lot more increase in the cost of electricity than simply the spot prices might suggest.
Importantly, these issues will become much more of a problem, and electricity prices will be forced disproportionately higher the more wind capacity is added to the NEM.
Hi Peter,
Wind IS proportionally lower during peaks, about 20% less on average in SA. It can seem like more because the older wind farms are non-scheduled and their production is subtracted from demand. In NSW wind production is more closely correlated with demand. This can be seen by the dispatch weighted prices that wind farms receive. In SA a WF gets about 20% lower spot prices than the average, in NSW it’s only a few %. Seems fair given my first statement?
There are no capacity credits in the NEM. In WA where there are capacity credits WFs receive a much smaller credit. The credit was originally equal and the IMO reduced it after review.
In fact WFs are lowering the wholesale price of electricity. This is necessarily true of any new generator that produces energy. In SA wholesale prices have dropped over the last 6 years while they have risen in all other NEM states, during which time the only new generators installed have been WFs. Also of interest the frequency of very high prices has fallen more than the frequency of very high demand days. Sometimes the wind does blow at the right time, however this makes for a boring headline.
This whole discussion about who did what is rather silly. The only conclusion I take from this is that a diverse mix of generators is ideal.
“Gas power does what it is supposed to”
“Wind power doesn’t do what it wasn’t built for”
Unfortunately all of this partizan bickering is making the average Australian even more confused about energy. Thanks for the wrap up Paul.
I was curious to see if it would become so hot that many wind farms would lose generating capacity due to high temperature shutdowns (which often occur around 40 degrees celsius for many models of turbine).
Though it’s hard to discern from generation data exactly what caused a shutdown, this didn’t seem to impact the fleet as a whole, which is good to see.
And almost every other CCGT power station in the NEM will go the same way in the next 12 months. Most operators will not formally announce mothballing and will say that the units are available for peaking duty, but as most of the plants do not have bypass stacks they will be useless as peakers.
Each gas-fired plant will have their own local considerations to work through, with factors including type of plant (e.g. cogen/not, combined/open/other, etc…) , contracted gas supply arrangements, location, portfolio dynamics and drivers etc…
We previously commented on Tamar Valley as just one example.
However in an environment where gas prices are rising (if current trends continue), and (again, if current trends continue) wholesale electricity prices remain capped by declining demand and increased non-scheduled or semi-scheduled generation, then the changing spark spread could be expected to have other impacts.
What’s shown in the NEM-Watch snapshot above above are boundaries between NEM Regions, as distinct from state boundaries (though in practical terms it’s pretty much the same thing at present).
Hence, it’s more a matter that Canberra is tightly meshed in the NSW region so does not need its own pricing region – moreso than the fact that its electricity demand is relatively small.
For many years at the start of the NEM there was a Snowy Region between NSW and Victoria – as captured in this snapshot on 7th December 2005. Snowy was abolished from 1st July 2008 with part allocated to NSW and part in VIC.
Hey Tom,
Dr Mark Diesendorf told us for years that a fleet of geographically dispersed wind farms will
displace base load coal-fired powerstations. Policymakers are still proceeding on the basis of his
now-famous “The Base Load Fallacy”. Remember it? So, let me correct your statement above here
to now read: “Wind power doesn’t do what it was built for, which was to generate base load
power in place of coal”. And to add, “What’s more, it doesn’t matter how many wind farms are
built, it never will.”
Nobody is “bickering” here, Tom. Peter Lang is merely stating the facts as they are. He is
absolutely correct when he says that wind does not show up to work when it is most needed.
Wind is an abject, futile, hideously expensive, and dangerous-to-grid-security, failure.
May I too add: Thank you for the analysis, Paul. It is most informative.
Regards,
Paul
Renewable energy primary justification is that it is expected to cut CO2 emissions. But it is ineffective at doing so and hugely expensive.
Renewable energy proponents like to make claims about how much capacity has been built, and the rate of growth of capacity. But these claims are highly misleading. They should talk only about how much firm capacity has been added; the cost per kW of firm capacity; the proportion of firm capacity compared to the total firm capacity of the grid; and the cost per tonne CO2 avoided.
About 5%-10% of wind capacity is firm capacity and that will decrease as the proportion of wind capacity increases in the grid.
The proportion of solar PV that is firm capacity is also very low, and will decrease as the proportion of PV in the grid increases.
Wind and solar are about half as effective at avoiding CO2 emissions as is claimed by their proponents. And they become less effective as their proportion of the electricity generation increases.
Because of the low abatement effectiveness, the abatement cost ($/toinne CO2 abated) is about double what is claimed by the proponents.
I agree with Tom. “Unfortunately all of this partizan bickering is making the average Australian even more confused about energy.” The ridiculous claims and twaddle by the renewable energy proponents should stop.
I think you are missing the other option that will be relevant in many locations. Community ownership can be considered a mix of private and utility ownership where the term “distributed” really comes into its own – and local smart-grids can implement the integration of all relevant options in the most economical way.
Local smart-grids (community or utility owned) will need to operate like the internet, where various levels of operation will occur to meet current conditions. As far as end-users are concerned, their power supplies should be seamlessly selected from the available supplies – via the transmission system or from local sources depending on availability, security and cost considerations.
There could be others evolve, as well – in the NEM there are a couple of Demand Response Aggregators operating in a variety of forms. It may be them that take the step of owning and dispatching the resources.
Hi Ian,
I think you raise a good point, local mini grids have many advantages and I think there is definite potential for these, particularly in rural areas. The business model will be much more of a complete energy provider, supplying both generation and distribution (gentility?). There is the potential for a large cost saving with this model, and I think it will become more common as the price of storage drops below distribution infrastructure costs.
Jono.
Hi
just had to say that a co-op and that is what you are getting at will not work (sorry) no co-op works that good.
Stand alone is the only way to go. Generate your own power,store it in battery’s and use it when you need it.
Now generating power and have it on hand is the hard part so it seems I must admit that it could not be done however I have seen a genorator that works off a 12 volt system and genorates it’s own power running an inverter that runs this persons home. He made it in his shed.
It was something I had never seen before and I have been round for 1/2 a century. It really did work how I do not know! Will he make another one not sure! Will he sell it I can give the answer to that NO I tried to buy it but no way and I had to make sure I did not tell anyone his name or where he lives.
Thanks
David.
“No co-op works that good” – Have you heard of Fonterra? It is the largest company in New Zealand and one of the largest dairy companies in the world which is a cooperative owned by dairy farmers. WIth $18B in annual revenue and $700M in profit I would say Fonterra “works good”
I would have thought coops for electrons would suffer seriously with small scale operation due to the problem of peak demand and all electrons not being of the same value at varying times. The coop would clearly need to engage itself with demand management- ie smart meters and peak pricing and the investment and admin of that on a small scale may well prove the greatest hurdle, absent the economies of scale of Big Gen now.
As well we live in a mobile world and attaching a lien to my home via a street, block or whole suburb power coop would be one I’d have to carefully consider buying into. After all if I sell up and move, there’s no guarantee the next buyer wants to be shackled with my home’s power coop obligation. That would leave most to consider stand alone system economics only and their risk return should there be a change of plans and any forced move in future. Again Big Gen has the edge over any small scale coop there.
This is great news. I can now replace my very dated and carbon intensive charcoal BBQ with a lower carbon emission alternative. Coincidentally, I had just run out of coal anyway.
Interestingly, my entry was based on AEMO’s 2013 NEFR forecast, but as we all know that doesn’t forecast NEM wide maximum demand, and provide probability of exceedance numbers rather than a single number. So a lot of crafting went into this, including finding reasonable regional diversity numbers (we were in a ENSO neutral period, so I went back and looked for similar years as guide) and I assumed it would be a fairly hot year based on the temperatures up till around Christmas and picked a number closer to the 10% POE than the 50% POE. Finally I then subtracted the forecast level of demand side participation during high price events.
There was a diverse range of entries – ranging from W.A.G. up to very scientific approaches (such as yours).
Well done – and thanks for the entry, and for sharing your (formerly) secret methodology. We’ll look forward to seeing a greater number of close entries next summer.
Hi Paul,
I assume those prices aren’t adjusted for inflation?
$35 in 2002 would be equivalent to $47 in 2012 (3%). I guess it’s still not a huge difference.
I would be interested to see the increase in demand over that time period. Did the dry conditions also drive higher consumption?
Prices shown are not adjusted in any way – they are published spot prices from NEMMCO/AEMO.
As many have noted, wholesale spot prices have not really escalated over that period, but instead can be understood by supply/demand dynamics (i.e. higher prices when short generation, lower prices when long).
With respect to your other question about demand levels, you’d be able to easily perform your own analysis with a licence to NEM-Review.
This is not impartial. It is clearly pushing for renewable energy. There is not a mention of the much cheaper and batter fit for purpose technology to cut emissions, i.e. nuclear energy. Why not?
Nuclear power is better than renewable energy for meeting all the important criteria. Renewable energy cannot be justified, on a rational basis, to be a major component of the electricity system. Here are some reasons why:
1. Nuclear power has proven it can supply over 75% of the electricity in a large modern industrial economy, i.e. France, and has been doing so for over 30 years.
2. Nuclear power is substantially cheaper than renewables
3. Nuclear power is the safest way to generate electricity; it causes the least fatalities per unit of electricity supplied.
4. Nuclear power is more environmentally benign than renewables.
5. Material requirements per unit of electricity supplied through life for nuclear power are about 1/10th those of renewables
6. Land area required for nuclear power is very much smaller than renewables per unit of electricity supplied through life
7. Nuclear power requires far less expensive transmission (much shorter distances and much smaller capacity in total because the capacity needs to be sufficient for maximum output but intermittent renewables average around 10% to 40% capacity factor whereas nuclear averages around 80% to 90%.
8. Nuclear fuel is effectively unlimited.
9. Nuclear fuel requires a minimal amount of space for storage. Many years of nuclear fuel supply can be stored in a warehouse. This has two major benefits:
• Energy security – it means that countries can store many years or decades of fuel at little cost, so it gives independence from fuel imports. This gives energy security from trade wars and military conflicts
• Reduced transport – nuclear fuel requires 20,000 to 2 million times less ships, trains etc per unit of energy transported. This reduces shipping costs, the quantities of oil used for the transport, and the environmental impacts of the shipping and the fuel used for transport by 4 to 6 orders of magnitude.
There is no rational justification for renewable energy to be mandated and favoured by legislation and regulations.
Peter,
Nukes don’t make sense for Australia. Capital costs are too high, water consumption is high, peak consumption in Aus. is high. By all means it’s great for the eurozone and the asian power economies.
Addressing your points:
1. France has less than half the peak load per capita of Aus. and significant import/export capability with neighbours which helps counter nuclear inflexibility. Their baseload is also higher per person. Nukes are perfect for them but the comparison doesn’t work for us.
2. Nukes are only cheaper where labour is cheap. The Brits tendered for a new nuke plant recently and the best price they were quoted was the equivalent of $170/MWh. Significantly more than solar or wind in Aus. and they have better nuke experience and easier procurement. BREE reckon $100/MWh with no decommisioning costs and a 60 year life. 60 years is significantly longer than the typical 40. Either way it’s still more expensive than $80 for wind. You’ve probably also noticed that PV is cheaper than retail, can’t see anyone getting residential nukes…
3. True but this includes hydro which has had some disasters. Renewables and nukes are both much, much safer than coal. The difference between them is not significant.
4. True but again we are comparing to coal which is much worse. Slight differences between the clean techs is nothing compared with getting rid of fossils.
5. Sounds right. We are running out of uranium faster than steel though…
6. True, even though mine area is large. This is why nukes are so great for Asia/Euro. Land is not scarce here and wind, solar and all other industry is much more productive per hectare than agriculture (by a few orders of magnitude).
7. True but this is only a small part of customer’s costs. The reduced cost of renewables compared with nukes offsets this.
8. No one is building breeder reactors. They cost too much. There’s no point turning cheap fuel into very cheap fuel if it increases capital costs. Also people get worried about plutonium.
9. True but again this is an advantage for countries with no resources. We are not threatened by a lack of energy supply.
Why am I making a big deal about this? Because advocating for nukes distracts from actually reducing emissions from the electricity sector in Australia. Australia is too small to build nukes at scale, or profit from fuel processing. Let’s continue supplying the U to the rest of the world, they need it more than us.
Thanks for putting up such a carefully considered and insightful set of comments, including those in relation to the role and uses of market modelling. I don’t have time to respond in any detail, but I will correct you on your recollection about the original purpose of the RET Act. Here’s an excerpt from the 2003 Tambling Review of the operation of the original Renewable Energy (Electricity) Act 2000:
1.1 In November 1997, the Prime Minister, the Hon. John Howard MP, announced
a package of national greenhouse response measures, Safeguarding the Future: Australia’s Response to Climate Change, which included the following initiative: Targets will be set for the inclusion of renewable energy in electricity generation by the year 2010. Electricity retailers and other large electricity buyers will be legally required to source an additional 2 percent of their electricity from renewable or specified waste-product energy sources by 2010 (included through direct investment in alternative renewable energy sources such as solar hot water heaters). This will accelerate the uptake of renewable energy in grid- based power applications, and provide an ongoing base for commercially competitive renewable energy. The program will also contribute to the development of internationally competitive industries which could participate effectively in the burgeoning Asian energy market. 1.2 This initiative, now known as the Mandatory Renewable Energy Target (MRET), is one of a number of Australian Government greenhouse gas abatement measures and has a strong focus on renewable energy generation and industry development. This focus recognises that emissions from fossil fuels used in electricity generation comprise around one third of Australia’s total greenhouse emissions, and that increases in market penetration of renewable energy can help displace fossil fuel emissions and improve Australia’s greenhouse and renewable energy industry performance.
I think that makes it clear that a core initial policy objective of the RET and its successors was reduction in GHG emissions from the electricity sector.
Thanks, especially, for pointing out 1 of the mistakes I made in the post (it’s almost certain there is more than just that one, so will look forward to other comments helping me see other slip-ups).
Curiously the one you note illustrates the error in just relying on my memory. Obviously my memory is not what (I remember that) it once was!
On “the age of entitlement is over” I dont agree that the RET can be classified as an entitlement. When the MRET was introduced, and through subsequent amendments, successive governments invited companies to invest in renewable energy infrastructure. Whatever the objectives behind the scheme, it was an invitation to commercial entities to invest and help achieve a national goal. It should not be confused with the car industry for example, which effectively would give an annual ultimatum to governments – “would you like to subsidise us or should we go?”
“Entitlement” is such a loaded & divisive word – bit like an obsessive focus on “Rights”. Certainly achieves an effect for Treasurer Hockey in its use, but what are the long term consequences?
Regarding the RET, it seems that there’s at least 2 categories:
1) Capacity already on the ground and that has been put there on an expectation the RET would continue into the future; and
2) People who’ve invested in projects not yet on the ground (shovel ready, or still on the drawing board).
If the RET is changed (which seems likely) what are the implications for (and compensation options for?) each.
Just one more example of the policy mess the NEM seems to be increasingly burdened with …
{but it’s not specific to energy – being a software company we’re also reading commentary about changes in the Budget about changes to funding/grant arrangements there, and the effect that this will have on sunk investment}
As a quick PS to my article above, this description of the RET on the Government’s website is doing the rounds on Twitter and would have been useful for me in helping me avoid the mistake I Allan pointed out above.
The “Notes of LETAG meeting” which were leaked and posted on the Australia Institute website in 2004 clarified some of the thinking around the RET. It was ostensibly intended as a response to Kyoto, but was supposed to provide a kick start for innovation and Australian industry – it was never intended as a subsidy for imported wind power. The difficulty is that, once enacted, it created a whole industry of corporations, consultants and lobby groups who relied on the RET since the self-supporting industry development simply didn’t occur. As I recall, in 1997 Robert Hill took a proposal for a $10 carbon tax to cabinet but was rejected. Somewhat ironically, this would have been a far more economically efficient policy and the NEM would have determined the most efficient level of wind and solar, whatever that would have been.
To add to your list of recent RET modelling I would add RepuTex’s analysis from April 2014 that looked at the implications of a downgraded LRET on the shape of the electricity mix in the NEM.
RepuTex used a bottom-up least cost dispatch model that mimics the bid and dispatch process of the NEM with a resolution of 200 periods per year. Important assumptions included in this modelling exercise were the continuation of a fixed penalty price for the RET, the ending of carbon pricing from July 2014, and inclusion of existing and committed new entrants. Scenarios considered included the existing RET, an updated 20% RET, and no RET from 2015.
One of the main findings was that once the CPM is assumed to be repealed, changing the LRET has much less impact on the resulting renewable capacity that is built. This suggests the combination of carbon pricing (even at EUA equivalent levels) and the LRET are important polices that work together to transition the NEM to a cleaner energy mix.
As for longer term implications, this study also found that new capacity will continue to favor renewables over coal and gas on a financial basis. This suggests that if NEM baseload demand continues to grow to exceed the capacity of the existing mothballed coal plants, developers will choose large-scale renewables as the most cost-effective capacity to build in the future.
Thanks Paul – your questions provide a salient reminder of the context behind how we got to the current situation on RET in the first place.
If there is mainstream political intent to retain RET as it was intended to work after the expansion in 2008/9, it does seem that a floating target, based on a % of actual load, is going to be a necessary, regardless of the level or the profile to get there. The uncertainties on load forecasts, through PV take up, home efficiency gains, customer price responses, gas pricing, etc, are much greater than they were back then. The idea of predicting those outcomes this time around with a fixed volume target that gets a ‘good’ outcome seem highly optimistic. A ‘good’ outcome means avoiding unintended costs of things like 1. over build of new renewables that don’t assist the demand supply balance and create unintended value transfers eg. between solar PV owners and others, or 2. under building renewables through too low a target that leads to more high emission plant being built and creates emissions levels above the Government’s (or the community’s) policy objectives.
The justification for policies to encourage PV is to reduce CO2 emissions. Therefore, arguably the most important questions that need to be asked are:
Q5) How much CO2 is being avoided per MWh of PV energy generated? (To work that out we need to know the amount of PV generated at about 10-15 minute intervals in the NEM and the amount of CO2 emitted by each fossil fuel generator unit – as is the case in Ireland, France and other EU countries)
Q6) What is the CO2 abatement cost with PV? (To calculate this correctly all costs that should be attributed to having PV connected to the grid must be properly attributed to the various generators)
Hi Paul,
thanks again for another interesting article.
Let me declare an interest: I have a 5Kw system and I am on the 44c rate for the rebate.
The following is a series of ‘post project review’ comments on planning and installing panels, off topic I know but bear with me please.
1. There are enough houses with panels on them now (is it in excess of 1 million?) that the roofing suppliers and the panel installers should be talking to each other about providing wind-rated fixtures so that panels can easily be installed at the optimum angle.
2. Some installations of panels are just a plain eye-sore, not if’s or but’s allowed. Not only that, they are an inefficient eye-sore, from the generation point of view. The panels industry needs to step up and ensure that people get value for money from properly oriented and installed panels.
3. Just as all diesel engines should run turbos, all domestic panels should be controlled at the micro inverter level, ie panel by panel, again so people get an output which approximates the nameplate figure. Large commercial panels don’t need this as much.
4. People on the low Fit’s, 8c and the like, will really appreciate the storage systems which are not really far away, say 5 years max, with the feverish work that is going into improved storage. The widespread use of efficient storage will be another major upset for the generators and the retailers. Storage will certainly decrease the evening peak; it will depend on the efficiency of the storage as to whether the morning peak is flattened as well.
5. The political response to domestic storage of electricity will be interesting to watch. With conservative governments in place in most jurisdictions, the response to rooftop panels has not demonstrated even an acknowledgement of the facts, let alone an enlightened approach in any sense. Hasn’t electricity reticulated from central generation been a political weapon/football over quite a number of decades? Now, in terms of metropolitan domestic generation, is it coming to an end? What will its death throes be like? To mix even more metaphors, is the ‘domestic panel generation’ tail is going to wag ‘large coal-powered generator’ dog?
6. There was a beer ad where the blokes built a combined shed where their properties met, and made it a ‘man cave’. What if the boys decided to pool their beer money, get a solar system spread across part of each of their properties, and then each take some of the output. I wonder what (being a Brisbane resident with Energex the local ‘poles and wires’ authority) Energex would say about the boys wanting that connected to the grid?
Whenever Mike Swanston from Energex contributes to this debate, he is always worth listening to. Thanks for this contribution, Mike.
Many of the people who can’t install solar would live in strata-title properties – even if they own the unit / villa / townhouse they don’t own the a roof. The individual property owner has to negotiate with their strata committee to install any solar panels, making things that much harder.
I’d be surprised if old Mrs Windsor had an interest in something as arcane as Demand Response – though I had heard that she takes a keen interest in switching off lights in that big old house of hers.
Grrat discussion on DM Paul.
There are many parties in the NEM who dont want to see DM succeed that’s why we have had such a long process regarding the Power of Choice review.
The other improtant role that DM plays in electricity netowrks and its good to see the AER incentivising DB’s for DM solutions; howver there should be a lot more of it.
Some methods of Demand Response are already succeeding in NEM (for instance, spot exposure for large industrial energy users – a method we have experience with).
The proposed new “Self-Scheduled Negawatt Mechanism” (bit of a mouthful – maybe the “Self-Scheduled Demand Sellback Mechanism”?) will be another method available, in the energy market, should it be introduced.
You note about the network side – this has not really been our focus, to date. Hence we’re less able to comment on how much of that is being used.
Without some form of dynamic network pricing, I suspect this will be problematic to introduce equitably – but with the debate going on about the see-saw of network flows at a household level, perhaps there will be other reasons to move towards network charges that are:
1) More related to kW than kWh; and
2) Change reflecting the loading on the network in that particular part of the world.
Mike Swanston talked about part of this in his “size of pipe” analogy in the forum I posted about here. Click through to the podcast.
A late comment: you may want to check on how the “Mr Blue” and “Ms Green” data series were compiled by Energex.
Obviously there has been very strong uptake of solar PV across the period shown in the second chart. So lots of Ms Greens transformed themselves into Mr Blues over the period by installing PV.
If the average annual usage data at each time point on the chart is based on all those in each segment (ie PV vs no PV) at that time, then the changing averages could in theory be driven solely by the diffusion of customers from one segment to the other over time, not by changes in the actual usage of individual customers across time.
If (hypothetically) the average usage of those who left the no-PV segment was:
a) higher than average for that segment – quite plausible as higher users with larger bills may be more motivated to install PV, and
b) roughly equal to the average of those already on PV (less clear why this would be the case)
then that could explain both the (apparently) falling average usage in the green segment and the static average usage in the blue segment, without necessarily implying any changes in the “underlying” usage patterns of individual customers. If the customers installing PV were tending over time to be increasingly large users, then that could even mask a declining usage trend for customers already within the segment.
I guess my main point is that it is very difficult to hypothesise about what charts like this one really show – let alone any underlying reasons – without fully understanding how they are constructed.
Thanks for a thought-provoking post! I have a couple of clarifications (as requested), then some comments.
Significance of the capacity market to DR
In part 4a, you asked where EnerNOC gets its revenues from if only 2% is subject to Order 745. All the other sources you mention play a part, as does our growing Energy Intelligence Software business, but capacity payments are the most significant. However, I don’t agree with your inference that this means that we’ll see correspondingly low uptake of the DRM in the NEM.
Where there’s a capacity market as well as an energy market, peaking resources (both conventional generation and DR) derive almost all their revenue from the capacity market. If it’s not mandatory to participate in the energy market (in PJM, it’s not for DR), then they many not bother with it.
If you don’t have a capacity market, then peaking resources have to make do with the energy market and associated derivatives markets. So long as the market is functioning correctly (I realise that’s a big caveat), at equilibrium they should get the same amount of revenue this way. The NEM isn’t quite a perfect market, so I wouldn’t expect participation to reach as high as it ideally would, based on fundamentals, but the potential is still significant.
Huge consumer savings from DR in PJM
In part 4e, you asked whether the reference to PJM saving $11.8b, or 40% of total costs, through DR participation was a typo or a calculation error. It is neither. The independent market monitor calculated this by rerunning the 2013/14 auction without the bids from DR and EE suppliers — i.e. procuring the capacity needed from generation alone:
“Based on actual auction clearing prices and quantities and make‐whole MW, total RPM market revenues for the 2013/2014 delivery year were $6,708,567,045. If no DR or EE had been offered into the auction, total RPM market revenues for the 2013/2014 delivery year would have been $18,535,847,876, a difference of $11,827,280,831 compared to the total based on actual results.”
This is from p.52 of Monitoring Analytics’ Analysis of the 2013/2014 RPM Base Residual Auction Revised and Updated. PJM’s auctions are run three years ahead, to give time for proposed resources to be built if they clear. Hence we know the corresponding figures for the next three years: $9.6b, $13.7b, $10.1b. Pretty soon, you’re talking real money. These savings show how much cheaper it is maintain reliability if you involve the demand side, rather than relying purely on the supply side.
Comments
As I hope I made clear in my talk last week, the kind of price-responsive demand with which you assist your customers is by far the best option for the small number of very large customers who are sophisticated enough to cope with the associated risks. By buying at wholesale prices, avoiding paying a margin to the retailer for risk management, they can come out a long way head.
The purpose of the proposed DRM is to allow many more customers to participate by introducing an alternative approach that doesn’t require them to take market price or volume risks 24x7x365 just because they want to be able to provide some demand response during extreme events.
I don’t think that introducing the confusing “negawatt” term, or suggesting that it involves the creation of a separate market for a separate commodity, really helps. There’s only energy involved. As you noted in part 3b, there’s no incentive payment; it’s not an incentive scheme.
One thing I would add: ideally, demand response should be a scheduled resource. This makes it much more useful to the system operator than price-responsive load. I haven’t seen the final version of AEMO’s rule change proposal. However, in the working group discussions before AEMO went away and wrote it, it was clear that the DRM was proposed to be non-scheduled only because AEMO considered it too hard to alter the NEM’s dispatch systems to support load participation in the near term. The intention was that this would be revisited once they had some initial practical experience with the DRM.
AEMO’s phased approach makes sense. It takes a while to get the implementation details right, just as it does for the market to adjust to broader participation, and for customers to take advantage of their new opportunities. This is why it is important that we start soon, rather than waiting until the next time it’s obvious that we’re paying over the odds due to the low level of customer participation.
Will think through the rest of what you say later, but want to ensure I understand the savings noted in the PJM capacity market, as that’s what really caught my attention!
Using the numbers you have quoted, and an installed capacity in PJM of about 180,000MW (think that’s about right – should be right enough to see that I understand the orders of magnitude):
Under the Reliability Pricing Model (RPM) we have two scenarios:
Scenario 1) Real = Total payment of $6.7b for a year across 180GW is $37,000/MW p.a.
Scenario 2) A re-calculated “no DR or EE” scenario = Total payment of $18.5b for a year across 180GW is $103,000MW p.a.
These numbers are reasonable – in the ballpark of what I’ve seen of the annual price for capacity in Western Australia, for instance.
Am I correct this far?
Can I assume that there’s about 20,000MW of Demand Response and Energy Efficiency in PJM? I think it’s less than that but the number should be good enough to allow me to get my head around this.
Assuming 20,000MW of physical generation pushed out at the top of the bid stack by the DR and EE, this means that there was about that much that bid the difference between the two total revenues.
This means 20,000MW bidding, collectively, $11,827,280,831 – or $591,000/MW p.a.
Given it’s a capacity market, I assume this can be peaking capacity, but even if we assume capacity is built for $1,000,000/MW (expensive for a peaker, I would think, even with price escalation) this would mean that there are generation developers out there pitching for a 2 year payback on investment.
The maths works out a bit differently as it is not a pay as bid auction, everyone is paid at the clearing price.
The additional cost is spread over all 180 GW of generation, not just the 20 GW in question.
There will have been 20 GW of generation bidding between $37,000 and $103,000 /MW p.a., which changes the price paid to all cleared bidders, results in increased costs of 180,000 MW x (103,000 – 37,000) $/MW p.a. which is $11.8 billion p.a., making the 20 GW of DR pretty close to the mark.
The payback period would then be close to 20 years, depending on discount rates etc.
If those same 20 GW of generators were bidding at up to $591,000, then that would be the clearing rate, and total costs would be ~$120 billion
That makes more sense – apologies for not figuring it out myself.
Are bid prices made public, after the fact? I would be interested to look at the bid stack, to understand more …
160,000MW of physical plant to bid at or below $37,000/MW p.a. – at 100% utilisation this is only $4.22/MWh capital cost (at most), unless I have made another mistake? That seems way too low to cover fixed costs.
If this is true, then the approach seems to be to under-recover in the capacity market and recover more in energy (unless the entire fleet is depreciated away)?
{please excuse the questions, if they are naive – we don’t have a capacity market to work with in the NEM}
Hello Paul, generators don’t run at 100% utilization, but the capacity revenues are still a small portion of their revenues compared to the energy market revenues. For DR, however, which is only called a few hours a year, capacity is the vast majority of its payment stream.
Unfortunately, PJM doesn’t release bid data, even after the fact. I believe this may have something to do with the fact that their auctions take place over several rounds, with participants able to adjust their bids at several fixed times.
In relation to your comments about Capacity revenue matching up with fixed costs and energy revenue matching up with variable costs, I don’t think many generators will approach it from that perspective.
A peaking generator will earn exactly as much capacity revenue as a baseload generator rated at the same wattage, but will actually earn less energy revenue, because it is only on and earning energy revenue a fraction of the time, whereas baseload is earning energy revenue during low price periods as well as high price periods.
Peaking generators tend to have lower fixed costs than variable generators, but higher running costs.
So a peaking generator, with its low fixed costs, high variable costs and low energy revenue may end up approximately covering its fixed costs with its capacity revenue, and then the energy revenue is profit on top.
A baseload generator however, has higher fixed costs but the same capacity revenue – they likely won’t come close to break even there, and they don’t particularly intend or need to. The energy market is (as Brett points out) where they make the big money.
DR, which earns energy revenue a lower fraction of the year than a peaker is even more reliant on capacity revenue (again, Brett got to this point before I did)
Fascinating to cross-reference your post with the estimated contribution from solar (http://pv-map.apvi.org.au/), and the demand charts on AEMO’s website. In Queensland on the 17th solar contributed about 500 MW at midday, or about 50% of the solar installed capacity, and there was a strong slump in demand co-incident with the solar contribution. However in Victoria a cloudy day caused a maximum solar contribution of 15%, and there was little midday slump in power demand.
On Sunday and Monday night this week there were strong winds overnight at a time of low power demand, and data from AEMO showed a South Australian price of $20/MWh compared with about $30/MWh in Victoria. About 100 MW were being exported from SA. Thermal power was still being generated in SA from Osborne (gas cogeneration, with heat sales to a nearby factory), Pelican Point (gas combined cycle) and Northern (coal). Each of these were generating about 160 MW. Some questions
1. Why is there such a large price differential of about $10/MWh between SA and Vic ? (The same occurs when SA is importing from Vic, except the SA price is $10/MWh higher than Vic). Yet much larger flows of energy occur between Vic, NSW and Qld on much smaller price differentials. Is SA’s Electranet using its market power (and wind power generators who have no-one else to sell to) to extract larger price differentials than in the eastern States ?
2. Is this pricing behaviour what was intended in the NEM ?
3. Why would the gas power station at Pelican Point continue to generate into such a low-priced market ? If it curtailed its production to below 60 MW, so there was net import into SA, then the SA market price would rise to parity with Victoria plus the $10/MWh margin Electranet build in. So by reducing production from 160 to 60 MW the owners could increase the SA price from $20/MWh to $40/MWh, and only pay for one third as much gas.
Q1) You ask “Why is there such a large price differential of about $10/MWh between SA and Vic ?”
The differential shown here ($38.92 in SA and $41.13 in VIC) is only $2.21 – however the question (about the price differential) is a common one, in any case.
If there were no constraints binding in the NEM, it would be a single generator setting the price for all 5 regions. This does not happen too often.
In the case above, the Heywood (AC) interconnector is bound (in NEM-Watch the arrow is red), meaning prices can separate between VIC and SA. AEMO is forcing power to flow west (the flow limits in NEM-Watch are explained here), meaning that the large volumes of wind in SA are suppressing just the SA price – even more today.
Q2) You ask “Is this pricing behaviour what was intended in the NEM ?”
Yes, it is. It means that, in the absence of constraints, the least-cost dispatch pattern (based on generator bids) is realised – and, with constraints bound, the “next best” alternative is used.
Q3) You ask “Why would the gas power station at Pelican Point continue to generate into such a low-priced market ?”
Other people (including ourselves) might speculate that:
(a) it might be to do with the contract cover they had sold, and/or
(b) because their short-run marginal cost is below the current spot price, meaning the revenues being generated are contributing something towards their fixed cost of operation; and/or
(c) for other reasons…
By the way, ElectraNet is the transmission company operating in SA (see our Power Supply Schematic) so does not have the involvement in the market you’re thinking of.
If this does not fully answer your questions, maybe next week I can come back to this…
Most of the volatility was in the South Australian market, which has the highest penetration of wind capacity, and has seen the greatest off-lining of thermal capacity. We could expect this volatility to reduce once the Heywood inter-connector is upgraded, and provide another 190 MW of transfer capacity to a much larger market. But this depends on whether the extra capacity is used to provide a relatively constant import or 190 MW, or whether the capacity is kept to smooth out the price peaks. It can’t do both at the same time.
To me the surprises in these data are
– why do the thermal generators (such as Torrens Island) not seem to be using the forecasts of wind power production to set the number of generation units kept on the boil ?
– do the generator operators not trust the forecasts ?
– spikes in the Tasmanian market should be easy to smooth out with some experience in managing quick-response stations such as Reece, unless the bidders want to intentionally generate price spikes
The next point of interest in volatility will be when the LNG exporters start bidding at the end of the year and gas prices start to climb. It would make it much more expensive to keep generation units at Torrens Island boiling ready for price spikes.
Some numbers I have seen indicate there is a lot of dumping of gas, with generation from the Darling Downs up significant despite warmer winter. I wonder if the other combined cycle plants are lifting production in Queensland. Is this not short term at least until the LNG plants clear the surplus.
Yesterday there was certainly a sizeable volume of gas generation in Queensland – as I noted.
What I noticed, but did not include in the post, was that the spot price for gas in the Brisbane hub was below $2/GJ – and again today ($1.88/GJ) which provides one reason why it’s being burnt in volume. See GasWatch for more.
It would be interesting to see this look at teh bid data repeated for the 17 July repeal date. My hypothesis would be the muted response from 1 July could be indicative of mixed sentiment around likelihood of the repeal bills passing the senate. However, once made certain there should be a “second wave” of bidding in the lower SRMC band.
Thanks Paul,
Interesting. How much money did Infigen make on their dispatched plant? I assume they didn’t curtail any capacity after the 5 minute spike?
Why was there a price difference of $3/GJ between Victoria and Queensland prices, when there was adequate pipeline capacity to move gas between the 2 markets ? Could it be that the pipeline owners charge this to shift gas south over this distance ? Once the LNG export plants start operation, could we expect a similar $3/GJ difference to move gas north from Victoria to Queensland, so perhaps we might see a $9/GJ price in Queensland and $6/GJ in Victoria ?
‘This article has been written in an environment where (it seems to us) two opposing views of the future of the electricity market have been verbally jousting from ever-more-entrenched positions. As I have noted before, we strive to remain technology agnostic – focusing, rather, of the market impacts of generation, whatever the source ‘.
It’s rather quaint that you don’t take a stand on the use of coal for electricity generation. It flies in the face of science as we know it surely.At least 2/3 of it has to stay in the ground for goodness sake.
Thanks Neil
I appreciate that that’s your perspective – as it is for many others as well. Our focus is on how the market is operating, under whatever the rules are at present.
Paul
“How much of this cost should be paid now, and how much later – based on assessments that technology costs will decline, and calculations about net present value using a variety of discount rates?”
If you take a discount rate larger than your rate of growth the right time to solve a problem will always be right before it’s about to happen. This is the problem with economists like Bjorn Lomborg saying that we should be dealing with the problem later. The size of the issue is too large for any one generation to deal with alone, the costs have to be spread out over many decades.
Also note that technology costs are declining because of scaling not research. Solar and wind technology is virtually unchanged since the 80s. Some improvements have been made with research, but most of the improvements are due to building a bigger factory, getting normal financing rates, and experience with installation.
Thanks, Tom
At the high end of the scale some might want to use commercial target rates of return – which would tend to favour deferral. At the other end of the scale, my vague recollection is the Stern report used a discount rate of 0?
Just one part of “a diabolical problem”.
Paul
Outstanding work. I commend the approach of always looking for an agnostic position and (although I am inherently biased) I have strived to remind readers that the energy industry is not evil per se. We need each other and your point on “what happens if” is most valid; the merit order effect is not perpetual! Good on you for sharing your incredible depth of knowledge to this important debate.
The issue at hand (IMHO) is quite simply are the non renewable sector (and consumers) simply being “gamed” by the incumbents because they can?
Clearly, the entire energy sector needs a tweak and a tune but in the meantime, its hard to know which mechanic to believe.
Great outline of the issues. I would just note that the NEO is not adequate to define what Australia needs from its electricity sector. That’s not to say the NEO should necessarily be expanded to include carbon reduction goals but that we need to recognise that carbon reduction is central to electricity policy and therefore can’t be ignored if we want long-term policy stability. The question that needs to be answered, or at least discussed more fully, is – how much carbon reduction do we need from electricity? And following from that – when, how, and who pays?
Some useful references on the “social cost of carbon” (SCC) are:
– The US government’s interagency working group, which ended up recommending four different price paths in recognition of the debates around discount rates, quantification of impacts and scientific uncertainty. “Three values are based on the average SCC from three integrated assessment models (IAMs), at discount rates of 2.5, 3, and 5 percent. The fourth value, which represents the 95th percentile SCC estimate across all three models at a 3 percent discount rate, is included to represent higher-than-expected impacts from temperature change further out in the tails of the SCC distribution.” The average 3% price is generally used as a central estimate. See also the US EPA’s page that looks at how the SCC is incorporated into regulatory impact analysis. http://www.epa.gov/climatechange/EPAactivities/economics/scc.html
– closer to home, the Climate Change Authority’s report on Australia’s carbon budget and emission reduction targets. This gives a thorough explanation of the long term emission reductions needed internationally and nationally, the role of the power sector in contributing to this, and a carbon price path that can deliver appropriately. http://climatechangeauthority.gov.au/reviews/targets-and-progress-review-3
I’d add that these 6 enduring challenges still remain, though perhaps the first one now looks like “generation capacity retirement” instead of development, especially if the RET continues in force.
Great piece. Just one correction the Earthquake and Tsunami in Japan was 11 March 2011 not in 2004. 2004 was the Indian Ocean Tsunami which occurred 26th December 2004.
A couple of points. There is much discussion in this piece about spot prices but what is the financial impact of forward contracts? The discussion on the Merit Order Effect and the “very polished” animation seems to ignore them.
The “2030 onwards” discussion about the AEMO report, where indeed they suggested that 100% renewables was technically possible, but it came at considerable cost – between $219 to $332 billion plus land cost. I published a piece on this which might be of interest to your readers:
Some believe that the spot price will lead the contract price – and some believe the opposite. Reality is somewhere in between. Oversupply of capacity, and continued decline in demand, will drive both down – until the supply side responds (or new methods of demand – like EVs – are introduced).
I listed the AEMO 2030 study because someone else would have. Did not get into costs – thanks for your link.
I notice many have made mention of the potentially large additional land cost of the 100% renewable study from AEMO. Most of the land requirements from this study is related to wind generation. The leasing cost of the land for wind generation is tiny compared to capex costs for the turbine. ie, ~$10k per turbine per year leasing cost, compared to the ~$5 million capex cost of the turbine. It is pretty much in the noise band in terms of cost.
I suspect this will be true of most of the renewable technologies.
Further to David’s point:
The cost of leasing is part of the LRMC, it’s not an externalised cost. Also the turbines and roads take up a tiny fraction of the total land area.
Even for large scale solar the cost of land is minor.
Thanks Tom
There are a number of different ways of assigning companies to generation assets – NEM-Review has three:
1) Owner – in many cases not the developer, or the company trading in the market. Some are just infrastructure investment funds. Our “Power Supply Schematic” Market Map also highlights these.
2) Bidder – which I used yesterday – takes account of the entity interacting with AEMO, and in most cases is the owner of the energy generated. But it excludes Non-Scheduled.
3) Trader – which I used here.
I understand that slicing the numbers different ways provides different results. The core finding today is that AGL is not insignificant in contributing to growing wind’s market share.
Paul
Paul, this is a result of government legislation not goodwill on the part of AGL. Their current opposition to the RET highlights this. No RET = no “AGL” windfarms. Development of them and on-sale also provided a great short term earnings hole plug via development fees. Maybe you could do an analysis of just how far out of the money those deals were – with the PPA prices published so inflated, seemingly to create up front development fees that they are now suffering “losses” on their green products – refer this weeks profit waterfall diagram….
Just for your interest; AGL is the third largest equity investor in Australian renewable energy. Infigen is top with 1,500m., Meridian is a smidgen above AGL both with about 600m. and there are another dozen players with 100m. to 500m. invested. Note that none of the other vertically integrated companies are on the list. (Bloomberg collated data)
Tom is absolutely right, most of their portfolio they developed and built but sold and signed a PPA with the current owner. Their suggested portfolio is nothing other than “greenwashing”.
There are plenty of operators out there who own more wind farms than AGL, this is just a nonsense look at what AGL has contracted. AGL sold out of its wind farms at huge profit, and now they’re trying to destroy the RET. What a joke of an article – it really reduces the credibility of Global Roam as a modelling house. Maybe disclose how much money AGL have paid to you in the last 12 months and how much you hope to get out of them in the next 12 months next time you post your next sycophantic article that attempts to gloss over AGL’s terrible renewable credentials in recent months.
Thanks for the comments. You seem to make 4 claims:
1) That asset ownership, not contracted position, is a truer measure of greenness
I understand this perspective. There are a number of different ways at looking at who’s greener than others – each has some value, so long as the writer, and the reader, understands the measures used, as I noted here. In this post, I just chose another measure.
2) That there are larger portfolios (in terms of ownership)
Some assets, for which AGL still “owns” the market risk (for both electricity and RECs) have been sold to infrastructure funds. This is not uncommon – for wind farms, or other forms of generation as well.
We’re currently finalising the 2014 issue “Power Supply Schematic” Market Map, which will enable us to collate up totals owned. Perhaps this can be another post at some stage in the future.
Happy to have a look at your data, if you can share with us (tel 07 3368 4064)?
3) That GLOBAL-ROAM Pty Ltd is a modelling house
We’re primarily a software company, not a modelling shop. Our website explains more. I think you have us confused with ROAM Consulting Pty Ltd, a completely separate company (see here for some background).
4) That this article was “sponsored” in some way by AGL Energy
This is not the case. In total over the past 12 months, AGL has paid us the royal sum of Aud$1,600 (exGST). I do not expect the following 12 months to be any different.
Regardless of what they (or any of the other 100+ clients we serve – green, black & indistinct) pay us, that will not influence what we post about on WattClarity.
Interesting article. Yes those ownership stats are based on Bloomberg New Energy Finance data, published in a recent note we released to clients on exposure to assets that could potentially be stranded by changes to the RET. We track ownership and investment in renewable assets worldwide. For more details you can have a look at http://www.bnef.com or contact us at sales.bnef@bloomberg.net if you’d like to know more.
Hi Paul
Chris, Ben and Waldo from Powershop and Meridian here.
As you know we are committed to protecting the current level of the RET. You can see many of our comments here.
GetUp and its members have a similar commitment to the RET, and went through a process to identify retailers with an aligned point of view. They didn’t find many.
GreenPeace’s guide to which you refer, and Giles Parkinson’s RenewEconomy article , show that there are a number of retailers out there who (regardless of what their marketing teams say about their renewable credentials) work hard at undermining the RET and the renewables industry. Even the more “renewable” amongst the cohort give at most faint praise to the RET, with none coming out hard and unequivocally calling for the RET to be left untouched.
We, and GetUp!, are encouraging customers who favour renewable energy to switch away from Big Three to send an unequivocal message of support for the RET. Whether they switch to us, Diamond Energy, or one of the other renewable retailers highlighted in the Greenpeace guide is not the point.
We understand that these payments made to GetUp! are going exclusively towards growing the Better Power campaign to save the Renewable Energy Target. The “owners” of GetUp! are the thousands of members who are participating in the campaign, and it’s a not for profit organisation that campaigns to support causes that its members believe in.
It’s no good to buy renewable energy (GreenPower for example) from a company who is looking to undermine or weaken the renewable energy industry in Australia. It may make you feel better, but it won’t help the future of renewables – because GreenPower only works to support renewables in an environment where renewables are supported by policies like the RET.
Thanks for the detailed comment. Helps me to understand the nature of the commercial relationship between Powershop and GetUp.
I think I understand the reasoning behind your comment:
It’s no good to buy renewable energy (GreenPower for example) from a company who is looking to undermine or weaken the renewable energy industry in Australia. It may make you feel better, but it won’t help the future of renewables
i.e. buying Greenpower now from someone achieves a 1x benefit (i.e. you buy a kWh you get a kWh) however you’re saying that you buy Greenpower from someone who advocates for the RET as well and there’s a multiplier in there (in terms of supporting growth in renewables beyond just the kWh you pay for).
My sense is that everyone will have their own personal perspective there – cost per kWh of green would be one consideration for many I suspect.
We don’t know how much CO2 is emitted because NEM generators don’t measure or estimate with sufficient accuracy and precision at sufficiently fine time intervals (e.g 15 minute intervals as in Europe) the emissions data from the fossil fuel power stations. The AEMO figures are based on long term averages and applied to all units in a power station. They make no allowance for the effects of cycling, spinning reserve, start up and shut down, or operating at less than optimum efficiency.
It’s my understanding that the data submitted by the generators to the Department of Environment as part of their reporting requirements is indeed accurate (using international best practice standard).
This information is not published by AEMO – which, as you know, is a separate organisation focused on electricity as distinct from emissions.
Data provided in NEM-Review is an approximation based on emissions intensity by station – which, as you say, does not take account of variation by load factor, etc… However it is useful enough for these high-level comparison.
Thank you for your reply. But my understanding is your understanding is not correct in that Australia does not measure emissions per unit (as the USA EPA requirements [1]. [2]), nor estimate them as is done in EU and as per the example I gave with the EirGrid web site I linked in previous comment. You can see similar web sites for other EU countries too.
This is why we have only a limited understanding about how much CO2 emissions the intermittent renewable energy generators are achieving. On grids where they have reliable information, such as Ireland which has many important similarities to South Australia, peer reviewed analyses show that wind generation was only about 53% effective at abating CO2 emissions per MWh of wind energy supplied to the grid. That is far less than is claimed by the wind energy advocates.
“Thank you for drawing out attention to this. We will ensure that the website is updated in due course with the missing text, which is:
EirGrid, with the support of the Sustainable Energy Authority of Ireland, has together developed the following methodology for calculating CO2 emissions. The rate of carbon emissions is calculated in real time by using the generators MW output, the individual heat rate curves for each power station and the calorific values for each type of fuel used. The heat rate curves are used to determine the efficiency at which a generator burns fuel at any given time. The fuel calorific values are then used to calculate the rate of carbon emissions for the fuel being burned by the generator. ”
Australia has no processes that go anywhere near world best practice for calculating or measuring (as the USA EPA requires of US generators), CO2 emissions.
Peter Lang, in the end this all matters very little because the National Greenhouse and Energy Reporting System requires measurement of actual coal consumed and sampling analysis of the chemical composition of the coal coming from the field in which it was taken. This provides us with a quite accurate measure of the carbon burnt in total over the period of the year and therefore CO2. At the same time AEMO metering measures the amount of electricity sent out to the grid by the power station. So on an annual basis even if not on a fifteen minute by fifteen minute basis we have a very good understanding of the total annual CO2 emitted and total GWh generated and the average emissions intensity. This then gives us a very good hint as to whether there might be a material deterioration of efficiency at work over time. If this were the case then we might then want to investigate what the cause might be whether it be increased variability of output due to say wind farms (I know what you’re getting at Peter) or for example a change in water content of the coal or some problem with spoiling or a range of other factors.
Oh, I see, so we’ve called in the renewable energy advocates to help out on this so called technology site. Confirms what has been evident for a long time.
Tristan Ellis (Editor of ‘ClimateSpectator’ and a passionate renewable energy advocate), you are wrong on this as has been pointed out many times in the past. Measuring electricity output from each individual generating unit at 15 minute interval and getting a very rough estimate (contrary to your claim) of carbon burnt in power stations over years does not allow the emissions avoided by intermittent renewable energy generators to be determined.
If such rough estimates are good enough why to the US EPA require measurements of emissions at 15 minute intervals along with calibration of the instruments measuring the emissions and reporting of these to the EPA? Read the links I provided above.
And why does Ireland (an example of EU method) calculate emissions at 15 minute intervals as per the EirGrid methodology quoted above.
Also as stated above, EirGrid’s emissions data allow calculation of the emissions avoided by wind generation as just 53% of what they would be if 1 MWh of wind energy avoided the emissions of 1 MWh of the energy displaced from the other generators. It is likely that South Australia’s grid is even less effective than EirGrid.
And this information is not new or rare. The analyses that have been done based on empirical data (as distinct from modelling studies) show similar results.
They also show that the higher the penetration of wind energy the lower is the abatement effectiveness.
What all this means is the CO2 abatement cost with wind energy (and all intermittent renewable energy generators, is much higher (e.g. a factor of ~2 higher) than the renewable energy proponents would have us believe.
Renewable energy is high cost for no benefit – i.e. no benefit in terms of avoiding climate damages, which is the primary justification for subsidising and mandating them.
Because Tristan has decided to post a comment on WattClarity, it seems that you have inferred (without any other evidence) I asked him to post and ‘gang up on you’.
Yet you, yourself, have commented a number of times on this post – and you know that I did not ask you to do that. Given I did not ask you, and yet you still posted, I can only surmise that you believe you made your comments because you are my Aunty?
Furthermore I would conclude, following the same reasoning, that because Giles has chosen not to comment he must indeed be my Grandfather
Can I remind you of your prior comments posted only last night – that you later requested I delete when the truth became self-evident? This was another example of you jumping to conclusions in attributing motive
As I have noted before, we have a broad range of readers (including yourself, and Tristan) who choose to comment (or not) of their own volition. As you have seen, most choose not to – we don’t mind, either way.
Once again I remind you that we are carbon (and technology) agnostic. We post about it occasionally as it is of interest to some of our readers – including, it would seem, you.
In terms of my article, above, it is not clear to me what value estimations of emissions at 15-minute granularity would achieve in this case?
Leaving aside the inherent difficulties of actually trying to measure emissions on any basis (and especially on a short-term basis), I am unsure how such measurements, when added together across each of the years shown above, would actually change the main conclusion in the article (that AGL’s larger portfolio produces significantly more emissions than any other portfolio)?
I have learnt from experience that you have your particular hobby horses that you like to repeat (on WattClarity and other sites) ad infinitum, but it does seem a little off the topic – don’t you think?
Yes, the emissions calculations uses in NEM-Review are approximations, as we clearly note here.
Any power station performance engineer will be able to explain why any emissions metric will not be 100% accurate. Calculating estimates on a 15 minute basis such as what you suggest would offer some advantages, but also carry some disadvantages.
As noted above, NEM-Review is fit for purpose:
(a) it helps a range of clients compare trajectories of estimated emissions over time, and works with information in the public domain;
(b) it helps draw out some key points for readers of WattClarity, such as in this post.
If you want to suggest to Minister Hunt that there are better methods for measurement & reporting of emissions under NGERS, I’m sure he’ll be all ears. You can reach him here.
If you want to suggest to AEMO that the estimates they publish could be improved, you can find them here.
I know that some people at AEMO do read WattClarity (though they are probably not the same people dealing with their emissions approximations). I could almost guarantee that Greg Hunt does not read WattClarity. Hence, it’s probably the wrong forum for you to be promoting such suggestions.
If we ever did come to analyse & post about the emissions avoided by any particular generation source then we might continue this discussion.
… but for tonight I’ll leave it here and (instead) think about how I would have otherwise worked out the winning lottery numbers if I had not been investing the same time reviewing off-topic comments on WattClarity.
I don’t understand the purpose of that long rant. If you didn’t call up Tristan or ‘Anon NGERS worker’ to reply why didn’t you just say so? Instead you went into a long, mostly irrelevant, rant, as has happened before in response to my comments.
You say WattClarity is technology neutral. Yet it seems pretty clear, from you’re responses and the balance of posts over time, that you are an advocate and supporter of renewable energy. The fact you frequently jump in quickly to attempt to dismiss comments and papers that point out the poor economics, high abatement cost and lack of demonstrable ability to achieve the main justification for subsidising and mandating them also suggests pro renewable energy bias – i.e. not technology neutral after all.
Your comments in the past responding to the low emissions abatement effectiveness of wind generation (and therefore high abatement cost) suggest you tend to jump in quickly on this issue and make comments that show this is not your rea of expertise. Such comments require a response.
“it is not clear to me what value estimations of emissions at 15-minute granularity would achieve”
In that case I’d urge you take some time to read and study the literature on the empirical studies of abatement effectiveness of intermittent renewable energy technologies – or don’t comment on it at all on the subject. You might like to start by studying the Joe Wheatley paper “Quantifying CO2 savings from wind power” because there are relevant comparisons with the South Australian grid http://www.sciencedirect.com/science/article/pii/S0301421513007829 )
Stating you are carbon and technology neutral may be your intention but you cannot prevent your own biases showing through, just as Tristan and Giles Parkinson can’t help their biases showing through.
If you wanted to comment to AEMO about their processes for publishing their estimated CO2 emissons figures, then this opportunity has arisen that would might want to contribute to.
Submissions due 9th October, so would be ideal if we could help you through the power station course before then!
Gents, I’ve done some work on the marginal abatement I’d like to share with you. In a nutshell I found that due to interconnection, most of the abatement from wind farms tends to be on NSW black coal generators. This is due to the market size and the fact that NSW coal has a higher marginal cost.
Renewable deployment actually displaced the least amount of CO2 in QLD due to cleaner fuel and modern plant (0.92 kg/kWh). TAS and VIC had the highest amount of abatement at 0.98 kg/kWh due to the market power of brown coal in those regions. SA and NSW were in between.
This may not seem intuitive but it does line up with some of the fuel consumption info that AEMO or Reputex put out.
Power stations must estimates emissions and energy consumption and production in accordance with the NGER (Measurement) Determination. All medium and large coal and gas power stations estimate CO2 emissions using “method 2 or 3” which involves sampling and analysis of their fuel inputs to determine emission factors. This method results in an accurate assessment of emissions with relatively low uncertainty. Reporting is subject to the NGER framework which includes a system of audit and the publication of certain data (including additional information for designated generation facilities).
Data reported by power stations are incorporated into Australia’s National Greenhouse Accounts and are used to compile scope 2 emission factors which reflect the average emissions intensity of electricity consumed in each state and territory (published annually in the National Greenhouse Accounts Factors).
“which involves sampling and analysis of their fuel inputs to determine emission factors. This method results in an accurate assessment of emissions with relatively low uncertainty.”
What does “relatively low uncertainty” mean? Please quantify the uncertainty and state how it is calculated.
The CER figures do not take into account the variations in:
– C content of fuel being consumed at a point in time
– thermal efficiency of the different units operating
– thermal efficiency of the units depending on its output
– thermal efficiency of the units as their output is ramped up and down
Perhaps you could state whether all coal fired power stations are using method 1.4 (d) and at what interval are the emissions measured (if not all, what proportion by annual energy generated?):
” (d) method 4 provides for facility specific measurement of emissions by continuous or periodic emissions monitoring.” http://www.comlaw.gov.au/Details/F2012C00472/Html/Text#_Toc330214226
The USA EPA requires the emissions to be measured at 15 minute intervals and the measurement instrumentation to be calibrated and all this to be reported to the EPA and publicly available. We have nothing like that.
Without emissions intensities from each unit, either measured as required in the USA or estimated based on measurements as listed above, assertions about how much CO2 is abated by wind generation are so uncertain as to be meaningless.
I look forward to your quantification of the uncertainty and explanation of the methodology for estimating the uncertainty. I trust you have reliable estimates for the uncertainties on the assay values of the coal consumed by each unit at 15 minute intervals.
Peter,
The power station operators will have to take regular samples of coal quality so they know what the boiler will be experiencing. I don’t know about 15 minute frequency or data availability, but it wouldn’t be hard to report if you are already collecting the data.
I have read with some interest and some amusement.
I would like to make a few comments.
Firstly nearly all the reports I have read regarding coal fired power stations and emissions are calculating or estimating their emissions based on the power sold (as shown in the AEMO 5 minute generation data) The fuel used and emissions calculated are estimations reverse calculated, using predetermined efficiencies and moisture content of the fuel. These emissions do not take into account the inefficiencies caused by idling generators to be available to back up wind farms or generator failures. Power stations refuse to publish actual fuel consumption in the same 5 minute time slots that the AEMO publish the generation data for every generator. They have this data but refuse to publish it due to the exposure it would bring on how inefficient they are being forced to run, not only due to backing up wind farms, but also covering line losses pushing power across the countyry. I am told the losses now from Adelaide to Melbourne exceed 15%. AEMO site shows on 4 July 2013 at 14:55, a fault dropped 336 Mw of Macarthur wind farm off line. It was instantly picked up by 334Mw of Black coal fired power station in NSW and Qld. Not a flicker of a light bulb and no blackouts. The coal was being burnt anyway and the load was picked up thousands of km away. Macarthur was not needed so in a net effect was not abating any GHG at all, and was in fact increasing the country’s emissions due to its maintenance and construction cost, and the losses in the power lines across the country to back it up. This backup of wind farm stoppages happens repeatedly, and the AEMO have confirmed that 500Mw and greater backups are needed many times a year.
If comparison is made from tonnes mined at power station coal mines, where this can be seen on annual reports, and compare it to Mw sold from the same power station, then the numbers are not even close to what they are reported in the theoretical reverse calculations used to report emissions. This discrepancy has been widening steadily over the past 6 years as more wind farm capacity is installed.
And finally, I talk to power station engineers and to contractors working on the sites. Engineers tell me that their coal feed does not reduce and increase at the same rate as MW sold, in some cases they can be burning 20% more coal than what the Mw being sold. A contractor working on a Black coal infeed conveyor told me how they were under pressure to keep the conveyors running during a maintenance job, the coal was continually fed into the generator boiler for the 4 or 5 hours of the job. Even though the coal feed remained constant, comparing AEMO generation data, the output nearly doubled then again halved over the period. A lot of coal was not being converted in to saleable electricity, it was wasted.
Opinions, thoughts and estimations are wasting so much time and money, we need the power generators to openly publish real time 5 minute coal feed data for every generator, so this can be matched against the AEMO 5 minute generation data, then we can all see what is happening in an open honest and truthful manner. Then we can clearly see how much coal is really being burnt, and how small the net emission savings by wind farm are.
Problem is, most of them do turn out to be myths (though I am sure that Elvis is still out there, somewhere).
Whilst we could go to-and-fro electronically for a long time, and we might achieve a raised level of understanding eventually, it would not be a very efficient use of my time, or yours. Instead, I have another suggestion:
For a number of years, QUT has run an excellent Introduction to Power Plant Course that has helped to lift the veil on the magic that happens behind the closed doors of a power station.
Such a course would (for instance) help you understand how unit coal bunkers stand between coal supply from larger station bunkers and/or the mine, and provide some hours of storage to enable coal burn rate to be modified over short time periods without necessarily changing coal feed rate into the bunkers.
Such a course would (for instance) help you understand the difficulty of accurately weighing the amount of coal fed into a boiler (let alone the problems in accurately measuring CV and other qualities) on an ongoing basis.
It would help you understand how spinning reserve operates at a station, and how it provides the capability on the NEM grid to recover rapidly from frequency drop that would follow any large station outage (be it wind, coal, gas, whatever).
Both you, and Peter Lang, do seem to have made considerable effort to start from nothing and learn what you can about how thermal coal-fired power stations. Such a course would build on this base to help you to understand how one actually works.
In the interests of saving my time on WattClarity, and helping you both up the learning curve, I would be happy for our company (GLOBAL-ROAM Pty Ltd) to contribute $1000 each to the cost of you attending this course. Perhaps some of our other readers would like to chip in as well?
However note that:
1) There would be some cost remaining to you to pay (depending on how much our other readers volunteer to chip in). Given the concerted efforts I have seen by both of you make to contribute to the discussion about the NEM’s power system, I am sure you will feel this investment in completing your education to be very worthwhile indeed.
2) More seriously (and sadly), though, the person who had played a huge role in delivering this course over a number of years (Mr Graham Proud) was killed in a car accident only a few weeks ago. I am not aware of what QUT will do to find someone to step into Graham’s huge shoes, but will be happy to follow up with them to find out, if this proposal is amenable to either, or both, of you. If this course cannot be run, I will endeavour to find another one for you that will deliver real knowledge.
Please advise if you will accept this offer to assist in your education?
Hamish,
Whoever you are talking to is bullshitting you.
1. There are no formal reserves as backup for wind. Trips on thermal generators are far more common (see Liddell, Eraring) and yet still no problem with security. Also note that SA has no new fossil generators despite massive WF capacity. WF output can reach 90% of load served.
2. Transmission losses are usually 2-5%. If they were 15% it would be economically attractive to build power plants in the CBD.
3. Fossil plant burn less fuel at lower capacity. From thermo law 1: energy in = energy out for a system with constant internal energy. If the plant was burning the same amount of fuel for a lower output they would need to actively cool the boiler to prevent damage (i.e. increasing internal energy). It is much easier to feed less fuel in (not to mention cheaper). The efficiency at min load is similar to that at 100%. Peak efficiency is at ~80%. Note that even if efficiency drops 1% when you halve your output, you are still burning about half as much fuel. Thus if a WF’s output forces a coal plant to turn down (nearly always the case) you still get 98-99% of the savings.
In terms of spinning reserve, I noted here how cost of spinning reserve has not increased with wind – however I also noted that there is still a cost of intermittency (currently being manifest as a wealth transfer from existing thermal generators to wind farms). Seems to me that it’s only one instance – in many, on all “sides” – of “socialising costs whilst privatising profits”.
Trips to thermal generators do indeed occur (and trips to entire wind farms would seem unlikely, due to the scattered nature of them – i.e. off the top of my head, I could only see a whole farm tripping if the main connection substation (or control system) had problems, whereas in a thermal station there are many more potential failure points to take a unit offline).
However, I think Hamish is speaking about the more frequent intermittency, as seen through capacity factors (though perhaps he could not put this into the right technical words) .
Statements about % of energy served from any particular fuel source in any particular geographical location, or “no new fossil generators” in SA (whilst understandable from a political point of view) are largely irrelevant. That’s the point of our National Electricity Market.
2) About losses
Losses depend on a number of things, including the distance travelled. There are instances where losses can be up to 20%, say, in the extreme. The AEMO documentation on Loss Factors & Regional Boundaries explains more. Note these are marginal losses, not average losses (which will be lower).
3) About heat rate
Fact 1 – You have noted (as did I) that coal consumption varies, significantly, with output.
Fact 2 – You have also noted that heat rate also varies with output, which is true. It’s even more complicated than that, because heat rate can vary over time (due to super-heater tubes being fouled with slag, condenser tubed becoming partially blocked, and a whole range of other reasons) and does vary (even as-new) with different configuration of plant (e.g. for a certain output it might be possible to have 4 mills running, or 5 – both configurations will yield different heat rates).
From what I have read, some people seem to latch onto Fact 2, and overlook Fact 1 – whilst some people do the reverse.
It is understandable that an outsider might see Fact 1 and Fact 2 opposed.
Finally, Hamish, one of the main reasons some generators may not want to have such commercial details as coal burn & heat rate published in real time is because of the highly sensitive commercial in confidence nature of the information. If their competitors knew this type of information, they might be able to utilise it for commercial gain (and their detriment) in the market (in our software development business, we’d certainly be looking to further enhance ez2view to help them do this, if possible!)
Wow! What a pathetic, arrogant, gratuitous response. It demonstrates many of the signs of intellectual dishonesty (more below)
So Paul McArble is resorted to offering ‘shut up and go away money’ and for it to be paid by ROAM Consulting.
This smacks of similarities to corrupt union behaviour. What’s next, banned from any further comments? That’s how his evident allies, Tristan Edis and Giles Parkinson (name first mentioned here by Paul McArdle, not me), prevent comments they don’t want rather than admit they are wrong on an issue. Anonymous commenters also appear on Tristan Edis’s politically partisan, renewable advocacy web site to defend WattClarity. So there is an apparent allegiance between these passionate (but irrational) renewable energy activists.
Instead of admitting you made baseless assertions and you cannot support them you try to bluff your way around it. You’ve demonstrated this type of response on previous occasions too when you made comments that showed you don’t know what you are talking about on the issue of how Australia’s CO2 emissions are estimated and the uncertainties in the estimates.
You also admitted, on this thread, you don’t understand why measurements at time intervals of 15 minutes (or 30 minutes maximum) are relevant for estimating the CO2 emissions avoided by intermittent renewable energy generation. [I pointed you, again, one of many papers which shows why very clearly; apparently you never read it or didn’t understand it, or any of the others]. And you seem to go into a state of denial whenever it is pointed out that the empirical evidence from other grids where such information is available (on CO2 emissions per individual unit per 15 minutes taking into account the thermal efficiency at the level of output for that particular unit at that time) shows that wind generation is much less effective than the proponents claim at abating CO2 emissions.
One such study shows that Ireland’s EirGrid was just 53% effective at abating CO2 emissions in 2011 (Wind generation supplied 17% of electricity and many relevant features of EirGrid are similar to South Australia’s grid which is held up by the wind energy advocates and activists as the poster child for renewable energy in Australia.
That means that CO2 abatement using wind energy is about twice as expensive as the proponents would have us believe. That is relevant. Why are Paul McArdle, Tristan Edis and Giles Parkinson trying t avoid the issue. Why are they trying to bury it? Why is Paul McArdle offering ‘shut up and go away’ money – which is to be paid by ROAM Consulting.
Why don’t you just answer the questions and provide links to the 15 minute data, the actual consumption and the assay values of the coal entering the burners at 15 or 30 minute intervals? We know the reason – they don’t exist. You know it too and are trying to doge admitting you are wrong and made baseless assertions up thread.
I’m sorry to see that (for the 3rd time in a few days) you have jumped to conclusions – this time in assuming some nefarious ulterior motive in a genuine offer to help.
Your “offer” is not a genuine offer at all. It is an attempted diversion from answering the question to get you off the hook and appeal to your followers. If it had been genuine it would have been done quietly by email. Your comments are arrogant, insulting and rude. You haven’t a clue what I’ve been working on for the 40 years before retirement (including having been on and in the dams, tunnels and power stations of the Snowy Mountains Scheme since about 1952, aged about 6). Whereas you comment, and try to make out you speak with authority, on subjects you clearly know little about. They suggest you are young, inexperienced and gullible.
You still avoided the questions. They are relevant to the debate which followed from your post and my first comment “We don’t know how much CO2 is emitted because NEM generators don’t measure or estimate with sufficient accuracy and precision at sufficiently fine time intervals (e.g 15 minute intervals as in Europe) the emissions data from the fossil fuel power stations”
Which you really didn’t have to answer at all since you seem to know next to nothing about the subject.
Well, what can I say. You want to glide over facts, first hand from people on the ground (with no vested interests) and assume that no one but you has any knowledge or contacts that know more than you are prepared to admit. Some comments to correct your assumptions. The conveyor was feeding the actual boiler, not a stockpile, if you read my original comment closely. PowerStation engineers with 35 years service as I understand, with considerable first hand experience running the stations tell me excess stream is vented, to put it simply, so that capacity is available with live hot available steam at any time. You are probably unaware that I personally worked in facilities that provided standby and peaking power to the grid, and are probably unaware that I worked under contract for state and federal ministerial advisers on a couple of high level energy projects for their American, Japanese and Chinese friends. I could probably run a course that could teach you a thing or two.
As for “commercial in confidence” being the reason why they won’t supply the coal feed figures, you must be joking. Anyone in the industry would have a very accurate idea of what all other powers stations in the grid are capable of, and their efficiencies and capabilities are published in annual reports. There is no real excuse for not publishing live coal feed data, unless you are trying to hide reality.
And as for SA, they more often than not are backed up by Victorian Brown coal generators. With industrial and manufacturing downturn, Victoria has an enormous unused capacity available to be sent anywhere in the Eastern states. Even allowing for considerable losses.
Tom interesting that you said “Transmission losses are usually 2-5%. If they were 15% it would be economically attractive to build power plants in the CBD.” Firstly if you seriously believe transmission loss from Qld to SA is less than 5%, then I have little faith in your understanding of the NEM or the impact wind farms are having. The Singaporean company who own the 500Kva line that runs through to Portland, currently are looking at the maintenance or upgrade that is required, because it has hit the 15% l.oss figure. They are also costing and in early stages of planning permissions, to construct a line beside the existing one with twice the capacity. The drawings have already been done for the power station to be built at Portland as well. And it is not a wind farm and it is not fossil fuel powered. If you do a search on the speech made a couple of years ago by the then federal Labour energy minister, you will see that the Federal Government stated then, that when the country finally realises that wind will not work, then they can build the new power station and all the Green voters will accept it, and we will have built the grid and interconnectors on the pretence of wind saving the world, to enable the power station to run the country. However today I was told that it may not be built in Portland now that there is a mothballed desalination plant in Gippsland that just happens to be a good size fit for the desalination plant required for the proposed power station. This significantly reduces the cost required to build the power station. Also for Melbourne, prevailing winds and weather would be safer having the power station next to the desalination plant, rather than at Portland. I am afraid that wind farms are just a distraction for the masses, so the government can implement the goal with less fuss. Not a conspiracy, not a work of fiction, the drawings have been complete for some time now, and both state and federal planners are fully briefed on the plan. If you personally know people in these departments as I do, then they can confirm this to you. If you do not know people in these departments, and have not seen the drawings, then you have no right to comment on it nor cast doubt on the credibility of those of us that have. I do not plan to make any further comments on this matter.
Cycle steam is not vented except in emergencies. Demin water costs too much.
I agree, SA WFs are often backed up by VIC brown coal. While the coal plant is run at a lower capacity it is consuming significantly less coal. That’s the whole point.
Thanks for a really helpful explanation of the issues underpinning some of the warnings we’ve been hearing about supply security. The slow build-up of a systemic security risk resulting from skimpy maintenance is pretty convincing. And to me, as a representative of the “people focused on broader decarbonisation” it suggests that there’s a strong security argument for a clear policy driver of orderly plant retirement. Not that I’m clear on what that should look like, yet. But would be interested in your thoughts on how such a policy could be designed to best address security risks.
Hi Paul…pointing out more of a typo than anything. In the first graphic, it is probably more accurate to say wind is suppressing scheduled generation not scheduled demand. If only demand was scheduled, it would be so much easier to supply!
Hi Liam
Thanks for the comment.
I was focusing more on Scheduled Demand (which NEM-Watch highlights) more than Scheduled Generation, in this instance – though the wording could have been clearer.
We see South Australia is coloured a deep blue, being down towards the bottom of the historical range of Scheduled Demand experienced in the region’s recent history.
Paul
Why is there such a large difference in gas prices between Brisbane and Adelaide ? As your gas chart shows, it’s clearly not limited by pipeline capacity. Is this price differential what the pipeline owners charge for the use of their asset ?
2. EirGrid calculates CO2 emissions at 15 minute intervals. The total emissions per year reported to the EU and IEA by other government departments based on total fuel consumption and properties reconcile closely with the EirGrid calculations. The calculation method is described here: http://www.eirgrid.com/operations/systemperformancedata/co2emissions/ :
“EirGrid, with the support of the Sustainable Energy Authority of Ireland, has together developed the following methodology for calculating CO2 emissions. The rate of carbon emissions is calculated in real time by using the generators MW output, the individual heat rate curves for each power station and the calorific values for each type of fuel used. The heat rate curves are used to determine the efficiency at which a generator burns fuel at any given time. The fuel calorific values are then used to calculate the rate of carbon emissions for the fuel being burned by the generator. ”
This paper is excellent and shows how to estimate emissions saved by wind generation for grids that have reasonably reliable emissions estimates at 15 minute intervals: http://www.sciencedirect.com/science/article/pii/S0301421513007829
Here is a (free) pre-submission version: http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf
3. As far as I am aware, neither AEMO nor CER have estimates that are even close to the time spacing (e.g. 15 minutes), precision and accuracy needed to make a reliable estimate of the emissions saved by intermittent generators such as wind and solar power. Therefore, we don’t know how much emissions are saved by wind and solar power in the NEM. Our best guess is that the effectiveness of wind power in the NEM would be roughly equivalent to wind power in EirGrid – i.e. about 53% effective at 17% energy penetration (proportion of electricity generated by wind power). That means 1 MWh of wind power avoids just 53% of the emissions from 1 MWh of electricity generated by EirGrid without wind power. This is important because it means the CO2 abatement cost with wind power is around twice what is normally claimed by the wind power proponents.
4. Of course, if we took a different approach to reducing GHG emissions, there’d be no need for rigorous, precise, accurate emissions monitoring to the standards needed for commerce in CO2. The compliance cost of emisisons monitoring will become huge if we go down this route. There is a better way. Just fix the problem. The USA and IAEA could do that by removing the impediments that are making nuclear power far more expensive than it could and should be. As the costs come down nuclear would replace fossil fuels world wide with no need for carbon taxes, binding international agreements, police forces to force compliance, international trade disputes and emission monitoring.
Readers may be interested in next week’s Energy Users Association of Australia Annual Conference- New Energy Paradigm- Better Energy, Better Business. The program features an impressive range of speakers from the regulatory, research, retailing, networking, generating, policy and political arenas, with the focus on the new energy paradigm and how best to manage it, including the pivotal issues critical to Australia’s large energy users as you describe in your article.
New Energy Paradigm- Better Energy, Better Business will be held at Crown Melbourne on 13 and 14 October. For further information or to secure your space, please visit http://www.euaaevents.com.au
I agree entirely with what Michael says about needing to know how your company uses energy is critical to reducing the costs of energy and finding the various tools to be better able to identify ways to limit the costs incurred – so much of the energy cost paid by users is for someone else to provide the resources for what they think your firm needs and to manage the risks involved in sourcing and delivering the energy your firm needs.
What is just as important is making sure that the tools are available to allow each energy user to implement the various approaches noted by Michael. The work and expertise provided by Michael is possible because of the work that has gone on many years before to make these opportunities possible.
But this, and similar work, on the electricity market still continues and needs the specialised expertise that the MEU provides.
For example, the establishment of the gas STTM was a long term project that involved many years of work but it was essential for energy users that throughout this lengthy process for the needs of users to be always top of mind as the development proceeded. This is a role for groups of users, such as the Major Energy Users (MEU). Throughout the development of the STTM, the MEU was involved at all levels of the process and ensured that the STTM delivers the maximum benefit to gas users.
The MEU continues to be heavily involved in making sure the electricity and gas rules are revised and rewritten so that energy users have greater flexibility in how they access and use the resources of the energy market so that the interests of energy users are maximised.
Another story of note recently occurred on Saturday afternoon (1 Nov) when there was a long period of sustained negative prices in South Australia. There was high wind power production (~1000 MW), low demand (~700 MW), probably because of a high solar contribution behind the meter, low thermal power production from coal (Northern ~250MW) and gas (Osborne ~150 MW, Torrens Island 50 MW), and maximal export to Victoria (~400 MW). Victorian prices were not much better at about $7/MWhr, with a high wind contribution (~900 MW) and maximal exports to Tasmania and NSW.
Tasmania is another interesting story. Prior to the abolition of the carbon tax, it consistently exported about 400 MW to Victoria. This was a deliberate strategy, and was achieved by running down its dam levels. Since the end of the carbon tax, Tasmania has instead been importing about 400 MW from Victoria. The strategy made $140m profit in 2012/13. Its dams are now at only one third of capacity, and winter inflows have been poor. I expect it will take several years for the dams to return to their normal supply levels, and I hope they have budgetted for the costs of importing power over that period.
You bring up an interesting point – because of what some might see as a windfall gain for existing hydro plant coming from the carbon tax, hydro plant in Australia ran harder during the 24 months of the carbon tax – an entirely rational behaviour. We highlighted the first week in 2012 here.
That’s not the only opportunity for borrowing from the future to arbitrage time – another way has been to run hard one year and lean the next in order to generate above baseline production volumes and hence create renewable energy certificates (LGCs/RECs). I thought I had posted some analysis on this online, but could not find in a quick scan.
Existing hydro investments could be used for much more generation balancing to complement the volatility of wind than it currently is, particularly the Tasmanian hydro stations. The Snowy stations and Dartmouth are used in this way, but while the carbon tax was applied Tasmanian hydro did little more than provide baseload power across Bass Strait, selling into what was at the time a higher priced market. In the process it has run down the hydro dams to one third of capacity, and now imports brown-coal power from Victoria, which is now cheaper without the carbon tax.
The NEM covers two quite different markets with different economic drivers. There is the SA-Vic-Tas cluster, which has the majority of the NEM’s wind and hydro capacity, and all its brown coal capacity. There is also the NSW-Qld cluster, which is driven by black coal, gas, and a relatively small amount of wind and hydro. The linkage between these markets has a capacity of little more than 1000 MW, so when it is maxed out, there is no further capacity for generation balancing. To accommodate the intermitancy of wind in the SA and Vic markets, better use could be made of Basslink and existing Tasmanian hydro stations. Basslink has a capacity of ~500 MW, but this could go from -500 MW (Vic to Tas) to + 500 MW (Tas to Vic), thus providing a potential buffer of 1000 MW. Unfortunately now that Tasmania have run down their dams they may have little capacity to export into Vic at times of high power prices, so until their dams recover we are unlikely to see this desirable market behaviour. There are also peaking gas plants in Victoria and SA that can also cover the volatility, but as gas prices increase they will need a higher power price before dispatching power from their plants.
The Labour government tried to purchase some of the older coal-fired power stations such as Hazelwood in Vic and Playford in SA, but the asset owners appeared to be holding out for a higher price than the government was prepared to offer, expecting that power demand would continue to increase. I wonder how many coal-fired asset owners now regret them not having taken up the government’s offer. This would have provided a more orderly path to reducing coal-fired capacity than waiting for receivers to step in.
Re security of supply with volatile wind and solar renewable energy, existing hydro investments could be used for much more generation balancing to complement the volatility of wind than it currently is, particularly the Tasmanian hydro stations. The Snowy stations and Dartmouth are used in this way, but while the carbon tax was applied Tasmanian hydro did little more than provide baseload power across Bass Strait, selling into what was at the time a higher priced market. In the process it has run down the hydro dams to one third of capacity, and now imports brown-coal power from Victoria, which is now cheaper without the carbon tax.
The NEM covers two quite different markets with different economic drivers. There is the SA-Vic-Tas cluster, which has the majority of the NEM’s wind and hydro capacity, and all its brown coal capacity. There is also the NSW-Qld cluster, which is driven by black coal, gas, and a relatively small amount of wind and hydro (but that hydro has important capacity for security of supply – Tumut 3 has a 1500 MW capacity but only enough water for a 4% capacity factor). The linkage between these markets has a capacity of little more than 1000 MW, so when it is maxed out, there is no further capacity for generation balancing. To accommodate the intermitancy of wind in the SA and Vic markets, better use could be made of Basslink and existing Tasmanian hydro stations. Basslink has a capacity of ~500 MW, but this could go from -500 MW (Vic to Tas) to + 500 MW (Tas to Vic), thus providing a potential buffer of 1000 MW. Unfortunately now that Tasmania have run down their dams they may have little capacity to export into Vic at times of high power prices, so until their dams recover we are unlikely to see this desirable market behaviour. There are also peaking gas plants in Victoria and SA that can also cover the volatility, but as gas prices increase they will need a higher power price before dispatching power from their plants.
You should have a running tipping contest for which station is to be closed or mothballed next similar to your annual peak load contest – however you would probably have to give away too many BBQ’s to keep up over the next12 months. Or is everyone still pretending that “bid unavailable” and “peaking duty” are different to mothballing?
In both of the extreme cases wind is being punished; when it’s windy everywhere and price is low wind farms get paid less, when it’s not windy and price is high wind farms miss out on the super lucrative times.
Tom, “punished” is an emotive word that evokes thoughts of unfair treatment, but I wonder if that’s actually the case?
In the case of “low” prices
When it’s windy all generators who contributed to that low price (wind and thermal) are paid the low spot price. This means the wind farms that generate plus the thermal plant that still bid very low prices to run at least at minimum load.
It’s a separate commercial decision for each portfolio and it seems that, at least till now, portfolios continue to be “happy” to run at very low prices for transient periods rather than back off further, or shut down completely. This decision applies to wind farms, just like any other.
In the case of “high” prices
Why is it the case that a wind farm is “penalised” for being unable to generate when the market requires it (i.e. which, in this particular case in point, would be when every other wind farm is still)?
At times when the ability of groups of generators to produce is restricted, there is added opportunity to extract higher prices from the market – no matter whether this is due to lulls in the wind, or major technical problems at power stations, or unavailability of fuel for thermal stations.
One point that I notice on a weekly basis is that the limitation of wind, systemically, occurs far more often than it does for other groups of assets (no matter which other way I can think to group them) – this is the main point I make in the article (i.e. once a week, perhaps, for 2000MW wind to not perform, as compared with once every few years for a 2000MW station to not perform)
In this article 15 months ago, I was surprised to see a much stronger correlation between wind farm output at individual stations than perhaps I’d been led to believe, and noted that this highlighted an opportunity for:
(a) Wind farms in different parts of the NEM where the wind profile was very dissimilar (ideally inverse). Perhaps this would be in North Queensland – I am yet to see data to verify; and
(b) Of course, there is the argument for storage to level out the injections from individual plant – and to level out “market” demand, perhaps as well. However such storage would involve investing more money in a market that’s already oversupplied, with demand declining.
Another point to note is that the (relatively) high prices experienced when wind is low (say $35/MWh in South Australia) is commonly held to be still below a long-run marginal cost of operation for a large thermal plant with – and this seems to be an overlooked point – that long run marginal cost defined on an assumption that it achieves a capacity factor that might be higher than that which is being achieved at present. A lower actual capacity factor further increases LRMC required – hence widening the gap between what’s being achieved and what’s required.
A more neutral interpretation might be that the wind farms that have been constructed, to date, have not taken sufficient account of locational factors as they relate to the NEM (as distinct from how they relate to wind production volumes) and so are themselves achieving sub-optimal returns, and being part of the puzzle that’s driving returns lower for everyone else.
More generally
Finally, I recognise that your choice of “punished” might have been a spare-of-the-moment decision and that a more correct word might have been chosen to convey what you actually meant.
However in my limited use of social media there does seem to be a fair degree of victim mentality pervading the discussion on all “sides” of the debate. It seems to me that that’s not going to get us to a truly sustainable solution.
Yep, I didn’t mean wind is a victim, just that the value of production is self discounted. Given that capacity availability is rewarded through high spot prices, wind is not being overpaid for its capacity.
I definitely agree about the suboptimal geographical placement. Unfortunately SA average prices were and still are much better than VIC or TAS. And you’d be crazy to compete with Tas Hydro. No one wants their revenue to be defined by a competitor.
We are already getting close in SA where wind generation meets the regional demand. So one day I can see that wind farms will have to bid a realistic price for their output (rather than budding the minimum) in order for them to get their targeted return. At the moment they are depending on thermal generation to bid so that the spot price + LREC provides the return on capital the wind farms need. What would wind farms do if there is so much wind generation that the spot price is set by wind farms for many hours – presumably consumers get paid by the wind farms to use their output. Oh joy! Come the day
David, those would be strange days indeed – but somehow I don’t think that energy users will actually ever see that respite.
Trying to ascertain how wind farms should bid to maximise returns in an oversupplied market with declining demand (assuming no major closure of other capacity) would be a tricky challenge indeed, for reasons including:
(a) A fuel source that is uncontrollable, and very much aligned with the fuel source a sizeable number of your own competitors use (especially in South Australia); and
(b) The fact that you lose REC revenue, as well, if you choose not to generate.
A “Death Spiral” for wind generators, if you will?
It does seem that it will be necessary that strategies mature over time, unless we collectively invent other ways of socialising costs whilst privatising profits (e.g. by progressively raising the cap price for the REC to make up the shortfall + layering in some other support for storage).
The “saving grace” for wind farms in South Australia is the interconnector to VIC (or this problem would have been met years ago), so measures of % production from wind for any given region (or other sub-set of the interconnected NEM) are a bit of an academic nonsense, most times.
Two markets where the wind generators don’t sell into a low market are Tasmania and NSW. In Tasmania the load balancing is from hydro (which has its load factor limited by river flows), and the 2 wind power stations are not highly correlated, one being exposed to westerly winds, the other to easterlies. In NSW wind is a such a small proportion of the market that these problems haven’t yet become apparent.
The upgrade of the SA-Vic interconnector will reduce the problems of low prices during high wind periods in SA, but could have the effect of passing these problems on to Victoria. The upgrade is only about 200 MW, so if an additional 200 MW of wind capacity were added in SA this would nullify its effect.
So where are the next wind farms likely to go ? My guess is Tasmania and NSW.
Thanks for this comment, Malcolm (and a couple others I have not had time to respond to either).
The surprising thing I found in this analysis was that all wind farms that existed at the time (across SA, VIC, NSW and TAS) were still reasonably in sync, meaning that there were still quite a number of whole days when production was close to zero .
It hasn’t hurt the generators outside of SA as much, in terms of driving spot prices significantly lower, but this would suggest that it’s only a matter of time, as new capacity continues to be added.
Personally, I’d really like to understand how similar (or dissimilar) a wind production profile would be as far away as possible in the NEM – say in North Queensland?
Paul, Cup Eve Monday is a very commonly scheduled RDO in many industries, as well as commonly taken as annual leave; the Monday reduced demand pattern is therefore not unexpected,and clearly predictable as you have shown.
I am amazed that the proponents of wind project haven’t taken into consideration the effects of correlated wind speed of multiple sites on market prices in their business case. In the last few minutes I have tested a correlation between hourly wind speeds at Hamilton (close to the epicentre of Australia’s wind capacity, with MacArthur, Oakland and Portland wind farms in close proximity), and Armidale on the Northern Tablelands of NSW. This is about as far north in NSW as I have seen wind farms proposed. The two stations are positively correlated and the correlation is statistically significant, but the R-squared is only 4.5%. So for practical purposes they are not correlated. Armidale Airport has a mean wind speed at a height of 10 m of 4.6 m/s, and Hamilton is windier at 5.2 m/s, so it is easy to see why the Hamilton area was chosen first. The most favoured location near Armidale would be at the top of the Tablelands between Armidale and Glenn Innes, which coincides with the interconnector between NSW and Queensland.
Nearly all Australia’s wind capacity is within 100 km of a line between Port Augusta and the north-west tip of Tasmania. Most cold fronts almost exactly match the angle of the line, so nearly all the stations experience high winds together. Cathedral Rocks wind farm near Port Lincoln is a little earlier than the majority, and the NSW wind farms and Musselroe in Tasmania a little later.
Another consideration is the complementarity between wind and solar. Wind tends to predominate during winter and as lows pass over. Solar predominates in summer and when there are highs. While the right wing bloggers might chide wind for not producing at times, these are often the times when solar is producing extremely well. So an analysis of the volatility of renewables needs to include both wind and solar, and include consideration of the times of low demand at night when neither produce as much as during the daytime.
Hi Malcolm,
Market prices are definitely taken into account for the business case for a wind farm. Unfortunately production is the major driver as market prices are are not much better for sites with different profiles. For example if the wind speed at hub height scales for the Armidale and Hamilton sites you mentioned the Hamilton site will have 44% more energy available. Even the worst correlating wind profiles (for price) experience only a ~20% discount to average market price.
The wind speeds I quoted are from Bureau of Meteorology weather stations and Armidale and Hamilton Airports, both of which are located in exposed areas. The standard measurement height is 10 m. While the airport wind speeds are lower at Armidale, there are more local prominences along the top of the range between Armidale and Glenn Innes than there are near Hamilton, which could mean a well-located wind farm could be exposed to higher overall wind speeds than in the Hamilton area. For example, there is a range just west of Glenn Innes that is about 200m above the surrounding landscape. In the Northern Tablelands, the transmission line approximately follows the ridge line and therefore close to these promenances.
In calculating wind power, what consideration is there of (i) altitude and (ii) temperature? Some of the Northern Tableland sites are up to 1400m, so there could be some loss of power because of a thinner atmosphere. Frost and snow are also considerations for the mechanical operation of the wind turbines at this location. By contrast in the tropics, wind doesn’t feel it has to have the same “penetrating power” beyond the coast. Apart from cyclones I don’t recall the Queensland coast being particularly windy when I lived there. Is this anything to do with temperature thinning the atmosphere ?
I have lived in Armidale, Townsville and Hamilton, and Hamilton definitely feels windy more often.
Density does drop with altitude and humidity but not a whole lot. Usually the increase in wind speed with altitude is well worth the small drop in density. Colder temperatures at altitude also offset the pressure drop a little.
I also wouldn’t worry about frost. Armidale temperature goes positive every day [I used to live there too 🙂 ] so buildup of ice is not possible. Ice buildup is a problem with temperatures constantly very negative for weeks.
The Armidale region could be a decent location due to the excellent transmission but you’d have to find a good ridge. Hopefully the New England Wind Coop gets off the ground.
Good post. I get the distinct feeling that your general references to ‘social media’ are really a specific reference to a tweet I posted recently that compared a single instance of wholesale pricing and wind energy penetration in the NEM:
Given the constant assertion that wind energy is driving up prices, a single instance of high wind penetration and lower-than-average prices is a demonstration that that’s not a rule you can apply, in general, to the genesis of wholesale electricity price.
I agree with the anonymous comment on that tweet that broader conclusions can’t be drawn from a single instance (I didn’t draw those conclusions, but that seems irrelevant to the critics). I’ve also included a comparison, over two years, of spot price vs wind output.
As I’m sure you share the same antagonism towards the presentation of single instances, surely one could asses the ‘left’ and ‘right’ corners of your post above together, and look at the net effect of both instances?
IE: Wind is either high, or low. The net effect is that it’s there, sometimes. Compared to a situation where it’s never there, ever, there’s a net decrease in wholesale electricity price.
Is this the middle ground you’re looking for, when both potentialities are considered?
No, I don’t think that is the middle ground. At least not in the longer term. Adding more supply to a market has an impact on future investment (and perhaps disinvestment). So it’s not reasonable to assume that everything else stays the same. Over the long term, if there is more wind, there will be less of something else. This can then be expected to result in higher prices when there is little to no wind generation. Whether it will average out as lower prices overall is dependent on assumptions around demand and the timing of entry/exit of other generation. At high levels of intermittent generation, it becomes more likely that reliability settings such as the Market Price Cap will need to change to maintain reliability.
Other costs that may increase and feed through to end user prices include interconnector upgrades and ancillary services, let alone the certificate costs.
Higher end user costs is not necessarily the wrong outcome if the community wants more renewable energy. But this tradeoff is rarely acknowledged.
But I agree that cherry picking individual instances of high wind/low price or low wind/high price and extrapolating from them is not meaningful.
If you think it is relevant to state that adding wind, as long as nothing else changes, will decrease the wholesale price, then fair enough. But one could say the same about brown coal.
Re cycles of wind in sync with social media:
No, I had not seen that particular tweet – there are plenty of others that celebrate the high wind times and (it seems) a number on the other side that note when wind is at its lowest. To me it just seems illustrative that there are two “sides” shouting at each other with neither really listening in return.
Re drawing a linkage between wind and low (or high) spot prices
You comment “Compared to a situation where it’s never there, ever, there’s a net decrease in wholesale electricity price.” is true on the assumption that all else remains the same, both in terms of:
(a) How all the portfolios bid; and
(b) How assets are made available in the market. for that particular dispatch interval.
I think this is the point Kieran is making with his “one could say the same about brown coal.” comment – i.e. what would have happened if you added more brown coal production, instead of adding wind.
What I have been grappling with is the effect to which we might see changes in future – either
1) as a result of conscious decisions by portfolios to withdraw capacity from the market, or reprice it higher to earn a decent return – or
2) in a “doomsday scenario” if we have the same type of response thrust on us more randomly as a result of collective/systemic “run it into the ground” decisions taken with whole groups of assets not earning a return. Kieran notes the possibility of increasing the Market Price Cap further to provide a return for capital-intensive plant running lower capacity factors than designed, whilst others have talked of capacity payments. They won’t be the only two possible responses.
This type of market-promoted feedback mechanism is also sorely lacking in descriptions of what’s become commonly quoted as “The Merit Order Effect”, despite the impressive graphics – i.e. it assumes that new capacity can continue to eat “someone else’s” lunch (socialising costs whilst privatising profits) with no competitive response in the market.
This is all really good stuff. I work as an advocate for clean tech like wind farms and solar so it’s pretty valuable to get a more nuanced and technical side of things from those in the know.
I’m having some difficulty grappling with one particular thing – Kieran’s comment that adding (say) 1 GW of brown coal capacity would have the same wholesale-price-depression impact as adding 1 GW of wind energy capacity (for instance).
My understanding is that wholesale price is lower due to the presence of energy in the lower end of the bid stack – pushing the ‘price-setter’ lower. Wind energy sits at the lower end because:
– High installation costs are offset by the LGC scheme
– Fuel can be extracted with near-zero costs
Am I wrong in thinking that these are the mechanics of what’s known as the ‘merit order effect’? If, in the above scenario, we simply had extra brown coal, wouldn’t that leave price at the same level, as we simply have more generator bidding in the same price band (as opposed to greater volume in lower price bands)?
Or is the point that brown coal is cheap, and as such, also serves the purpose of filling out the lower price bands (and therefore triggering a reduction in wholesale price)?
The question really comes down to: Are wholesale prices depressed due to some characteristic of wind energy, or the simple addition of megawatts of new generation of any kind?
If the second is true, would the creation of 1,000 megawatts of diesel generation decrease wholesale price, regardless of the frequency or magnitude of its dispatch? Or, the creation of 1,000 megawatts of gas?
Again, thanks for the replies – this is very informative.
Reality is, oftentimes, more complex than any pretty animation (or any market model) can reflect.
A drop in price (i.e. Price A – Price B, comparing two scenarios for a dispatch interval with all else the same) occurs when added capacity in Scenario B is added below what the regional price is for that region in Scenario A. It mostly* does not matter how much below Price A the added capacity bids.
Hence for that hypothetical dispatch interval Price B would be identical, no matter whether the added 1MW of capacity was brown coal or wind.
However this is further complicated by thinking about:
1) What price the additional capacity bids.
(a) Wind might bid at some number below zero (typically it might be the negative of the REC price as “break even”)
(b) Brown coal will also typically bid well below what Price A is, most of the time. Without the carbon tax this might be in the range $0 and $10/MWh for much of its capacity (though in tranches).
(c) Gas and diesel a fair bit higher (reflecting higher SRMC). Hence they might not have the same moderating effect on price (they certainly won’t all of the time).
(d) (refer *) If the added capacity was bid just below Price A, the outcome will be subject to a number of nuances of constraints invoked at that particular time.
That software tool you use should enable you to have a look at real historical bidding behaviour & constraints (someone within your company might know how to drive it a bit harder?). If it does not, then our ez2view product can help.
2) The availability factor (capacity factor for wind) of the asset will also play a role (remembering that we’ve started to only look at a single hypothetical dispatch interval):
(a) 1MW brown coal (at 90% say) would have that moderating effect on price for 90% of the dispatch intervals on average; whereas
(b) 1MW of wind would have the full moderating effect a small percentage of the time, more often having a moderating effect of only a fraction of a MW – this might be thought of as a moderating effect, on average, of 0.4MW because of a 40% average capacity factor.
However keep in mind that thinking about this “on average” is a bit of a nonsense, for a dispatch interval – the capacity is either there, or it’s not.
3) Timing of the availability:
If the dispatch interval we choose is one where supply/demand is tight and Price A is high, then the price drop could be significant. However if there’s plenty of spare capacity the price drop could be quite modest (e.g. in the middle of the night, or when there’s already plenty of wind blowing, etc…)
This is why making simplistic assumptions about the effect that adding more wind into South Australia (i.e. where the wind pattern will be largely the same as all the other wind there) could well be erroneous.
480MW may not be much in the AGL portfolio, but combined with a couple of other gas fired generators in SA that are rumoured to be planned to be mothballed next year it could get interesting on a hot day with no wind in SA
Note that it’s not an immediate closure. It would coincide with an extra 190 MW of capacity from Victoria when the Heywood inter-connector capacity is increased, a drop in demand with the closure of Holden and other larger users, and an increase in gas prices as the Gladstone LNG plants compete for Moomba supplies. The Torrens Island plant is not a rapid-response gas turbine like Dry Creek, but needs to be kept hot. This makes it less competitive as a true peaking station. The Dry Creek plant has rarely been used recently even on days of high demand.
Hi Peter,
It depends on the state. The worst case is QLD with 0.92 t/MWh (due to modern black coal plant). At the current LGC price of $33 that’s $35.87/t.
QLD doesn’t have much wind though so a better representation would be SA or VIC with abatement costs of $34.73/t and $33.67/t respectively.
The displacement calculations assume that wind displaces the marginal unit (but not hydro units as they can hold their water until later).
Hi Peter,
I took the marginal unit per 5 minute from AEMO’s NEMPriceSetter.xml files.
My assumption is that if you added 1 MW of wind at any point in time that you displace some typical unit. The abatement effectiveness is based on the t/MWh from ACIL Tasman’s 2009 paper, this is weighted by the frequency that each unit sets price. This method isn’t great but it is more representative than using local generation as the smaller states are strongly affected by VIC and/or NSW generation. The result is that if you add renewable generation anywhere in the NEM you are most likely going to displace a NSW black coal generator.
I reweighted the values to remove hydro as that energy can be used later to displace fossil generation. Must-run hydro doesn’t set price and can’t be displaced. If wind was the marginal unit then the addition of more wind power abates nothing; that energy isn’t stored.
OK, so how do you know I’m not completely full of **it? Since 2010 gas generation dropped only 2% (0.378 TWh). Black and brown coal generation both dropped 20% (31.7 TWh total); mostly because of demand. Wind is up 74% (3.3 TWh). Hydro is ballpark after their run on the carbon tax. The weighted average emissions intensity from the change in coal generation is roughly 1.04 t/MWh. So either my method underestimates wind abatement effectiveness or energy efficiency is better than wind at abating.
I did this because Grant King kept going ON about how wind displaces CCGT. You’d think the guy would know something about gas generation wouldn’t you?
ACIL Allen came up with a similar cost of abatement for wind as referenced in pg. 20 of the CCA’s review http://www.climatechangeauthority.gov.au/files/files/reviews/ret/2014/review.pdf
They had a cost of $32 which would suggest that the abatement effectiveness is above 1t/MWh, as the LGC cost has consistently been above this level.
pg. 166 of the actual report https://retreview.dpmc.gov.au/sites/default/files/files/ACIL_Report.pdf
Please let me know why you think my estimations are optimistic and what wind would be abating if not coal generation.
I tried searching for an email address for you but all I got was fashion and publishing businesses. Let me know and I’ll send you an Excel file.
Daniel T. Kaffine, Brannin J. McBee, Jozef Lieskovsky (2012). ‘Emissions savings from wind power generation: Evidence from Texas, California and the Upper Midwest’: http://econbus.mines.edu/working-papers/wp201203.pdf
Herbert Inhaber (2011) ‘Why wind power does not deliver the expected emissions reductions’: http://www.sciencedirect.com/science/journal/13640321
This paper references many studies into the effectiveness of wind power at reducing GHG emission. All these studies used empirical data. Many studies were not used if they had used data that is not publicly available.
Could you ask Paul McArdle if he’d mind forwarding your spreadsheet to me.
Peter,
You keep using that Irish source. I get it, if you displace gas the emissions abatement is small. Same thing for CA and TX. We hardly use gas in Australia though.
Check out the abatement for MISO in the paper you linked – 0.92t/MWh (pg. 7). MISO has a lower proportion of coal generation than the NEM, but like the NEM coal is most commonly the marginal unit. Unlike Ireland, CA, and TX they don’t get half their electricity from gas/nuke. A much closer representation of the NEM. http://energy.gov/sites/prod/files/2014/09/f18/Remarks%20of%20Kurt%20W.%20Bilas%20-%20Exec.%20Director%20Government%20Relations%20-%20MISO_0.PDF
Seriously, do look at pg. 35 of the econobus link you have above. It has a nice graph of emissions saved by wind vs. coal generation fraction. This would be helpful when comparing between regions.
Regarding Inhaber’s paper, his abatement curve uses Germany as the example. Germany’s main priority is prematurely switching off their nuclear fleet, thus emissions from the electricity sector have remained flat as renewables are only keeping up with the reductions in nuclear energy. I would phrase this as keeping emissions from rising while removing nuclear.
However you phrase it the example does not apply elsewhere as other nuclear fleets are not retiring by force. You should not use his fictional equation for other regions.
Inhaber also cites de Groot on part load efficiency.
Total BS.
80% load, peak efficiency is 35% for subcrit coal.
50% load, lowest efficiency is 33%.
100% load is a bit below peak efficiency.
At worst that’s 5.7% more emissions per MWh but with 37.5% fewer MWh sent out.
Back up generation just has to be available. It’s not actually generating.
If a coal plant running at 100% has a SRMC of $35 and sells into NSW at $35 they are breaking even. If the same plant ramps down to 50% but consumes the same amount of fuel they will lose $35 per MWh. This ramping happen to the black coal generators every night. They would go bankrupt very quickly if you were right about backup.
OR you could suspend disbelief and realise that they just feed less coal in.
The Boauhinia rail electrification project has just been completed, so rather than diesel trains carrying coal from the Rolleston mine all the way to the coast it will be electrics, which will no doubt further increase electricity demand. Queensland now has a total 4,000 KW of electric locomotives, so if all locomotives were at 100% load factor it would be 4 GW of demand !
Another Queensland story of interest is the end increase in gas prices – today Brisbane trades are $4.21/GJ, which is higher than any other capital city. The arrival of the first LNG tanker in Gladstone Sydney prices to $4.00/GJ, but no obvious impact yet on Vic prices at $3.50/GJ.
So with higher Queensland gas prices, power prices from peaking gas plants would need to be higher. How much higher Queensland gas prices will go will be interesting, and probably not what the investors want to hear. The high prices for LNG landed in Japan now appear to be a thing of the past as new low-cost LNG producers such as PNG are joining the supply side, and there appears to be only modest growth in Japanese LNG demand.
“With the three projects each opting for electric compression of gas delivery to Gladstone, we’re looking to see a sizeable increase in demand for electricity as a result”.
This is not quite correct – althout QCLNG and APLNG are electrified using electricity from the NEM, the GLNG upstream project has gas turbine driven hub compressors and electric driven nodal compressors supplied by open cycle gas turbine generators so does not increase electricity demand on the NEM.
The excellent data available through NEM-Review tells me that a large new continuous load of around 400 MW came on line in Queensland on 27 October last. This suggests that the 2014-15 summer peak is likely to be considerably higher, for equivalent weather, than in recent years; this has already happened on 9 and 17 December. However, without knowing what the load is, it is difficult to be certain with such a prediction.
So what does this mean for gas powered generators? My understanding is gas can change output much much faster than goal which means some companies must be making a bucket load.
Do you think with more renewables gas will play a much bigger role in the future to supply peak demand when the renewables wane? I’ve been quite skeptical of the many green oriented articles stating that gas electricity will play a tiny role in the future of Australia’s energy market.
That’s one possibility, Craig – especially when thinking about the broader LNG supply chain, including upstream electric compression (as noted here). We’ll wait to see what our further investigations reveal, in terms of size and timing.
Several of the process diagrams for LNG trains show that the compressors are driven by gas turbines (costly?). Also, if they were consistently running the liquefaction plant could they store all the LNG? Surely someone can confirm or deny the rumours!
My personal opinion is that the unique G20 public holiday in November caused a step change in southeast Queensland customer behaviour. The pristine weather before the supecell hailstorm was just the right time to head to the Gold/Sunshine coasts and drive up the tourism load. Also the early holiday could have driven consumers into tropical summer holiday mode earlier than usual, turning on the air conditioning earlier than would otherwise have been the case.
Thanks for the great links Mike. They also built two 30 MW gas power stations in the area (pg. 17). It’s probably not too expensive if you run them on unprocessed gas and it would give them a bit more flexibility in managing gas supply vs. demand.
Pretty sure its just the start of summer, it has been hot and humidity has been up. There aren’t any regional loads that are up significantly apart from the gradual growth in the Surat. The real demand in QLD still has a while until it hits.
We estimate that solar installations supported by the Renewable Energy Target and energy efficiency activities supported by the various state based schemes will have contributed 1,877 GWh in 2014 or 89 per cent of the observed reduction in demand for 2014.
How was the estimate of solar PV generation done?
What is the assumed fleet-average capacity factor of PV installations that are actually still operating? How do you know?
What proportion of the originally installed PV installations are still operating? How do you know?
Why would they not be operating? The panels have a 20 year warranty and last much longer than that. If your inverter breaks it’s pretty cheap to replace.
If Green Markets ws genuinely concerned about reducing global GHG emissions, or Australia contributing to global reduction – as distinct from just pushing jobs, income, economic growth off shore – you’d be advocating for nuclear power not renewable energy.
1. Nuclear power is a much cheaper way to substantially reduce global GHG emissions than renewable energy.
2. Nuclear power has the capacity to provide all humans energy needs effectively indefinitely.
3. RE cannot sustain modern society, let alone in the future as per capita energy consumption continues to increase as it has been doing since humans learnt to control fire.
4. There is much greater capacity to reduce the cost of nuclear energy than renewable energy.
5. The issue with nuclear is political, not technical. The anti-nukes and RE advocates are the block to progress and they have been for the past 50 years.
You are a joke.
Hinkley C tender price was $170/MWh. That’s at close to 100% CF. If we went 100% nuclear we would have 60% CF and closer to $230/MWh.
If you wanted to use breeder designs it would cost even more again. Uranium from the ocean? Fuel costs go up 6 times.
“Much greater capacity to reduce cost” Just because it’s so expensive doesn’t mean it is going to get cheaper. Nuclear power has increased in cost over the last 50 years. Renewables have plummeted in the last couple of decades.
I can’t wait until all the old pro-nuke dinosaurs die out of the energy industry. You are a waste of oxygen.
The CSIRO ‘MyPower’ calculator shows that, even in Australia where we have cheap, high quality coal close to the main population centres and where nuclear power is strongly opposed, nuclear power would be the cheapest way to reduce emissions: http://www.csiro.au/Outcomes/Energy/MyPower.aspx
MyPower is an online tool created by CSIRO that allows you to see the effect of changing the national ‘electricity mix’ (technologies that generate Australia’s electricity) on future electricity costs and Australia’s carbon emissions.
Below is a comparison of options with different proportions of electricity generation technologies (move the sliders to change the proportions of each technology). The results below show the change in real electricity prices and CO2 emissions in 2050 compared with now.
Change to 2050 in electricity price and emissions by technology mix:
• For the same real cost increase to 2050 (i.e. 15%), BAU gives a 21% increase in emissions c.f. the nuclear option a 77% decrease in emissions (compare scenarios 1 and 3)
• For a ~20% real cost increase, the renewables option gives 62% decrease c.f. nuclear 91% decrease in emissions (compare scenarios 2 and 5).
• These costs do not include the additional transmission and grid costs. If they did, the cost of renewables would be substantially higher.
That site uses old BREE data. They have since updated.
For 2020:
nuclear – $100 to $170 (assumes plant lifetime longer than ever achieved, no decommissioning cost, free fuel infrastucture)
wind – $60 to $110
PV – $70 to $190
I agree that if you want to go 100% clean power you need nuclear, batteries are still in their infancy. We haven’t got the political will to even get 25% clean power. May as well choose the cheaper option which is clearly large scale wind and rooftop PV.
As usual you haven’t a clue what you are talking about. The BREE are projections for 2030 and 2050, and they are the updated data. You didn’t even bother to read any of it did you. And the plant lives are generous for renewables and penalise nuclear. you really haven’t a clue.
And we haven’t even included the cost of transmission and the risk (almost certain) that renewables cannot do the job the advocates claim
Furthermore, the differences between the 2012 and 2013 are trivial compared with the very large superiority of nuclear over renewables.
The MyPower links to information are broken. They used to say they used the 2103 BREE AETA report data. However, the linkt to “what are the numbers based on” now takes you to eFutrure and that data is based on 201
This statement may be incorrect. The links to the data source are broken. My recollection is that they were updated some time last year to the 2013 AETA report and the information links said so. There were discussions about it at the time. Now the only link (perhaps out of date) points to eFurture assumptions and that says it is based on 2012 AETA report. My recollection is that MyPower was updated to 2013 data, but I could be wrong. But the differences are still insignificant in context.
Right, so we should stop reducing emissions now and wait until 2050 when nuclear MIGHT be cheaper. Got it.
All transmission costs are borne by the generator. It’s the law.
This includes connection, frequency control, and reactive power support.
Maybe you’d know that if you worked in the power sector.
Again,, you demonstrate you don’t know what you are talking about. Transmission cost is not borne by the generator; it’s borne by the consumer. Wind farms pay only for the connection to the grid. But the cost of transmissions for a largely renewable energy generation system is huge.
Nuclear is far cheaper than renewables now. It can be far cheaper than renewables when the impediments on it – imposed as a result of 50 years of anti nuke propaganda and disinformation – are wound back.
Sorry, Tom, you haven’t got any understanding of any of this and your confirmation bias prevents you from doing any unbiased research. You don’t even Accept BREE and CSIRO figures when they don’t say what you want to hear.
You are an example of the people who have been blocking progress for the past 50 years or so.
The connection cost includes any transmission line additions required to connect to the grid. Any new customer or generator has to pay for line additions and they do not own them either. WFs also pay transmission charges like all generators. You are imagining that large new transmission lines are being built on the public dollar for the sole purpose of connecting a WF. In that case it would be expensive. But it doesn’t happen.
I’m not trying to stand in the way of nukes. I’d be happy for them to be included in the RET. I’m annoyed with you for spreading misinformation about RE.
You are misunderstanding. you clearly have not read the link I gave you. To greatly reduce GHG emissions we need a large proportion of electricity generated by low emissions technologies – such as nuclear and/or RE. The additional transmission cost for say 50% nuclear is very much smaller than for 50% of electricity supplied by renewables – e.g. rough estimate $37/MWh for renewables v $2/MWh for nuclear (this is addition to the existing grid). Very rough but gives you some idea of scale.
The additional transmission cost for say 50% elecgtricity supplied by nuclear is very much less than for 50% of electricity supplied by renewables – e.g. rough estimate $18/MWh for renewables v $2/MWh for nuclear (this is addition to the existing grid). Very rough but gives you some idea of scale.
You said: “I’m annoyed with you for spreading misinformation about RE.”
I don’t believe I am. I believe you don’t understand the relevant facts. You’ve made many unsubstantiated, and pedantic and irrelevant, assertions but you have not shown that the important, relevant facts are incorrect, let alone shown they are wrong by a sufficient margin to change the conclusions.
The three main components of the total system LCOE for a system that can substantially reduce GHG emissions are:
– LCOE of the technology (e.g. as per CSIRO eFuture and MyPower using data from BREE AETA)
– LCOE of additional transmission required (much greater capacity and much longer lines to widely dispersed areas)
– Expected value of risk that RE cannot do the job (meet the requirements of the electricity system economically).
You need to show that the numbers I’ve given for these three items are so far out that the conclusion is wrong.
If you can’t show your alternative figures to substitute for mine, and provide authoritative sources (relevant to NEM) for all of them, then you have no valid argument.
Nuclear power cheapest electricity with largest emissions cuts by 2050
The lowest cost way to generate our electricity and reduce emissions by 2050 is with a large proportion from nuclear power.
Here I use the CSIRO ‘eFuture’ calculator to compare two scenarios to supply electricity to meet the projected electricity demand on the National Electricity Market (NEM) in 2050 as well as cut CO2 emissions. The two scenarios are: 1) nuclear power not permitted and 2) nuclear power permitted. eFuture determines the generation mix that gives the least cost electricity for that scenario using the selected inputs. The scenarios compared here use the default (central estimate) for each user selectable input. The two scenarios are compared on the basis of CO2 emissions intensity and wholesale cost of electricity.
CO2 emissions for nuclear not permitted are 80 t/MWh versus 25 t/MWh with nuclear permitted. That is, if nuclear is not permitted emissions would be 3.2 times higher than if it is permitted.
The table below lists the LCOE (wholesale price) with nuclear not permitted and with nuclear permitted; the third column shows the ratio ‘No/Yes’ (nuclear not permitted/permitted)’. Cost items of common interest are itemised. Costs are in $/MWh.
Item No nuclear With Nuclear No/Yes Ref.
LCOE from ‘eFuture’ 130 85 1.5 1
Accident insurance 0 0.1 2, 3
Decommissioning 0.15 0.01 4
Waste management 0 1 5
Transmission, 50% penetration 18.5 2 6
Total LCOE 149 90 1.8
Without nuclear is 80% higher cost than with nuclear permitted. With a higher proportion of renewables in the default scenario, the cost of electricity would be even higher, probably more than double the cost with nuclear permitted.
Policy analysts also need to include in policy options analysis an estimate of the risk that renewables will not be able to do the job. We know nuclear can provide around 75% of electricity in an advanced industrial economy because France has been doing it for over 30 years. But renewables have not demonstrated they can or will be able to. Many practitioners think they will not. An estimate, in LCOE equivalent terms, of the risk that renewable technologies do not meet the hopes of the proponents is $54/MWh.
The risk that renewables will not be able to do the job is the major risk that should be questioned, not the costs of waste disposal, decommissioning, accident insurance etc. of nuclear all of which are negligible compared with LCOE and the risk that renewables do not deliver the benefits claimed by their proponents.
Just on the question of why AEMO’s MTPASA and STPASA forecasts seemed so different for the week you examined, the reason is that the MTPASA takes no account of the actual weather forecast – it is simply based on long term climatic statistics. The STPASA forecast does take into account the actual weather outlook.
The MTPASA demand forecast will almost always be significantly higher than the STPASA forecast unless the near-term weather forecast is predicting conditions (broadly) corresponding to a maximum demand day for an entire season, which is the the demand metric that the MTPASA forecast is attempting to represent (at 90%/50%/10% probability of exceedence levels).
The sole justification for wind power is to avoid GHG emissions. Therefore, I’d suggest displaying the emissions in real times (as is done in other countries). More importantly show the emissions avoided by wind in real time as well as:
CO2 emissions by technology (t CO2/h) and t CO2/MWh
CO2 abatement cost ($/t CO2 avoided)
I think the Pitt and Sherry “mysterious jump on Queensland demand” is answered in AEMO’s last two editions of Energy Update February 2015 edition page 2 and December 2014 edition page 3.
Quoting “Queensland experienced record breaking temperatures in late October and November last year. The November 2014 mean maximum temperature was the highest on record for the state, leading to some very warm nights for Queenslanders,” said Ms Pimentel.
Secondly, In addition to PV behind the meter there is also the price triggered demand side diesel generation embed in the distribution networks to consider. Some of these embedded generation sites can be quite large >1MW and there surprising number of of them out there. Looking at yesterday’s spot price some these embedded generators may have potentially run well into the evening. So if we were adjust for both solar PV and embedded demand side generation what is the underlying demand?
And finally is anyone tracking the number of failed and “abandoned in place” solar PV systems that the owner cannot afford or be bothered to repair. There are plenty of anecdotal stories of failed PV systems where the installer has gone out of business (no warranty) and home owner is faced with the cost of replacing the now ~5 year old inverter, if you haven’t yet recovered the install cost on the original PV system, then paying for a major component is bitter pill to swallow for many.
Cheers
In terms of the “mysterious jump” I think what AEMO has noted is part of the answer, but have not had the time to really dig further.
In terms of embedded generation behind the meter, solar PV has been (as much as I understand) the biggest growth factor in between 2010 and now. That’s one reason why we have focused on it, in particular – to provide a more integrated view in NEM-Watch v10 complete with that data.
In terms of price-responsive demand changes, either by diesel gensets starting or by actual demand reductions, there was some of that happening yesterday we know (some are our clients).
In terms of anecdotal evidence of solar faults, I have yet to delve into that – but I know that Tristan and others have commented. If you know of other sources of published data, in addition to the CER, please list them here so I can check them out as time permits?
My concern is that moves in this way might be described as “jumping from frying pan into the fire”…”
I agree 100%. And I agree with your comments about the RET. Simplifying enormously (for clarity in a blog comment), the rational answer seems simple. It’s to dump the RET and allow Direct Action to work to buy abatement at the least cost (based on reverse auction processes)..
However, as ACIL-Allen pointed out in their commissioned report to the RET Review Expert Pannel, simply dumping the RET and repealing the legislation sends a message of heightened regulatory risk in Australia (no worse that what some advocate to shut down coal power stations of course, and probably less economically damaging because all repealing the RET would do is to remove the guaranteed sales of all the RE generators produce and the subsidy of the RECs. UIf they are cheaper than fossil fuels, as some claim, they don’t need any production subsidies. Let’s see.
ACIL-Allen suggested that closing the RET to new projects would be preferably and gandfather the exisiting projects. I am persuaded this would be the best approach. Let’s give the Direct Action approach a chance: Reasons to support Direct Action (and close the RET to future projects) are:
1. It’s much lower cost than the Labor-Greens CO2-e emissions restraint policies
2. It’s highly flexible – so it can be adjusted to fit with whatever policies the world agrees to
3. Carbon pricing is highly unlikely to succeed [1], [2], so its a totall waste of time energy and resources continuing to advocate for it
4. The US is implementing direct action, and has no intention and not realistic prospects of implementing CO2-e emissions pricing
5. Direct action has demonstrated it is by far the best way to reduce emissions – e.g. France’s CO2-e intensity of electricity is 10% of Australia’s and it has near the cheapest electricity in EU.. That’s not due to CO2-e pricing. Its due to direct action policies – as are virtually all the countries that have nuclear power substituting for fossil fuels.
6. Countries with the highest proportion of renewables have the highest cost electricity and high emissions intensity from electricity. The countries with the most nuclear have the lowsest cost electricity and lowest emissions intensity of electricity [3]
On your post [1], you are assuming that the cheapest options are excluded first which isn’t necessarily so. Electricity is a sector with low abatement costs and is easy to measure.
Both direct policies and taxes have their benefits. I previously worked on an energy efficiency scheme where businesses wouldn’t accept money to upgrade their equipment. As soon as the threat of the carbon tax came along those same businesses were desperate to upgrade even though the rate of return was lower without rebates. People are more afraid of loss than they are excited for gain 🙂
Direct action nuclear takes capital costs out of the electricity sector and puts them on the tax payer. You could just as easily do that with renewables and have free energy.
Your point 2d) is why I’m against capacity credits. The benefit of a high market cap is that the rewards for actually delivering are high and the rewards for not delivering are zero.
You could probably justify building a OCGT if you could capture 10 hours of $13k per year. If supply was an issue people would knock up some peakers in no time.
Alternatively, if the Market Price Cap actually reflects the value of lost load than there will be a greater incentive to bring capacity to the market, make available additional ramping capacity and extend asset life. Unfortunately, there are strong political reasons not to allow the MPC to reflect VOLL that are counter to the markets and consumers long term interests.
>”Gas also faces competition from renewable energy, with increasing renewable capacity, mainly wind power that services the base load demand. The penetration of solar PV systems with energy storage pose as a potential competitor to GPG in the peak electricity segment in the medium to long term. The ability to store electricity would enable households to change consumption patterns by charging batteries during off-peak and using stored electricity throughout peak period. This would see falling demand for grid-electricity during peak times.”
Does Core Energy’s projection of solar PV systems assume the massive subsidies (and hidden cross subsidies) will continue indefinitely? What is Core Energy’s projection for the scenario where the subsidies ceased (as clearly they should be if policy was based on rational analysis)?
Peter this is an area of material uncertainty. If solar becomes competitive without support mechanisms and storage becomes a cost competitive reality then gas faces a major challenge in the water heating and room heating space. It is simply too early to make an informed call. It is an area Core is analysing closely as it is has major implications across the energy chain.
If? PV is the cheapest domestic fuel, subsidies or not. You can get a system for $2.3/W before STCs these days. With 5% interest you’re laughing as long as you consume at least a third of what you produce.
If gas stoves weren’t nice to cook on domestic gas would be dead.
I don’t know what figures you are using, but I cannot understand why people advocate solar power. It doesn’t seem rational to do so. not only is it many times more costly than fossil fuel or nuclear generated electricity, it is also not sustainable – that is it requires fossil fuel of nuclear to support it. If you add the cost and energy required for energy storage it is even further from being sustainable. The fact it is not sustainable is a slam dunk.
If the issue is GHG emissions reduction then why not advocate fro nuclear. It is far superior against all the important requirements: energy security, reliability of supply, cost, health and safety and environmental impacts?
Peter,
A typical offgrid system has an EROI well over 10. How is that unsustainable?
Solar is now cheaper than nuclear*. Rooftop is much cheaper as you save on distribution costs. You can’t just go out and buy a 10 kW reactor.
*Hinkley C $180/MWh, Chinese new reactors $100/MWh, Saudi Solar $60/MWh, undiscounted unsubsidised Aussie rooftop solar $100/MWh
Thanks for the insightful article and analysis. However I think your conversions in table 1.1 might be out. Total gas of 171 PJ equaling 18.2 GWh doesn’t feel right. Should it be 18.2 TWh?
One area where more analysis is required is whether solar will be effective in meeting winter heating needs in the residential and commercial sectors. Has anyone seen quality research/analysis addressing this topic?
It depends on your system, climate, and insulation. Offgrid systems often tilt their panels up towards the winter sun to give a more uniform annual profile.
With respect to gas, does it matter? A reverse cycle aircon at peak rates is still cheaper than gas.
Related fun fact: German domestic gas is half the price of Australian domestic gas. Their wholesale price is about twice ours.
Great article!
You could also charge for capacity and energy separately like for large customers. Many homes already have interval meters or smart meters. Such tariffs have encouraged offpeak heating and cooling systems to be built in the US.
Interval meters don’t have to mean TOU. You can measure customer load over an interval and charge based on average kW. It’s not as good as peak kVA but it’s a start.
A part of the ‘big issue’ is that networks in some States have been built much larger than the largest load. Better pricing is important, but the AER putting the clampers on high costs and over-investment will deliver the first and biggest wins for all consumers.
A certain margin is sometimes necessary, in the CBD in Brisbane a few years ago it was fed by a few old cables whose status was not fully known. We were running a risk like Auckland which was not fully restored for up to several weeks. Thus instant capacity compared to peak load (particularly for CBD) is not a good measure of a gold plated network, it is likely to be driven by reliability standards.
That is much less true for suburban areas and congestion pricing is much more necessary.
When ever I talk to networks companies about this issue they almost inevitably tell me the problem is “lack of granular data”, IPART said the same thing a few years ago. They either don’t know or can’t manage the data fast enough to offer genuine or dynamic pricing signals; so they smear things in a crude way.
In fairness, they are under huge pressure to reduce costs and so new data systems and analysis doesn’t seem to be on their agenda and of course, the current regulations don’t really incentivise or encourage innovation.
Seems to me (happy to be corrected) that rethinking the regulations and incentivising granular data collection is fundamental to changing the status quo?
I’ll be very interested to know the average wholesale price of power from the solar farm. Royalla has contracted to sell electricity at $183/MWh (or there ablouts from memory). That’s about six times the average LCOE of existing coal plants (according to BREE AETA, 2012). and over three times the average spot price.
Of course all that is because the very large hidden subsidies and cross subsidies are included. What an unbelievable boondoggle solar and wind power are. And for what purpose? What effect will they have on the climate?
Talk about being short-sighted. By very definition we’ll have to phase over to renewables eventually, the sooner we start the better.
Clearly you don’t live anywhere near a coal fired power plant. The number of environmental hazards they pose is extensive, that’s excluding their massive CO2 output. Coal is cheap because there are unpriced externalities. It’s that simple.
As for solar subsidies, a recent (Nov. 2014) ABC News article stated that: “Australian coal, oil and gas companies receive $4b in subsidies”.
There’s no hidden subsidy or cross-subsidy here, it’s pretty overt.
Also, that contract price is the same as the cost of nukes, except it’s a fixed cost not rising with inflation.
Peter
Next time you see something like that please take a screenshot so we know what to investigate – it’s a bit hard for the customer service team to know what they are investigating otherwise.
Paul
Looking at the lack of correlation between wind and solar on a 30 minute basis, it is worth mentioning that a lack of correlation is a lot better than a positive correlation. It certainly means that there are going to be much fewer instances of both wind and solar being simultaneously at near record lows (or highs) than just wind or solar alone. This will likely have implications for any estimates of storage requirements (or spillage from over-generation).
I’ve noticed that when looking at monthly averages of wind and solar, that there is a very small negative correlation. This is because wind tends to perform a bit better during winter than summer (and obviously solar performs better during summer than winter).
Regarding your questions about solar’s capacity factor.
The APVI website indicates that the overall average CF achieved at the PV peak on March 3 is a little over 50% (ie, a long way short of your preliminary estimate of 79%).
This would suggest that they are using an installed capacity of a about 3.3 GW, which would give an average capacity factor of 13% for this 7 week period. This seems a little on the low side to me, most reasonable systems I’ve seen are more like like 16%, however I have no doubt there may be many systems suffering from shading etc that would bring the average down.
Having said that, they are estimating 4.0 GW of installed capacity as of Nov 2014 (or reported installed capacity of 3.9 GW), so I’m a little confused with that discrepancy.
Given that it is rare for a single PV system to ever achieve close to 90% of it’s capacity (unless the system has an greatly undersized inverter relative to the panel capacity), having every system in Australia simultaneously averaging anywhere near 80% of capacity would be highly unrealistic.
Thanks for the comments, David – will review the information you’ve supplied.
In terms of degree of correlation, I would like to be able to understand how a big wind farm in north Queensland would match against the existing stock in the south-east of the NEM. Can you help me with that one?
I don’t have that info to hand at the moment, though I can tell you that when looking at 2014 wind output on a daily basis, the following average correlation coefficients between states apply (R value)
So you can see that (on a daily basis) there is essentially no correlation between wind farms in WA and the other states. And given the relatively poor correlation between wind farms in SA and those in TAS or NSW, I think we could guess that the correlation between SA and QLD is going to be pretty close to zero on a daily basis.
Total national PV capacity (includes ~50MW of large generators) as used in the map is 3,896MW, based on monthly CER installation data. That data is ~3 months old at present so it’s an underestimate: we’re certainly >4GW now.
however I remain a little confused. As 12:20 pm today (March 30), the live data page indicates that QLD solar is generating at 43% of max capacity, and also that it’s generating 497 MW. This implies a max capacity of 1156 MW. But this is in disagreement with what it has for QLD’s state total of 1249.7 MW of PV. Can you shed any light on this disagreement? Is it assuming a max capacity that is less than the generation capacity?
Clearly an extreme case – but it does raise serious questions about the realty of relying on electric vehicle battery storage for NEM backup. It also brings into question some of these “100% renewable energy” analyses that have circulated over the last few years.
As a scenario assuming onlywind supply, with EV battery backup, it’s clearly not a plausible scenario.
I’m not sure that it makes sense to leap to it brings into question some of these “100% renewable energy” analyses as I’d believe that those studies would have included more than just those 2 technologies.
This kind of analysis I’m doing as time permits, but next steps might include:
Step 1) Overlaying solar over the same period and scaling to see how the “unmet demand” gap might reduce (this is perhaps the most obvious next step as solar, along with wind, are the two renewable technologies being deployed at scale in the NEM already);
Step 2) Following this, I could look into how existing hydro capacity might be able to alter its operation (in this hypothetical scenario) to fill the gaps. Keeping in mind that most of this is not pumped-storage, but does have existing storage, it would seem to be a less expensive form of storage than anything new.
Though I do have a question about what this would mean for the functional operation of the market in this scenario – i.e. when other forms of generation are unable to supply, “storage” in whatever form, has market power and so the NEM devolves to periods of negative prices (when the wind blows and sun shines) and then periods of extreme prices (when the opposite is true).
Step 3) Looking into other technologies that others are working to try to bring to commerciality to see how it would work to fill in gaps.
I suspect, at the end of the day, that a Pareto principle will apply here as well – in that the cost per greenhouse abatement will escalate quickly as we move up the “% Renewable” scale. Hence absolutely “100% Renewable”, under all scenarios (e.g. even in drought years), might not ever be an economically desirable outcome for the majority – unless some new technology comes along.
1271 GWh. That’s a lot of batteries certainly but it’s very doable by other means such as hydro. Whilst they are not pumped storage as such, we’ve already got *individual* hydro schemes storing far more than that so it can certainly be done.
Just as we won’t likely see all energy into the grid coming from wind, so too we won’t likely see all storage in the form of batteries.
Dual axis tracking systems look amazing, but the cost increase far outweighs the improved energy harvest. The improvement in energy harvest is also affected by latitude. It only becomes appreciable at higher latitudes – talking >35N or S. There is a significant increase in BOS for dual axis tracking – motors, controllers, comms within the plant, not to mention the foundations. When the modules were expensive it was a great idea, now that modules are cheaper, the most cost effective way to increase energy production is to put in more fixed modules. Similar arguments can be used to justify preferencing flush mounting vs tilt brackets in residential installs.
Single axis is a lot simpler mechanically so the economics are subsequently a bit softer. The increase in energy harvest is reduced too – I’ve not had much to do with single axis trackers, so I’m not sure where the line falls.
The graphs are from the 19th – I’m not sure what the tilt angles of the fixed and tracking systems are, but if they’re close to latitude then these graphs will be of almost the ideal day – close to the solstice. Comparison of the peak output / installed capacity of each system will be interesting at the end of June. The capacity factor of the dual axis should remain relatively stable over the year, the other two will vary quite a bit.
The arguments for or against west vs north orientation are heavily dependent on the tariff and the site – unless afternoon tariffs are much higher or the west facing roof is the only suitable roof and its a choice between some solar and barely any, west facing will always result in a fairly large reduction in energy harvest. At typical residential roof slope it’s around 25% reduction from ideal. This reduction is less noticeable/important at low tilt angles (say 15 degrees or less), which does allow for a lot more freedom on large commercial/industrial roofs, and can allow for very high coverage ratios on space constrained ground systems.
The relative cheapness of modules may allow to offset this drop in output through increased capacity, but with no time of use tariff or a flat load profile, the timing of the solar peak is a bit irrelevant so you go for the best $/kWh.
Also, while orienting to the west (or east) does shift the timing of the peak, it’s not really noticeable until you get to really silly tilt angles and start to lose a lot of output due to atmospheric absorption and a lot of space due to mutual shading.
Sounds a good idea. Any idea when the costs will come down enough so that this technology will
be on peoples houses in Queensland? Or will the cost always be too high do you think?
Any idea what the following is (approximately)
Dual Axis Total Energy Over typical March day
——————————————————————————- = 120%, 110% ????????
Optimal Fixed Total Energy Over typical March day
Very good response from Nathan. Its true the best PV outputs for fixed tilt are often recorded around the equinox – in Melbourne at least, this is often in October/November or April, when the elevation of the panels allows the highest levels of radiation captured by the plane of the PV array.
The calculations for tracking are pretty simple – is the additional capex of the installation of tracking exceeded by the additional value of energy on an NPV basis? You would add to this cost the maintenance and risk of downtime involved with tracking. If you follow some of the bigger installations you may see that the developers have already sone these sums and decided not to track.
On the residential PV side, the depressingly small feed-in tariffs in Victoria have led installers to so something weird – orienting half of the modules east and the other half west. This is an export minimisation strategy as it pushes the generation into the morning and evening peaks and minimises export typically seen in the middle of the day. It doesn’t give you the best kwh harvest but rather the best $ harvest. So PV policy can give us some odd PV architecture. Note don’t try this by connecting these arrays together – they should have separate inverters!
As mentioned, no-one is expecting to go 100% wind, and I doubt many people think we can go 100% wind + solar.
The UNSW 100% renewable study indicated that solar and wind can meet about 88% of our electricity needs on an hourly basis throughout the year. The remaining 12% was obtained almost equally from gas turbines and existing hydro. The study required about 200 GWh of storage, primarily from molten salt associated with CST, though I suspect this is more likely to come from battery storage (and with the CST replaced with PV). Not insignificant, but achievable (10 million homes x 20 kWh, though I suspect a 3-way split between residential, business and distribution is more likely and sensible)
As a PS to this post, we’ve received the following suggestions (thus far) of what might have contributed to the reduction in the all-hours demand:
1) Mothballing of Port Stanvac desalination plant
2) Operation of industrial generation plant behind the meter
3) Closure of Penrice Soda
4) Reduced energy usage at mining sites, including Olympic Dam
5) Car manufacturing closure.
Note that I have not checked any of this, but just record as a starting list for further digging, when time permits.
With respect to other claims made, this reduction in all-hours demand is distinct from (layered underneath) the reduction of daytime demand due to embedded solar.
Could a simplified tracking system work with floating arrays, such as the recently-opened unit a Jamestown in South Australia ? A whole chain of floating panels mounted at the latitude angle could be moved between NE, N and NW during the day by a relatively simple automated pulley system. The UQ tracking systems would be expensive because they are holding the entire weight of the panels. If instead that weight is floating, a much simpler system could be used.
There’s quite a few unique things electrically in Tas.
First is that at the far off-peak times, heavy industry accounts for 80% of the total load and just one factory is one third of the entire load in the whole state in the middle of the night.
Second is the dominance of peak rate water heating, with a market share around 88% of residential hot water. Typically these are 160 or 250 litre storage systems (no such thing as instant electric hot water in Tas) with 2.4 or 3.6kW element. Hence the big jump in demand in the morning as people take showers. Off-peak has only 8% or so of the market, and virtually all of those are on the off-peak rate with afternoon boost – there’s only about 1000 or so customers using the off-peak night only tariff in total so it’s very minor. That said, gas, heat pumps and solar are slowly gaining increasing market share at the expense of the traditional electric storage HWS.
Third is that electric heating dominates (67% market share) with most of the rest being wood. Gas is sufficiently rare that if you have a house with gas and are selling it, the real estate agent really will mention the gas as a definite feature of the house.
Then there’s the means of heating itself. About 40 – 45% of homes use what are universally known as “heat pumps” in Tas – that’s a single split system air-conditioner in the terminology used in other states. They’re always installed for heating, with a heating calculation (but no cooling calculation) to determine the size and spend 99% of their operating time in heating mode. Typically, a 7kW or so output system in the main living area only and electric resistive heating in bedrooms etc is very common.
Then there’s those with resistive heating systems, and typically that’s a 6 – 10kW fixed heater in the living room. Yes, a 10kW single phase 240V electric heater – and they literally do glow red hot in operation (as they’re supposed to). Others have just a fixed fan heater, but still 3.5 – 7kW generally.
So there’s a lot of heating equipment installed versus relatively little cooling equipment. Hence cold weather can drive demand far higher than hot weather ever will. But heating is becoming more efficient as such, with heat pumps going from basically nothing 20 years ago to their current usage and a move away from resistive heating.
Also notable is that the traditional use of off-peak electricity in Tas is space, not water, heating via storage heaters. These are decidedly out of fashion these days and in terminal decline. Common a generation ago, before too long they won’t really exist at all.
Comparing past years, also of significance is the closure of the once-massive paper factory at Burnie. APPM (Associated Pulp and Paper Mills), or better known locally as simply “the pulp”, that place dominated the town for decades but it’s completely gone now and so too is its’ power consumption. At its’ peak it had 10 production lines producing paper, a pulp mill to supply the paper machines and another production line producing boards, slowly declining over 30 years before final closure in 2010. The separate Wesley Vale paper mill, with one machine, also closed at that time.
The remaining paper mill, Norske Skog at Boyer (near Hobart) has also reduced electricity consumption due to process and feedstock changes. Many still call that place ANM – Australian Newsprint Mills – and that’s what it used to make. These days one of the two machines has been converted to produce higher grade paper and the feedstock changed to rely solely on softwood which requires less energy (electricity) to process via the mill’s mechanical (not chemical) pulping process.
Then there’s a few small industrial co-generation plants, there’s a significant one at a vegetable processing plant and another one at a public hospital, installed in recent years which have also reduced grid demand a bit.
So overall, quite a few things have individually reduced load by small amounts but collectively it becomes a bit more significant. Also a lot of LED traffic lights are around now and street lighting is starting to change to LED’s too. And of course the move away from incandescent lighting in homes is another one.
All that said, so long as the “big 4” factories continue operating then overall demand can’t possibly change too much. The big 4 use roughly half the state’s power on average and two thirds at the far off-peak times.
I presume you mean Thursday 5 March 2015 (not the 6 March), when half-hourly operational demand reached a maximum of 8,831 MW at 17:00, 66 MW off the all-time record.
The maximum temperature on 5 March 2015 was 36.9 degrees and the mean temperature was 28.3 degrees (at Archerfield). While not doubting this was a hot day, it ranks around the 10th percentile in historic weather terms, which may be loosely associated with a 90 POE demand. This would imply that the actual recorded demand was significantly above AEMO’s forecast.
Thanks for the correction, Richard – I see I had it right in the image, but wrong in the transcription. Noted this above.
Thanks also for the clarification of where temperature was at on that day – but not sure I understand the which was around the 10th percentile (i.e. average or peak) and how this translates to a 90% POE. Apologies that I’m slow on that translation.
As per the chart above, that day was certainly different than the days either side!
UChoose is the only Unbiased Comparison Service(aggregator) – the only one in Australia which receives the same commission irrespective of which offer the Customer chooses. Plus we do not have any preferred partners or sponsored offers. True choice and unbiased.
Great post. The high correlation between wind output in SA and Vic pointed out in a previous post means that prices in both markets are low or negative at times of high wind output and low demand. Increasing the SA-Vic interconnector capacity through the transformer upgrade at Heywood will do little to alleviate this problem. However, correlations in daily wind output between Vic and NSW are much lower. Surely it’s time to re-rate the Vic-NSW interconnector limits. There are 4 high-voltage power lines, but they seem to have very conservative ratings. (The coal generators in NSW would hate this because it would increase the low-cost brown coal power coming from Victoria at times of low demand in the early morning.) The power lines are
1. Khancoban-Tumut 3 (330 kV)
2. Khancoban-Tumut 1 (330 kV)
3. Dedarang-Wagga (330 kV)
4. Red Cliffs-NSW (220 kV)
Khancoban and the Murray power stations are regarded as part of the Victorian system, even though they are physically in NSW..
Looking at the Google Streetview image of power lines just north of the Khancoban switch yard (36°13’0.55″S, 148° 8’29.84″E), there are 4 sets of power lines, 2 of which go from Khancoban to Victoria and can carry 1500 MW of power from the Murray power station into the Victorian market. The westernmost 2 power lines look identical to those going into Victoria, and should be able to take 1500 MW from Khancoban into the New South Wales grid. Yet the maximum Vic-NSW interchange at 4 am this morning was only about 1200 MW, when there was a price of -$10/MWh in Victoria and $23.96 in NSW. The Dedaran-Wagga line should also have a similar capacity of about 750 MW, and the Redcliffs line perhaps another 200 MW. There could be further rating increases with dynamic ratings, because much of the transfer would be required at night at times of high wind and cool temperatures that would keep the wires below their threshold of about 50 degrees C. So there could be a total capacity of well over 2450 MW.
The Qld-NSW interconnector has been progressively up-rated, and looks much less substantial than those at Khancoban, yet I believe carries up to 1000 MW, while the 275 kV SA-Vic interconnector via Heywood looks similar and will soon have a capacity of 680 MW. So the Vic-NSW combined limit of about 1200 MW for 4 major power lines indeed looks on the low side.
A correction to your comment about Tasmania having to conserve hydro supplies . There have been excellent inflows over the last week into the Pieman, Mersey-Forth and Derwent catchments, to the extent that 2 of their hydro dams are spilling. However the catchments with the greatest storage – the Great Lakes (for Poatina) and Gordon have received minimal inflows. So the catchments with good inflows will allow water in the Great Lakes and Gordon to be conserved.
There have been periods of negative power prices in Tasmania today – perhaps the combination of high wind, low demand and run-of-river hydro power stations.
To quote hitchikers’ “Don’t Panic”. As you pointed out, possible LOR are common but actual LOR are not.
The key issue here is less wind and more the scheduled outage of an interconnector.
The issue of interconnectors to SA and Victoria was a key part extreme weather impact on the NEM study conducted as part of National Energy Security Assessment (NESA). The study identified SE Australian heat waves as the type of extreme weather event likely to stress the NEM. Interconnector failures are among the most significant power system risks during such events, with South Australia at risk for both being at the end of the NEM chain and heat driving demand.
The most recent january 2014 heatwaves pushed SE Australia to the limit, but while Reliability and Emergency Reserve Trader (RERT) contracts were formed they were not invoked.
A key issue is network congestion in Victoria affects what the interconnectors can carry to/from SA, and whether it is frequency synchronised. Each jurisdiction has its own preferences for responses to such challenges (ie how much political flak for large blackouts).
The rise of renewables has changed the paradigm of ‘spinning reserve’ – but the rise is gradual, and it is not the only possible response: already the NEM has a range of special schemes in both Tasmania and South Australia. As noted, demand response and storage are increasingly becoming economic options.
The short answer is that the future is likely to be a more complex mix of generation and demand tools than the historical ‘more of the same’, but ultimately it is very unlikely the lights will ever go out.
What is missing in the market now is pumped storage hydro to level out the market, particularly the 1 am to 5 am periods of negative power prices when there is high wind. According to Wikipedia, the Murray power station has all of its generators capable of pumping, so if true it should have the potential to soak up to 1500 MW from the Victorian system at less than $10/MWh and sell it back to the grid a few hours later at over $30/MWh.
In the NSW system the Tumut 3 power station is pumped storage, and there is also a pumped storage system operated by Sydney Water, so a higher Vic-NSW rating capacity would provide access to these systems.
The NEM report files do show a “Snowy pump” of 200 MW operating during some of this time. Not sure if this is on Tumut3 (NSW system) or Murray (Vic system). Now that intermitancy is becoming a problem, surely one of the best power station investments would now be in these pumped storage systems.
The role pump storage hydro might play in balancing increased intermittent supplies is one of the intriguing questions in this all. I wonder what type of consistent price spread there needs to be (between pumping and generation) to make it work, given that (I assume) it would have to be running much more frequently than any of the 3 existing schemes do now.
Paul, my concerns with reporting that “1/4 of electricity was produced by wind” based on a 5 min DI is just the kind of story that the “greens” run with to suggest this is now the norm. I see it’s down to 8% in the current interval.
You’re a frequent reader of WattClarity, so will understand that:
1) I do try to remain objective; and
2) I have posted about the highs of various types of generation, along with the lows.
As I mentioned on Twitter, I’m perplexed by the choice to focus on wind energy as a causal factor in this event, despite it playing a minor role in its occurrence. As you may know, a lot of people are regularly seeking to find reasons to lay the blame on wind farms for a very large range of adverse impacts – health, grid reliability, electricity price increases, mobile phone reception, etc.
Most of these are predicated on an inability (or an unwillingness) to consider causality – if something bad happens, and it happens at the same time that a wind turbine was generating electricity, that doesn’t mean wind turbine operation was a necessary or sufficient (or both) factor in that thing occurring.
I understand you’ve included AEMO’s correction at the bottom of this article, but someone motivated to implicate wind energy in problems that would have happened anyway would be well served by the headline in this article.
I might append a further example – Tony Abbott said in 2013 that ““renewable energy targets are also significantly driving up power prices right now”
The RET scheme contributes ~2% to 4% to residential bills at the moment. So whilst his statement of contribution is true, it’s not a reflection of the major cause of price increases – network infrastructure spending.
I will follow up further the comments made on Twitter and LinkedIn and investigate further. As my understanding increases, I will post more commentary here.
I do understand that there are different people who view wind farms very differently, just as there are people who view coal seam gas extraction very differently, just as there were many inquiries in years-gone-by about the effects of living in close proximity to high voltage power lines. Controversy is not a new thing in the energy supply industry.
Another contributing factor to the negative power prices in Vic at times of low demand and high wind production is that the brown coal generators don’t appear to adjust to the situation with lower output. To them, brown coal is “free” because the operation of the coal mine is a fixed cost, whereas in NSW it is a variable cost (because black coal can be sold on the world market). This market behavior differs from when there was a carbon price was in place and there were variable costs associated with brown coal use.
But their profitability is now so low that another coal-fired power station (150 MW) will soon close.
I can certainly tell you, it’s been a windy week down here. Gloomy too, unfortunately.
As for what the market could do if the improbable occurred… I take it the abrupt loss of transmitted wattage from SA would potentially jeopardise the rest of the NEM? But apart from that, isn’t there a whole heap of Victorian and NSW hydro effectively on standby, judging by those generation numbers?
Paul,
You mentioned credible contingency. In cases like this the NIL_MAXG constraint will be active limiting wind farms and the interconnector and requiring other generators to increase generation. If the interconnector trips (or Northern/Pelican Point) both wind and gas can ramp up.
We’ve been subsidising solar for 45 years at least. The subsidies for production should cease. It cannot be justified.
I recall Dr Mark Diesendorf, Dr David Mills and Dr Stephen Kaneff in about 1991 arguing that “solar is baseload capable and cheaper than nuclear now, if the stupid government would just give us more money to demonstrate it.
That was 24 years ago. But still solar is not viable. Solar is still hopelessly uncompetitive and all investment would stop if not for the subsidies and regulatory incentives to prop it up.
Don’t make me laugh. We’ve been subsidising nuclear for even longer, and at higher cost, with worse results.
Residential solar is the cheapest consumer fuel -without certificates. That should be your argument to cease subsidies.
Nuclear costs more than solar even when run at 95% capacity factor. When nuclear does load following it can’t compete with diesel let alone anything else. Leave nukes to those without any other resources.
As usual you don’t have a clue what you are talking about. if you want to compare subsidies for different technologies you have to normalise – i.e. show the subsidy per unit of energy supplied through the the plants life. The subsidies for solar are an order of magnitude higher than for nuclear.
We got a return on investment with solar, look at the second graph. Nuclear subsidies per MWh are rising.
Nuclear $180/MWh, Solar $100/MWh. What don’t you get here? It costs more to subsidise nuclear.
Having a hard-on for nuclear power doesn’t constitute some form of expertise.
There’s been a fascinating sequel to this high Tasmanian demand. On Thursday I recall the Victorian spot gas price hit $10/GJ during a cold snap, and peak power prices up to about $90/MWhr. There was an almost complete absence of gas peaking generation across SA, Tas, Vic and NSW. Wind was a strong contributor, providing up to 1000 MW each in SA and Vic, and peaking being provided almost entirely by hydro, despite very low dam levels at 2 of Tasmania’s largest power stations. The only gas units operating over this period were Osborne and a couple of units simmering at Torrens Island.
Then on Sunday a fire shut down SA’s Northern power station, but its contribution wasn’t missed because there was about 1000 MW of wind power in SA. This was more than the market could absorb, because there were extended periods of negative power prices in SA on Sunday afternoon and much of the Monday public holiday. Surprisingly, during this period of negative power prices there was minimal export of power to Victoria. Is a failure of the market bidding process ?
Point 2 needs further development, “The physical limitations of the transmission network and how it is represented in the NEM algorithms to allow competition across locations”.
There are proposals for a few transmission lines to move to a dynamic rating (eg Ballarat-Bendigo), but not the main interconnectors. I suspect the ratings on many lines are set conservatively low. However Basslink can exceed its static rating by 25% for up to 5 hours. Why only Basslink ? A dynamic rating would reduce the spikes that occur in only one State.
The temperature of a transmission line is a function of its amperage, plus weather parameters of wind speed, air temperature and solar radiation. I work in agricultural science, where we can predict the temperature of apples using these weather parameters to within 3 degrees C. Surely it’s not too difficult to work toward a dynamic rating system updated hourly from the BoM forecast grid.
Wartsila has the gas power generation technology (smart power generation: start to full load in less then 5 minutes and high open cycle efficiency) that can give operators the possibility to suppress price spikes better than any other technology current on the market.
We have a rigorous study made by a consultant company that prove the point about our technology. If you are interested in the report, please send me an email, I will be happy to,share it with you.
Paul,
You should probably mention that gas generators are withholding capacity to drive up price. Considering that wind is a smaller player it seems strange to blame them for the opportunism of others. I noticed that brown coal did not increase production at all in response to demand… 😉
I have not looked in this instance, but it would be a natural part of the energy-only market design if generators are “repricing capacity” at times of tight supply/demand balance (a topical activity given this debate going on at present).
On other occasions other readers have noted that generators don’t physically withhold capacity – so in the past I have used the term “economic withholding”.
Paul
PS in terms of brown coal in South Australia not increasing production, I think it has other things keeping it occupied at present…
Hi Paul,
Much is said about the merit order effect reducing average wholesale electricity prices. It occurs to me that prices must be lower when the wind blows, hence wind turbines must receive below average market price. Conversely, as you suggest in your article, scheduled generation can achieve above average price by withholding output until prices are high.
Have you thought to use NEM Watch to determine a wind output weighted average price in SA? My intuitive feeling is that if the time weighted average is around $35 per MWh then the wind weighted average might be closer to $25 and with increasing share of wind this might drop further. This bodes badly for wind investment in SA as the $65 penalty will be cheaper than a PPA with a wind farm.
1) Your hunch is correct – wind farms do tend to show lower diversity than I was initially envisaging (even across regions) and hence collectively “shoot themselves in the foot” in terms of spot price outcomes. Other posts on WattClarity that look in this area include the following:
2) You (or other readers) can crunch the numbers yourself with the help of the NEM-Review package, which (as the name suggests) is specifically aimed at this sort of historical analysis: http://www.NEM-Review.info
Whatever the effects on spot price the effect of wind farms is essentially to reduce the amount of fossil fuel burnt per MWHr consumed which might be important, don’t you think, in a few years. Our children and grandchildren may even thank us for our efforts, at least in SA.
P.S. if you really want to see ugly power infrastructure check out the Dry Creek Power Station at the corner of Grand Junction Rd and Churchill Rd Dry Creek in Adelaide. These roads carry thousands of commuters each day and the stacks of DCPS have been waiting for a paint job since at least 1992. Perhaps we should start a brickbats & bouquets award for infrastructure as they do for architecture. DCPS Coords: 34°50’52.65S, 138°34’56.59E
Paul,
the electricity supply industry has never been one to consider the visual impact of its infrastructure on the landscape or the suburbs, largely because no one in government ever insisted that they do so. Admittedly, there is some improvement round the ‘burbs (Brisbane) where the Energex substations are now somewhat more pleasant to the eye.
Unfortunately, much of LV distribution network has been put in above ground and its remaining useful life is probably 40+ years. So there is not much change possible there.
Other owners of infrastructure in SEQ (Queensland Rail, Brisbane City Council) are making significance steps in decorating what they own, eg Milton Railway Station, Coopers Railway Station in Brisbane, plus the traffic signal boxes in Brisbane. They are even involving the community in achieving this!
So there is some infrastructure which can be made pleasant to the eye, if the owners can step out of their old ways.
Is that enough for Joe Hockey and his leader? Who knows? No holding your breath now.
There are precious few power system stabilizers in South Australia. If Northern closes there will be even less. What happens on a high demand, windy day, and a few big lines trip?
A colleague (leaving the utilitysector for greener pastures) once said “everyone wants the product, but no one wants the packaging”. That pretty much sums up the above.
I understand the sentiment, and (it seems to me that) there is some truth to this.
However it also seems to me that there is also other layers – in that we’ve seen there are things a large number of people do want, but some have not wanted to pay their “fair” share for these. Some obvious cases in point being:
1) Historically, the mass roll-out of residential solar PV to hundreds of thousands of people, supercharged on the back of (initial) overly generous feed-in tariffs (of which I am one of 400,000 Queensland beneficiaries, for instance).
2) In the current day, a rocky road ahead in the move to more cost-reflective network tariff structures (that’s not being helped by the understandable claims that the self-interested network companies are only making moves now that demand is declining, whereas this should have happened 10 years ago when the take-up of air-conditioning was accelerating – i.e. that they are only concerned because its affecting them now, not when it was “only” a matter of one group cross-subsidising another).
Re stability of the SA grid, there is a relatively new gas combined cycle power station that has hardly produced any power for the last few months. Pelican Point has a capacity of 475 MW, and I’m sure its owners would be watching the latest developments with Northern and Torrens Island A with optimism.
Unfortunately Pelican Point runs on natural gas and once the Queensland LNG plants are commissioned there will be no gas to run base load generation anywhere in the NEM
Here is one answer, and an obvious one. I am unable to buy roof-top mounted solar panels as I live in a flat, currently rented, by the way. (Refer the correct comment by Edward Boyce, above. ) Strata title buildings must secure unanimous agreement to make a move. It strikes me as inequitable, indeed, a failure of democracy, that government provides incentives not equally available to the entire population. These solar rooftop incentive schemes represent a transfer of wealth to the better-off. There are other reasons, some architectural, that inhibit the take-up. You do not broach some other complex issues about sunshine accessibility in built-up areas,. e.g. If a high-rise building or tree overshadows someone’s’ collection area, is there just cause for complaint? While the installation of roof-top solar circumvents network distribution costs – the larger part of a consumer bill – the more equitable alternative is large-scale solar (e.g. AGL’s NSW power plants) providing the asset owner is prepared to accept the commercial risks of wholesale competition. Your comments make the assumption that the network utilities remain dangerously asleep and are unable or unwilling to improve their efficiency. Maybe, union-dominated bureaucracies will prove too lazy to save themselves, but I do not buy the death spiral story yet.
The renewable energy advocates have blocked substantial progress on cutting CO2 for the past 20 years and are continuing to do so. Locking in the RET will delay substantial progress for another 20 or 30 years.
Only nuclear can have a substantial on cutting the emissions intensity of electricity. Anyone who has crunched the numbers knows this. Rational analysts have know this and have been showing it for at least the past 25 years. But the RET is forcing renewable capacity into our market and it will block the possibility of substantial emissions reductions from electricity for 20 or 30 years. Even once the public and voters see through the renewable energy ideologue’s dogma, it will take another 20 to 40 years for the impact of the irrational policies that have incentivised renewable energy to wash out of the system. That’s how much damage the green ideologues are doing. They either don’t realise or don’t care.
Thanks you. Excellent post. Very informative. From my perspective, the sooner governments sot trying to pick winners (e.g. pro-renewables, anti-nuclear, market interventionists) the better for everyone (other than who like to live off government imposed subsidies).
Sounds like you really need a copy of NEM-Watch (www.NEM-Watch.info) or NEM-Review (www.NEM-Review.info) so you can explore your question, at your own leisure.
There are ups, and there are downs (or is there a question behind your question about specific downs?)
Perhaps I’m missing something, but there seems to be a problem with the final figure.
Since 2005 (but excluding 2015), Australia has installed over 4 GW of PV, and just under 4 GW of wind. That’s an annual average installment rate of about 400 MW each, or 800 MW in total.
The columns in that chart are clearly much less than this (perhaps 300 MW average?). Even it it excludes PV, it still seems to be wrong.
Hang on, so SA’s decrepit thermal plant is unavailable and the state has a dearth of fuel reserves but wind is to blame for higher prices?
The fact is that if none of the wind plant had been built prices in SA would be even higher. The reason so much wind capacity was rolled out there rather than in VIC or TAS is precisely because of the high average prices.
I am in Ireland. I’ve estimated the cost to power Ireland entirely with wind turbines and pumped hydro energy storage. The estimated capital is 400-1500 billion Euro Their electricity consumption has been 26 TWh per year for the past 3 years.
Thanks Paul, very interesting.
On wind farm location: unfortunately SA price is still very attractive. Weathering the occasional bout of low price is still worth it.
Last month dispatch weighted price for some SA farms; Snowtown $40, Lake Bonney2 $49, North Brown Hill $43. Average SA price was $73. “A rising tide floats all boats” etc.
In NSW Boco Rocks got $38 and NSW avg. was $39. Portland got $32 and VIC was $34. Musselroe got $31 and TAS avg. was $35.
I note you didn’t have time to investigate individual behaviours. I think that’s where the biggest stories were in relation to that volatility. Fortunately AEMO has done it all for you.
It is going to be interesting to see how the combination of PVT’s and Tesla Powerwalls are going to influence energy useage. I came to your site here to see if there was a relationship between the daily timing of PV and wind, and as expected wind lags solar with the daily wind peak being in the late afternoon. I’d be interested to know if the APVI solar data includes self use energy or is it the total rootop energy production? The powerwall is going to blur the data as it takes the dc from the panels and the inverter only sees the energy as it is drawn from the system. It is at the inverter that the wifi link reports the energy output from the rooftop system. So the powerwall will serve to better represent energy consumption timing. However as thermal energy from PVT’s can be estimated from the electrical output from the panels, the Powerwall will mask that energy throughput. As the thermal energy from PVT’s is more than double that of the electrical output of rooftop systems this is a significant influence which will then only be visible in the drop in offpeak energy draw from the grid.
Disclosure: BilB has been banned from commenting at websites such as BraveNewClimate and Catallaxy
I think David Osmond has covered much of the territory.
But. Remeber that we are only at the beginning of the energy innovation revolution. It was just a few months ago that Elon Musk announced his Powerwall, and although there are none in Australia yet we are already counting the impact that this new technology will have. Similarly PVT’s are on in their infancy, and practical affordable backup power generation has not yet emerged. I am championing the Liquid Piston 70cc rotary engine out of Connecticut as being the disruptive technology that will fill this gap for residential and small business. This engine fueled by natural gas (methane) will produce a maximum 3.5 kilowatts, but is ideally suited for a steady 2.5 kilowatt output. So for low solar periods this engine will bridge the electrical energy gap feeding DC energy directly into the Powerwall and its consequent waste heat at 5 kilowatts directly into water heating filling in for the thermal output from the PVT’s. Where energy can be used in this balanced way the Liquid Piston engine is operating with 80% and above efficiency from methane gas. My estimation is that such backup will be used up to 600 hours per year in the average residential or small business premises.
The question naturally arises, “where does the methane come from if we are to have a 100% decarbonised economy?”. Good question. as it turns out, according to Victorian government figures, each Victorian disposes of 2 tonnes of waste material per year, more than 50% of which is waste paper and cardboard ie cellulosic material that comes from trees. This amount of material, if gassified and recombined into methane gas produces enough gas to generate 15,000 kwhrs per year per household of 4 people. One can derate that figure in as many ways as one likes and it still provides the complete answer to the low solar period problem, and from a renewable sustainable source.
It is my expectation that by 2030 rooftop solar will be predominately 4.5/9.0 kilowatt PVT’s Powerwall and Liquid Piston engine backed up to the extent that there will be sufficient over capacity to smooth out all but catastrophic climate events. There is one other variable that is yet to come into play, and that is the EV, or from my expectation the plug in hybrid PHEV. Such vehicles fit well into the 4.5/9.0 PVT household as such a system has the capacity to charge 2 such 8.5kwhr batteried vehicles 365 days a year. How ever in the event of a major low solar energy period there will be the option to not charge the vehicles and run on their fossil fuel or even their more sustainable bio diesel fuel.
The short and the long answer really is that zero wind for a number of days is a problem for now, but not in the long term. Perhaps the real issue is how will industry, high rise residential, and high rise commercial resolve their energy problems?
“One of the things that’s been puzzling me for some time has been the apparent willingness of wind farm developers to use development sites that deliver a wind farm with an output profile that appears very aligned to the output of other wind farms in the region.”
Just a guess – but perhaps it is due to how the projects are hedged. For example, if hedging using the exchange, the price at contract expiry is what matters (which is both unrelated to when you produce, and is sort of ‘balanced’ by the non-windy prices). Having said that, I understand that the liquidity in the South Australian futures market is pretty low – and wind farms are pretty unlikely to hedge with the exchange (for other reasons).
Related to this – a poorly designed / implemented CfD (and/or PPA) might have a similar effect. Simply going for the lowest price CfD (rather than lowest cost) would likely encourage wind farms in regions with good wind resource / high capacity factor and favorable planning laws (i.e. South Australia). The price might be the lowest here – but that says nothing about the difference between this price and the value in the market (and hence cost). For example, a higher priced CfD in NSW that was perhaps anti-correlated with wind in SA might end up costing less much, due to a price premium in the wholesale market.
This is all speculation of course – and presumably the people who design such things have already thought of this!
Moving through my fourth decade in the industry…, I see new energy technologies and energy market uncertainty compelling both residential and business consumers to respond by taking increasing control of their own destiny. In doing this these consumers become, what we call ”Proactive-Consumers” or “Prosumers”.
We see this evolution every day. Over the past 8 years our company focus has been on helping businesses enter contracts for grid supplied energy and arranging the best available network tariff, but of late our business is increasingly about helping businesses run the maths over alternative energy supplies, solar PV, battery storage, negotiating energy buy back agreements and in some cases designing and costing complete grid disconnection.
Even in Tassie roof top solar now amounts to over 80MW installed capacity, so consumers have invested probably $150M over the past 10 years. This tells us consumers are taking increasing interest in, and control of, their energy choices. Therefore overtime the traditional energy supply business model and political intervention will increasingly loose, or at least drastically change, in its ability to influence consumers.
When we discuss Network tariff reforms or investment in more centralised generation, we also can’t help but wonder if some incumbent businesses are planning for future scenarios which may not ever materialise. The challenge we think is rather to describe likely future scenarios and then plan around these alternative futures.
Cheers
Marc
Only in my second decade in and around the NEM, but have also seen the same cycles. The basic problem is that the NEW capex is set at the start of the process based on forecast demand… and forecasts are always wrong. Having agreed that an investment is “required” the regulator can’t very well turn around and say “oops sorry, we were all wrong and now you don’t get to make a return on that capital”. On the other side, regulated prices are intended to prevent networks exercising monopoly power in setting prices. I think it is a fair trade off to allow networks to recover a a “fair” return on those “good faith” investments that all sides have agreed are likely to be needed. Yes if it was a “real” market the investment would be stranded and not earn a return, but on the other side if left to that same market, network providers would address that stranding risk by shortening the payback period on that capital expenditure – resulting in higher network charges anyway. You essentially can’t argue for regulated cap to protect consumers on one side and then argue against that same regulatory system protecting the return on capital for the investor on the other. Benefit sharing is as good a mechanism to “correct” these forecast errors as they become evident, as the operator is incentivised to not spend capex/opex they don’t require as that (lack of) need becomes apparent.
Thanks Mervyn.
What’s the process of engineering variation into switching times?
How would you propose that SA’s controlled load peak (sometimes the daily peak load), be flattened?
A very similar pattern will emerge in Qld with solar PV (farms) and as they start to ramp up and build more capacity towards their 50% target. Spot prices here are already feeling effect of PV, and as growth in PV continues (both small and eventually large-scale), spot prices will continue playing less of a contribution to revenue-earning side of the project.
Re Alan’s comment, for PV + battery there is one recharge-discharge cycle per day to recoupe a battery or pumped hydro investment, and combined with wind or coal for night charging there are 2 cycles. Periods of high wind occur on approximately a 3-day cycle, so it would take much longer to recoup a battery investment.
Queensland already has one pumped-storage hydro station that could reduce some of this volatility.
Re the article, some of the South Australian discounts will spill across to Victoria when the Heywood interconnector upgrade is completed (190 MW of additional capacity). Since the main Vic and SA wind farms have highly correlated wind speeds, this will increase the discounts for Vic wind farms.
“Since February 2014 I find that I’ve published 24 posts on renewable energy here on Energy Matters (linked to in order of appearance at the end of the post) . In them I’ve written about wind, solar and tidal power, hydro, biogas, hydrogen and methane, CO2 emissions, interconnectors, exports and imports, energy storage, load management, backup capacity and ramp rates, the UK, France, Germany, Norway, California, remote islands like Eigg and El Hierro, the world as a whole and even mythical places like Atlantis and the Island of Denmark. I’ve reviewed “energy future” plans formulated by others, such as DECC, the National Grid and France’s ADEME and even come up with some plans of my own. I can’t think of any stone I’ve left unturned, or for that matter anything more I can write about for the time being.
So, time to sum up.
The ongoing transition to renewables – is it leading us into a clean, green, sustainable energy future, or will it leave us freezing in the dark? Or will it do neither?”
This post, and some of the other more recent ones listed in the 24 previous posts may prompt so thoughts you may not have otherwise had.
1) Wind in the W1 (i.e. current for 2014-15) totals 9,406GWh
2) Solar in the S1 (i.e. current for 2014-15) totals 5,068GWh. Note the correction I have added above to clarify that it’s actually just the small-scale series I used as “solar” to avoid the bump late in the FY when large-scale begins.
Nice one Paul. It seems to me that we are likely to start having significant amounts of battery storage in the coming decade, particularly in homes with PV. As a first guess, say 1 million homes x 10kWh = 10 GWh.
It would be interesting to see how the flow duration curves would change if you model this storage. A simple battery algorithm such as store whenever energy would otherwise get spilled, and release as soon as generation reduces below consumption. Since spillage only starts around 6W+6S, you’d only have to model this scenario upwards.
Perhaps repeat with 20 & 50 GWh of storage.
Of course, some of this could also be done with existing pumped hydro instead.
Yes, this type of “next step” is already underway.
There are virtually an infinite number of permutations for how storage might be used (especially considering all the different actors who will have a finger in the pie, and their various incentive). I had already opted for a simplistic approach, such as what you suggest, but realise that it will be quite different than how it will play out in reality.
Another simpler suggestion is that when looking at larger wind capacities, you use a much larger scale factor for NSW than for SA. It is certainly unlikely that SA will increase capacity by 10x, but that may be quite possible for NSW (and you could also use NSW as a proxy for QLD additions too).
This will help increase geographic diversity, which is currently dominated by SA & Vic
Because my focus is on “NEM-wide” in this analysis, I had already decided not to do this kind of thing. Fundamentally it would open up another degree of freedom in the analysis that would make it even harder to produce some useful results in the time available.
If you scaled the renewable generation as a function of state demand, then you wouldn’t add an extra degree of freedom. ie, if NSW has 35% of annual NEM demand, then scale NSW wind so it provides 35% of wind generation and likewise for solar. Unfortunately that still leaves a problem of how to model QLD wind, perhaps using NSW wind as a proxy may the best you can do.
Hi Paul, you’ve mentioned what the spillage is in terms of MW, and % of time. I’d also be keen to see what it is in terms of GWh, or even better, % of renewable generation (GWh spilled / GWh of renewable generation).
Love how you tied solar & wind to the baseload label, well done.
Seems to me that the accuracy in weather forecasting will be the key factor in not just the future electricity market, but also in industry (if they adjust to the peakiness of the market).
I would add another factor for the Alinta decisions: a rather poor (and unjustified) design choice. Specifically 50 MW capacity with 15 hours of storage. That is terrible configuration to asses at this point in time. The same size field / plant (albeit with a large power block and smaller amount of storage) – say 150 MW with 5 hours of storage, would be much more sensible (and would cost a similar amount).
As a side note – the recently constructed solar reserve plant (U.S.) has a 110MW/10hr configuration, which makes more sense in that market, (it was able to secure a PPA for $135/MWh as well).
Perhaps this would still be non-viable – due to the landscape considerations Paul mentioned. But given the price premium for being for being able to sell in the evening peak (or at times of low wind, and/or sell more on demand energy relative to wind) – it would be quite close.
Paul
How right you are about the need for better ramp rates in the “new electricity supply world”. In its response to the AEMC review of the AER ramp rate rule change, the Major Energy Users commented to the AEMC that the AEMC should be looking to the minimum ramp rates needed for the future market rather than setting ramp rates that could be provided by the existing fleet. By mandating a higher ramp rate requirement on generation plant, the AEMC would recognise the need for more agile plant in the future but could also provide a mechanism for allowing exemptions for existing low ramp rate plant.
But the AEMC persisted with a low ramp rate minimum in its preferred rule change
Some of our rapid-response fleet has come from hydro, and some from gas. An ideal gas plant would have its own storage facility nearby so it is not at the mercy of the spot gas market. The only one like this I am aware of like this is Mortlake, which has a nearby gasfield as its storage facility a start-up time of less than 5-minutes.
An interesting scenario is emerging for this coming season for drought-affected Tasmania. Their 2 main storages – Great Lake and Lake Gordon – have shown virtually no net replenishment since their power stations were run hard until the end of the carbon tax in July last year. Has the owner of Hydro Tasmania (the Tasmanian government) been asking too much in dividends, rather than sparing these dams by more imports from Victoria ? If water levels get close to the low supply levels on the intake structures, they may be in some trouble which they may blame on the drought, but is really because of their own management. Meanwhile Snowy Hydro has managed its resources much more conservatively.
Off topic, but an issue of the Portland Observer newspaper last week reported the transfer of a 250 tonne transformer from the port of Portland to the Heywood switching yard. This is a key part of the Heywood Interconnector upgrade, which will add 190 MW of capacity. According to the AEMO website, it is due to go live in July 2016, but if the transformer is almost in place, perhaps it will go live early.
This upgrade is a key reason for March closure of the coal-fired Northern power station, because it will enable additional cheap brown coal power from Victoria into the South Australian market. It will also enable additional wind power for SA into the Victorian market at times of excess generation.
That’s my understanding also – however I did not know, for sure, if that information was public knowledge hence was not going to be the first to state so.
Th e story portrayed by the aluminium industry is that they couldn’t possibly participate in any load shedding at any time, because then all their pot lines would need to be emptied out of solid aluminium. Yet when when the problem is their side of the meter it doesn’t seem to be a problem.
Paul,
I think that you have incorrectly characterised the issue that Snowy are flagging in their Rule change.
You have suggested that the the delta between CLEAREDSUPPLY and INITIALSUPPLY represents how much demand has varied from what AEMO expected over that 5-minute period. This is not the case – in fact the difference between these two variables is exactly equal to the amount of demand change that AEMO have forecast for the 5-minute period.
The problem that Snowy have highlighted is that after AEMO issue the 5-minute price and dispatch targets, non-scheduled loads (and generators) can suddenly change their consumption/supply, which is seen by the NEM as an unexpected change in demand. The scheduled generators however are required to continue to ramp to their 5-minute dispatch target.
As an example, suppose that AEMO forecast that the demand in a region is expected to increase by 20MW in the next 5-minutes, and calculate 5-minute price of $300. AEMO issue a dispatch instruction to the marginal generator in that region to increase its output over the 5-minutes period by 20MW to meet the expected demand increase. The marginal generator then commences increasing its output to achieve a 20MW increase by the end of the 5-minute period.
However, other non-scheduled entities note the high 5-minute price ($300), and choose to reduce their consumption in aggregate by 50MW. This demand reduction was not anticipated by AEMO, and despite the fact that the demand is now 50MW below what was expected, the marginal generator needs to continue increasing according to its dispatch target (or risk being declared as non-conforming).
AEMO need to balance the power system supply and demand, and they do this by reducing the output of the generators assigned to frequency control.
At the end of the 5-minute period, AEMO then “capture” the fact that the demand reduction of 50MW has occurred, and then ask the marginal generator to ramp back down again.
The intend of the Snowy rule change is enable AEMO to anticipate the 50MW demand change, and prevent the unnecessary and inefficient ramping up and down of marginal generators and frequency control services, and the resultant inaccurate pricing forecasts.
Do these calculations include the impact of reduced Tasmanian hydro inflows on the supply of certificates ? We have now had 2 years of low inflows, and over the next 5 years or so the system will need to be run much more conservatively to allow storages to build up from their current 26%. It may mean that an additional ~500 MW of wind capacity (over and above the forecasts above) would be required simply to replace the missing hydro certificates.
Thank you Malcolm, that is a good point you’ve raised. The basis of the future requirement is calendar year 2014 data, which includes half a year in which carbon pricing encouraged strong hydro outputs.
It would be reasonable to assume lower future hydro outputs in the near term as you note; how it varies as we get nearer to the higher LGC targets in 2020 is unclear. If the 2014 performance is higher than long term hydro outputs, that would make the target more challenging and need further new capacity as you point out.
What you say makes it even more challenging than I first thought. While the carbon tax was in operation, Tasmania typically exported about 400 MW of hydro power to Victoria on close to a 24/7 basis. This was achieved by running down their storages, particularly the Gordon Dam. I can’t remember what the net flows were between the end of the carbon tax in July 2014 and the end of the 2014 calendar year, but for simplicity of argument, lets assume the net flow was zero. So for the whole of 2014, this would be an average of 200 MW. Now Tasmania is importing about 400 MW from Victoria on a 24/7 basis. We therefore have 200 + 400 = 600 MW less hydro generation than in 2014. To make up for the reduced certificates from this hydro using wind with a 33% capacity factor would require 600 / 0.33 = 1800 MW of wind capacity over and above the new capacity foreshadowed in the article.
Snowy Hydro has also been generating less than normal because of low inflows, which may require further wind capacity.
I’m sure there are arguments with the HEC as to whether the reduced inflows are cyclic or a more permanent reduction through climate change, as has occurred for rainfall in south-west WA. There would now be an imperative of security of supply, guarding against the possibility of a further year of low inflows in 2016. Tasmania can only import about 40% of its electrical energy requirements on Basslink, 10% is from wind, and there needs to be enough water in storage to supply the remaining 50% for perhaps a 2-year buffer period. So I expect the high level of import from Victoria to continue for at least the next 2-3 years, resulting in a further a shortfall in certificates.
The ACT target of 90% renewables is only going to make it more difficult for everyone else. LGCs from the ACT’s PPA sponsored generators are to be passed to the government and then cancelled in the same way that green energy commitments are met. ACT retailers will still need to purchase a number of LGCs. Renewable energy activists in the ACT have made sure that the target was additional to the amount of renewable energy generated through the RET scheme. If for some reason the ACT doesn’t end up with sufficient sponsored generation the government intends to buy LGCs on the market to make up any shortfall and thus achieve its target. ACT electricity consumption will be about 2.8 million megawatthours in 2020. The target also includes that PV electricity generated within Canberra.
Paul interesting analysis.
I would only add that everyone is paid on the half-hourly price and so may respond after a high price event. How do the graphs look if the trigger price uses the settled price rather than the dispatch price?
Been away on leave so no time to respond. The 5/30 issue does complicate this consideration in a number of different ways:
1) You note that generation (or negawatts) might respond after a 5-minute price, so long as it’s in the 1/2 hour. This is one permutation.
2) Another is that those participating in this way might try to pre-empt spikes by dispatching based on predispatch prices, gut feeling, or some other method.
I don’t have time to delve more presently, but will see if I can return to this in future…
Not sure the LNG sector will consume 20% of QLD’s electricity given that the three downstream plants on Curtis Island and GLNG upstream have their own gas turbines and are not connected to the grid
The LNG plants will certainly have their own off grid power for liquefaction process. I think the is expectation that some of the power required for upstream gas production and transmission may be taken from the grid.
Demand spiked, Angaston (diesel) dropped 37 MW, and Ladbroke (gas, price setter) was constrained down due to network limitations. Wind dropped 5 MW the first event and 18 MW the second time. Small changes compared to diesel and demand.
Hi Paul,
Note that Tomago (800MW) was taken off just after the first $13800 5 minute price, just before the weather change. AGL ability to do this for a couple of hours is a significant tool to reduce VOLL in the market. Also may affect who wins a barbie!
Great web site.
Regards,
Rod
My understanding was the wind bids at the bottom of the stack (essentially -$1000/MWh) due to it being non-despatchable as per market design. That is to say that wind is a price taker, not a price maker. Its low operating costs have little to do with its position in the merit order for generation in the setting of the price in the market.
Paul – it may be worth a commentary on Tasmanian electricity pricing. With Basslink down for the next three months there has been significant RRP volatility.
You refer to the uptake of rooftop solar being a transfer of wealth from the lower to the middle and upper classes. It needs to be noted that it is the lower income households that are installing PV at a greater rate. Multiple reports/studies/actual statics analysing and comparing regions of PV uptake have found that it is the locations with lower income ratios that have the higher uptake of rooftop solar PV. Especially in recent years where there is low or no FiT for PV owners exporting surplus energy back to the grid, the transfer of wealth is actually the other way. These households pay $1,000s of their money for PV that has long payback in most cases, but has multiple positive externalities, e.g.social, environmental and economic benefits.
The increased supply of energy in the grid is contributing to maintaining a stable cost of energy, through negative pricing events and when aggregated and with large-scale RE generators have a depressing effect on wholesale prices -merit-order effect.
It is too simplistic to claim they are free-riding and having a (negative) wealth transfer.
I agree with the mandatory introduction of dynamic tariffs. How about making the transition attractive to home owners?
This could be done by mandating dynamic tariffs based upon coincident peak demand, instead of maximum site demand. Residential peak demand is less correlated to network coincident peak demand.
Or, by mandating that demand charges are uniform for all customers, e.g. 10 c/kW daily charge, irrespective of the site (residential, commercial, industrial)? The rate could be adjusted each year to ensure the networks make their regulated returns.
Or, by mandating that dynamic tariffs only be introduced if fixed (daily) charges are eliminated? This way, end-customers would be incentivised to install solar + storage systems that reduce their impact upon the grid. A reasonably efficient single family residence with a correctly sized solar + storage system could draw zero load from the grid during summer, and only 1 kW during winter (limited to off-peak times). The home owners could have a zero bill during summer, and an almost zero bill during winter. In the meantime, the contributors to coincident peak demand would pay for the grid, as they are the ones who determine how expensive the grid needs to be.
The effect you seem to have missed as a discrete element in the equation, I believe, is the increase in the rate of change of dispatched generation required late in summer afternoons as the peak demand arrives and the solar input decreases meaning that scheduled generation is required to respond more quickly. I think this would be more apparent if you looked just at SA demand where there is less industrial electrical load as a percentage, due to gas substitution, and the high impact in SA of air-conditioning demand which rises trailing the temperature increases. This means that late in the afternoon the demand rises more quickly as solar output falls away. Fortunately in SA we have relatively flexible generators due to the high penetration of CCGTs, OCGTs and even our gas fired steam generators are relatively flexible compared to large coal fired base-load generators. The Vic-SA interconnector also has a major effect but fortunately, due to the relatively small import required, does not tax the rate of change of interstate coal-fired generators.
In winter of course any solar generation will be all gone by late afternoon having reduced the output of the scheduled generation to a minimum or even to shut-down in early afternoon and then the evening peak will be entirely filled by scheduled generation and possibly wind. This will really test responsiveness and once NPS is OOS show the value of CCGTs.
Hydro Tasmania must have been under a lot of pressure from the Tasmanian government to produce short-term dividends rather than let the dams fill up. With the Gordon Dam so far below its maximum head, each unit of flow is generating much less power than normal. To operate at its maximum rather than its current head would be equivalent of another 120 MW wind farm, with no additional capital or operating costs.
It is understandable that the big users who have been asked to cut demand would take a hard-nosed bargaining position. Hydro was willing to take short term profits at the expense of the longer term security of supply, so why should the large consumers forgo profits too ? Perhaps they have agreed to short term reductions in return for extension of their low power purchase prices.
Hi Paul, thanks for a well-researched summary of the current Tas energy situation.
It may be worth noting the environmental consequences of the requirement to draw the storages down so low and keep them there – many are now below the environmental risk thresholds developed by HT itself and may be sufficient to cause long-term ecological changes. Do you know if the environmental flow for the Gordon River (world heritage area) is still being delivered with Lake Gordon storage levels so low?
Lower output per cumec at low levels increases the costs, as well as potentially damaging turbines as sediment loads increase. Nail-biting times for HT, the Government, industry, the Tas population and some of Australia’s most beautiful places.
It looks like now they have had to lay a new separate return layer now to try and get cable working again to full capacity.
Any comment for above relative to basslink? ie is it same technology, both have a fault date 10 years after entering service and a similar looking fault type.
I think the work of Amory Lovins and RMI would be of use here; I’m thinking of “Small Is Profitable” and “Natural Capital”.
Small is Profitable covers off how distributed resources can add value to an existing network so maybe a bit off the end game. “Natural Capital” spends some time discussing the concept of energy as a service, which us how I would see a massively atomised electricity market playing out – even if the energy is free there is still a cost to managing it yourself and there will always be a capital cost. Its not realistic to think even the engaged prosumer will be across the technical details of managing and maintaining their energy source and storage system… there will be a big role for those who can provide the service, trouble free. Whether thats on a unit basis or flat rate like rent will depend on how these things scale.
Given that concerns about climate change were mentioned in 3a, I would propose a Q6 section: “Would it address the need for strong climate action?”
I’ve only been doing energy/climate for a few years but I’ve noticed a severe lack of analysis around the prevailing and near-totally unspoken assumption that intermittent renewable sources + storage = effective climate action. Emphasis is always on the liberation of the consumer or breaking the stranglehold of utilities, with perhaps mention of how clean it would be. Well, how clean exactly?
Earlier this year I finally found some work from Stanford. http://gcep.stanford.edu/pdfs/events/workshops/Barnhart_NEA%20Stanford%20April%202015.pdf As you can see on slide 27 the projected lifetime emissions intensity of solar power supported by the front-runner of distributed storage technologies – lithium ion – is pushing 200 gCO2/kWh (I think “kg” is a typo). There are examples of conventional national and regional electric grids which operate at far lower than this intensity.
So a “rapid decreased in lifetime emissions intensity of Distributed Solar and Storage” would ideally be added to the list of assumptions here.
An interesting question. I expect electricity won’t be “free” for many years, apart from short periods when there is excess from renewables, because there is too much market inertia in the system. A fully off-grid house needs access to sufficient sunlight for house power and probably to charge an electric car. Pay for an up-front capital cost, and the rest is “free”. Suitable for rural and some suburban areas, but there simply isn’t enough sunlight to do this for most apartments. There is little incentive for these sorts of investments in rental properties. Furthermore, residential demand is only about 20% of Australia’s total, so the real action is in the commercial, retail and industrial sectors. These have their own paradigms, and while there is already a good return on investment for solar panels within such businesses, most organisations aren’t set up with the internal investment structure and empowered internal proponents to make this happen. So I would expect a slow gradual decline in grid power demand.
Some of our readers will find this article by Blad Plumer on Vox Energy & Environment provides an excellent discussion of some of the factors that need to be kept in mind relating to forecasts for how much, and how quickly, solar PV costs will decline, as it gains market share.
Great to see the report on the closure but its a shame the Port Augusta solar project hasn’t got up yet. Fingers crossed for those workers to jobs in the new economy.
Prof. Ledwich, you forgot to mention the role of Energy Efficiency.
Much better Building Energy Efficiency Standards than the prevalent 6-star standard – which is not checked after the build by the way – would help a lot to bring down the peaks. 8-star homes should be mandated and insulating homes should be incentivised again.
There are two main factors contributing to the peak:,
The first is aircon use to bring down the temperature in these inefficient ‘glorified tents’ as they were also called. An efficient home like a 9-star home needs minimal cooling, meaning a much lower peak.
The second culprit is cooking. Unfortunately with cooking nothing much can be done to reduce the peak, except for mandating efficient appliances. An efficient fridge is using 100 Watts whereas an old inefficient one can use up to 400 Watts. The same goes for ovens etc.
I agree that control of hot water and pool pumps during peak times (or mandating efficient pool pumps) is also necessary. Hot Water could mandatorily be heated during the day which the ‘duck shaped demand curve’ above shows is reasonable. Unfortunately the Newman Government also got rid of the heat pump or solar hot water requirement for new housing.
I disagree that lighting and TV use is contributing much to the peak
They would possibly contribute a maximum 600 Watts to the peak, depending on the house of course (except for plasma TV’s – not very common nowadays). LED downlights bring that value down quite a bit as well.
Cooking and aircon use however would draw 5000 W and on average around 2000 Watts.
Another factor is diminishing the willingness of people to save at peak times.
Because daily network charges are so high now – at least in Qld – saving electricity doesn’t really save the customers a lot anymore. Going back to low daily charges would be a lot like TOU charges because every kWh used would in this case signal that it costs them more to use more without having TOU rates. This isn’t the case anymore with high daily connection charges. My bill is over 50% made up of these charges.
Economists normally value the impact of such an event from the “resource cost” implications, i.e. the additional input costs to the industry (such as fuel and hired plant), plus the forgone value of any lost sales. For an event like this one, the resource costs will be much lower than simply multiplying the marginal spot price inflation by demand.
Your comment reminds me of the distinction between “Market Benefits” and “Customer Benefits” that have perplexed many who’ve looked to justify an upgrade of QNI over the years through the Regulatory Test.
I appreciate the distinction you make, however believe that the broader community will think of the cost of the Tassie crisis in terms of increase in costs paid collectively – independent of whether an economist might class some as “wealth transfers”.
– Liddell issues early April had quite an signicant, extended impact on supply, combined with some other unreliable baseloaders (Eraring in particular) and a heavy outage season (majors at Kogan, Darling Downs and Vales). Note the PDC for VIC reflecting separation overnight from northern states, overnight was particularly affected by low baseload availability (particularly with some participants bringing down plant at their lowest position eg Eraring clinkering overnight, similarly of late Loy Yangs with coal issues conserving during the day/overnight such that they can run morning/evening peak)
– Repricing of TIPS/SA gas. The Brooklyn compressor issue earlier in the year limited participant ability to fill storage and the impact is most notable in the gas market. Whereas LNG has capped the market in the past we’ve seen a number of prices into the $20 and even $30 range. The TIPS pricing in the $400 range is more than the spark spread but suspect they might be attempting to conserve gas to last out winter and a combination of transport considerations
– And as everyone has been talking of, the impact of LNG trains in QLD. There has been an increase in load from upstream compression, combined with intermediate gas generation largely limited to running only morning/evening peaks
A very important question and I wonder who has the most to lose and who has the most to gain?
I wonder if this is fair statement
Traditional suppliers/ Distributors = Big loss
retailers/customers/country = Big Win
I don’t know much about the energy markets but this seems like a likely equation. The question is what will the retail market look like, clearly it will need to be more IT savey.
The information in these graphs is really interesting particularly the Distribution of Prices for each State. Have you put together similar graphs for the Q1 prices or Annual prices. I looked through the Archives but could not spot anything.
Ben’s comment about resource costs is well made (as is Paul’s about the opportunity cost of consumption foregone by customers who reduced their usage). The other thing to remember about simple spot price (uplift) x volume calculations is that they miss the impact of financial contracts between buyers and sellers (predominantly Hydro Tas) which would substantially mitigate the net purchase payments of buyers. In looking at the effective additional financial cost to buyers (as a whole) one would really need to know their netspot electricity exposures and multiply those significantly smaller volumes by the assessed spot price increase. Of course there is no publicly available data on the size of these net exposures but it would be safe to assume that they are much smaller than gross demand volumes.
Interesting comparison would be WA, which has a portion of its gas allocated to domestic users (as have all other LNG exporters in the world except on the Australian east coast). The WA spot power prices on the AEMO website are reminiscent of a year ago on the east coast.
There’s something interesting I’ve noticed – someone will pick a few large-sounding SA price intervals, then present the penetration of wind in that state, and then conclude that the high prices are caused by wind.
Considering this strange phenomenon is occurring to a similar degree in all NEM regions, is it fair to assign the penetration of wind as a necessary factor in this phenomenon?
IE: if the penetration of this specific fuel type was lower or zero, would this phenomenon cease to exist? A hypothetical worth thinking about, considering how this has now taken the form of an unquestioned assumption, in some media outlets.
Surprisingly good analysis.
He states the cost of curtailment as fact though, without any supporting information.
Curtailing a few % of potential production makes very little difference to lifetime cost of energy. See the maximum generation periods in winter and how little duration there is.
‘Innocence’ and ‘guilt’ are such odd concepts, when you think about it – but they’ve been created for this issue due entirely for the need for a collection of media outlets to campaign in a way that protects incumbent fossil fuel interests.
I try to think about the logic of the claims. If we had a system of all gas, and no wind, the problem would still exist. If we had a system of some wind and some gas, the problem exists.
So, wind is necessary in the emergence of this phenomenon – as you point out, periods of low wind result in the dominance of gas, which is expensive, and so price goes up. But gas’s contribution to the issue – namely, high cost, is both necessary and sufficient – if the fantasies came true and we took an axe to any generation facility that doesn’t rely solely on pollution-intensive fuels, we wouldn’t be living in a wonderful land of low prices and bountiful supply (and emissions would be high, which is a bad thing).
Hence, my tweet pointing out that laying the blame solely on a [necessary but insufficient] factor, and entirely ignoring the [necessary and sufficient] factor is wrong.
As you say, there’s no getting around the fact that there are times when wind output is low across the entire South Australian fleet. As you also say, gas prices are high. So, the most constructive way forward would be to openly and honestly deal with *both* of these components, in a way that doesn’t mean a return to carbon intensive fuels, simply because we can’t be bothered solving this problem.
I guess it’s worth pointing something else out – spreading misinformation about the national electricity market is a direct contributor to problems that can arise out of the system, as decision-makers are influenced by this information.
In old media, statements are declarative, no evidence is ever included, and challenges occur over the course of days, long after the original declarations are assumed to be correct. In new media, even if misinformation springs up, at least these can be challenged or criticised (as you’ve done with my tweets above).
A major factor in the flurry of falsehoods that have emerged is traditional media outlets failing to understand the nuances of the system they’re talking about. The end result will almost certainly be a worsening of the problems they’re concerned about, with the added effect of destroying a new Australian industry. I feel that this impact is under-discussed – everyone just seems to shrug and accept it.
Actually, I don’t think Hornsdale will help very much. The problem is not when Hornsdale can produce, all the other wind farms can produce then and there is plenty of electricty. The problem is that when Hornsdale is not producing all the other wind farms don’t produce either and there’s not enough electricity in SA.
It’s the fundamental problem with intermittentancy that people have been carping on about for a long time. Putting more wind farms in SA won’t fix it.
I see this as a self-inflicted problem, with the best solution simply being to not inflict the problem, i.e. distributed generators should simply reduce local demand, they should not be allowed to feed any surplus into the grid. Any surplus can be stored in the batteries that renewable enthusiasts insist are not merely myths.
Energy price remains in the negative for SA, probably due to all the wind. Pelican point is running, my guess is as it can make money on the FCAS support? Osborne is off which is the first time for awhile but QPS5 (also Origin) is now running. I have a vague recollection that QPS5 is registered for FCAS but not Osborne, but not 100% sure
As a quick PS to my article this morning, I’ve included the following comments from an anonymous (and more learned) reader.
The reader emailed these comments in to help me, but I thought they would be more broadly useful. Thankfully they have agreed to let me post:
“Your layman’s definition for regulation service was pretty good … for contingency services.
The regulation services basically do 4sec corrections for demand/supply balance to keep the frequency around 50Hz hence keep time error as low as possible (a lot of manufacturing and computer systems use the frequency as an internal clock/counter).
Therefore, you can have a slight mismatch between demand and supply, and it will create time-deviation accumulation that can be corrected over a period of mins/hours in a day (you sometimes see through market notices additional regulation services enabled to correct time errors), but it is not the end of the world
The contingency services you described are what the 6sec, 60sec and 5min enablement markets are used for – continuous preparation for the trips of generation, major load or transmission elements, with each service helping to restore the frequency back to normal operating bands progressively as the time after the event continues. In SA though, there are relatively few contingency service providers hence they have a significant number of load tripping schemes because they know that the contingency services will not be enough (this was established many years ago, well before intermittent generation became such a hot topic).
In the case of the local service currently enabled in SA, it could be argued it is effectively a quasi-contingency service they are enabling with the 35MW local regulation due to the co-optimisation that occurs with the 5min FCAS, the settlement of which is something that the current dispute between AEMO, Origin, CS Energy, Stanwell, Alinta and the Coalition are arguing about in front of the dispute panel:
It’s even better/worse than that: SA Power Networks and ElectraNet are aware that during the day, tripping certain feeders with high PV penetrations might exacerbate the frequency imbalance, as some of these feeders have ‘reverse’ power flows on them. The problem is they don’t know which load points this is happening on, and it changes during the day!
One note – a shortage of FCAS-contingency services in SA (or any state) could be lessened by the AEMC approving and furthering the ‘ancillary services unbundling’ rule change, which they are due to comment on by 1-Sep.
Today loads (demand side) are technically able to offer the three FCAS Raise contingency services, but must be offered to market by the load’s retailer. Due to the complexity of a retailer administrating such a service and short retail contract durations, almost no demand side loads participate in the FCAS market today. AEMC notes that “with the exception of an aluminium smelter and two pumped hydro loads, all of the 20 registered participants to provide FCAS services in the NEM are generators.”
The A/S unbundling rule change would allow demand-side specialists (“DR aggregators”) to recruit and offer aggregated FCAS (raise), which should democratise the FCAS market, increasing demand-side participation and supply of FCAS, theoretically putting downward pressure on FCAS prices. In the context of SA, provision of increased FCAS from the demand side may lessen incidence of load shedding (UFLS).
I guess something worth not skirting around is the fact that challenges will be presented as insurmountable, catastrophic and civilisation-ending by a range of parties keen to ensure that decarbonisation does not occur.
In that context, it becomes extremely important to channel the inspirational and genuinely wonderful spirit of engineering and science, which is legitimately technology neutral and looks to correctly diagnose and remedy these quirks in a fairly optimistic and logical way. Which is why what you’ve highlighted above is so important – the difference between a ‘challenge’ and a ‘problem’.
Thanks very much for the analysis. My intuitive guess is that this correlation penalty is going to get worse as more Wind is added in Victoria as well since there will be less opportunities to export. My other intuition is that wind generators will increasingly suffer as residential solar + batteries takes off.
Fossil fuel generators have the capacity to tailor their production, at least to some degree, in accordance with predictive weather patterns but Wind generators will just keep competing with each other. Their only option will be substantial investment in their own storage technology (whatever that may be) but it’s not clear the CAPEX would be worth it.
It seems that the ACT electricity consumers are taking on all the risk of the PPAs signed with Horesdale Wind Farm, set to provide LGCs to the ACT government as an offset to the non-renewable supply fed into the ACT. The 1200GWh will represent about 40% of the ACTs electricity use. In order to lift the nominal renewable use to 100% the ACT government will surrender these LGCs.
John,
Your analysis of the ACT wind subsidy is spot on (if you can excuse the pun). Minister Corbell is spinning the line that ACT consumers are protected by a rising average spot price which closes the gap in the contract for difference. However the wind outputed weighted average spot price is declining (if it wasn’t then merit order effect would be another fantasy). ACT consumers will pay a much higher subsidy to SA wind farms than the government is predicting.
The media in the ACT is accepting the Government’s propaganda. Several op/eds have praised the canny Minister’s scheme.
I suggest it’s a combination of high labour costs and the increasing time and cost in bringing the aging generation units back into production after failures that are killing Hazelwood. For the last 3 weeks of September it produced little more than 1000 MW, or a 55% capacity factor. According to The Age article, there are 550 direct employees and a about 300 contractors, which is a lot of people to support from just 1000 MW. For comparison, AGL’s Loy Yang A has 500 direct employees, but in September it produced an average of about 1600 MW. Furthermore, its output has ranged from 1000 to 2000 MW, tending to be lower when total Victorian wind output was higher. AGL also operate the MacArthur wind farm (plus others), so they would not want to depress spot prices in a way that affects their other generation assets. For this reason, Loy Yang A would tend to be achieving a higher price for its output than Hazelwood, but also has the benefit of lower labour costs and lower maintenance costs.
The owners of Hazelwood (GDF Suez and Mitsui) also own the newer Loy Yang B, and would be acutely aware of the lower operating costs of the newer power station. They would be beneficiaries of less competition in the market for base load power after the closure of Hazelwood.
Federal and State Government can help the #LatrobeValley community in #Gippsland through a “Just Transition”
We are yet to meet with Josh Frydenberg (Federal Minister for the Enviroment and Energy).
We have already met with;
Daniel Andrews (Victorian Premier),
Lily D’Ambrosio (Minister for Energy, Environment & Climate Change, Suburban Development),
Wade Noonan (Minister for Industry & Employment, Minister for Resources),
Jaala Pulford (Minister for Agriculture),
Harriet Shing (MLC for Eastern Victoria), local Councilors and other Community leaders.
This has new in an attempt to help save this region from the uncertainty of the BrownCoal Energy industry transition specifically relating to Hazelwood and Yallourn Power Stations and Mines.
The initial feedback is good but the pressure will need to be maintained to ensure commitment of funds and resources to our area of Latrobe Valley (Morwell, Traralgon, Moe, Churchill, Yinnar, Glengarry, Tyers, Trafalgar, Yarragon and Warragul).
1. Transition Framework is needed to facilitate existing power workers moving from a closing power station or mine to an active one.
2. Create a BrownCoal Transition Group in Latrobe Valley to co-ordinate the closures in a planned way. The group membership needs to consist of Federal, State, local Council, Unions and Industry representatives.
3. Build a Genuine World Class, High Speed Rail (HSR) that operates between Sale – Traralgon – Warragul – Pakenham. Create a transfer Hub at Pakenham to enable integration and lower costs.
(Not Melbourne to Sydney that draws the negative involvement of the Airline Industry lobbyists and adds costs.)
3. Ensure local residents are given priority for jobs created by this process.
This will help maintain local housing prices and also council revenue through Rates which will be needed to help our community.
4. Build a new Gas pipeline from Sale’s Longford Gas plant next to the old one, through to Hazelwood and Yallourn Power Stations to allow conversion from Coal to Gas generation. This was fully researched in 2008.
This helps employment at both ends of the pipe and prevents the need to do coal seam gas or fracking.
5. Develop a Green Factory Centre dealing in Wind Turbines, Blades and other technology in Morwell behind MidValley where an existing rail siding exists and integrate it with the high speed rail (HSR) to be used at night and during off peak periods.
This has been done successfully elsewhere by speaking with OEM’s and including in the signed deals.
This can’t be due to insufficient capacity at this time of year, maybe loss of a frequency control unit, not enough synchronous capacity available to deal with that, maybe due to some being offline at a time of plentiful wind power?
Great site by the way , there is very little out on the web that is not extremely biased.
I do enjoy reneweconomy, they are slanted way too green and hence not an accurate resource but are a good source of news.
Their green assumptions slant their data but I still enjoy their mostly accurate direction most of the time and their stories are relevant.
While I may work in the coal industry I also have 40 solar panels on my roof of 10kw running on APS microinvertors. The reality is that with the 5c Feed in Tariff it’s a joke and is disappointing so my next step is to maybe connect an EnPhase battery system for night use and store my daily generation. Price point and are still an issue.
It was fascinating here in Adelaide watching from the hills some suburbs in darkness while some had just had their power restored.
My main question is why it took so long to restore power, from about 4 pm until 7:30 pm at the earliest? There is enough inter-connector capacity to power the entire southern half of Adelaide, and there had been no damage to transmission infrastructure in the southern part of the State. Indeed, this is what occurred with the first restoration of power to my area, that all but about 100 MW was from Victoria. They apparently had a meeting – was to make up for not having pre-agreed contingency plans for black starts? Were there contradictory positions that the various market players were arguing about ?
Is the apparent slow staged restoration of power about maintaining stability and control of frequency? Not that I fully understand the complexities of electricity generation, but I thought it interesting that the ElectraNet boss on radio this morning mentioned that wind generated electricity cannot be used for ‘black start’ processes, that they need to get full control of load and frequency before introducing wind..
black start with windmills? – that’d be “interesting”
I hope that an honest and thorough engineering led evaluation of the failure is performed – then we’ll see. The moment you see pols pronouncing on the matter – you’re being managed…..
Good clear article Bruce and thanks also to Paul McArdle for getting his analysis up so promptly.
A couple of very minor corrections to Bruce’s numbers – there was ~300MW of wind generation in the South East at 16:15 AEST – the other 80 MW was from Ladbroke Grove (gas turbines). Torrens Island was producing 274 MW from three (not two) B units (200 MW capacity each), I’m not clear how Bruce knows the 4th Torrens B unit (TORRB2) was synchronised. According to bid data it was not offering any capacity to the market.
ElectraNet says that “3 of the 4 transmission powerlines between Adelaide and the north of South Australia” were damaged by the storm – this would refer to the four 275kV lines running from Davenport / Robertstown to Para and Mannum respectively (see Figure 2-2 in https://www.electranet.com.au/wp-content/uploads/report/2016/06/20160630-Report-SouthAustralianTransmissionAnnualPlanningReport.pdf) These lines run on 3 separate corridors almost all the way, so losing three of them in a single event is extraordinary. That said it’s not yet clear exactly where the failures occurred and why that would necessarily lead to the loss of all supply from generation north of Adelaide – we’ll have to wait for fuller information to understand why.
Given that loss of supply though, and with ~700+ MW already flowing from the South East there is no way the Heywood interconnector and transmission to the South East of Adelaide could have securely picked up the slack (I understand that Heywood interconnector flows briefly shot up to over 1000 MW at which point protection mechanisms tripped it off), nor could TIPS – or any other plant – ramp anywhere near fast enough to meaningfully address the supply-demand imbalance. The question then is why the South Australian Underfrequency Load Shedding scheme did not trigger and reduce load in a more controlled way before the system went black – again we have to wait for a detailed analysis of events before jumping to conclusions.
So nothing to do with SA’s second largest subsidy farm, Snowtown North stage 2 reaching 96% capacity, then auto-furling and 140MW of supply disappearing in an instant, just before total grid failure? http://i65.tinypic.com/24onc3o.jpg
(Actually 2670MW lost from two Snowtown subsidy farms combined at the same time).
That moments later, the destabilised grid suffered a cascade collapse must be just a coincidence, or something to do with some transmission towers, or something. I’m so glad paid Big Wind shill Bruce Mountain-of-it cleared all that confusion up.
Massive power black outs happen all over the world. New York City completely blacked out. Central London where Scotland yard was completely lost. The power grid between Canada and USA where power line easements were ignored and trees shorted out the whole system. Risk management in the power industry looks a 85% reliability. When the 15% failure occurs all hell breaks out. My belief is learn from North Queensland and buy a domestic generator from Bunning’s. North Queensland has always had blackouts from natural disasters. For the academics.
V=IR.
V stands for power stations. Adding pressure to the system.
R stands for transmission lines a pathway for the energy.
I stands for current or energy flow. lose any one and lose the lot. Vectors, Engineering maths, unreal numbers etc etc. Queensland had a similar event in 1986 when Joh ruled and coal was the bees knees. Australians have become an ignorant population. Commercial TV and the owners should be classed as those who broke an ideal way of life for the sake of “”Who is the Greatest”.
The 70 km/hr wind speed at Whyalla Airport could well be magnified several fold on the hills that the transmission line crosses. There may also be some harmonics that set up by strong winds on the line.
BOM reported 100+ kph winds at Snowtown
28/03:37pm W 50 104
28/03:34pm WNW 54 104
28/03:30pm WNW 46 104
28/03:28pm NW 35 104
That is windspeed followed by gust.
AFAIK those BOM values are ground level air speeds. What sort of speeds would you expect at 30m above ground (where the towers are)?
From BOM: Sep 28 UPDATE Severe Weather Warning, locally DESTRUCTIVE wind gusts about 130km/h are possible KI, York Pen and Mid North.
So on the face of it, I could believe that the towers were hit by local gusts in excess of design load.
The question is – is a design goal of 41m/s in fact adequate?
I find this quite interesting. I live in Casper, Wyoming, USA. Today we have a high wind warning—meaning approximately 50 to 65 kph sustained winds and gusts to 95 kph. At times, we will have such winds for 2 or 3 days at time. Transmission lines sometimes go down, though usually it takes more than the 95 kph to bring them down. Some towns have had 160 kph wind gusts repeatedly. I realize that the high winds are not usual to South Australia.
Wyoming does have some wind plants, but most of our power comes from coal plants. The wind turbines often shut down due to the high winds, but no one seems to notice. I have to tell people to watch and see that the turbines don’t turn. I’ve seen them shut down in wind—turning one minute, still the next. People just don’t understand how wind turbines work and how variable and limited the output is.
“The ultimate outcome of this would be that thousands of battery storage, solar, community wind and CHP units would be incentivised and optimised for grid stability – aligning DG owners and the grids’ interests while at the same time lessening volatility on the wholesale market.”
Nice powerpoint. I sat through many of these types of presentations in a long IT career. Everything would be integrated and optimised into a smoothly working environment. Never really happened. Now looking at the evolving power industry and the desire to have vastly distributed, non standardised equipment purchased by individuals, community groups and companies all interconnected, playing well together and optimised for grid stability. I do believe the extent of challenge is understood, or perhaps its not acknowledged because it doesn’t fit the desired narrative.
Small example, we have had our house with smart meter for 4 years. Even though one organisation controls the equipment, the standards, the deployment etc and gets penalised for non delivery, they still have not deployed remote metering (coming real soon they say). That is a relatively simple project , albeit with large scale. Integration of 1000s of disparate systems, with fine control to deliver grid stability, when you don’t control all the levers seems to be a courageous goal Minister.
Hi Paul
I am still a Learner so I am chasing information rather than voicing my undeveloped opinion.
Can you explain the AEMO spot price for electricity in more depth please.
Alternatively point me to some resources.
I have been informed that Wind Farms have negotiated a set price per KW for 20 years.
Does Snowtown Wind Farm really get paid $112 per KW for all the electricity they produce?
Thus the spot price does not affect them at all.
Is it not true that large industrial electricity users, eg ALCOA, have negotiated a fixed rate (tariff) and so are not included in the spot price.
Is it not true that large power stations bid for and get a locked in amount and price 24 hours earlier
So the spot price only reflects the small amount of shortfall or surplus electricity requirements for that 30 minute time frame.
Sometimes, not often, the AEMO price is negative. As in -$2.34.
I am to believe that all electricity providers are paying some one to take their power.
A very poor business model I would think.
That’s a few questions there. I’ll try to respond in more detail in future posts (weaving into other commentary) but here are some starters:
Q1) Explain the AEMO spot price?
The AEMO has run educational courses for many years. I’ve actually never been on one myself, but we have sent other team members on some from time-to-time. More recently they have created this “General Public elearning – National Electricity Market (NEM) Overview” which, at only $220, is probably high value for money.
Q2) Do wind farms (and other generators) have long-term contracts in place?
The NEM is what’s known as a gross pool, meaning that all big generators actually receive the spot price for the energy they generate. The spot market also (largely) dictates how they are dispatched. Hence they all care about what happens in the spot market.
That said, most generators have a range of hedges in place to provide (partial) protection from the volatile revenues that come from spot. Some hedges are in the form of long-term PPAs (power purchase agreements) like you mention.
These hedges are not complete protection, so all big generators should still be concerned about what happens in the spot – though in reality some don’t realise this until they get burnt one way or another.
Q3) Do large loads have long-term contracts in place?
Because the NEM is a gross pool, ultimately all energy users are buying with reference to the spot price. As is the case for generators there are (most times) hedges in place that have the effect of smoothing the pain (or the benefit). These hedges can be of various durations – though long-term ones (e.g. 20, 30 years) are legacies of arrangements in place prior to the current market starting in 1998.
Q4) How does bidding work?
Generator offer prices (they are called “bids” in the NEM) are fixed the day before for the current trading day, but volumes in price band “buckets” can vary up to real time. Hence, dispatch volumes are only decided by AEMO in real time.
Hence the spot price really, really does matter.
Q5) Can prices go negative?
Yes, they can drop to a floor price of -$1,000/MWh – and sometimes do (like yesterday in QLD for a dispatch (5 minute) interval). When this occurs, yes, suppliers are paying the AEMO (on behalf of wholesale customers) to take their power.
I’m someone who actually follows a few utility market blogs like your own, so I’m reasonably more informed than most people out there. This is literally the first time I have ever heard of this program. I’m now going to have to go back to the quarterly novel that my power company sends me and see if they ever mention these plans.
But here’s what I see when considering this:
1) First of all, I never knew about it, and I just asked a few co-workers none of whom have ever heard about this.
2) The charge for 10% is $1.10 per week and the charge for 20% is $1.80 per week. Those are pretty damn affordable, but at such low percentages they make me feel like they’re not really making a difference. This also makes me suspicious that this is just like other green schemes where you’re paying money for the promise that someone has done something, while I wouldn’t feel confident that they actually did. What is their overhead on this plan, how much goes to admin, versus just paying for the electricity?
3) The charge for 100% is 5.5 cents per kWh. That’s a great percentage, though it would literally mean roughly $2 per day for me which is $182 per quarter. Which is a fucking huge jump. That’s $14.30 per week.
4) If I’m the only person doing it and take up rates are that low, then they definitely aren’t spending this money on alternative power sources. In a hypothetical situation imagine no one in my state actually took up this plan and I was the only person who did. Do they turn on my own special wind turbine for an hour to supply my yearly needs? Of course not, so it’s not like I’m actually paying for something physical. I don’t even feel certain that by voluntarily paying more for my power that these companies would then invest in that technology more. I see these more as placebos.
5) Why does it step from tiny amounts (10% and 20%) to 100%? Why can’t I pay for 50%? If you can pick percentages at all, why is there such a disconnect between almost nothing and absolutely everything? Things like this make me uncomfortable with why it’s structured like this.
6) I don’t like to think about my power bill. You’re forcing a complex choice which requires some extra thought on top of picking my provider, plans, billing period, and similar. All of this just adds to further anxiety / analysis paralysis when picking these things. If you really want to see who doesn’t want this, make it the default setting that everyone pays for 100%, then have people opt out. You’d probably see these percentages reversed.
This is why I believe there is a disconnect, it’s not because people don’t value it. It’s because the way in which you’re marketing it and selling it to me doesn’t make me feel like you’re actually doing anything different. This isn’t a vote with your wallet situation because it’s the same person promising me that they won’t just take the extra money.
But then again, maybe I think differently than everybody else.
Just to clarify – when you say “… the way in which you’re marketing it and selling it to me doesn’t make me feel like you’re actually doing anything different…” that I am not one of the ones offering to sell you the GreenPower – just someone on the outside, looking in and trying to understand.
Think I understand your dual concerns about “additionality”, and “auditability”. My understanding is that the Clean Energy Regulator address both of these in how they audit the production of LGCs, and allow for voluntary surrender (for GreenPower, and other schemes like the ACT PPAs).
The reason I don’t pay for green power is because the electricity companies have to buy green power whether or not I pay extra for it. I thought about this for 10 seconds a couple of years ago and stopped paying for 20% green power because I feel like I’m ust subsidisng everyone else and the power company.
Thanks for the explanation of your situation, Mike
The way I understand it, if you had purchased GreenPower, it would have meant additional LGCs purchased from the REC Registry (hence more green power supplied, over and above the MRET target).
I understand the way it would feel like subsidising others, though.
I think a lot of people are willing to pay a little extra but they don’t feel it is fair if they are the only ones paying. That and the previously mentioned points about general awareness of the scheme.
Government Accredited GreenPower is actually a highly regulated mechanism. When you read the actual rules on how it works, you will find that both Mike and Uriah’s concerns are adequately dealt with. The certificate schemes means that the premium is definately spent on green power and that they only have to buy the green power if their customers are signed up.
The real problem is that people don’t know about GreenPower. At sustainability forums, you can ask who supports renewables, doesn’t have solar but has GreenPower, many people don’t know about it.
One way the disconnect between the many who support renewables and the few who purchase green power could be decreased is by rolling out a default for GreenPower for new customers who then have to opt out if they don’t wish to pay the premium.
Paul there was some analysis a couple of years ago that when times are tough people stop paying for the green power option. I don’t think that it has ever recovered from that, It had a good take up then when it was shown on the bills how much it was people simply said no. It costs extra and most won’t pay for it so agree words over actions.
I think there is a difference between supporting renewable energy and deciding to pay for additional green power. Most people agree that emissions must be reduced, and that renewable energy is a means to this end. However, few people feel that their individual contribution will make any difference. Contrast this to roof top solar wherein a tangible individual benefit is perceived.
The ACT government has recently cancelled all their green power purchases for government business because they thought they could do more good for ACT ratepayers through investment in energy efficiency.
Climate change mitigation requires national and united national effort. Worthwhile climate change adaption can be achieved by individuals through choices that affect their built environment.
Don’t know about you but I’ve had a number of Indian reps cold call offering to change my light globes, shower heads and install chimney balloons, door seals and power boards- http://easybeinggreen.com/energy-efficiency/
so many consumers would have taken up the free offer and as such believe they’ve done their bit, particularly with dramatically increased power prices. We all know there’s no such thing as a free lunch and so the cost of such programs is in our power bills.
Now I’m in Adelaide with an older Tudor home and I can tell you within a few weeks 2 of the 3 chimney balloons deflated, the TV power boards crapped themselves, the family moaned so much about the shower heads I changed them back and I already had mostly CFL globes but the Indian guy was happy to give me a heap of them and sign his form to say he’d installed them (they have someone from head office ring you on his mobile there and then and you nod and agree all is kosher, like lots did with the pink batts scheme when they already had insulation.
Most folks I know have experienced all this and figure they’ve done their bit, particularly as they know they’re getting screwed for all this dodgy rubbish via their power bills.
1. I’d be encouraging bad policy, making Australia less economically competitive and less efficient. Renewables reduce our productivity.
2. Renewables are hugely expensive, requiring massive subsidies *(well over 100% subsidy) which is forcing competitive power suppliers out of the market
3. Intermittent renewables are a very expensive way to reduce emissions. At 20% penetration, wind is about 50% effective at reducing emissions. Nuclear on the other hand is greater than 100% effective (because it tends to replace baseload coal).
4. Wind and solar are not sustainable. Their ERoEI is insufficient to power modern society as well as replace themselves. They are entirely dependent on reliable power sources to support them.
5. They are supported by a purely ideological agenda. This has a short life.
6. After 60 years of solar PV and and 100 years of solar thermal engines, solar supplies just 1% of global electricity, wind 4%. And there’d be almost none if not for the ideological agenda forcing governments to massively subsidise them.
These are my reasons. I expect many others have made a rational decision and decided the risks and possible rewards are not worth the costs.
My widower father was in his dotage living alone in a unit and I increasingly looked after his financial affairs, a major problem being doorknockers and charity collectors. When I opened an electricity account to deal with I was surprised to find he’d switched to dearer Green Power with Origin and I enquired as to how that came about. It was then he produced the ‘presents’ he’d got posted later via the nice young fellow knocking on his door. A nice blurb from Origin, a sower head and some CF globes and a miniature green Origin footy to go with the dearer power bill.
So I’m straight on the phone to Origin to complain about sales staff preying on pensioners like that only to be told not to worry as the Green Power deal was off and all customers would be refunded anyway. Seems the Govt had knocked back their claim to Snowy Hydro being flogged as Green Power after Origin had been the winning tenderer in its privatisation.
If you read my submission http://www.greenpower.gov.au/Business-Centre/Program-Review/~/media/4488FFC5C5B04BACAEA881E393F33BB8.pdf or summary of recommendations http://www.greenpower.gov.au/Business-Centre/Program-Review/~/media/6DDD9A53908E49AA9BE6A0AE098154AA.pdf … on the Review of the GreenPower Program undertaken last year by UTS, you may appreciate the technical and legal issues associated with GreenPower and the simple fact that there is no legal framework that allocates the ‘renewable use’ attribute or the ‘reduced emissions’ attribute to those customers paying for GreenPower. Neither the Renewable Energy (Electricity) Act or the NGER framework transfers these attributes to a paying GreenPower customer. These attributes are shared across all grid customers. Consequently, the pricing structures for GreenPower are not fair. More customers understand this and it contributes to why the scheme is failing. GreenPower Marketing guidelines guide the customers to claim reduced emissions whilst those emissions are already allocated across all other customers under the NGER Framework and National Greenhouse Accounts Factors..
So “GreenPower pricing should be based around the cost of producing and delivering renewable electricity rather than as a penalty above standard electricity.”
Well it’s like this. If you pay homeowners 54c/Kwhr for any unreliable leftover solar power they’re not using, when reliable baseload costs maybe 8c max, we’ll fall all over you to be green and squeal like stuck pigs should you want to take away all our compassion for saving the planet.
What’s the real problem with green power pricing? It needs to be costed on the reliable base load it can always provide, just like thermal always was, but has now been bastardised by fickle renewables taking the cream off the top of all the milk. The only way to redress that is to enforce renewable generators to tender their power on the basis of the maximum base load power they can guarantee to the marketplace at all times. Naturally they’d have to partner deals with thermal power providers in order to do that and price accordingly.
Encouraging “clean” electricity without capacity payments for “dirty” electricity is a recipe for disaster, because the “clean” variety tends to be AWOL during heatwaves. Who would build a new conventional generator when the govt is intent on shrinking its market?
Australia needs to take care of and replace its conventional generators, the only component of supply that can deal reliably with broken equipment, and the low-wind days of heatwaves, but is it possible to change the tide of public opinion and the adverse economics before widespread power outages does that the hard way?
A question that arises with the Victoria and SA being short energy over the 17/18 summer. Does this result in the SA-Vic inter-connectors being under utilised as NSW/Tas can not import enough to flow through Victoria onto SA to meet demand?
Seems the regulators are going to have to guarantee via the market, the returns of Torrens Island and Pelican Point by all accounts-
“The COAG meeting discussed an interim report from Chief Scientist Dr Alan Finkel on energy security, which was commissioned after the September 29 blackout.
Dr Finkel told The Advertiser that measures had been put in place to safeguard SA’s power supply, including new procedures adopted by the Australian Energy Market Operator. “Within the last week, AEMO has put forth a requirement that there has to be a minimum of two synchronous generators operating at all times within SA,” Dr Finkel said. “That provides voltage support and system strength.”
The synchronous generators are gas-fired turbines, which are more reliable than solar or wind power and stabilise the network.”
So much for Australia’s phony national electricity market where all electrons were supposed to be equal but anyone with any technical brains knew some are more equal than others.
“What game are these union heavies playing, one does wonder…”
Gotta get the redundancy payouts fully bedded down with all these Labor/Greens about. Sort it now or we’ll give the punters a bigger taste than SA of what you’re up to with all your unreliables.
The problem is you’re trying to model an extremely variable and complex system when it’s obvious you should stick to computer modelling a simple system like climastrologists do.
As someone responsible for a fair few modelling exercises I concur with a lot of what you say Paul.
Two additional observations:
1) Modelling is not the same as forecasting. Non-modellers in particular too frequently seem to believe that anything coming out of a “complex” model must be a forecast, and far too often go on to treat the results as holy writ.
You’ve clearly pointed out the plethora of challenging assumptions needed to feed any model, and the real world uncertainties and unknowables that models cannot adequately capture. I’d argue that these difficulties have increased in uncertainty and complexity in recent years (consider the future price / availability of gas for power generation as just one example).
Given this, treating any single set of model outputs as a prediction of the future, rather than just a scenario reflecting a particular set of inputs and modelling method, is naïve at best and quite disingenuous if done deliberately.
2) Modelling and model results are too often employed to advance particular policy positions. This is related to your point about modellers talking their own book, but I guess my observation is that once every policy position is advanced and buttressed with the support of sophisticated (but often opaque) modelling, presumably with the aim of rendering it immune from debate and criticism, it actually devalues the modelling process in general.
I think it might be a mistake to totally discount solar. One of the behaviour changes that will improve the benefit to solar owners is if they can load shift to times when power is available to them from their solar panels; eg rather than overnight water heating instead heat to full capacity in the middle of the day.
The other impacts that are likely to be significant are: –
Shift to LED lighting, including by local councils which must substantially reduce overnight lighting loads,
Greater awareness of parasitic loads and a greater portion of consumers actually turning off devices overnight, and
General improvement in the efficiencies of appliances, fridges and air conditioners
I’m rather disappointed at the one sided comments from you about workers when it was actually AGL that threatened Victoria’s electricity supply by stating they would lock out all workers, even those not involved in the 14+ month dispute.
The real threat to the Energy Market and the Government’s quick response came from AGL’s escalation of a dispute and a lack of reserve in the NEM to cover an individual company decision to not just lock out the Power Station workers, but also the Mine Workers which are not even involved in the dispute.
AGL’s decision to lock all workers out was simply done to cause an AGL induced coal supply shortage to ENGIE’s Loy Yang B station. This was a legal manoeuvre to affect a 3rd party company, so as to force an arbitrated outcome rather than negotiate in good faith bargaining.
I’ll repeat the workers view, that it has been 18 months, where ALG would not even meet with the unions to negotiate. For this to be the first threat of industrial action from the CFMEU after the 14+ months lack of good faith bargaining by AGL. It’s very rich of you to side with the Boss.
Maybe you should stick to the facts and not speculation, like you normally do. Basking the workers and insinuating they are smoking something and want a longer Christmas break is hardly impartial considering that it wasn’t balanced with comments about expected multi million dollar profiteering by the remaining generators. I would have thought by now that you especially would know that the electricity generation sector is a 24/7 365 day a year operation and that each and every shift worker has spent many a Christmas Day, Easter and public holidays at work and away from their families.
Contrary to the AGL propaganda the dispute is not over pay raises, it is all about job security, as AGL wants to cut worker numbers as evidenced by their forced redundancy clauses.
You should really be reporting about how all the Generators are simply looking to game the market and maximize their profits at the expense of all consumers.
ENGIE owns Pelican Point Gas, Hazelwood and Loy Yang B, so when they close Hazelwood they are looking to make a killing from the market. Some of the Hazelwood Units could run until 2020,2019,2019 and 2018 without any extra expenditure but are being shut down for profiteering and ideological reasons, without a second thought for the Workers, Latrobe Valley Community and consumers.
To put the announced closure into perspective, ENGIE hired young workers 2 years ago and promised the a job until 2025 with a potential of 2033.
ENGIE’s previous announcement of redundancy payments were a negotiated requirement of the EBA they signed 2 years ago and not a cent more was offered.
I’m one of those directly employed 500-750 Hazelwood Workers that are about to be made redundant in March because of this corporate profiteering.
How about you stop the union bashing and look at the problem for what it is, a political football between State and Federal Governments all based around the Generators manipulating the AEMO/AEMC rules for maximum benefit (To the shareholders of course).
I for one appreciate your more impartial and fact driven articles, that’s why I read and often refer people to http://www.wattclarity.com.au rather that the one sided reneweconomy bullshit.
I hope you take my criticism constructively and aim to deliver more facts than speculation.
AND there has been a decided move away from off-peak water heating (less heaters replaced like for like) by a combination of policy makers encouraging gas heaters as less emissions intensive and higher less competitive rates for controlled off peak tariffs. As mentioned by Michael, LED replacements for Street Lighting would be significant. This would also be the case in NSW so not sure if a similar impact would be evident there also?
It surely can’t simply be a coincidence that 28 September 2014 happened to be among the warmest for that time of year (including an all-time September maximum record being set at Avalon Airport). http://www.bom.gov.au/climate/current/month/vic/archive/201409.melbourne.shtml Could this be indicating that some sort of sweet spot is being hit between space heating and cooling demand? Unusual pre-heating of water going into electric-powered hot water systems may also be a contributor?
As you completely fail to mention Australian Standard AS4755: Demand response capabilities and
supporting technologies for electrical products, it appears that you are unaware there are technical solutions to the problem you introduce in your first line of your Joke Punchline article.
If you are aware of AS4755 then it’s disappointing that you have not informed the readers of your article the following information. WattClarity readers are not your average dumbed down blog and deserve more detailed information.
The solutions are not without consumer backlash though.
It’s called Demand Response Enabling Devices or (DRED) and there are various states of operation from DRM1, DRM2, DRM3 and DRM4.
DRM1: shut off, or operate at minimal load, for the duration of the event. In an air conditioner, this means the compressor must not operate, but fans and controls may continue to do so. During normal peak load events the utility (or a pre-programmed DRED) would ‘cycle’ air conditioners to operate for, say, 15 minutes and then shut off for the next 15. In emergencies, a constant DRM1 signal could switch appliances off for the duration of the emergency.
A pool pump controller responding to DRM1 must switch the pump off.
An electric, solar-electric or heat pump water heater must cease the use of electricity for heating water (whether by resistance elements or compressor operation) but water circulation pumps and controllers may continue to operate. This makes DRM1 a technically superior method of control for solar and heat pump water heaters than traditional off-peak, which de-energises the entire unit.
DRM2: continue operation at not more than 50% load for the duration of DR event.
DRM3: continue operation at not more than 75% load. This was differentiated from DRM2 at the request of electricity utilities, on the grounds that it will be easier to enrol customers in DR programs if they can be assured of having at least 75% cooling during DR events.
DRM4: switch on even if timers or user settings would not require operation at that time. This is intended to allow appliances to come on at times when electricity price or CO2-intensity is low, eg when renewable generation availability is high. It would apply to swimming pool pumps – by bringing forward but not extending daily running hours – and to electric water heaters – allowing additional energy to be stored by temporarily raising thermostat settings. Utilities may also use DRM4 to pre-charge solar-electric water heaters on winter mornings, to prevent their boost elements from operating during the evening peak when space heating, lighting and cooking loads coincide.
.
Now here is the real Joke Punchline..
For an appliance to be AS4755-compliant, it must be able to respond to only DRM1(Shut off or Minimum load), which is not really consumer friendly.. An old adage used to be that the customer is always first.. This adage iis lost when you prevent a customer from using the air conditioner that they brought for the sole purpose of cooling themselves down on a hot day, all because it’s a hot day and Lack Of Reserve LOR exists due to insufficient reserve in the National Electricity Market (NEM)
The real problem!
Who is managing this whole debacle that is centred around closing existing power plants like Hazelwood Power Station without cheap reliable alternate energy sources ready to take up the Lack Of Reserve LOR created… NOBODY!
Well allegedly it’s the Australian Energy Market Operator (AEMO), under the control of the Australian Energy Market Commission (AEMC), which was actually created by the Council of Australian Governments(COAG).
Currently the entities of AEMO, AEMC and COAG are the equivalent of a rudderless ship being dragged by the undercurrents of the energy Generators.
Mark,
The availability of demand response options does not negate Shane’s main point – the cost of funding infrastructure needed to satisfy the increasing peak demand due to air conditioning.
There is a limit to how demand response can be used to manage this. Turning off lights, swimming pool pumps, and other miscellaneous power uses can shave a bit off the peak. You can’t really turn off your fridge and freezer on the hottest Summer days. In order to make a substantial reduction you would need to turn off or ramp down air conditioning. How would consumers perceive this? They spend money on a home air conditioning system only to turn it off during the hours they need it most. If forced into such action, they may start to wonder why they purchased air conditioning in the first place.
There are solutions in the form of better building designs however anyone involved in the building industry will understand how difficult it is to promote these sorts of design improvements. Construction standards in Australia are poor by comparison with other OECD countries. The industry is geared up to churn out cheap lightweight structures at very low prices. Any sustainability initiatives struggle to gain acceptance in this environment. When consumers are prepared to spend more money on their homes it tends to go to amenities such as granite kitchen bench tops rather than sustainability features and initiatives.
Note also that the Building Code does not help. It contains energy efficiency provisions based around minimizing annual energy consumption. This does not lead to the same outcomes that would be achieved with a focus on minimizing peak electricity demand. There are some perverse outcomes from this when NatHERS star rating software encourages designs that do not minimize peak demand. For instance:
– there is limited credit for deep overhangs and other external shading features
– dark coloured roofs are sometimes favoured over light coloured roofs
For most of the 200 or so years of European settlement in this country people have survived without air conditioning. For locations in the Southern part of the country, it is still possible to provide a comfortable home without the need for air conditioning. The critical factor is thermal mass. Lightweight homes can never be adequately comfortable without air conditioning.
If the retail electricity sector were properly deregulated so that consumers paid the true price of power at the time they use it, uptake of air conditioning would be greatly reduced. This should include time based transmission and distribution fees – not just for energy consumption costs. Such deregulation would then expand the market for well designed homes. Unfortunately, those who had already invested in poorly designed lightweight homes would suffer (financially, thermally or both).
Note also that cost increases resulting from the closure of coal fired power stations are a separate issue. Even without these closures, the increasing uptake of air conditioning was always going to drive cost increases – in generation, transmission and distribution. These issues may be compounding however it is important not to conflate them.
You are correct – AS4755 provides some mitigation of the issue, but as the article pointed out, I don’t think this standard really came into effect (with any real impact) until well into the air-con roll out. The point of the very brief article (not initially written for the readers of Watt Clarity) was to highlight the slow, reactive nature of regulators in an era of fast-paced change. Whatever your view of the rapid uptake of solar (and likely uptake of storage), it will have consequences to how energy markets behave. I didn’t make a big issue of it, but over here in the West, we have been bogged down by a 3 year “Energy Market Review” which has not really achieved much and hasn’t even addressed the looming issue of solar + storage. I can’t help feeling that in a couple of years, we will look back and wish we had the foresight to implement some market reform that might mitigate some of this impact – where I specifically, albeit briefly, focused on the capacity market and its inadequacies.
“Who is managing this whole debacle that is centred around closing existing power plants like Hazelwood Power Station without cheap reliable alternate energy sources ready to take up the Lack Of Reserve LOR created… NOBODY!”
Govt is managing it alright by intervening in the marketplace with subsidies and legislative fiat to create a whole bunch of suppliers to engage in the purist form of dumping one could imagine. The only problem is the nature of supply via the dumpers when they’ve driven the dumpees out of the market. Still, there’s no doubt blackouts and brownouts will reduce the emissions they want to and that’s the payoff apparently.
OTOH to stop dumping in its tracks Govt would only have to legislate for a level playing field (well aside from the subsidies and other specific legislative fiat which would be small beer under anti-dumping legislation) All it needs is for legislation that only permits suppliers to tender to the grid up to a maximum they can supply 24/7 all year round and bingo! naturally they’d have to cut deals with thermal providers to do that but bang goes their desired emissions payoff.
I should add the obvious caveat that it assumes our historical and ongoing inability to store energy which is rather pitiful, apart from in the form of calories and pumping water uphill. Yes I know the cheer squads place great faith in electochemical storage, but all I’d point out to them is we’re still using essentially the same lead acid battery in our cars Henry was plonking in the Model T.
Roger S,
I know there is a real potential to store electricity via a molten salt energy system.
Specifically by initially using off peak or hydro grid electricity as well as other heat sources to heat a molten salt energy storage system. This stored energy would later be used to heat steam for a turbine. Ideally this would be used in a NEM price peaking scenario and would likely be competing against gas generators. In some states the option of a solar heating array could be implemented. http://www.solarreserve.com/en/technology/molten-salt-energy-storage
I find it odd that the Office of the Chief Scientist should declare something to be “unstoppable”, which sounds much more like politics than science, and deciding the outcome in advance is the antithesis of good science. Predictably the marketing departments of the “Transition” industry are in raptures about “unstoppable”. The notion of consumers eagerly embracing “The Transition” is also somewhat less than convincing to me, eagerly embracing free money is probably much closer to the truth.
Both solar and wind will never have much direct impact on peak demand, but why does the Chief Scientist fail to mention this fact? It is already widely known that solar PV in South Australia has pushed peak demand times towards dusk, so adding more capacity will have very little effect. Similarly the large wind capacity in SA has pushed peak demand on non-wind generators to times of very low wind power, which hobbles any further wind capacity.
Moree has single axle tracking while the Broken Hill and Nyngan plants have fixed panels. This might be the reason why Moree gets up and running before Nyngan does. It is also several hundred km north east of Nyngan
Can I suggest that a good step can be taken towards a More Mature Discussion by dropping the ubiquitous word “challenges”, and calling it for what it really is, problems, problems, problems.
Ironically Tasmania, only recently beset by a major problem, now looks like the best place to be for electricity consumers, because of its relative immunity to governments setting targets.
PS one of our shyer/shier readers has emailed in the following explanation:
“the reason for the staggering of outputs of NSW plants is the stretch of plants across the timezones. Nyngan (102MW) is clearly in the middle of the timezone, Royalla (20MW) and Moree (56MW) more towards the East. Broken Hill (53MW) is actually in the South Australian timezone even though it is located in NSW, so is close to half an hour behind in terms of when solar noon occurs.”
Presumably the 2940 MW is a 5 minute scheduled demand figure – from what I can see the lowest half-hourly scheduled demand on 2 Jan was just over 3050 MW.
Interestingly if you use scheduled + non-scheduled (intermittent) generation as the demand measure then the minimum value was just over 3300 MW, the Dec 2016 minimum was not much higher at just over 3400 MW, and way back on New Year’s Day 2000 the minimum was 3377 MW – although back in 2000 AEMO didn’t report total non-scheduled generation, however this would have been a pretty small number before Victoria gained any non-scheduled windfarms.
Interestingly, in a research report published by Morgan Stanley yesterday and titled “Inside Industrials: Electricity: Demand -1% but Supply -23% (03 Jan 2017)”, the following passage was included:
“Of particular note is the 3.5% drop in Victorian electricity demand. The
Portland Aluminium Smelter has been running at ~27% since a power supply
interruption in early December (and may continue to do so for up to 6 more
months, according to the smelter). However, allowing for a ~0.1TWh demand
reduction for the smelter and the hotter 2015 Victorian summer, we estimate
Victorian demand is still down ~2.5-3.0%, which we attribute to continued growth
in rooftop solar, and a spike in commercial energy efficiency activities (the
Victorian Energy Efficiency Target targets 5.9mtCO e of abatement in 2017, up
from 5.4mtCO e in 2016).”
(my emphasis added)
Postulate a distribution of eleven household energy consumption groups around the mean shown in the graph for years 2009 – 2013. Annual kWh :-
11250
10500
9750
…
3750
Year 2009 average 7500 kWh.
Remove the greatest household energy group and recalculate the average for 2010. Result 7125kW.
Remove the next greatest group and recalculate for 2011. Result 6750 kWh
For 2012 result 6375 kWh
For 2013 result 6000 kWh
The ‘Average Annual Domestic Consumption (kWh)’ graph is generated.
Now take the removed groups to form the ‘Domestic (with solar)’ average.
For 2009, the greatest group without solar was 11250 kWh but 7000 kWh with solar, so solar average generation for this group was 4250 kWh.
For 2010, the next greatest group was at 10500 kWh without solar but the consumption with solar was 5000 kWh. This requires a 5500 kWh solar generation.
The average consumption of the two greatest groups (7000+5000) with solar is then 6000 kWh.
Solar generation in 2011 of 3500 kWh for the third group and solar generation of the fourth group in 2012 is also 3500 kWh and the final group requires 2500 kWh solar generation.
The ‘Domestic (with solar)’ graph is generated.
The hypothesis is that the highest electricity consumption groups in the first years buy full sized (5kWh?) panels, presumably off-setting the air-conditioning load in large houses. The third and later groups buy smaller panels or fit in less favourable orientations. The falling graph for those without solar is a statistical trick caused by the largest consuming groups leaving this group. The remainers do not have to change their behaviour at all.
There was an article that briefly appeared on The Age website that the Wonthaggi desalination plant had not yet produced any of the water that had been ordered back in March 2016. The order was for 50 GL of a 150 GL capacity, to be delivered by June 2017. The website of the plant’s owner, Aquasure, still shows that no water has been produced. One would think that in order to minimise power costs, the plant would be run at 100% capacity for 33% of the time between July 2016 and June 2017, utilising the half-hour periods with the expected lowest third of power prices. According the The Age article, there has been a commissioning problem with the underground cable from the Cranbourne Terminal Station (perhaps akin to the problems with Basslink). At full capacity, the plant uses 90 MW, so we can expect a boost to Victorian power prices when they finally resolve these problems.
The desalination plant is part-owned by Engie, who also part-own Hazelwood. It would surely be in their interests for the plant to be operating before Hazelwood shuts, firstly to obtain a lower power price for the desalination plant, and secondly to boost to the price of power sold by Hazelwood.
The 50 GL of desalinated water come at a high cost, whereas about over 80 GL of water flowed down the Yarra River past the pumping station for the Sugarloaf Reservoir in September and October 2016, because there hadn’t been sufficient investment in capacity to pump this water to other reservoirs.
Despite the afternoon price spikes Wivenhoe pumped storage power station still hasn’t turned on, (but it was generating in the evening peak on Thursday). Is its ownership by a company that has some major coal-fired power stations leading to a less than optimal market outcome ? It is owned by CS Energy, a State-owned entity that also owns Kogan Ck, Callide and Gladstone power stations. If Wivenhoe were to bid during the price spikes, it would reduce prices to spot-exposed consumers, but lower the price received by the coal-fired power stations in the fleet. One would think that if owned by another entity it would offer its full capacity of 500 MW during price spikes and evening peaks, and recharge through its pumps at night when spot prices were below $50/MWh.
These crazy prices must surely help the business case for additional large-scale solar farms in Queensland, and the GenEx pumped storage station. The spot gas prices were crazy too, at $13.73/GJ at the gas supply hub.
The high demand in Qld is presumably due to the LNG activity and the high temperatures, however, I’m curious to know how big of an impact the high temps are having. In other words, due to the LNG trains the demand in Qld increased, and due to this increase will even small-to-medium increases in temperatures (and thus load) lead to significant price jumps?
seems the 50% QRET is quiet timely, for daytime loads anyway (PV).
Not sure how closing Hazelwood is a win for emissions reduction, the same amount of electricity will be needed after it closes, some sources will get to increase their outputs, and there is only one type of generator that can do that useful trick.
Because no plant is more emissive on a t CO2/MWh metric than Hazelwood, so even if you are assuming its output will be replaced entirely by other coal plant, emissions will be at least marginally lower overall.
Great to see PV heavily utilised and reducing the carbon emission through most of the day, certainly the only bright spot. Such a shame most of the generation was through coal.
On a half hourly basis the scheduled demand peak looks to be 9369 MW. This is about midway between AEMO’s 50% and 10% probability of exceedance forecasts for this summer, and more than 400 MW higher than the previous high in 2010 (also on 18 Jan – how about that!).
An interesting point of comparison is that the 2010 peak occurred in the half hour 14:30 – 15:00 (trading interval 30) while yesterday’s was for half hour 18:30 – 19:00 (trading interval 38). This shows how solar has – by reducing scheduled demand during the sunlight hours – pushed the time of system peak to much later in the afternoon. In 2010 scheduled demand was over 500 MW below its maximum by TI38, so had the same underlying consumption pattern applied yesterday, in the absence of solar you might have seen a half hourly demand approaching 9900 MW!
Actually 18 Jan 2010 wasn’t the highest half hourly scheduled demand for Qld prior to yesterday – that was surpassed by 9105 MW on 1 Feb 2016, my mistake 🙁
But the points about later time of peak and maybe another 500 MW of demand in the absence of solar PV still stand.
And this is probably labouring the point, but on first inspection someone might look at the 6:30-7:00pm peak, with solar PV producing virtually nothing by that time of day, and say “well, solar is doing nothing to reduce peak demand”. However this is incorrect because in deferring the time of scheduled demand peak, solar PV has moved that peak to a time of day when underlying consumption has fallen (ie some industrial / commercial load would have reduced by then), leaving a lower maximum to be supplied from scheduled generation.
But once the scheduled peak has been consistently shifted to that time of day it is true that additional solar PV capacity will not further reduce the size of that peak.
One thing AEMO seems to neglect to do is adjust actual peak demands for weekend reductions, there are many holidays as well at this time of year, so its quite possible that previous years would have had similar demands if the previous heatwaves had fallen on working days.
The same thing happens with wind power, once it gets “substantial” it pushes times of high peak demand on non-wind sources to wind lulls, and ADDITIONAL wind farms have very little effect on those peak non-wind demands. South Australia passed this point a few years ago.
At some point in the future, when you may be looking for something to do, it would be interesting to see what the $ impact on the market was due to the solar output. In other words, quantifying the avoided cumulative cost by the depressing nature PV has on the spot price (merit-order). Conservatively, this may assume the existing spot price, i.e. spot price during solar output. This would be instead of trying to estimate a new marginal generator and raising spot price.
With the current availability of data it is quite impossible to determine the actual total output from small scale solar systems. The APVI data is based on such a small number of the installed systems, and even that sample shows gross distortions in system output, that expansion of the data to provide a whole state estimate is almost ludicrous.
Even the systems that are reported through PVOutput.org are more likely to be recently installed (more likely to have WiFi outputs) and from systems installed by those who are more likely to follow regular maintenance routines. So the base data is likely to be heavily skewed to the “better” performing systems.
I would be very keen to see calculated confidence intervals for the APVI data, but my guess is that they would be so wide as to make the estimate of “peak” output from solar systems very difficult.
As the AEMO data is quite similar it is likely that they follow a similar procedure as the APVI data; i.e. relying on a very small number of systems actually reporting output.
I therefore suspect that picking peak usage by relying on this data is very dangerous. But that’s not really the point. There are all sorts of methods for controlling the actual peak demand on any one day, uncontrolled solar input being just one, but there are also many methods which would provide some control over the peak.
The real issue for consumers, industrial and commercial, is the price and reliability. From your graphs it is clear that the behaviour of the small scale solar units causes a narrowing of the peak demand on conventional generators. To maintain reliability we still need conventional generators capable of covering the actual peak (in the event that solar output is low). So we have not saved any capital but have pushed down the operating hours over which this capital can be recovered and so sending up the price. We can see that in the peak pricing as the solar systems start to fade.
So small scale solar system are not really helping the peak demand issue, they may in fact be making it worse. Irrespective of whether they actually reduce the peak or not.
On the possibility of west facing solar panels – it is very interesting to see the additional output from single and dual tracking solar systems. But all of this, including batteries, just keeps making solar a more expensive, rather than a more competitive power supply option.
Michael, the diseconomy you have pointed out of running base-load power stations over a narrow peak period underlines the unsuitability of coal or nuclear baseload power as a backup or complement to renewables. What we need is not baseload but dispatchable generators, like gas, hydro or solar thermal with heat storage. From this perspective it is extraordinary that the solar thermal option is routinely ignored in discussions like this and, of course, by government..
Although an enthusiast for PV solar, I can’t see the point of installing large PV arrays over flat, agriculturally useful land. There are very large areas of commercial shopping centre, warehouse and factory roofs as well as carparks and walkways near existing powerline infrastructure where PV panelled roofs would provide shade and power at zero land cost.
All investment in land-covering solar should be focussed on solar thermal with energy storage.
Hi Paul,
the spot market electricity price in Qld seems to be spiking more and more in late Jan/early Feb. I can’t recall ever seeing such extensive spikes. What is going on ? and what can we expect this to do for the next round of power price rises from the retailers ?
Allan
These analyses are really appreciated
They provide a clear and concise explanation of what has occurred.
What is missing (and this is not your fault) is why the generators are playing games and spike the prices to MPC and not offer prices more in keeping with their costs. There have been a number of attempts to prevent the generators using their market power (MEU – economic withholding, AER – ramp rates, SA gov’t – bidding in good faith) but they still do it. I am aware that generators will seek to maximize their profits by whatever means are available, but if there were increased competition then we would see the market more reflective of marginal cost bidding rather than gaming the rules
Thanks for your comment, I appreciate the feedback.
You’re correct that deriving explanations for the underlying bid structures / strategies of generators are outside the scope of these quick event reviews. But obviously they are important and of great interest.
As any number of commentators have observed, the NEM market design effectively requires the possibility of occasional “contingent market power” where generators are able to have offers priced very significantly above short run marginal costs dispatched. But the issue is how frequent these possibilities ought to be in a well-functioning market? Impacting on this will be basic supply-demand balance, industry structure, and the extent of financial hedging (both explicit via contracts and implicit via vertical integration). I think a general difficulty with rules seeking to govern bidding behaviour may be that the real issue is not behavioural but structural – eg supply side concentration.
It’s probably also the case that as we have seen increased vertical integration, the interests of retailers and generators in the prevailing level of and volatility in spot prices have converged.
All that said, an overarching control is meant to be that as the level of spot & contract prices reaches or exceeds long run marginal costs of supply (LRMC), the real or implied threat of new entry increases. For a long time in the NEM we became used to seeing prices arguably well below LRMC – entirely appropriate for an oversupplied market – so with generation retirements and increasing fuel costs, as well as strong load growth in Queensland, the recent increases in price level and volatility are not unexpected directionally – I’m not going to offer a judgement here on their level. But if they prove to be sustained, I’d expect to see even more discussion about new entry – both large and small scale.
Can you please explain how the reserves in various LOR1 notices for NSW have been calculated and how that panned out in reality. 57241, 57238, 57234, 57232, 57222, 57219, 57217 , 57216 (on 4th Feb)
Also how did notice 57218 on reference temperatures materialize?
Why was NSW coal generation so low? It can go up to 9,500 MW but did only 8,500 MW on 6/2/2017 at the critical time. Was there a plant, transformer or transmission problem? Heat-related?
Why did imports from Victoria suddenly drop between 13:00 and 18:30? A spike of almost 2,000 MW at 19:30?
Murraylink is exporting at twice the limit the constraint is set to. Doesn’t look like any spare generation in SA at the moment. LOR2 already declared, waiting for the LOR3. Perhaps we will dodge it?
We didn’t dodge the LOR3. Up to 100MW rotational load shedding implemented. TIPS A1 unit was offline, I assume for maintenance. That would have saved us but shows the supply situation is very tight on hot days. TIPSB1 has a blip and went back to 150MW for awhile as well, it hasn’t been outputting it’s nameplate (200MW) for awhile.
Pelican Point 50% mothballed still. That would have saved us this time. Fortunately the load shedding was less than an hour but it is disappointing we have available generation offline due to essentially a market failure in my view. (The most efficient gas generator unable to run economically)
High efficiency of a combined cycle plant is only useful if the plant runs as a baseload. With high renewable penetration in SA, a combined cycle plant would have to ramp-up and ramp-down to accommodate carbon-free energy from renewables. Such ramping-up/down would not allow a CCGT plant to operate at its most optimum leading to higher operating cost for the owner of the plant. No wonder, a highly efficient CCGT plant is not despatched by the owner in a market with high renewable penetration.
A reciprocating gas engine power plant is ideally suited for numerous start-stop, ramp-up/down application without impacting the maintenance cost and such a plant can also achieve high efficiency (46~47%). Such a plant can be built in 100MW chunks and allow efficient dispatch of just 10% of its capacity due to its multi-unit configuration.
I’ll be happy to provide more information.
Sorry not to respond earlier to your comments, afraid I got busy digging deeper into the events of yesterday – shows how in the above post I focussed on Thursday’s outloook and completely missed the risks just around the corner. If I lived in Adelaide I suspect I would have realised that AEMO’s predispatch was well on the low side.
Should be more about Wednesday on WattClarity this afternoon, but your observations are right, Murraylink exceeding constraint limits especially.
I wonder if AEMO under-forecasting has anything to do with it, they don’t seem to take weekends/holidays into account. Maximum demands were quite low in SA during the last 2 summers largely by chance, no heatwaves culminated on full working days (Mon to Thu), but this year they have done so in both QLD and SA.
AEMO modelling of wind power is more subtle than stated in the article. For one purpose they use a nominal capacity factor, but for the Reliability Standard and the ESOO they use actual wind power data for the previous 6 years, adjusted to 2016/17 levels.
Given that the recall time on Pelican Point’s second unit was obviously a lot shorter than some earlier statements had indicated, I think a higher demand forecast from AEMO either on the day or a day or two ahead may have made a difference on Wednesday. It still would have come down to AEMO issuing a Direction early enough to get the unit online for the evening peak – as they did on Thursday.
I’d be very surprised if AEMO’s forecasting process ignores weekday / weekend differences but the absence of mid week extreme heat in the last two summers would have made calibrating and tuning their models more challenging.
Thanks for pointing out the actual wind-trace methodology AEMO use for ESOO and reliability standard analysis. For the purposes of this quick writeup I felt the simple capacity factor “benchmark” was an appropriate point of comparison, but yes they do adopt a more sophisticated approach in full market modelling.
Allan, you are my new Font of Knowledge on the grid. Your articles should make it into mainstream media. At least they should be on the daily clipboards of our pollies for digestion.
Some random related comments and questions….
Is AEMO running a corruptable system? Have you seen evidence of the generators being pulled back / taken off line in the face of an impending tight spot? Just to play for the $14k? Do AEMO have people who check the bonafides of any off-lines? Just wondering.
I’m staggered at the increase in demand from hot weather. There must be opportunities for controlling this excessive demand as a cost effective way to avoid load shedding.
Is it true that NSW were about 2GW short of power on the 10 Feb 17 during the worst of their power peak? And that Tomago smelter pulled consumption to avoid cutting power to mums and dads? Did NSW still have to swallow a $14k spot price during this “almost outage”? In Adelaide it was reported as one big ‘back patting exercise’ after people reduced power consumption and averted load shedding, in response to a call from their Premier.
Your article refers to ROLLING load shedding being ordered, but the news casts in Victoria suggested those without power had no power for that whole period.
Were those reports true?
If so why was the withdrawal of power not shared sequentially over various consumer blocks as the term ROLLING implies.?
This tweet from SA Power Networks certainly describes the load shedding as rotational ie rolling. AEMO and the relevant authorities in SA (and in every other jurisdiction) would have a pre-approved list of loads / areas to be shed sequentially in this type of event. The duration of such shedding in any particular area would normally be about 45-60 minutes, then power would be restored and the next scheduled area disconnected, and so on. As this event lasted less than an hour, I imagine that only the first load block on the list was affected. AEMO’s media release above states that load reduction applied between 19:03 and 19:40.
As always there will be blame games, South Australia’s problems are home grown with renewable energy and poorly performing gas markets have taken their toll there.
But all in all, it is just too darn hot and AEMO have a tough 3 days ahead. Good luck.
Why is it a “remarkable outcome” that we are on track to break the 2011 record. How many more people are there now in NSW than in 2011? Probably half a million at least. Yes there is the penetration of Solar PV, but there are also many small and subtle power hungry developments , such as the continuing increase in average house size, the increase in size and number of appliances, huge televisions, coffee machines , computer equipment etc. Its surprising it hasn’t happened sooner.
Is the Tomago aluminium smelter back on again? With how much load? Is or was this included in the scheduled demand? A smelter should be the last to be included in load shedding. Why not disconnect the air conditioned office blocks of those decision makers who are responsible for our energy mess?
Hey Paul – thanks for the update. In regards to Solar data – am sure you though about this – do the timezones of the APVI data line up with NEM-time? It also seems a bit strange the small scale capacity seems to line come online an hour or so (ish) earlier..
100% agree Paul! The sooner the industry infighting stops, the sooner solutions will be delivered. If it doesn’t, political solutions will override correct, ecominc and efficient outcomes.
Hi Paul
really big guess here, but could the more rapid decline in small scale solar be because its predominantly roof top mounting which may be more prone to afternoon shading from neighboring homes and trees, which large scale solar may not?
I think that would certainly be a contributing factor.
Another thing to take into account is the three major large scale solar PV sites in NSW are spread out geographically east to west (Moree, Nyngan and Broken Hill) compared to small scale rooftop PV (heavily concentrated on the east coast) so generation is more evenly distributed. This also helps explain the later start time for generation for large scale PV as Moree is the furthest east yet still 400km from the coast. The Moree solar power station is also single axis tracking which has a smoother output curve over the day than fixed position installations.
with the temperatures Qld has been experiencing, I can’t see how Energex’s Peak Smart initiative would be too popular. It seems to me that if it is hot then the last thing people would want is for the network owner to throttle back their airconditioners.
The little blip at the top of Qld’s demand curve seems to indicate some load control but it is so small now relative to the MD that I wonder whether it is worthwhile messing around with load control devices at the domestic level. I wonder how much hot water load is actually switched off during these hot afternoons. Not much I would have thought.
And while we are on the subject of hot water, how many homes in Qld have solar PV and electric hot water ? Such craziness astounds me.
Most of the many large-scale solar farms soon to move into construction across the NEM are proceeding with horizontal single-axis tracking, and will be generating large volumes of power facing west during the afternoon peaks (just as the Moree Solar Farm has been collecting lately with NEXTrackers, last Thursday being a prime example). It’ll be interesting to see the extent to which just 1GW+ of these projects impact the shape of the duck curve.
Solar data on the Aneroid Energy website shows that the Moree Solar Farm was still producing at more than 80% of capacity until shortly after 6 pm (I presume this is local time, not NEM time). So all the new solar farms with tracking should be producing well right through the summer peak hour. However in the winter its decline in output commenced at 4 pm.
So systematic market failure where the market does not provide incentives for efficient large generators to run when they are needed. Instead of trying to come up with an extremely elaborate market mechanism to balance all of these competing interests, we should nationalise them and take the profit motive out, so that we have the capacity when we need it. Natural monopolies like all utilities markets, can’t be engineered into a competitive market.
The fact that this is also happening in NSW and QLD shows it is an issue with the market structure and likely also rational self interested actors gaming the system.
I find it odd that small inefficient generators in SA can continually run making profit but large efficient generators can’t.
If we nationalised, we would just be facing government failure instead of market failure. Why do you assume a nationalised system would work better given governments’ track records? Why do you believe that government’s don’t face their own financial constraints that will influence how much they are prepared to invest in the system? How do small scale alternatives to grid electricity like solar and battery fit in? should we nationalise these too? If not, why not? And given solar is already competitive and some argue that in SA at least, batteries are too, then perhaps we are no longer in the situation of a “natural monopoly”.
I don’t assume that you won’t have issues with government failure, nor that they won’t have their own constraints. More so, they would have LESS constraints than an ordinary business. Solar and battery are independent, like running a generator for your own home, you’re trying to increase the scope there by likening a 5KW generator to a 50,000KW generator. The market is still a natural monopoly, particularly with distribution.
Every issue you’ve brought up is a problem under a market system as well, except under a market system you’ve also got the additional issues of regulatory capture and gaming the system.
“Solar and battery are independent, like running a generator for your own home..”
True but that’s a rare situation wrt the communal grid whereby there are lots of rooftop solar generators with no storage capacity bludging off the insurance of the grid and paying no insurance premiums. Quite the contrary as many are being actively subsidised to bludge off the grid like mine is.
Putting aside the public / private ownership question, a different issue worth considering is the generation technology mix in South Australia (and elsewhere in the NEM). While large scale combined cycle gas plants are thermally efficient, they are not nearly as flexible as some other forms of generation particularly with rising gas prices and increased amounts of variable renewable generation pushing them towards peaking / reserve operation. Here I mean “flexible” in an economic sense as much as an operational one (however CCGT plant is generally less flexible in terms of startup time, minimum run time etc than open cycle turbines).
We can see this in the decision by Engie to reduce the quantity of gas contracted for the station. Contracting gas supply and transportation for the full plant capacity would involve committing to higher fixed costs, with an uncertain likelihood of recovering these costs from a limited number of opportunities to run economically at full capacity (due to high running costs on gas which has become a much more expensive fuel than it was when Pelican Point was built, versus the very low marginal cost of renewables).
Regardless of who owns Pelican Point this underlying economic shift creates a difficulty in economically optimising the operation of a station like this and maximising its value to the system. For true peaking / reserve operation in an environment of high gas prices, the best form of “gas-fired” plant might actually be open cycle turbines running on liquid fuel – the lower fixed costs of both the generation plant and fuel supply outweighing the higher costs of relatively infrequent running. Hence “small inefficient generators” running “profitably” and Pelican Point at only half capacity!
Some people are pointing fingers at the wind power lull, but it is the wind power peaks that have killed the concept of “baseload” in South Australia. Generators like Pelican Point and Osborne operate in baseload mode, maintaining steady output, between infrequent step changes. In recent years baseload generators have been often kicked off the system by wind power surges, so its not surprising that some have closed or been mothballed. Ordering loss making generators to be on standby does not seem like a complete solution.
A major long-term problem for SA is keeping Torrens power station in operation, it is the workhorse load-follower, surely the most important electricity asset in the state.
Thankyou for an excellent and easy to read analysis of what happened with lack of supply from Pelican Point.
So, at the end of the day, if a generator can’t guarantee that it will make its money back on providing electricity to the grid, it won’t, unless they are directed to by the AEMO in which case any excess charges will be covered.
Sounds fair to me – they are not operating as charities, or even as public services.
Also sounds like a failure of the AEMO to predict usage.
How much of the planning and foundations of the NEM took the impact of secondary suppliers’ pricing into account when the national model was developed? Are there just too many links in the profit chain?
“Under the National Electricity Market (NEM) rules, generators can not bid plant into the market if supply can’t be guaranteed”
What a laugh that is with renewables and until the AEMO has the power to restrict any tenderer of electrons to the communal grid to the maximum they can guarantee 24/7, all year round this Govt sponsored form of domestic dumping will continue without a level playing field that consumers require. If that mandate was implemented, then renewables would either have to partner with thermals to lift their average tender rates and/or invest in storage, whereas at present they’re freeloading on thermals as insurers and not paying them their just premiums. What a crock that is and the ACCC would normally have the power to jump on such dumping practices should it occur.
A simple analogy would suffice if say a Labor Govt wanted to privatise the Fire Brigade and proudly announced the CFMEU tender was the lowest priced and has been accepted. Just a wee problem in the fine print folks. There won’t be any fireys on duty when the sun goes down and/or there’s no wind or wind over 90km/hr. Yeah right!
So you’re saying small gas generators that have low start up costs by higher operating prices should not be allowed to turn on in peak periods and off in low periods? You’re saying when gas prices go up and they need to shut down, they should be forced to stay on?
I’m just drawing a comparison with your standard for wind and applying it to gas as well. Small generators have a benefit over large generators in some regards, so the large ones are “insuring” the small ones. This isn’t unique to wind.
Fluctuations happen in resources and these have flow on effects. You can debate over their size, but that’s about it.
“So you’re saying small gas generators that have low start up costs by higher operating prices should not be allowed to turn on in peak periods and off in low periods?”
No I’m not saying that at all because clearly those gas peaking plants, short of a breakdown, can produce up to their maximum output anytime demand and pricing calls them forth. OTOH they couldn’t match a base load cogen gas plant like Pelican Point that would have long term gas contracts for the whole of their plant and it would be up and running and rather than 6 hours lead time to fire up the idle generator that could easily ramp up output with both gas generators already operating. That’s because they’d either be insuring fickle wind and solar and getting their just insurance premiums, and/or the unreliables would be paying for battery, hydro or molten salt storage in order to compete on a level playing field.
We need high priced peaking gas plants too but what we don’t want is fickle generators dumping in any part of this grid equation.
“I’m just drawing a comparison with your standard for wind and applying it to gas as well.”
Actually on that point there’s no reason why it shouldn’t also apply to large base load plants like Pelican Point too. ie under their 24/7 all year round maximum they wish to guarantee to the AEMO they’d have to in turn partner with fast start gas peaking plants to insure them. Nevertheless creating a level playing field would largely impact the unreliables who are at the mercy of the elements and rightly so if we as consumers aren’t to be.
Err guys, try ruling an average 30% average line through this graph and then imagine the wind turbine industry not being able to tender any power above the line and having to make up any power below the line- http://anero.id/energy/wind-energy/2017/february
Get the picture?
Not to worry folks the CFMEU Fire Brigade quote includes an ironclad guarantee only union fireys will be employed, whereas the other nasty capitalist tenderers won’t guarantee that 😉
Roger, I think you make some good points. Its a pity you use such a belligerent style — it detracts from your essential points. That we pay the same $ per MWh for dispatchable energy as we do for semi-scheduled and non-scheduled energy seems to not be a proper recognition of the value of the energy offered. Because of the variability of both sources it increases the daily range and variability that the load following plant has to account for, and its more expensive for that plant to operate in that environment. As we see here in Allan’s excellent analysis of Pelican point. It seems inappropriate to me that the generators that contribute to the creation spikes by their variability, also enjoy the high prices. Because wind and solar gets “dispatched” first, and gets paid the spot price price, there is no incentive for them to invest in storage of some sort. If there was some discount from the spot price to reflect their non-dispatchable nature it would encourage storage.
Mate, I get belligerent when the lights go out because too many technical experts have been too nice for too long, while belligerent watermelons have cowed them into shutting up and going along for the ride as they foist their theories and unreliables on us all as consumers. To put it bluntly to these leftys who turned their red coats inside out to wear the green on the outside when the Berlin wall fell and their ideology was in tatters, all electrons are equal, but some are more equal than others. Ipso facto I gave you the only rational, sensible way to make them all equal, whatever the world’s omniscient guru thinks should be the same tax price for all to price their input CO2, in order to get the world’s atmospheric CO2 back to the level the omniscient guru knows it really should be. I am not that guru but I sure as Hell know unequal electrons when I see them and the time for pleasantries is over.
The NEM is not a creature of the environmental movement, but of economists. The environmental movement has been advocating markedly different strategic policies in this sector for more than 40 years. Governments have preferred to take their advice from economists and the conservative press and have failed to predict or to plan for the train wreck that is now coming upon us at full steam.
A coherent policy has been made difficult if not impossible by the fragmentation forced upon the sector by vocal advocates of competitive free markets in the conservative press and conservative think tanks.
Question for you: Who is able to effect fundamental reform of the National Electricity Market? Who is in charge of the regime? It certainly ain’t the environmental movement.
While it’s true that within any given half hourly settlement period all generators earn the same spot price for their output (before adjustment for locational network loss factors), the average price that different generators earn across a given day, week, etc can vary widely depending on how each generator’s output moves as the spot price changes.
Taking just the 8 hour period midday – 8pm in SA on the day of load shedding, wind generators’ output fell steadily through the afternoon, while thermal generation ramped up to fill the gap and meet (not quite, unfortunately) the increasing demand (see https://wattclarity.com.au//2017/02/initial-analysis-sa-load-shedding-wed-8-feb-2017/ for a chart showing this). Prices in this period rose from a few hundred dollars per MWh up to the Market Price Cap of $14,000/MWh whilst load was being shed.
The demand-weighted average spot price over this period was $4,642/MWh. Wind generators as a group, with much more of their output occurring before the price spiked and much less during the period of extreme prices, earned an average price of $1,584/MWh, a 66% discount, while thermal generators as a group earned $5,206/MWh, a 12% premium. Within that thermal group, generation that ramped up specifically for the extreme price period would have earned a much larger premium.
This is an extreme example, but it clearly shows that the spot market – as intended – does send price signals rewarding dispatchable generation able to contribute more during periods of tight supply-demand balance. Previously on WattClarity, Paul has posted some interesting articles including this one commenting on the increasing discount to average spot prices being earned by windfarms in South Australia.
Aside from the energy price, there are other mechanisms – principally the markets for Frequency Control Ancillary Services – that can reward / penalise generators for ability or lack thereof to contribute to stabilising system frequency etc, again related to the “controllability” of those generators’ output. This is an active area of debate and development in the NEM, as the need for and nature of the services required changes within a transitioning system.
Thanks for taking the time to explain that not so obvious effect of averaging, and that it does create an incentive for dispatchable generation, and therefore for storage. Thinking about the main thrust of your article, the day ahead bidding requirement makes it quite awkward for generators that are market exposed for their fuel cost. I suppose this would be similar for pumped hydro, because their fuel, the power to fill their upper reservoir is also market exposed. For a large pumped hydro with a daily charge/recharge cycle, the day ahead bidding would require them to bid to sell energy they haven’t secured yet. To charge the upper tank they would be bidding as dispatchable load, and if their bid mean they did not get dispatched as a load, they then would not be able available to generate. That seems to place them in the same predicament as Engie, having uncertainty at both buying and selling. On the other hand, smaller unscheduled generators don’t have to commit ahead of time, and can respond the actual market price at the time, a much safer alternative, than having to commit in advance. The playing field seems quite in need of some levelling.
G’day Paul, hoping you can help resolve a bit of confusion. The first graph is headed ‘Trended Aggregate Solar Output (small scale and large scale), while the annotation is “the aggregate numbers from the AEMO (historical small-scale solar data from AEMO…) (my emphasis). If the latter is *only* small-scale solar, why are the numbers almost identical to what would be obtained by simply reading off the graph? Shouldn’t they be appreciably smaller, if they omit the large-scale input?
I have a 3.8kW PV array on frames to reorient them N on my WNW roof slope. The reason for this is to locally power the washing machine in the morning on days when the solar dryer (clothesline) will then dry the wash that afternoon. By such load management our usually 2-person household pays nothing for electricity, only interest on investment. Energy exported at 25c/kWh for the first hour and 8c/kWhr thereafter pays not only for power drawn (at 25c/kWhr) from the grid at night and cloudy weather but pays for the line rental as well. The solar hot water helps considerably of course.
It does not matter that the peak output does not coincide with peak airconditioning demand because we don’t have an air conditioner. Our Hardiplank/brick/Colorbond house is well insulated and we open all the windows at night (in summer) and close them and curtains during the day. But even if we did have an air conditioner we, like everybody else, could effectively match our demand to the PV output peak using a timer on the air conditioner. With a sensible pricing scheme utilities could make it worthwhile for everybody to run their air conditioner during the N facing PV peak and shut it off as demand rose for the evening cooking demand peak. Heat and “coolth” represent storable energy and if the house is cool when people arrive home from work it can stay cool through the evening.
Storage batteries and pumped hydro would be useful for other load mismatches but we don’t need them for hot days. Thermal mass does the trick.
-Flexible generation already gets a price premium and wind already gets a price discount.
-“AEMO supply reliability leads to interconnection” what?
-AEMO’s capacity contribution assessments had nothing to do with supply shortfalls in SA
This guy is just trying to drum up business by chucking all the keywords in one place.
Here’s one fix: Bipartisan policy then get out of the way.
What we need is proper planning for load shedding. Cutting supplies to an aluminium smelter should be a non-no. Strict measures to reduce demand and to stop demand growth, e.g. passive solar houses instead of energy hungry residential towers. There is also need to cap power consumption of households at the meter during peak periods.
High pool price spikes suggest system has reached its limits and/or there is market manipulation.
As for the 5 min bid stacks imagine we had auctioning of petrol at our filling stations every 5 minutes!
I agree with some of your points; however there is now a need to move towards a capacity market to ensure that capacity is always available.
This could be done by AEMO invoking its Reserve Trader Powers but they seem to have been missing in action on this to date.
I do not agree with raising VoLL.
if anything VoLL should be reduced. it adds a huge risk premium to the market and it has yet to achieve its stated aim of being a signal for new generation.
It should be abandoned in favor of a capacity market which values capacity, not just energy which may or may not be there when you need it.
“It should be abandoned in favor of a capacity market which values capacity, not just energy which may or may not be there when you need it.”
There are 2 parts to that overall emergent problem. The traditional but growing one of peak demand whereby aircon use is demanding 20% more generation and distribution capacity for a few weeks of the year in summer and that’s particularly correlated with Adel and Melb weather. In that respect it makes power bill sense to price accordingly i.e. consumers with aircons can choose lower prices if they sign on to having a smart meter that can demand manage their aircon compressor for 20% of the time or stick with full user pays. The other biggie now for consumers is getting reliable electrons anytime, summer and winter, regardless of smart metering in summer, but that’s where my level playing field policy comes in. Under that regime the AEMO will know exactly what amount of reliable electrons are available in total and so will all the market players with future planning and investment in mind. Not so with the current disastrous regime.
“There should be no need to throw out the energy-only market design, but there is opportunity for market-saving innovation, perhaps by introducing a price premium for output from scheduled generation, paid for by a discount on the output from semi-scheduled generators (mostly, wind) that only ‘turn up when they can’.”
The notion that some omniscient public servant can work out the optimal price premium and discount between reliable and unreliable electrons tendered to the grid is to believe I’m the second coming, which sounds fine to me if I’m paid according to my new station in life. Much wiser for mere mortals to set the level playing field and let the multitude of players work it out between themselves. Ipso facto that means EVERY tenderer of electrons to the communal grid can only tender up to a maximum they can guarantee 24/7 all year round and let the market work it out.
I don’t know what sort of cross transfers would eventuate with the only obvious level playing field outlined, but clearly fickle wind and solar would have to pay reliable thermal partners their just insurance premia in order to up their average tender quantities (wind had an average of 29% of installed output only in 2016 but what about all their dips below that line?) and/or invest in storage to achieve it. Let them work it out over time between themselves, or else vertically integrate their generation mix. All I want is reliable, despatchable electrons, with appropriate compensation if I don’t get them and judging by the political ruckus at present, I’m not alone.
The free-riding is easily demonstrated looking at wind output as a proportion of installed maximum output- http://anero.id/energy/wind-energy/2017/february
and astute observers will notice the below zero graph lines and perhaps be interested in switching the graph to MW top right.
In 2016 the average output in Australia was 29% which marries well with an oft quoted figure around 30% of installed output for wind energy in general. Simply draw a horizontal line at 30% and look at all the dips below that line and their negatives that wind generators would have to pay a despatchable thermal insurer to cover them (fast start peaking gas perhaps?)or else install storage to be able to use some of those peaks. Notice also that to the extent they rely on thermal to backstop/lift their tender levels that reduces thermal 24/7 all year round tender capacity by the same amount, but the bonus is the AEMO and all players then know the absolute level of reliable electrons that are possible and can plan investments against any rising overall demand with confidence. Not a hope in Hell of private enterprise doing that now with the wild roller coaster they’re on, but if you think public servants can do it better with our taxes, then you’re part of the problem. The problem being you aren’t looking at the big picture, but are instead stuck within your blinkered speciality like so many- https://wattsupwiththat.com/2017/02/20/climate-science-and-climatology-specialization-and-generalization-forest-and-trees/
Unfortunately a national electricity grid knows no such bounds.
Yeah no. The spot price you are referring to isn’t the actual price experienced by the customer. 90% is contracted at a significantly lower rate, more like 5c/kwh rather than the low 8c given to the ‘poor’ home solar people (excluding transmission and retail of course). Perhaps we should estimate the massive waste of taxpayer money incurred when home solar owners were reimbursed 60c/kwh gross (irrespective of whether they used that power or not) when the average price for generated power was 3c/kwh over the 3 or so year period the scheme was in place. Which one do you think actually hurt your wallet more?
Thanks for your comment. The analysis in this article relates to the wholesale (spot) electricity market. All retailers pay the spot price to AEMO, and conversely all generators receive the spot price from AEMO. Most market participants also hedge their exposure to spot pricing in order to firm up revenue (for generators) or cost (for retailers). Hedging instruments are settled against the spot price and hence their price is largely determined by the market’s expectation of spot pricing. The current futures pricing for NSW baseload Q1 2017 is about $120/MWh or 12c/kWh.
Residential consumers pay a fixed rate to a retailer. Of course, this fixed rate incorporates the costs to the retailer as explained above.
As for the solar bonus scheme (which has been discontinued), its purpose was to encourage the uptake of solar PV in the early stages of adoption in the Australian market and it has certainly achieved that goal. Yes, 60 c/kWh was a high feed-in tariff, but remember that solar panels are a 20 year asset. We will be reaping the rewards for a long time yet.
The owners of sparingly-used assets such as the Port Lincoln and Colongra power stations must be wondering why when those assets were required, they failed and were unable to bid into a a high-priced market. Inadequate testing prior to forecast demand peaks ? Perhaps the market operator should require twice yearly testing of their systems before the demand peaks in later summer and mid winter.
Paul
Wattclarity is an amazing source of information
While I might not agree with all of the conclusions reached, the process is clear and the information provided is accurate so we can use it to help us better understand the highly complex NEM
Please keep up the good work
Thanks for keeping it going
David
I realise this is largely a rhetorical question, but I’m going to answer it anyway. Presuming you mean ‘x works in the industry, therefore rather than x being an expert whose opinion should be respected, x is a vested interest who should be ignored’, I reckon I’ve been noticing this sort of sentiment in these circles for nigh on ten years.
Hi,
I note Jamey’s comment (below – from todays topic) that those on the $0.44c feed in tariff are still eligible for a further retailer feed in. Is this in fact the case. If so, I would really appreciate some info re the application process.
Q2) What do you think, about not being eligible for the 44c Feed in Tariff?
Jamey: It seems obscenely unfair. Those who were lucky enough to sign up before the deadline for the government tariff are still eligible for a retailer feed in meaning that most are receiving more than 50c per kW,…
In my case, I am one of the lucky 400,000 in QLD on the 44c FIT and I do receive a bit more than that (50c or 52c in total, can’t remember specifically) with a particular retailer whose name I won’t mention. I believe pretty much all of them do it, so shop around.
Apparently it wasn’t only the media getting skittish Paul – AEMO directed additional Pelican Point capacity into the market that afternoon despite no sign of forecast LOR2 on the horizon. There was an actual LOR1 declared but reserves were still over 450 MW before the direction.
Re the 44c FIT. The people who received this had a higher exposure to finance costs of construction, a higher exposure to risk (knowledge of the life of PV installed was lower confidence) and made commitments in a time where government *had* to encourage activity to build a market.
If you compare people who squeezed in ‘under the wire’ at the end of the 44c FIT to somebody who applies shortly after its cancelled, yes, it “looks” unfair. I think the government could have done a sliding scale tail-off, or a number of other things.
But this is really no different to somebody who gets first round shares, before float, and somebody who buys at market rate post float: there are winners and losers, there advantages and disadvantages, there are risks and rewards.
I don’t have solar. I value PV and wind immensely, and I am delighted that part of my bill goes to a social offset and equalization to people who have invested in PV and yes, that includes the 44c early adopters. I’m glad 44c ended: it wasn’t sustainable. I am sad 11c has been dropped to 6c and lower, I think its disincents people, from long term capital investment in PV to all our benefit.
But this really isn’t “fair” and “unfair” -it just feels like it is. If you want “fair” and “unfair” then lets talk about how people get to invest in coal and gas, and are not exposed to the longterm remediation costs because of systematic rorting of the cleanup cost.
Perhaps the emergency capacity for the Heywood inter-connector should be up-rated. It’s already had last Friday’s stress test and passed. Currently the emergency level is specified on AEMO’s website as 701 MW for summer, 717 MW for autumn and spring, and 762 MW for winter. Surely a more sophisticated set of temperature-controlled limits could be applied, as on the Vic-NSW inter-connector.
I have only just signed up to get you email which is very informative and interesting.
While i have read what you say you do it is hard to get my head around someone wanting to do this without some kind of angle – you know the old saying that there is not free lunch.
Would you like to tell me a bit more about why you do what you do, which as I said I am very interesting.
Cheers,
Ken in Adelaide.
Drop in load was probably due to “load relief” – as network voltages fall, so too does demand – its the common Np and Nq indices power engineers are familiar with
So far as the “missing” load is concerned I wonder if consumer appliances, specifically modern inverter type air-conditioners, may have tripped on account of a sudden drop in voltage and then re-started a few minutes later?
I don’t know how these systems respond to sudden voltage fluctuations but if the load did indeed go “missing” and it wasn’t due to load shedding or a large industrial load being tripped then there aren’t many other possibilities given we’re talking about 300MW or so.
Losing four operational units to one transformer fire does not seem like ‘single failure proof’ design.
If no distribution level load shedding occurred, then the shed system load may have been directly connected to the HV grid and shed either automatically or manually by AEMO?
Directly connected HV would normally be big industrial loads. Does SA have ~300MW of those? Reconnecting those loads could have been done gradually as additional generation synchronised and loaded.
The most significant reason probably is that the other generators both synchronous and non synchronous did not drop their skirts and run away. There was a lower percentage of non synchronous generation running than for the September event and the majority of the non synchronous generation was remote from the event and probably did not see any severe voltage or frequency fluctuations anyway. So it didn’t all trip as in the September event and the synchronous generation close to the event would be more forgiving.
The second is that the measurement is probably not correct – mismatched instantaneous readings from a transient event – and you would need to examine carefully the actual voltage and frequency fluctuations at various points in the network.
The indicated drop in power is not all that great at just over 12.5% of the load immediately prior to the event. There obviously was a significant drop in voltage – reports of lights dimming – so all resistive load will have immediately stopped pulling a fair bit of power.
Likely we would probably also be very surprised by the actual load that was shed and then reconnected quickly. Every UPS in the state would have flipped to battery (and all have probably had a service/new batteries since September had just identified all the flakey ones). Every capacitor in every electronic power supply gave up a bit of it’s spare store. Quite a few sensitive large loads probably did disconnect and then immediately reclose without any real impact on their connected plant. Some loads probably stayed off but have not made the press because they regularly do that during such events. The reasons that these did not become “press events” is that users could immediately go about restarting them so there was no great damage – it just becomes one of the regular unscheduled events that operators are trained to handle
So in summary there probably was a fairly significant drop in load but it did not become an “event” because the network stayed up – probably thanks to the greater percentage of online synchronous generation than September. And the remoteness of the non synchronous generation meant it was more isolated from the event.
The more you look at the September event the more you begin to wonder if a few nutters distributed though the management systems (AEMO included) were looking to the upcoming wind event as some sort of opportunity to push the barrow of wind capability. It begins to look as if they were almost trying to break some record for the percentage contribution of wind to the SA network – perhaps trying to get to 100% of SA power from wind and took their eye off their prime roles of actually keeping the power on. Even before the event everyone would have know that too much non synchronous generation was a risk – but everyone seemed to ignore that risk. And high winds are always a network risk even if they only bring down a few isolated conductors and tree limbs.
But in September those in the know seemed to deliberately ignore the risks and plodded on blindly with a commitment to loads of wind. This recent event shows the stark contrast in the outcomes. Although they didn’t have the choice this time as the output from the wind and solar resources was relatively low at the time.
A high voltage fault at a large generating station is one of the worst due to fault infeed availability. Hence, depending of the nature of the fault, strange protection system operations can happen. Losing multiple generators during this type of event is not unheard at all. A fault becoming a 3-phase fault with possibly delayed clearing is very extreme.
All I know about this event is that some piece of equipment probably failed explosively. It happens. With a close-in unbalanced 3-phase fault, generator neutral currents can cause their protection systems to operate. If the failed component impaired protection systems, resulting in a slow clearing of the fault, the event will become quite severe over a large part of the grid. Compressors in air conditioners and fridges will see a voltage loss and kick in restart delay timers/breakers for a period of time. Other voltage sensitive equipment and synchronous motors disconnect and also have a delayed restart. Back in the day, we used to call it “load shaken off” for the lack of a technical term, so you get the idea. Unplanned load cycling.
Power systems cannot be designed for clean, controlled protection for every type of event. I am unfamiliar with the SA protection requirements but in North America, for instance, most areas are designed for proper protection for either a 3-phase fault or a single phase fault with delayed clearing. Hence you can have events that exceed design capabilities but that have a very low probability. Perhaps this was indeed a 3-phase fault with delayed clearing.
Thanks everyone for the comments, some great insights there.
Some specific responses to people’s comments:
Malcolm – I’m not quite sure which piece of data you are referring to (a reference / link would be good), but the numbers look close to the per-circuit thermal ratings (MVA) on the two 275kV Heywood (Vic) to South East (SA) transmission lines. They might represent the highest levels AEMO would be prepared to schedule Heywood interconnector flows to in an emergency (and only for a very short time, << 15 minutes) but there may be other limitations and in any case would not leave the system in a secure operating state. In general the maximum secure transfer capability of the interconnection is quoted at 600-650 MW but frequently other constraints limit imports to considerably lower levels.
Ken – yes there’s a commercial angle to this but it’s primarily about trying to demystify the workings of the market, and you can see from the other commenters’ input that I learn plenty from doing it as well.
Andrew – that form of very short term load relief on frequency / voltage dips certainly exists and assists with riding through the immediate aftermath, but doesn’t persist for the 10-20 minutes we saw last Friday. I think what happened may be more along the lines suggested by Shaun, Michael & Tom.
Shaun – I think you, Michael, & Tom are probably on the right track – it will be interesting to read what AEMO says. Your comment about inverter-connected loads also made me reflect on the large amount of inverter-connected PV generation – had there been a large scale dropout of that generation last Friday then supply-demand balance would have been that much harder to sustain over the interconnector!
PaulD – I don’t believe there is that much large, direct connected load in SA available for contingencies like this (whereas in other mainland states smelters can play this role), and in any case I’m sure we’d have heard about it if (say) Olympic Dam had tripped.
Michael – Great comments, thank you. Relative to the September event, while there was certainly a lot less non-synchronous generation running on Friday, remember that in September it was the loss of nearly half that generation that led to the collapse. Last Friday it was well over half of the running synchronous stuff that fell off. If you take the pre-event September outputs by generation type, then adjust for the loss of 465 MW of wind, you end up quantities of thermal and large-scale non-synchronous output that are actually pretty close to the mix in SA last Friday once the three TIPS units and Pelican Point had tripped – around 400 MW of online thermal, 4-500 MW of non-synchronous wind, and nearly 1000 MW over the interconnectors. On my reading of the AEMO System Black report it wasn’t then a trip of the remaining online windfarms that precipitated collapse but tripping of the Heywood interconnector on loss-of-synchronism protection. I know what happens in the immediate aftermath of a big system disturbance is nowhere near as simple as my MW output comparison here but it does point to the load response last Friday (and clearly no loss-of-synchronism event) as being pretty important. And thanks for the insights into load behaviour after a big voltage dip – which would have been much closer to the load centre last Friday than it was in September.
All I’ll say on your thoughts on operation of the system on Sep 28 is that 1) there had already been plenty of previous occasions where wind output exceeded SA system demand so I really doubt any “nutters” were going for some sort of record and 2) I remain amazed that more than 5 months after the Black System event there has been almost no sign of any independent review of how the system was operated on the day.
Tom – thank you also, some of your comments address matters far outside my range of technical competence, but again the insights into behaviour of connected loads are really helpful in understanding why the system was able to ride through.
Below is a link to the file of AEMO line ratings I referred to. My reading of it (rows 11229-11246) is that there are 2 circuits between the South-East terminal station in SA and Heywood. The summer rating for each of 2 lines is 597 MW under normal conditions, and 701 MW for emergencies.The normal rating for the Heywood transformers is 3 x 370 MW (rows 3552-3569). Both of these are ~1100 MW and much higher than the publicised target rating of 680 MW.
In the file, the ratings seem to bear little relationship to the underlying physics or appearance of the line. In some State ratings are based on seasons, while in Victoria it is temperature. Some lines with few conductor cables seem to have disproportionately high ratings (eg Ballarat-Bendigo 347 MW at 20 degrees). The ratings seem to have a stronger roots in history than physics, and managed with excessive conservatism. I suspect that the best investment in the NEM would be to re-rate the inter-connectors based on sound physics, with dynamic ratings calculated from air temperature, wind speed, and transformer temperature. Re-rating won’t necessarily be cost-free, because it may require stress tests with helicopters flying the length of the line using thermal imaging cameras.
Allan, the numbers I have got for the September event has Wind providing 45% and Gas 10% of SA load prior to the event. For this event you have Wind only providing 18% and while Gas is at 42% prior to the event – quite a different generation mix. Solar at 13%/18% (last time vs this time) is pretty much the same. Imports at 529MW/613MW (last time vs this time) did give more headroom this time for imports.
I don’t think that September can be split into two separate events such as failure of wind and then separation from the NEM as separate independent events.
Michael – yep I accept that it’s artificial to split the September event into wind failure followed by Heywood trip; however the 456 MW of wind lost did occur in two main tranches – about 200 MW triggered by a voltage dip seven seconds before the interconnector opened, then another 260-odd MW less than a second beforehand, after a sixth major voltage dip. AEMO’s comment was “Voltage collapse began after clearance of the sixth voltage disturbance and sustained generation reduction of 260 MW associated with three wind farms (Groups B and C). This confirms system voltages were stable and would have remained stable if angular instability had been avoided” – see p39 of the most recent System Black report. So I think the event could be at least be split pre- and post- that 6th voltage dip (which was due to another transmission fault – see Table 6 / p26).
As the same report makes clear, the windfarm runback / disconnection was due to programmed control system actions following the number of ride-throughs exceeding pre-set limits, rather than failure to ride through the voltage dips themselves. I think that may be important in thinking about how that generation handled the primary faults on that day.
Your percentages for gas and solar generation prior the black system don’t look right at all, see p18 / Fig 1 showing 330 MW of thermal (~18%) and p24 where solar PV is estimated at 50MW (<3%).
Finally and most importantly, gas may have been 42% prior to last week's event but with more than half that tripping off, the percentage available to then respond was well under 20% of pre-event system demand. I still think that if demand had not fallen substantially, the outcome last Friday may have been quite different.
Thanks Graham for a good and concise explanation of the issues.
The hard part is to get action now so that the large gas users don’t disappear. Then again, if the large gas users do disappear, so does much of the domestic gas problem. There would be lots of spare gas and lots of spare electricity too! But if that occurs, so do future investment and jobs disappear and that increases the current account deficit and narrows the export (or import replacement) base,
“Thanks Graham for a good and concise explanation of the issues”
With all due respect Graham is mainly pointing out the symptoms of the patient’s fatal disease- http://anero.id/energy/wind-energy/2017/february
I would have thought somewhere along the line someone would have called out the emperor has no clothes on. That’s because it’s a fundamental axiom of engineering that you can’t build a reliable system from unreliable components and that’s why there are lemon laws around for simpler cars. A second year apprentice mechanic would understand it. No matter the emperor knows best about disproving that fundamental axiom with no countermanding precedent existing in history so here we all are with everyone looking every which way or donning dark glasses. The emperor is starkers gentle folk!
@Allan,
It looks to me like they have fitted UVLS (note: not UFLS) to the big industrial loads since the September blackout. It seems to have saved the system.
You state :-
‘SA Power Networks seeking an explanation – the latter has confirmed that there was no distribution system level general load shedding’
Perhaps the next question for SAPN is, “Did the total distribution demand vary over the timespan of the incident and by how much?’
Do we have to postulate that the solar PV stayed attached and the air-conditioners, fridges and freezers tripped because of a voltage dip? I doubt you would notice a ten minute break in air-conditioner operation; it would just look like the thermostat operating. However, …
… UFLS and UVLS:
The system frequency in SA would not vary much during the incident because it was attached to the rest of the NEM grid. It is this fact that makes frequency response in SA ineffective with the Heywood interconnector operating.
I know nothing about UVLS systems in SA but maybe large loads have them fitted? This looks relevant :-
‘Kill switch’ for mines to save electricity grid – The Australian http://www.theaustralian.com.au/national…for…/9f99512ea64c98b7ef6975f49e050235
4 Oct 2016 – BHP Billiton’s Olympic Dam mine in South Australia. … “The under-voltage load shedding scheme … has been installed as one of the interim …
[PDF]SA Events – update report – Australian Energy Market Operator http://www.aemo.com.au/-/media/…/AEMO_19-October-2016_SA-UPDATE-REPORT.pdf
19 Oct 2016 – Report – Black System Event in South Australia on 28 September 2016 and …… Availability of under-voltage load shedding at various locations, …
‘the grid manager has been directed to trip a ‘kill switch’ …
Looks like it works!
https://stopthesethings.com/2016/10/04/south-australias-wind-power-debacle-killing-miners-mineral-processors-scarce-jobs/comment-page-1/
Alas. The latest ‘plan’ from its vapid Premier, Jay Weatherill is aimed at keeping the lights on in Adelaide (where the voters are), while cutting power to its miners in the North of the State.
In order to prevent the grid from collapsing every time there’s a total and totally unpredictable collapse in wind power output (see above and our posts here and here), the grid manager has been directed to trip a ‘kill switch’, that will cut power to SA’s two biggest mining operations: BHP Billiton’s Olympic Dam and OZ Mineral’s Prominent Hill.
“The under-voltage load shedding scheme … has been installed as one of the interim measures that will eliminate the risk of voltage collapse spreading throughout the transmission system,” the report said. “This risk if not addressed had the potential to cause widespread power outages under certain network conditions.”
Well with today’s Advertiser article kicking off with-
“Largest power station at ‘end of life’
THE owners of South Australia’s largest power station have likened it to an elderly person sitting in a retirement home…” we can expect gas fired Torrens Island to be closing down soon like Hazelwood and Port Augusta before them. There’s only so much you can wring out of sticky tape and string as the recent fire at TI attests. We’re living on borrowed time with no reinvestment in thermal power stations as the unreliables free-ride on that insurance with their own negligible marginal cost of production. It remains to be seen how consumers react when they’re stuck with too many unreliable electrons with no more backup.
The remedy is obvious but the political fallout to administer the correct medicine now means waiting until the patient is in emergency intensive care on life support.
Although there was no distribution level shedding, I suspect there will have been automatic HV underfrequency relay trips at the transmission level.
The big difference between this event and last September was, because the interconnector had some reserve capacity and remained on, the frequency held up sufficiently to allow the underfrequency relays to operate and shed that required 300MW or so.
In September, the frequency fell 7HZ in one second, too fast for the underfrequency relays to shed load.
Graeme – I’ve seen some high resolution data on system frequency – not at all SA locations admittedly – and it doesn’t seem to show falls to anywhere near as low as 49Hz, which is the trigger point for the first blocks of UFLS. Unless you mean direct-connected transmission customers (and I believe there are < 100 MW of such load), other loads on the transmission system are distribution offtake feeders etc, and had any of those been tripped by UFLS action then SAPN would certainly have been aware of that. SAPN's comment to me was that they believe the load reduction was along the lines of what Tom referred to above as "load shaken off" by the system voltage disturbance.
But we'll have to wait for a detailed report on this to be sure.
As an experiment, I tried flicking the supply switch on my running heat pump. It stopped, unsurprisingly, and did not restart for several minutes. It did not wait for the house air temperature to drop to its normal minimum temperature, so the restart was not the normal thermostat operating.
The pump manual says :-
Frequency ON-OFF Max 4 times/h
Inching protection 10 minutes
ON and OFF interval 3 minutes
Power source voltage +/-10%
A sufficient (exact magnitude unknown) voltage dip that reached the extremities of the system would seem to be able to stop all the running air-conditioners and the automatic restarts are postponed for several minutes by design. Would this help to explain some of the ‘missing demand’ and the subsequent demand recovery? The solar PV presumably remained attached.
Perhaps it was unwise for the gas industry to fight against a domestic reservation policy. Domestic reservation is not good economics, but very good politics, because it means the gas industry would have some political allies. As it is, who are their allies ?
– the local manufacturing industry hate them because gas prices are going up so much
– taxpayers hate them because the gas industry pays no royalties
– environmentalists hate them because of high fugitive emissions of methane
– farmers hate them because of damage to their land, fed by stories of how the industry has behaved in Queensland
– the State LNP don’t like them in Vic and NSW because the farmers hate them
So for anyone not part of the the gas oligarchy the answer is … much more solar (including storage), some LNG imports, and partnering with juniors such as Strike Energy (as Incitec is doing). I’m sure that’s not how the gas industry planned it to be.
Where is the gas to replace coal and/or to solve problem of intermittent renewables? Apparently conventional gas on the East Coast is peaking. The whole LNG export from Gladstone was a bad idea.
In 2011 my main concern was that gas – our only alternative transport fuel – is wasted.
The APLNG and QCLNG projects developed enough new upstream gas to meet the needs of their downstream LNG plants – it is GLNG that should never have been built because their upstream fields are not capable of meeting the demand of the downstream LNG plant. Therefore they are reliant on third party gas which has resulted in existing domestic gas being sold to GLNG and the mothballing of gas fired generators.
“Therefore they are reliant on third party gas which has resulted in existing domestic gas being sold to GLNG and the mothballing of gas fired generators.”
I would have thought more a Catch22 at work here. The gas fired generators (ie particularly Pelican Point) with subsidized and mandated unreliables impacting their bottom lines are not confident enough to negotiate long term contracts, so the gas suppliers and infrastructure developers natural look elsewhere for predictable returns. Again it is all against the backdrop of unreliables free-riding on thermal producers to insure their fickleness and now we have a State Premier demanding it of them and promising to build another thermal competitor with taxpayer funds to really ice their cake.
Although as a humble layperson and power consumer I should really leave professional analysis of the situation to my betters in these matters- http://nla.gov.au/nla.obj-158793567/view
I find the article and discussion very intriguing. Particularly intriguing is the very plausible notion that there is a high level of “invisible” load shedding provided by inverter powered compressors. Something that occurs to me is that one of the benefits that’s claimed for large battery installations is rapid voltage support and this may be contrary to this load shedding effect. I wonder if the relatively slow response of synchronous generators turns out to be an advantage here… because it allows enough time for the inverters to sense the disturbance and switch off. Of course the inverters of the battery installation can be set to whatever is needed, and it seems it might be good if they are set to match a synchronous generator response, rather than aiming for something “better” that would prevent voltage collapse. There’s probably a middle ground here of enough voltage support so that it does not collapse too deeply causing generation and transmission to drop out, and yet enough collapse that inverter powered devices can can sense the drop and switch off.
I suppose the action of the inverters is designed to protect the device, not the grid, but it seems like there would be a good opportunity to use this phenomenon deliberately to protect the grid.
@chris
I have given thought to your suggestion that inverter-driven heat pump tripping is a useable affect. IF that was demonstrated to be the case, it could be an answer to SA’s present woes that are basically driven by heat pumps (OK, wind power generation collapse matters as well) in hot weather and in particular by large numbers being turned on at much the same time of day.
Because all the houses are hot by late afternoon in summer, starting a domestic heat pump makes it run continuously at high load until the house internal temperature drops to the minimum set point. It is a crazy way to run a heat pump. Pumps should be left on all the time on thermostat control with the set point altered a bit when the the house occupants are not there.
That is the way I run my heat pump. It is switched on in September, off in April and runs under thermostat control, drawing 270 to 350 watts when running; overnight the set point is dropped 2C. Done that way a small pump suffices to heat (in the UK) a whole house, except for when outside temperatures dip to freezing. A small pump is not only cheaper than a large one but also more efficient, smaller and less noisy and critically uses far less peak time electricity.
Your suggestion would be a form of fast-acting demand management with a guarantee that the pumps do not restart for a few minutes. Nobody should be that put out that their heat pump is stopped for a few minutes, provided it is not damaged by this being done repeatedly. I suppose the awkward squad might fit surge arresters?
http://digital-library.theiet.org/content/journals/10.1049/joe.2014.0180
In this study, results from testing six heat-pumps are presented to assess their performance at various voltages and hence their impact on voltage stability.
…
Note that all units continued to operate down to a voltage of 0.7 pu or below.
Heat-pump performance: voltage dip/sag, under-voltage and over-voltage
Table 1
under-voltage cut-off, pu
good (0.7)
low (0.52)
low (0.55)
moderate (0.6/0.68)
very low (0.24)
low (0.58/0.54)
Analysis of results obtained indicates that all units have a reasonably well-defined under-voltage cut-out (and cut-in) level. However, for many of the models this is set at a surprisingly low voltage.
All five units fell to low power operation after the transient and resumed their previous operating conditions after their normal start-up delay (typically 1–3 min).
AEMO recorded a 400MW drop in energy demand because of the incident.’
Surely, a trip of solar PV generation would be seen by the grid as an INCREASE in demand? Try working the problem again and suggest it was the air-conditioners that tripped and restarted later under micro-processor control?
I have checked the underlying report and the problem is there. One of the recommendations is to check the solar PV performance under grid transients but no mention of the air-conditioners.
Fault at Torrens Island switchyard – AEMO https://www.aemo.com.au/-/media/…/2017/Report-SA-on-3-March-2017.pdf
FAULT AT TORRENS ISLAND POWER STATION SWITCHYARD AND LOSS OF MULTIPLE GENERATING UNITS,. 3 MARCH 2017. Australian …
As PaulD has picked up, AEMO released its initial report into this event a few days ago.
It provides a lot more useful detail on the nature and sequence of the faults that led to the TIPS B and Pelican Point trips. It also confirms that there was no underfrequency load shedding, but that there was a sharp drop in demand immediately after the first faults / trips, and it was probably this that prevented another trip of the Heywood interconnector which would inevitably have led to another System Black in SA.
As it was the interconnector went uncomfortably near breaching the (revised) triggers for Loss of Synchronism protection – see p12: Based on the above, AEMO believes this event came very close to tripping the Heywood Interconnector.
Strangely there is no discussion in the report of the reasons for the 400 MW loss of underlying demand – offset initially by what AEMO believes was loss of 150 MW solar PV generation yielding a net 250 MW demand drop (see Figure 3 in the AEMO report). I’ve seen subsequent comments attributed to “an AEMO spokesman” relating the underlying demand drop to the “natural response of modern electronic equipment” to low voltages caused by the faults, and stating that “AEMO will develop better modelling of load and rooftop PV response to severe voltage disturbances, to fully understand the implications for power system security.
Just as well – if the numbers had been the other way around, 150 MW loss of load and 400 MW loss of PV generation – then the lights would definitely have gone out!
Love your analysis as always Paul. One comment about your conclusion in C1. The following is a simplistic analysis assuming that all electricity is traded on spot market, but hopefully it has enough truth that it demonstrates my point. If you look at periods in the last 12 months where demand in SA exceeded 2.6 GW (43 half hour periods), price increases very steeply with demand. It’s not a great correlation, but a linear fit indicates a $23/MWh increase in spot price for every MW increase in demand. A 100 MW battery could potentially reduce prices at this extreme end of the spectrum by over $2000 per MWh. That’s a saving of over $6m per hour, or $137 million over the 43 half hour periods. That represents 10% of the total value of the SA spot market in the last year, and goes a long way towards justifying the cost of the battery.
My understanding is that the 100MW gas turbine at Quarantine failed to start under load, and that is seen by AEMO as the straw that broke the camels back
The protection system for other power plants and the interconnector often uses an I^2t rule which depends on the thermal time constant of the systems. It might be set at a number like 100 so if the over-current is say 10% the system will stay on line for 100/(1.1)^2 = 80 seconds. If the load went to 160% of rated load it would shut down after 40 seconds.
100MW of batteries could have allowed three things,
a) a lower peak load on the interconnector therefore a longer time before trip
b)some controlled load shedding
c) some useful governor response on the gas plants that were running
b) Ramp time for Quarantine and reciprocating plants so that would then have been stable enough to take up the load.
I don’t know the exact numbers but 100-150MW of batteries (including grid controlled behind the meter systems) may have just halted the slide
The situations for September 28 and March 3 still seem very similar, so why didn’t the system go black on March 3.
Fig 7 of the March 3 AEMO report shows a very different impedance trajectory between the two events and that is why Heywood didn’t separate but why the difference in the trajectory?
There was slightly more headroom on the Heywood Interconnector for March 3 but peak flows looked very similar.
It is clear that during the March 3 event there was considerable load shedding that seems to be just a natural function of the connected devices. But why didn’t the same shedding happen during the September event. Was there something in the way the wind farms went off that masked the voltage drop at users and so the same unintentional load shedding did not occur?
The other interesting thing appears to be the reaction of the solar PV systems. The solar systems are happy to drop off but leave the load they were supplying still on line. This seems to have generated an instantaneous hit of 150MW to the load which you would have expected to make the whole state system more susceptible to tripping. This doesn’t seem to be a good idea.
It’s frustrating that AEMO weren’t able to comment on the reasons why significant load tripped on Mar 3 but not on Sep 28. However the relevant charts in AEMO’s incident reports show that the voltage dips near Adelaide during the March incident were a lot larger than those on the black system day. The nature of the loads would have been different too (more aircon load in March).
On the solar PV question, it’s possible that (some) load associated with those PV systems may have actually tripped – AEMO’s report doesn’t say either way. If you are saying that solar PV systems ought to be set up so that the inverter tripping causes the entire customer supply at that point to disconnect then that seems a pretty blunt approach – nor likely to be practicable.
I would be very interested to understand what caused the difference in voltage responses – September 28 lost 456MW of wind generation and March 3 lost 600MW of thermal, but the system response was so different. What caused that? The March 3 trips were at transformer level while the September 28 trips were at generator level – does that give us a different response?
On the PV response – if I was a non PV household in Adelaide I would be very pissed off to find that my neighbour (who had installed PV to save money on his power bill) shut down his PV system to protect it and decided instead to pull their whole load from the “high priced” network and that additional load helped pulled the whole network down. I would have thought it would be relatively simple, and reasonable, to trip a householders whole supply in the event that their PV system thought it was unsafe to be connected to the network. I sort of expected that would be the normal response if the PV system was feeding into the network.
It is also a matter of equity. Should users who are trying to distance themselves from centralised generation and network distribution (and save money) be given equal access to that network in the event of an emergency to the extent of compromising that network?
15:05 Heywood 636MW TIPS B2 starts runback
15:05:15 PP ST tripped -70MW over 89 seconds
15.07:34 TIPS B2 tripped -130MW over 02:34 minutes
15:10 Heywood 861MW, then 2nd peak at 918MW shortly afterwards
It was close to one second from incident start to the maximum load on the Heywood interconnector. There was a second, lower maximum on Heywood just after 15:10.
The first maximum (a 962.9-495=468MW rise) must therefore have been caused by :-
1: the grid transient,
2: the loss of TIPS B4 and PP GT11(a 280MW fall in generation in total) and
3: any loss of solar PV in the first second [1: suggests a delay to about two seconds], together with
4: any loss of demand in the first second.
It is possible/probable that 3: and 4: saved SA from another statewide blackout. Such a result would be pure serendipity. Fault 3 and the trip of TIPS B3 were too late to affect the first Heywood maximum load.
What is missing is an exact time for the 400MW loss of demand and any PV detachment; the report states ‘near simultaneous disconnection’. Was it before or after the Heywood maximum load at 15:03:47.300 or even spread either side of the maximum?
Fig4 has a horizontal scale of one hour and has not captured the first, sharp interconnector load peak of 962.9MW at 15:03:47.300.
It could be that much of the solar PV stayed attached for close to two seconds i.e. after the first Heywood interconnector peak load [1:].
Is this delay what saved the SA grid from another blackout?
How fast does the solar PV reconnect?
‘According to IEEE1547, DERs must wait 5 minutes before restarting automatically after disconnecting due to off-nominal voltage and frequency, provided that frequency and voltage have been restored within tolerance. … no provision for staggered restart’
1: Performance of Distributed Energy Resources During and … – NERC http://www.nerc.com/pa/RAPA/ra/…/IVGTF17_PC_FinalDraft_December_clean.pdf
distribution system, and, in particular, with distributed wind and photovoltaic (PV) resources. ….. system are likely to trip on under voltage within ten cycles.
Includes references to IEEE Standard 1547 must-trip requirements e.g.
V<50% Max Clearing Time 0.16sec
50%<=V<88% MCT 2.0 sec
Fig 1: refers
Fig 8: shows A/Cs (air-conditioners) tripping on thermal protection and coming back.
Thanks very much PaulD, I hadn’t previously noticed that level of detail.
I can’t access the NERC paper you referenced at the moment but will keep trying.
In your dot point 3 are you suggesting that some of the PV systems could have disconnected within 0.2secs (10 cycles) and made up much of the difference between the rise of 468MW on the interconnector compared with the loss of only 280MW of generation. So 188MW of apparent extra load unaccounted for.
This is not shown on Fig 3 of the AEMO report but that could simply be because the timing of the SCADA could not pick up these rapid changes. This could mean that the 150MW rise we do see on Fig 3 is some of the remaining PV systems falling off. After all it appears there was about 420MW of solar on line at the time and so far we only have 188MW plus 150MW of “phantom” load.
We might get some idea of how fast the PV systems reconnected by also looking at Fig 3. The 150MW rise in load that AEMO has attributed to the PV systems, gradually disappears over the following 2 secs and the load keeps falling, but more slowly for a further 2 secs. So this could be PV systems coming back on rather than additional load simply falling off the system.
But all of this just makes the question of what was different between the March 3 event and the September 28 event all the more perplexing.
May I suggest an addition to the problem list?
Problem #7
Black Start
Solution: A big battery and an OCGT.
===================
If we think back to the September blackout, there was an issue with the black restart arrangements; they failed to work and SA’s grid had to be restarted with a long ‘jump lead’ from Victoria. In the absence of the interconnectors, SA needs a new battery to cold start its first OCGT to re-energise its grid. That allows wind and solar to reconnect. It makes sense to me.
It makes for an interesting operational issue. Do you run down the battery to try to prevent an interconnector trip or do you save the charge for the black restart? If you are leaving the NEM, you definitely need a big starter battery with a full charge at all times.
The battery’s primary job is to cold start the OCGT but it may also be good for supplying inertia/voltage support into the system during grid transients for very short periods of time. I would not expect the battery to be regularly deep cycled, so a long life is in prospect. The OCGT is available to run when prices are high enough for it to make a profit; that might be the end of $14,000MWh prices spikes. Run at minimum load, the OCGT would provide some system inertia, reactive power and spinning reserve, although spinning reserve is useless in SA until after the instant when the Heywood interconnector opens.
If I was rebuilding a grid system for full independence, this is where I would start.
The author’s unambiguous ‘no’ to the question whether wind and solar will make the system unstable is not correct. Sure, power systems are designed to deal with variability, but traditionally that variability has arisen from the demand side, as well as contingency events. Intermittent wind and solar add an additional source of variability. The evidence from overseas market is very clear that intermittency requires additional high frequency reserves to be held.
A substantial amount of spinning reserve (inertia) will be removed from the national electricity market with the closure of Hazelwood.
It would make for an interesting comparison of say frequency drop rate of change before and after Hazelwood closure upon an event such as a 660MW’s separating from the grid followed in close succession of another 660MW’s
I await the next Bayswater type separation event for comparison.
MrCyberdude
Would be great if a few more frequency drops with greater variance were shown in a graph to get a real feel for the differences between FCAS response from different thermal generators and renewables, especially considering the ever increasing renewable sector.
MrCyberdude
That was one of the harder articles to get through here at watt clarity, for me at least.
While the author concludes that a disproportionate amount of the FCAS costs have been allocated to the wind generators, I would also say that frequency control (FCAS) was not an issue in the past decades before wind generators came alone and made frequency such an issue.
If wind generators brought the problem then they should offset it either as part of their regulatory compliance which would probably mean gas cogeneration at their sites or they should simply pay the lions share for what they have created.
I had never seen so many frequency swings in both directions before wind generators were connected to the grid in quantum.
MrCyberdude
Wind is hardly causing the high requirements and prices for FCAS regulation services. Please see “Fast Frequency Service – Treating the symptom not the cause?” for a technical discussion on how synchronous generation, excessive local requirements and in my article poor AEMO forecasting is much more important than the use of wind as a generation technology. It is simplistic and disingenuous to put the blame on wind generation.
I would add that FCAS became a problem when the synchonous generators relaxed their frequency regulation a few years ago because AEMO penalised one for curtailing when that curtailment was made to assist frequency. So the AEMO decision directly signaled to the market to not prioritise regulating frequency. FCAS goes up. Also aging coal fired generators having problems has increased FCAS. On the other hand Telstra* big battery has reduced FCAS costs.
Looks like Pelican Point is back in the market at full capacity, as previously modelled by AEMO as a result of Hazelwood closure. Currently outputting approx 420MW. Haven’t seen any official announcement yet.
thanks for the article and also to the site for assisting a non EE person to gain a little understanding.
Dr Mackenzie, does your comment lead to a requirement to revert to the rules on frequency control that applied prior to 1999 as described in the ” symptoms not cause” article as a first step?
Not necessarily. I do think that there needs to be a rethink about causer pays factors generally, and especially for the intermittent generators but there is some justification for the concept. It is the formulation, where the existing dispatched unit causer pays methodology was simply used for wind and solar without really thinking through all of the implications, that is the issue.
OK thankyou for the response. As a person who worked in continuous manufacturing dependent on steady electricity supply for 36 years I do not understand why govt. would allow mechanisms that reduce stability or reliability either from technical or market aspects.
No need to respond to the comment above, just an observation.
Yours and other articles show that there is much that we as the broader public do not know about how our system operates so please keep up the good work.
Thanks for the analysis Paul. This transition really does feel chaotic and would be nice if some level of government could decide what they were doing and plan into the future just to give us some indication of future employment, if nothing else!
We are already seeing some announcements that would partially offset the loss of Hazelwood, such as the additional unit at Pelican Point and large solar in SA (Bungala 200 MW, Riverland 330 MW, Kingfisher 120 MW, Tailem Bend ~100 MW) and northern Victoria (Mildura area 320 MW, Gannawarra 60 MW).
South Australia has tended to be an energy importer from Victoria, but would likely turn to being a net exporter on account of its better solar resource, and Pelican Point being the most efficient gas-fired power station in SA/Vic. It has among Australia’s best solar resource in locations close to areas of grid strength. This would help balance out wind, because days of low wind usually have good sun.
Should also note that while Hazelwood closed, there were 3 large coal units at Loy Yang A and Yallourn. When these are returned to service there would be an additional 1220 MW of coal capacity.
The use of the word transition to me infers an ordered change of state. Instead we have a bunch of fairly random decisions taken on the basis of ideology, politics and virtue signalling.
It seems absent any thoughtful and informed work on a forward plan and the right mix and sequencing of power sources. Right up to this day, I have seem nothing that looks like anyone has any kind of plan beyond what they need to do for the next State or Federal election cycle.
I am in VIC and will be installing what I need to protect myself over the winter period. Probably a reasonably sized generator and transfer switching.
Its great to have such concise analysis; I’m surprised there hasn’t been more discussion since publication. Kate refers to a paper published on 03/02/2017; is it possible to have a link to that? It sounds extremely interesting.
Thanks,
Shane
Ever since I’ve become addicted to watching AEMO’s NEM Dispatch overview, I’ve come to a few conclusions rightly or wrongly;
1) SA spot price appears to be consistently higher than a number of States and almost always higher than Qld to a significant extent
2) the contribution towards meeting the power demand in SA by wind turbines is surprisingly poor given all the political hype. It appears highly variable and unpredictable (unsurprisingly) and at times reaching extraordinarily low levels.
3) associated with 2) above, there seems to be a high reliance on the interconnectors with Victoria to supply “coal fueled”) power into SA thus challenging the notion that SA thrives on “green” power.
4) it appears that SA would be in all sorts of trouble if it were not for the ready availability of power from Victoria
5) it appears that due to the relatively low amount of synchronous generation along with various other generation issues, the Heywood interconnector is subject to unhelpful constraints at some crucial times
6) power flows to and from Tasmania together with the constraints applied to Basslink don’t seem to make a lot of sense sometimes when considering the value of premium energy stored in Tasmania especially when Tasmania has been experiencing unusually low rainfalls in the past year or so. It seems that the temptation to make some opportunistic profits because of the price differentials between Tasmania and the mainland might be just too much.
Are my observations reasonable or am I missing some vital clues ?
“Why don’t we ensure (via appropriate rules and technical standards) that all new generation technologies contribute to frequency control and adjust the market to remove penalties that currently exist for generators when they act to provide good frequency control?”
You mean like also ruling that no generator can tender anymore power to the grid than they can reasonably (ie short of mechanical breakdown) guarantee 24/7, 365 days a year for a level playing field market? Now you’re talking but who’s going to point out to our current overlords that it’s a fundamental tenet of engineering that you can’t build a reliable system from unreliable componentry? Unfortunately we just have to wait for the train wreck for them to finally comprehend a national power grid doesn’t run on emotion.
I am hopeful that a well designed system will emerge from the current unsatisfactory situation, and that a reliable system will be built from reliable componentry – but only if people pay more attention to the physics and a bit less attention to the fashionable economic theories of the day.
Well in defence of market economics I’d say it can provide some useful tools for the greatest good for the greatest number but that will depend on the underlying assumptions.
For example it may be the science of Catastrophic Anthropogenic Global Warming alerts the economic practitioner to some social cost that is not being reflected in the much lower private cost of fossil fuel use and there’s a simple remedy. Just call the guru who knows the additional price that needs to be added in order to bring the level of CO2 in the atmosphere back to the level said guru knows it should be. Having imposed that on all players in say the business of producing reliable, 240V, 50Hz electricity as required by consumers is then another set of issues which we would probably agree needs more physics rather than political interference. The economist would say set those basic level playing field parameters and the various players will then come up with the economic solutions for us.
But back to the CO2 assumptions and that guru’s estimate of social cost. With only around 150 yrs of reliable thermometer records our guru has had to rely on such temperature proxies as tree rings, ice cores and varves along with a host of computer models to form an estimate of CAGW and there is some debate about the scientific veracity of all that. However there is one measure of global temperature that clearly rules them all and that’s sea level rise since sea level rises with temperature for only two reasons. Thermal expansion of oceans or runoff from ice melt on land and the polar caps.
Well you wouldn’t want to rely on a tide gauge in Venice or the side of a volcano in Hawaii but here in our own stable ancient weathered continent we have a couple of very reliable sea level gauges. Take the one at Port Arthur(one of the longest man made in the world but certainly in the SH) and it shows us 13.5cm sea level rise from 1841 to 2000 or an average sea level rise of 0.85mm/year. But there is another at Hallett Cove Conservation Park in SA (so proclaimed by none other than our Premier Jay Weatherll as Environment Minister in 2010 because of its world geological significance) and that shows there was an ice age 20000 years ago but around 15000 years ago the ice began to melt and the sea level off SA rose 130 metres until 6-7000 years ago, which could be an average increase of 16.25mm/yr for 8000 years.
So what conclusion did I logically draw from that? There are either some very very clever scientists and their computer models that can extract the anthropogenic CO2 signal from the current Port Arthur temperature proxy, or else aboriginal cooking fires and traditional burnoffs to flush out game have an awful lot to answer for with that Hallett Cove temperature proxy. Now where were we? Oh yes, economics and the underlying assumptions.
I agree with you that market economics can provide useful tools which can be used to benefit society as a whole. I will leave it to others to determine for themselves whether they believe it has actually been applied to this end.
My main point is that some things can easily be traded – e.g. commodities like food, minerals, fuel and (specific to the point of this article) power. For these things designing markets is easy.
However, there are more esoteric, less easily defined benefits that derive from concepts that cannot be easily turned into commodities. In the power system world this consists of all types of ancillary services. These things are not easily expressed in a market paradigm.
Markets that are set up to deliver these types of services easily lead to unforseen consequences which we have seen playing out in the frequency ancillary services, so that now we have worse frequency regulation and a less secure system than we did in 1990. This contributed (albeit only peripherally) to the first state wide black out we have had in approximately 50 years.
With respect to your comments on climate change, I will just treat that as commentary. I am happy to reply to any questions related to Power systems which is my field of expertise and what the article is about. Climate change is not my field of expertise and therefore I suggest you refer such comments to a climate scientist.
Climate change is not my particular field either, but facts are when I go for a stroll through a Conservation Park and my interest is aroused as to just what are the broader facts of the matter. As a result I naturally continue to believe in climate change, just not CAGW anymore after previously reserving judgement on some worrying critiques of computer modelling of lesser proxy variables- https://www.ncdc.noaa.gov/news/what-are-proxy-data
Science and the scientific method don’t fit in neat little compartments or become the preserve of anyone. The facts speak for themselves until they’re contradicted or otherwise interpreted.
Nevertheless particular fields of endeavour can require certain expert technical expertise to fully comprehend the bigger picture for the policymaker so with that in mind can I ask your thoughts on this interpretation of where we’re headed on the current path with the grid- https://www.youtube.com/watch?v=gJtv7gkuh1s
There is a legitimate role for good Govt overseeing areas of market failure with externalities (the tragedy of the common and the like) but there is a sense with our national grid that no-one is in charge anymore with a history of incrementalism and the science of muddling through. Whatever level playing field social cost we put on CO2 and bearing equally on all inputs it should (concrete or steel for thermal or wind generators), that still leaves facilitating reliable voltage and frequency at the flick of a switch for consumers and policymakers need to understand and address the obvious shortcomings now. In that respect you’re talking of frequency, I’m talking of reliability of supply while others are beginning to talk about the voltage- https://www.getinvolved.qld.gov.au/gi/consultation/3703/view.html
There must be strong incentive to produce all three at the same time with no beggar thy neighbour outcomes on any score.
Re point 2, the contribution of wind is higher in winter than summer, and higher for the newer wind farms such as Hornsdale that has a higher hub height. Hornsdale has averaged a capacity factor of 41% since the start of January, and most new-build wind will be like this. Once there is more large-scale solar in SA, this will partially balance wind output, being higher in summer and under high pressure cells, and lower as lows pass over with their windy conditions.
Re point 3, now that Hazelwood is closed, Engie’s Pelican Point has been operating at higher capacity factors, leading to much more frequent exports from SA to Vic. This is the most efficient gas power station in SA/Vic. Until about February, most of the load balancing in Vic was from the Snowy. Now it is Mortlake (gas) that seems to be doing some of the load balancing, and some from Newport (gas). The coal generators generally run at close to 100% or not at all (when down for maintenance).
Re point 5, the Heywood inter-connector seems to have much more than its fair share of problems. For example, last Sunday (9 April), export from SA was limited to ~290 MW from 11:35 until 15:00 on Sunday 9 April during a time of high wind output. This caused SA prices to crash to minus $45/MWh over most of this period. At other times of the day, export from SA was never more than 500 MW, and SA prices held up well. Supposedly the inter-connector has an export capacity of 650 MW via Heywood and 220 MW via Murraylink, totalling 870 MW, under a project that finished in March 2017.
How much of this wind correlation penalty arose from SA becoming a net importer to a net exporter of electricity ? The SA market appears to go through 4 stages prices depending on the electricity deficit or excess.
1. When demand exceeds the inter-connector import limit, price is determined by the SA bid stack.
2. When demand is below the inter-connector import limit, and SA is a net importer, the price is ~$10/MWh ABOVE the Vic price. The $10/MWh is probably a margin made by SA Power Networks.
3. When excess is below the inter-connector import limit, and SA is a net exporter, the price is ~$10/MWh BELOW the Vic price.
4. When net supply exceeds the inter-connector export limit, price is determined by the SA bid stack, and is generally negative.
So at most times, the Vic price provides at floor to the SA price, but there is a $20/MWh price difference between net import and net export.
Since the post was prepared, Hazelwood has closed and Vic prices are determined by the marginal cost of gas-fired electricity, first at through Mortlake then at Newport, both of which are more flexible than was Hazelwood. Furthermore, the inter-connector capacity has been increased. Previously during times of high wind production, Hazelwood would keep up full production, leading to a depression in prices from the combined SA and Vic wind production. On Sunday 9 April was the first instance of high wind and low demand in the post-Hazelwood market. Loy Yang A reduced its production and provided stability services, while Loy Yang B and Yallourn W continued unabated, but prices subsided only a little. So these market changes are likely to reduce the problems of the wind correlation, at least until more wind supply is added. Now we need more balancing solar to supply the market under anticyclonic conditions (high pressure cells).
Thanks for writing this up Paul – an interesting situation.
There are a number of things going on here that are worth pointing out.
The transmission line which went out of service, causing the sequence of events, runs between Mortlake Power Station in western Victoria and the main part of the Vic 500kV grid. It is one of two lines that supply the Portland smelter and also the interconnection with South Australia. So with this line out, a trip of the second 500kV line on this corridor (the Moorabool – Tarrone line) would isolate South Australia + generation and load in the south west of Victoria (principally Mortlake PS and the smelter) from the rest of the NEM.
In this situation AEMO invokes a couple of key constraints (amongst others):
1) FCAS constraints requiring 35 MW of FCAS regulation services (“regulation raise” and “regulation lower”) to be sourced from South Australian generation – so that if the second line trips, there will be some generators already providing regulation services in SA to help manage frequency in the “islanded” South Australian system (actually SA + SW Vic in this case). While the DC Murraylink interconnector would still connect the two regions, it cannot currently transfer regulation services due to its design – presumably this might be another technical option for provision of FCAS in SA along with the possibilities you mention of wind or solar generation being modified to provide some of these services. Whether feasible or economic I don’t know.
As you point out the very limited FCAS bidstack for these services in SA plus the requirement to source this locally under the above constraint led to pricing of these raise and lower regulation services shooting upwards. There is clearly insufficient competition for provision of these services within SA to drive prices down to reasonable levels when they are required locally under conditions like those that existed last Tuesday. (Ironically, technically Mortlake PS could provide regulation FCAS within the SA + SW Vic island, and had relatively cheap bids in for these services, but the way the NEM’s regional boundaries are defined, it is obviously not considered a South Australian generator for the purposes of procuring “local” FCAS when this constraint operates! – I wonder if AEMO could make the relevant constraint a little bit smarter so that under the appropriate precontingent conditions Mortlake could be included in the generators from which “local” FCAS is procured – but possibly not, due to the way that regional pricing and cost recovery for FCAS operate.)
2) The risk of islanding also leads to a constraint being imposed on net exports from the SA + SW Vic region, so that if the second line trips there won’t suddenly be such an excess of generation in the island to cause a frequency rise too large to be handled. The limit is 50 MW but because the island would include Mortlake generation and the Portland smelter, the constraint doesn’t simply go onto the SA-Vic (Heywood) interconnector, but includes Mortlake generation and Portland load as well.
At the time of this constraint being invoked, Mortlake was running at 551MW, very close to full capacity and well above Portland smelter load (which appears to have been about 220 MW), so the constraint forced NEMDE to dispatch additional flows westwards on the Heywood interconnector, at the same time ramping down Mortlake generation as fast as possible (nominally 26MW per minute across the two units or 130 MW per 5 minute dispatch interval), in an attempt to keep the net flow to the rest of Victoria from the SW Vic region to under 50 MW. This had the effect of forcing generation levels and dispatch price downwards in SA due to the increased Heywood imports, and briefly upwards in the rest of the NEM (probably due to the rapid ramping down of Mortlake). Hence the “counterprice” inter-regional flow mentioned in your writeup.
Interestingly, within a few dispatch intervals and as Mortlake continued to ramp down to about 300 MW, SA reached the position of being a net exporter to Vic (with Murraylink flows eastwards exceeding westward Heywood flows) so prices in SA being lower than those in Vic were in the expected relationship.
The direction would have been to ensure two large synchronous generators were online. As Pelican Point was offline, and only one TIPS ‘B’ unit bid in to the market, I believe the direction would have been for AGL to start a second Torrens Island “B” unit. The minimum requirements for online synchronous generation have been published by AEMO previously but I couldn’t find the link.
The initial direction very early on Tuesday morning was for Hallett GTs (8 units) to start. These ran until early in the afternoon when a second TIPS B unit came online, again directed by AEMO.
The paper on minimum synchronous generation requirements in SA is at this link.
What impact has resonance on grid stability? Had a look at this guys video where he states that integrating solar and wind into a grid has serious issues. Thought you electrical engineers could explain it to us laymen? https://www.youtube.com/watch?v=gJtv7gkuh1s
Wind power in SA tends to peak strongly at night, probably due to “gully winds” in the hills near Adelaide, where most of the wind farms are located, but it will still drop dramatically during heatwave afternoons, no matter how many wind farms are added. Wind power is SA has killed the concept of baseload, without providing much in the way of firm heatwave capacity, maybe there are some benefits that compensate for those terrible characteristics.
An excellent article Allan, makes it a lot clearer to the casual market observer like myself. I’d be interested to understand the impact of the intervention pricing on generators, I assume if AEMO constrains them they would be compensated?
Intervention pricing attempts to calculate what the market price would have been in the absence of AEMO’s direction(s), and is the basis for the prices actually settled financially with spot market participants.
Participants who are directed by AEMO are entitled to be compensated under the market rules, which also contain provisions for adjustments to settlement amounts for other participants whose dispatch was (indirectly) affected as a result of a direction. The details of how these amounts are calculated are complex, to say the least.
Why is this not in the news? I know it is a budget week, and I am watching from QLD, but does the market assume that parties will come to the table prior to the 11th, because as a lay outsider, reading the AEMO notices, there could be some pain in store on the 11th onwards.
I can understand that other generators may be hesitant to contract for gas, in order to bid if at the 11th hour AGL could turn around and cancel the outage (I am guessing they can do this via the market rules).
Two units suddenly shut down at 2:30pm and 3pm today, with what looks like the final one off from 8:30pm. But perhaps coming back for tomorrow evening’s peak (based on predispatch Vic availability). Strange.
If readers are wondering why it’s such an industrial rocky road.
FairWork is broken when it comes to industrial relations and essential service workers, and it is this reason that it is likely to get worse as time goes on. Unless there is a mechanism to deal with scenarios where a company fails to negotiate in good faith then terminates agreements then threatens to lock out workers, therefore effecting 3rd parties at their industrial whim, it will likely escalate on both sides.
You should do a follow up for the effect on the NEM of lack of wind in NEM in recent weeks – zero wind generation in the NEM as I write this (with some sites showing negative?). I am not anti-wind but it has seen some radical pricing with no wind in recent weeks post Hazelwood closure.
Simon Benson reports in The Australian, June 3 on the Finkel Report –
“Significantly, rule changes to the National Electricity Market will include the requirement that all new large-scale renewable projects have storage capacity such as batteries to address ongoing reliability issues.”
That’s a start to unwinding this mess with unreliables bludging on thermal and engaging in dumping but what about all the existing ones creating the train wreck at present? http://anero.id/energy/wind-energy/2017/may
And what politician is game to tell all the mums and dads with rooftop solar that their surplus electrons are not wanted in the communal grid unless they can reasonably guarantee them 24/7 all year round? To do that is to admit they got it badly wrong with half of the Green power equation by not factoring in storage costs to maintain equivalence with dispatchable thermal generation.
There are a couple of models for competition in duopoly / oligopoly markets – two different versions include Bertrand comptetion – where you competing on price, not output, and another is Cournot competion, where you basically compete on quantity, rather than price.
Two very different modes of competition, and behaviours – but both theoretically profit maximising (though in different ways). Maybe it would be too wonking or obscure for the Energy Minister to say “We are going to direct them to behaviour in a manner consistent with Cournot competition”.
…and if they did that, change the mode of competition, CS energy would also be forced to respond (assuming they are also economically rational / act strategically, and try to maximize profit given different behaviour from Stanwell).
Complete speculation – and theoretical. But there is – at least arguably – some logic to it.
Regarding point 2b, it may be more accurate to say “the high prices paid *by retail consumers* for electricity are something that should be celebrated”.
A large proportion of industrial consumers are still insulated from the extra costs associated with supporting the Energiewende, and, apparently, levied with the intent of dissuading electricity use.
I doubt the energy transition can be achieved with a significant presence of wind power. Seems to have been amply demonstrated multiple times , with Germany living through similar doldrums earlier this year, but happily being able to lean heavily on neighbours.
You point 2b is well made. These weather patterns are not new, so plunging into high % wind with 100% backup of whatever type is at the very least irresponsible.
Since solar lulls are never going to go away, and wind power lulls must now be counted as credible contingencies the word “transition” is surely inappropriate. If people insist that the country has to be carpeted in solar panels and wind farms then their output will only ever be the alternative power system, there will have to be another power system that can provide almost full demand when required. Thus, the only “transition” is one from sanity to insanity.
Thanks Paul for a very timely and appropriate review.
As a resident in Victoria with clients in SA and Tasmania, we have all noted that the weather patterns in May and June this year particularly have been dry and regularly still (or low wind).
The BOM has noted this in a number of weather reports and points to a a changed weather pattern where there has been effectively a ridge of high pressure over the southern states where most of the wind farms are located. This high pressure ridge has meant that there has been lots of still and low wind days.
But what has caused this to occur is beyond this simple engineer but I suggest that a question to the BOM might lead to a much more detailed answer
Regards David
BoM reported a couple of weeks ago that the low pressure systems which normally traverse southern Australia at this time of year are (for some reason) further south, closer to the Antarctic, than usual. Hence less wind than usual.
Look up the Southern Annular Mode (SAM) on the BoM website. It has been in a positive phase, which has meant fronts well to the south of the continent, and weak winds within the latitude of most of the wind farms. The SAM shows persistence but can’t yet be predicted. The wind industry isn’t the only one suffering – grain farmers on Eyre Peninsular have had less than 10% of their normal June rainfall. A more strongly positive SAM may be an outcome of climate change. For the power industry, offsetting investments include solar, which would tend to be higher during high pressure periods, and wind farm sites such as Hornsdale that have good exposure to easterlies. Power demand has I expect also been down because of mild weather.
In agricultural research it has been common practice to test new management practices on simulations based on 30 to 100 years of weather data, with the anticipated effects of climate change. Perhaps this should become standard in the power industry. Studies such as that of Andrew Blakers only seem to go back to the start of the NEM, rather than imposing NEM power demand on a longer period of weather data combined with known climate change effects.
Good point using A longer climate record is always a good idea for planning, however I suspect agricultural research is much more concerned with rainfall which in a farming nation like Australia was obviously a huge focus of our observational stratergy. Perhaps wind wasn’t recorded with such diligence?
I believe SAM is also related to ozone good work out of Monash university on this.
Dear Paul
I would certainly be interested in knowing what impact of the recent change in weather patterns has had on the off shore wind data for the same period.
It would also be great to have the MWh total per month of wind which has been rejected due to oversupply in the winter months.
Based on this recent event, numerically, I can see nobody better qualified than wattclarity to estimate the TWh of storage necessary for the East Coast of Australia.
There are no off-shore wind farms in Australia – and only a couple mooted at this point. We don’t currently have access to off-shore wind speed data either. Hence sorry we can’t help on this.
Thanks for the kind words, by the way. In terms of battery storage, you might find our other site of some value: http://www.batterystorage.info/
Dear Paul
It goes without saying that there are no off shore wind farms in Australia.
I am surprised to hear that wattclarity do not have access to off shore wind data.
The storage site referenced is not practical.
Kind Regards
Jonathon Sarah
When we say global warming causes climate change, what that actually mean is climate belts shift. Regions between 30° – 45° will see climate shift by 100-150km per degree of warming. If climate belt has associated wind elements, in the southern hemisphere, these will be moving southwards.
In other words the roaring 40s become the roaring 42s.
Climate belt shifts generally aren’t linear, but rather occur as a step change – a great example of step changes in climate is found in the drying of sw WA and loss of stream flow – https://twitter.com/ProfRayWills/status/877335681038655488
Step changes are of course also points on a bell curve – perhaps this is the first iteration of moving wind resource, and we should expect to see more. In the longer term as climate belts shift, some wind farms may be at risk of being left behind…
Of all the forecast flops, the biggest howler has to be the Frontier Economics wholesale forecast that made its way into the AEMC’s retail price trends report last year.
It all makes you wonder whether the structure of the NEM is not appropriate. While it seemed to work reasonably well when demand and consumption was increasing with little impact from intermittent generation, the outcomes of Paul’s assessment is that we are seeing a seismic shift that warrants a look at whether the NEM as structured is fit for purpose.
While agreeing with Paul that there is no one cause, what he has left out is whether the degree of competition (or lack of it) that now exists is contributing to the problem. After all,the NEM is predicated on vibrant and vigorous competition.
Another option might be whether the government owned vertically integrated model that existed before 1995 might have resulted in lower costs than we see now
Hi Paul McArdle |
Absolutely excellent report and summary.
Your Someone who understands (or at least may 80% of a complex electrical system). Way more than most .
I’ve been following the AMEO dash Board for some time and predicted that soon as they started shutting down coal stations the price was going move up and much more that the experts said.
The people making these decision seem to have little idea of how this system works and how no or delayed decision are going to effect very body. There seems there was no transition plan mapped out or thought given to spinning reserve requirements. would have though NEO could have just lifted the price for auxiliary service some year or so before . The Governments had chances to leave the major coal station/s (now shut down)on operation stand by until the newables/ storage were bedded in. I note the reason the history SA northern station was build was to give SA security and control of their state Elect Costs from Vic coal supply (strikes), Sounds like they don’t read the history , what’s that old saying , “those who don’t learn will repeat the mistake.” SA even took generation out of the hand of private companies make it state owned,
Maybe run a check how many companies
own all the generators? Can analysis gaming the market?
keep it up
Thanks Simo
How is it NOT the RET that’s causing the lack of dispatchable SUPPLY and the higher prices? It’s the RET’s subsidies to the most inefficient power source that are killing off the base load source and causing shortage of SUPPLY…IMO
The gentailers see the huge RET subsidies and favoured status for RE inexorably making coal and gas unprofitable…. and seeing the writing on the wall , they hedge by investing in intermittent sources and putting FF… their coal plants especially…on death row…announcing that they’ll be shut down soon because they can’t compete with the subsidised undispatchable wind and solar..
The privatised gentailers cannot be forced by government to keep unprofitable FF generators operating because they have shareholders and are bound by company laws.
Therefore FF generation shuts down …or is marked for shutdown and SUPPLY dwindles.
Hazelwood closed because Engie saw that writing on the wall…and its closure diminished SUPPLY.
Greens , weak governments and government disregard for the concerns of landowners closed down CSG in NSW and Victoria…diminishing SUPPLY of gas…but had there been no RET ….no VRE with their huge subsidies…and without climate hysteria running Australia’s energy policy there would have been no need to resort to CSG IMO.
Climate change hysteria and the divestment campaigns…and claims that ALL FF will be phased out in favour of RET-susbsidised VRE…. caused a drop in new gas searches and subsequent shortfalls in conventional gas SUPPLY.
The wind drought caused lack of wind electricity SUPPLY.
Gentailers are in the box seat..having covered themselves for all eventualities…and can price the scarce SUPPLY as they wish.
AGL and ORIGIN appear to be running the show now-AGL favouring herding solarPV owners into virtual power stations…to MANAGE their demand…a venture they and their Californian partners SUNVERGE in whose operation AGL is the largest investor at $20million…expect to be very lucrative.
Both AGL’s Vesey and AEMO’s Audrey Zibelman are American RE zealots whose energy plans for Australia are exactly in sync with Labor’s not LNP’s.What sort of government except one seeking to deceive its constituents lets that happen?
The Americans apparently see Australia as an opportunity for their RE experimentation…having not a care in the world that unlike America, Australia has no other country ‘to run an extension lead to’.
Zibelman told her alarmed New York State property-owners in her former job….that they could always run a lead from NYState to Canada if NY became energy-insecure due to her policy there.
They seem to have no knowledge and don’t give a damn that Australia is COMPLETELY ALONE in being the only industrialized country in the world that’s without alternative energy…no massive hydro or nuclear power…and yet is killing off the only energy…coal …that can keep us industrialised and competitive.
There is NO OTHER first world country that’s without hydro and nuclear…. that can do without COAL…NONE!
The shutdown of Australia has already begun as Malcolm Turnbull fiddles.
The LCOE for 100% renewable supply at current component cost and absolute best finance charge is $700/MWh. The prospect of that getting lower is not in sight.
As more renewables are forced into the NEM the cost of wholesale electricity will approach the $700/MWh.
The NEM is a network that takes advantage of economy of scale. It suits coal fired generation. It is inappropriate for a ubiquitous energy source that has no benefit of scale.
The concept of transition to renewable generation supplying the grid is fundamentally flawed. Renewables should be all local supply sources feeding their dedicated loads; never attached to the grid.
Thanks Paul,
Always good to have some insight into what’s driving our weather now that our energy markets are so interrelated to weather.
What about Son of Sam or even Uncle Sam?
But the bigger picture is gas prices have risen due to Green enforced demand policies globally and lack of coal supply locally.
As for correlating price rises with renewables installed the previous point is also relevant in that we have a national grid so a spike in renewables in SA simply lifts all boats with backup expensive gas generation, while SA grabs the localised benefit of the cheap marginal wind production. Basically beggar they neighbour policy which we can’t tell from the simplistic graph. Needs work.
2017 pricing to-date are truly scary and this doesn’t bode well for those of us who want an orderly transition to a renewable grid (or even for industry or the grid itself to survive).
I agree that there appears to be a systemic shift in the market, which is not obviously explained with off the cuff simplistic explanations tossed around (without any data). As Simon above mentioned a non-functional market must at least be one of the possible explanations.
I think all the bidding data is public but maybe not easily accessible. Has anyone done a comparison of the bidding stacks for say 2017 vs 2015? OR is there a public forum where I could contribute analysis?
Discussions on topics like – how has the average price tripled in the middle of the night, when there is ample excess coal base load capacity – would be very interesting. Further, it would at least be worth illuminating the bidding behavior of certain market participants.
“Wholesale prices and those charged to businesses that buy directly are up an extraordinary 60 to 70 per cent. It’s renewables that are doing it, along with the closure of Victoria’s giant Hazelwood coal-fired power generator, according to Tony Abbott. The truth is simpler, and speaks volumes about the appalling way we’ve handled energy planning.”
Unfortunately this is a case where there are other factors at work, in addition to just the high gas prices (including both the Hazelwood closure and the intermittency).
It’s not more simple than Tony Abbott’s (incorrect) portrayal – it’s more complex.
Your chart of utilization has a buried story, because on the far left is the coal, and on the far right is the Wivenhoe PHES which is co-owned, and thus was elected to be less profitable by the coal-power-station owner. Who is in turn owned by the state.
The LCOE of the PHES element is pretty good: its amortized over the requirement for Wivenhoe to exist for flood mitigation. Its only the incremental cost of the PH part, the S part was there anway. So your chart shows how something which a really low PHES, can be turned into a non-earner, if the owner doesn’t want to use it.
Why doesn’t the owner want to use it? Because its only role would be to displace coal power. If it was either split off, or chinese walled, it would have run far more.
Some of our more learned readers might want to also add in their thoughts, but you raise the valid comment about portfolio effects in operational decisions.
That’s one reason I am intrigued by the possibilities of shifting Wivenhoe into the mooted “Clean Co” by the QLD Govt. This would potentially deliver both:
1) Market benefits, in terms of increased competition (which, as noted yesterday, has declined in QLD); and
2) A valuable opportunity for collective learning about a real world case study into the use of large-scale storage.
With particular respect to Wivenhoe, it’s important to remember that a true cost of production calculation would need to factor in both:
2) Round-trip efficiency
On top of this, my understanding is that the round-trip efficiency of the pumping-to-generation process is not as great as it might initially seem.
Hence (1) and (2) add up to a significant (and variable/hard-to-predict) short-run marginal cost of operations for Wivenhoe.
As you point out with electricity supply and demand are inextricably linked so LCOE is only a useful measure if all are levellised. That means generators should only be permitted to supply levellised electrons, namely those they can reasonably guarantee (ie short of breakdown) 24/7 all year round.
LCOE as a measure was clearly only relevant to fossil fuel sources in the past but instantly became irrelevant with the enforcement of unreliables on the system. Now while unreliables may be able to outbid fossil fuels for pumped hydro under an enforced levellised supply regime they would still have to contend with the opportunity cost of wasting some water for alternative irrigation,etc with the Wivenhoes and Tumuts or invest in pumped seawater.
Truly levellised electricity with guaranteed despatchability would be agnostic as to their incorporation into any portfolio mix, unlike the situation we have now and LCOE would return to its rightful place in long term investment decision-making.
I’m afraid that’s not correct at all, so perhaps I didn’t make the point clearly enough in the article, or you are too focused on the “dispatchability” issue.
Different generation technologies with different balances of fixed and variable costs can yield changing LCOE relativities depending on assumed capacity utilization. The fact that LCOE is usually quoted for a single, assumed capacity factor for each technology makes it anything but “technology agnostic”.
Have you ever wondered why coal and CCGT are generally quoted as having much lower LCOE values than OCGT (or reciprocating engines), but in real electricity systems there is often plenty of the latter built despite this apparent cost disadvantage? It’s because OCGT economics are much more favourable for the low utilization roles which are inevitable because of the peakiness of demand and the need to cover other plant outages.
There seems to be a common misconception that OCGT plants are built primarily for technical reasons like fast start and ramping capabilities, but in large scale systems these are secondary to the economic driver which is lower capital cost – if something is only going to be used a few percent of the time you want the cheapest capital cost you can get and variable fuel costs are far less important.
LCOE should have a very limited role long term decision making since it unrealistically assumes a fixed or guaranteed capacity utilization.
I understand where you’re coming from here but in judging demand utilisation investment decisions fossil fuel plant investors know they have the capability at their fingertips to respond to various demands anytime. Unreliables don’t but have the imprimatur to dump on the system with extremely low variable costs.
As consumers we don’t want to know that a Tesla battery plus Hornsdale wind might be able to supply 100MW of power for an hour or so but what can Hornsdale supply for a whole week if called upon, let alone 24/7 all year round? Without enforcing levellised supply criteria we are headed for a Green fallacy of composition whereby supply is at the mercy of the elements. My ancestors weren’t so stupid when bright sparks like Tesla and Edison came along to show them the light and the way.
The real issue that our market operators are yet to address is the need for firm capacity to back up renewables and a mechanism for charging for capacity.
A capacity market will do this and will not favour one technology over another.
People often point to the WA capacity market as a failure and yes it was as their capacity market mechanism could not work when most of the generators offering capacity were owned by one entity and their IMO had a completely flawed methodology for determining the cost of capacity.
You don’t have to back up renewables if you simply rule that all tenderers to the grid can only tender up to a maximum they can guarantee 24/7 all year round and leave them all to work it out. Obviously the unreliables would have to invest in storage of some kind to lift their average tender or else partner with thermal generators and pay them their just premium for that.
Yes I know it will need some phase in for existing unreliables to comply with that level playing field and to break it gently to all the mums and dads with their rooftop solar but somebody has to stand up and scream out loud about the elephant in the room. It’s been the greatest moronic Groupthink that got us here in the first place and if you want to get rid of thermal completely then we aint seen nothing yet with the cost of going all renewable with the storage necessary to make it reliable 24/7.
You just look at the graph and ask yourself what sort of battery or pumped hydro storage would they need to be able to tender 10% of installed capacity reliably 24/7 all year round and think about the cost- http://anero.id/energy/wind-energy/2017/june
One of the price tipping points for Victoria, which flows through to SA and Tasmania, is that there is now insufficient coal-fired generation to support even the night-time load. So even at night, spot prices are set by open cycle gas (Mortlake and Newport). According to the Aneroid Energy website, there are currently only 3,700 MW of coal-fired generation operating in Victoria, with Loy Yang A4 and Yallourn W S2 not operating (with a combined capacity of 900 MW). Night-time load is about 4,200 MW. When Hazelwood was operating, its low bid prices would feed through to low night-time prices in SA and Tasmania. These low-price periods no longer occur.
Another new feature is AEMO’s constraint of wind output in SA to 1200 MW if there are only 3 gas generators operating on the 275 kW network. This has led to curtailment of ~200 MW on recent high wind days (during both night-time and the morning peak), and good prices (~$70/MWh) for non-curtailed wind farms, whereas previously at times of high wind output and low demand, prices would become negative. So again, no tail of low-priced periods.
I would have thought that anyone familiar with electricity costs would already know that only the full system cost is a valid measure. Why does the matter need to be raised? Obviously because there are many cost claims circulating that are simply fake. I hope the next article exposes them. And by the way it is not valid to ‘factor in’ externalities like the damage cost of carbon emissions by adding them to the LCOE. The ‘externality adder’ has different purposes, especially in helping decide what are the best investments to make in reducing damage.
“..it is not valid to ‘factor in’ externalities like the damage cost of carbon emissions by adding them to the LCOE.”
Well it can be that we choose to impose a CO2 tax as a truer social cost on the system provided it bears equally on all tenderer’s CO2 inputs but the level of tax for such a level playing field is one of considerable conjecture.
I’d argue that also bears implicitly on your your other observation- “I would have thought that anyone familiar with electricity costs would already know that only the full system cost is a valid measure”. That’s why I’d argue you can’t possibly have a full system cost allocated fairly amongst all the tenderers without the level playing field of despatchability as well and so far that’s not the case. Finkel recognised that with his recommendation about despatchability for any future unreliables but I’d suggest he was politically silent about all the existing ones for the obvious. Too many red faces admitting they got it wrong.
Lets assume a very simple system and look at the effect of the RET
In the country of Utopia electric power is produced 50% by wind power and 50% by black coal over a 12 month time period
.
Black Coal produces electricity at $55 per megawatt, but due to the RET they have to pay the wind power producers $85 per megawatt, for a clean energy certificate, so their cost is $55 + $85 + 10% profit, + $154 per Megawatt.
Wind power gets the going rate at that time plus the $85 from the Black Coal Generator.
Thus the power end users Manufacturers , Commercial users and Householdsare paying higher prices than they did previously
Wind powers producers are very happy because they get paid when ever the wind is blowing and then get paid again by the Black Coal Electricity Generators
Not what I would call a ‘level playing field’
Please show me where me I am wrong, it is “doing my head in” and my Electricity Bill is just going up and up and up and up.
A timely article, Paul. I am wondering if anyone has done any research comparing price forecasts with actually observed spot prices. As the price pattern changes, forecasts based on historical trends should be getting less relevant
In terms of mitigating the “Crisis” in both price and supply what can be done to rectify the inadequecies of the current process?
Should existing coal fired stations be preserved until the process is improved and the alternative system in place and proven?
I visited the AEMO site today, and screenshotted what i saw. Over $300/MW in NSW! (25/07, 17:45) The other mainland states were not far behind! I certainly hope this does not become a trend! There didn’t seem to be any rhyme or reason for it. Generation figures appeared to be about normal?
The bottom line is, if governments leave critical infrastructure to the market then being free it will move in any direction. Mostly up to protect and reward investors.
The nuclear argument is now essential to the future of our electricity network as it appears ‘new coal’ is almost as unacceptable as nuclear.
Paul
Thanks for the post – as usual topical and to the point
The 6th point made by Sims was about the “loosening”of the network regulation rules about 10 years ago to maintain or boost government revenues. We are still living with the result of this and will do so for many decades to come unless something is done.
But we also need to reflect that the electricity wholesale market is becoming less competitive and this has allowed the wind and coal fired generators to sell their product at prices twice what they used to sell it at profitably, and the benefit of the differential between the cost of wind and coal fired generation and the selling price is not going to consumers
Gold plating would be going on with events like the Victorian bushfires and the SA blackout and mandated compensation in mind. Politicians respond to more quickly to the outraged twitterati.
Had an example of kneejerk compensation when the BIL enquired as to whether I had had a blackout in suburban Adelaide like he had but couldn’t remember having one but the $450 compo cheque said he did. No for me but then I reminded him there’s been a large country blackout including his shack at Mannum and that was it although he regularly enjoys positive cheques from his solar FIT there given its useage.
As for gold plating in general does it also have a lot to do with voltage and frequency management of widely dispersed variable power sources and can they even distill out those costs? Are there any estimates of that around or does it go deliberately unanswered through political censorship of enquiry?
It is unconscionable that the world should be turned inside out like this…but for Australia itself the potential for existential damage is of a magnitude that dwarfs that for all the other first world countries …because we have no alternatives after the Left kills fossil fuel.
Every other first world country has base load alternatives such as massive hydro and or nuclear and connectors to other countries to fall back on in the case of energy insecurity…Australia has none of that!
Australia will be ALONE…out on a dangerous limb as this RE juggernaut demands it commits suicide….the only country in the world being told to commit economic and social suicide !
For the amount of money the ANU’s 100% RE electrical system for Australia plan would cost….all of Australia’s coal and gas plants could be progressively replaced by HELE coal plants and any better versions to come…and there would still be money for the upgrading and/or replacement where necessary of the Australian grid.
There would still be a cut in emissions that’s demanded ….augmented by the elimination of the fugitive methane emissions from the CSG mining that would no longer be necessary.
Any Australian business people who are not deeply secretly invested in the RE juggernaut would have rocks in their heads if they let this insanity go one step further….especially as they should be au fait with the complete unsustainability of it all in the long-term with regard to EROI.
Unless of course they just don’t care about all that…they’re just in it to make their money quickly before it all hits the fan and Australia goes belly-up.
As I understand it, 4 gas generators are required to be online for Wind to be allowed greater than 1200MW but I believe Osborne is not “counted” as it is only connected to te 66kV network. But it does seem surprising PPCCGT is only classed as one unit for this purpose? I cant find the actual exact requirements for Wind to be allowed greater than 1200MW. Is it documented somewhere?
Given PPCCGT has been running at full load for the last month (2x GT and 1 x ST) your point is not relevant. If you did some research you would know Engie announced it was returning both units to the market a few months ago.
The next question is what AEMO are doing to update this constraint, given that there are another 3 large wind farms under development. They are supposed to follow processes that involve contestibility, but this appears to be an emergency procedure. Can the new big battery gain accreditation to be part of the “3 generators” rule ? Does it have enough synthetic inertia to count as 2 generators ? Can batteries in the new Riverland Solar Farm be included based on their synthetic inertia as measured at the 275 kV network ?
AEMO have also been developing plans for a new 275V AC “Horsham link”, but again there doesn’t appear to have been any contestibility of approaches. The existing AC link via Heywood is limited to well below its 1150 MW thermal rating because of frequency issues, so would a better investment be to upgrade it to DC between Tungkillo and Heywood, which could give a capacity of ~1500 MW with wire return or 3000 MW with earth return ?
There are currently 440 MW of new wind capacity underway – Hornsdale 3 (109 MW), Lincoln Gap (212 MW) and Willogolecha (119 MW). Perhaps one saving grace of these new ones is they are more exposed to easterlies than many of the older wind farms, so while they will produce during the westerlies of low pressure cells, they will also produce during the easterlies of highs, provided the high pressure cell is centred far enough south (as it normally is during summer).
How on earth can thermal possibly plan and predict their costs facing a wind sector varying between 3-85% of installed capacity we typically see here- http://anero.id/energy/wind-energy/2017/august
Feast or famine and like Pelican Point they’re supposed to be able to enter into cheaper long term contracts for gas. When they couldn’t possibly offer that to the new gas field developers to defray their investment risk, there were plenty of Asian buyers only too happy to fill the void and leave the locals to the vicissitudes of the spot market. Laughing all the way to the bank now with global gas price rises.
As if that’s not hard enough, now we have the SA Govt throwing taxpayer funds at a Tesla battery and diesel generators that will ostensibly be creaming the top off any power price peaks further compounding their planning and contractual input dilemma. Never have I seen so much folly with lack of comprehension of the big picture here and the impending disaster.
There are parallels here with the NBN thought bubble. Utter Groupthink bedazzled by the chimera of ultimate speed and totally oblivious to the elephant in the room with bandwidth/congestion costs and now they’re doing the same again with power despatchability as they’re mesmerised by the lure of ‘free’ power from Gaia. The Devil’s all in the despatchability and he’s about to claim his dues.
Metro7 comment on 1/8/17 deserves repeating………
“It is unconscionable that the world should be turned inside out like this…but for Australia itself the potential for existential damage is of a magnitude that dwarfs that for all the other first world countries …because we have no alternatives after the Left kills fossil fuel.
Every other first world country has base load alternatives such as massive hydro and or nuclear and connectors to other countries to fall back on in the case of energy insecurity…Australia has none of that!
Australia will be ALONE…out on a dangerous limb as this RE juggernaut demands it commits suicide….the only country in the world being told to commit economic and social suicide !”
Paul
Thanks for the post – it started me thinking and developing some thoughts I already had. The argument is that the current high wholesale price for electricity is being driven by wind (see above post) and gas.
But where is the benefit of the lower cost coal and wind reflected in the wholesale prices? Using NEMReview v6 I drew up a chart on a quarterly basis of the last 4 years of the total contributions of black and brown coal, hydro, wind and gas and matched these to the total NEM demand. (I tried to paste this into the dialogue box but it won’t paste!)
Interestingly the share of wind and coal to the total consumption have barely changed, ranging from 77 to 84% and the last quarter being 81%. The contribution from wind has also been a fairly constant proportion too ranging form 3% to 5%.
The cost of coal fired electricity has not moved much over the past 4 years but the cost of electricity has doubled. While I agree that the cost of gas fired electricity has increased, the volume of the gas generation hasn’t risen by much but the futures prices for electricity now seem to be based on the current high cost of gas as is the spot price.
So where has the benefit of the low cost coal and wind fired generation gone? The 2017 annual reports from Origin and AGL both show significant increases in contributions from their electricity sectors over the past 12 months.
So maybe he NEM rules need to be reviewed because they may be the cause that has allowed the gentailers to gain significant benefits.
A friend has been helping a medium sized business plan for their energy needs. Despite the fact that they are in Outer Metro Adelaide it appears that their best option is to buy and operate number of diesel generators for n-1 redundancy and disconnect from the grid.
As has been observed previously, “We should start by noting that no-one actually KNOWS, in aggregate, how much small-scale solar PV systems are injecting into the NEM (or indeed in any grid).” (https://wattclarity.com.au//2016/06/the-opacity-of-distributed-small-scale-solar-pv-output/). Given this, how actual are the ‘NEM-wide actuals’? Isn’t this this more just one model converging with another?
There seems to be minimal innovation in retail pricing agreements. For example at present there are substantial price declines in the futures market. An innovative retailer should be able to offer a lower price for say 2 years, but with a switching fee. This way the retailer could purchase on the futures market.
energymadeeasy eh? Now I’m pretty good at maths but trying to feed in a service charge, various charging tiers for summer and winter along with off-peak and to cap it off solar FIT scheme credits and the retailer solar bonus credit and on top of that split charging a couple of times a year with price rises part way through the 90 day billing periods and it’s a hulleva task trying to compare offers.
Did it once with a pushy door to door salesman I finally convinced to leave me his you beaut pricing/savings and it was $10 annual difference over my non-contract price with timely settlement discount. Complete waste of time when no domestic contract will guarantee you won’t get a price increase anytime after you sign up.
I don’t know if Paul answered your query separately, but let me have a go if he hasn’t.
As you no doubt know, the wholesale market price is set each half hour at a price which draws on enough supply to satisfy the demand over that half hour (“the market clearing price”). AEMO calls the cheapest sources first and then goes up the cost curve until it has enough supply to match demand. The last generator on that finally satisfies demand sets the market clearing price.
As coal plants close (eg Hazelwood), gas fired generators are increasingly becoming the generator that is last on and sets the market clearing price. A number of moth-balled gas fired plants are being restarted in response to the changing supply situation and the higher prices provide sufficient price incentive to do so.
So as gas prices have risen, so has the price that gas fired generators require to turn on and respond to market demand. This is what is driving up the wholesale price – the rising price of gas. (There may also be some pricing uncertainty as well due to the insecure nature of supplies as coal exits the market and there is no resolution on a national emissions policy).
The coal fired generators and renewables (when they supply) are probably doing very well as their costs are not changing but the price they are receiving from the market is much higher than it has been over the last 4-5 years.
Addressing the gas situation (and landing a national emissions policy) would go a long way to addressing the trend in electricity wholesale pricing without having to tamper with the NEM rules, which in itself creates uncertainty.
Why is it so hard to find what the southern annular mode is doing on the BOM website?
To me it is the most important predictor of rain in southern Australia particularly in winter west of the divide.
The cause of the problem is vertical market integration. Emergence of 3 giant gentailers killed the competition. There should have been some type of “cross media laws” in place which shouldn’t allow generators to buy retailers.
Hi WattClarity,
I have spent the past 73 years thinking and trying to understanding how things work on the basis of the facts and the truth. The notion of “truth’ goes back to at least the Greek thinkers but on and off for the next 2500 years a lot of people have thought that the truth was a worthwhile goal in life. After the Scientific Revolution or may be at the same time as the Industrial Revolution these ideal seem to have been part of progress. Then came the Post-Modernist period and now we have moved onto the Revisionist Period. The “TRUTH” has gone – one must be “progressive” all facts if they get in the way of a point of view can be “reviewed” and adjusted. Take careful note and listen carefully about the augments about why one should vote “YES”. This rot started many be 50 years ago with advertising. We let people tell us that they product was the best. They all said that. One brushed that aside on the basis it was only advertising what would one expect someone to to say about their product. But letting such talk go unchallenged was dangerous. Look where we have got to. You seem to think that if you can show someone to be wrong with facts you might win an argument. That was once true but we now live in very dangerous times and such arguments don’t now carry much weight any more. The Unions used to talk about “Direct Action” they knew how to win an argument. May be now words are no longer mightier than the sword!!
At times when the spot price in SA is zero, is there no incentive for VIC to import this very cheap energy? Would this not raise the price in SA materially? Or does the wind supply in SA exceed the export capacity to VIC in these circumstances?
I watched wind generator output in Southern Aust in May and June and it was pathetic. Oh that’s right we did not have our laughable 100MW energiser battery. Thermal powerstations did all the heavy lifting and ensured reliable supply something which wind and solar will never achieve. I am all for renewable energy but let’s stop pouring taxpayer money into this sector. We are told the technology has improved out of sight so let the proponents put their money where their mouth is without sucking the taxpayer dry. The debate is really ramping up and so it should. The public have been conned long enough and had a gutfull of the climate change religion. Unfortunately many people and the Government have woken up too late.
How does your weighted spot price average track against similar measures for gas prices in Australia?
In Ireland there is a very tight linkage between the wholesale average price and gas prices. Gas prices explain 80-90% of electricity price variations, http://www.ey.com/ie/en/industries/power—utilities/ey-future-energy-and-consumers
Readers may be interested in an analysis I did on USE recently, the main conclusions being that it is a poor single/zero-figure measure of a complex issue, and that more wind power in SA will have very little effect on load-shedding:
Thanks for your link – I’ve posted a comment on it at your blog. In summary I agree that additional information on the results underlying the USE measures would be very helpful and make the ESOO much more useful to readers.
The trick is how to capture and harness this highly variable energy source through effective storage technology, and that’s what our best people are working on
It is regrettable that the energy price debate is inevitably tainted with loose statements. It is even more regrettable that “delivered energy” prices are discovering dizzy new heights. Having said that though, by my calculations, Australia CPI in that period is just shy of 2.5% p.a compared to an implicit 5% compounding increase in wholesale electricity prices. So the generalisation is not entirely untrue…
Hi WattClarity,
When someone asserts that something cause causes something almost always the logic is flawed. WC is correct to challenge Hicks in the SMH, however I would suggest the rebuttal is equally flawed. Correlations or lack of, are a good starting point in discovering how inputs alter outputs but one needs a good deal more than a positive or null correlation to draw a conclusion one way or the other. Hicks suggests the NEM caused the price rise while WC produces good figures to show many years passed after NEM was created before the price rose. But it is equally valid to ask were the NEM to have never been created in the first place would that have prevented the very recent rise? If the answer might be “yes” then it is not wrong to suggest that the NEM has caused the rise. Trying to analysis this matter any further one gets bogged down with the inadequacies of the English language to explain such matters in a few words. The single word “cause” without careful definitions in particular circumstances is always dangerous. Positive or null corrections are an interesting subject. Looking for the causes of tornadoes in the US, one might be impressed with the very high correction between a certain observable single parameter and the frequency of tornadoes. The number of tornadoes per year and this variable tack one another over a 5 to 1 range over the past 100 years. One could well conclude this parameter to be the cause. The variable was the number of successful persecutions per years for having illicit distillation apparatus. The reason for the close collection was that both were tied to US Federal spending. More money meant more Federal Agents and more small weather stations to record tornadoes. Corrections by themselves are dangerous, almost every different sort of food has been shown to be both bad for you and good for you with correlations of one kind or another.
I would suggest that the NEM can be blamed for many negative outcomes; however whether better outcomes might have come from a world without the NEM must be consigned to What If History.
Great analysis Paul. Based on your estimate, LGC spot revenue foregone for those wind farms is roughly $5.2 million! This, plus increased SA spot prices for consumers makes it an expensive constraint for those involved. Be interesting to see what a cost-benefit analysis of the (seemingly arbitrary) constraint would show…
If the wind farm owners are waiting for AEMO to change these arbitrary rules in their favour, they will be waiting a long time because AEMO is under so much other pressure from government. Instead they should contribute and fund a trial, possibly with ARENA support, as Hornsdale wind farm are doing to test the new big battery and the wind farm’s capacity to offer FCAS. It is only through the dedicated short term resources of a trial that there will be enough evidence to convince AEMO that a new set of operating rules will work.
At the time some of the more recent wind farm investments received financial close, there was an AEMO/Electranet project to increase the SA-Vic capacity of the Heywood inter-connector to 650 MW. Yet AEMO continue to impose a 500 MW constraint, citing frequency problems. So they haven’t delivered on their project. Any other contractor would be in legal problems for under-delivery.
The 1200 MW rule is perhaps a boon for wind farms, because although their volume is constrained prices remain strong, whereas without the rule the SA market prices would crash because of the 500 MW Heywood link constraint.
One would hope that with more batteries in SA to maintain frequency AEMO would relax the 1200 MW rule, and allow the Heywood inter-connector to operate at the promised 650 MW, or perhaps even its line thermal capacity of 1150 MW.
Wind farms are both downward and upward dispatchable. On short time scales, a curtailed wind farm can be confident of its ability to move its output up. Likewise for solar PV.
I couldn’t see it linked in the article but earlier this month AEMO published a report providing some background information and details on the modelling done to establish the system strength requirements that underlie the 1200 MW wind constraints, and the conditions under which it will constrain wind generation.
It’s a good read. So the SA system needs to have enough synchronous momentum + instant demand response + battery to cover the tripping of the largest single gas generator operating at the time. As more batteries are added, fewer synchronous generators would be needed to back up the others.
There was no modelling of a 100% renewable system on a sunny or windy day, where there would be sufficient ramp up from batteries and curtailed capacity that no unreliable gas generator needs to be “covered” by others.
I’d like to draw attention to two interesting things about the AEMO report. The first is that despite Kate Summer’s numerous analyses of generator dead band issues, there is no commentary on this issue, and no modelling of how the synchronous generators might respond with tighter deadbands, and the role that might play in limiting oscillations. Secondly, they refer to the possible role of synchronous condensers in supporting system strength, and that they will model this in future. It seems to me that these issues are closely linked. Whilst inertia picks up the immediate loss of a generator, additional power has to be added to the system and this can come only from the governor response or across the interconnectors (or maybe even from the wind farms, because under the modelled scenario they are running curtailed) If the governor response remains slow, it seems to me that wide oscillations will follow. The modelling indicates that they used benchmarking for the synchronous generators that was developed for the system black event. Does this mean AEMO continues to overlook this issue?
Link to Kate’s article here https://wattclarity.com.au//2017/03/fast-frequency-service-treating-the-symptom-not-the-cause/
The term “dispatchable” used to (by default) refer to the ability to “dispatch” units to run or not run on command. The term was developed when all generating units had constant access to fuel. The wind energy’s co-opting of the term and bastardizing it’s ,meaning by parsing it from the concept of system adequacy contribution was clever but immoral in intent and in practice. Private enterprise excused from morality is cancerous and ruinous to societies.
This is useful in understanding the constraint. While noting the lost LGC revenue above, I’m curious as to how many of the constrained half hours occur in the wee-early hours, given SA is notorious for having the wind blowing overnight when demand is at minimum. As such how many have actually prevented generation at what would have been low (or even negative) energy price?
Thank you for this very well explained article re FCAS. Who said electricity is dull. Like any other piece of communication, it depends on how it is delivered Great article.
Stefano
“The aim is to put in place a robust scheme that delivers good system frequency control and improves power system security and reliability.”
Overarching number one macro rule- No tenderer of electrons to the communal grid can tender more than they can reasonably guarantee (ie short of unforseen mechanical breakdown) 24/7 all year round.
When you’re putting together a very large goods or freight train that has to get from A to B on time and handle all the inclines along the way there are certain basic performance requisites you require when calculating the number of engines you’re going to hitch up.
It’s a lot like my car when it doesn’t start or I push the accelerator and it doesn’t perform or stalls. I don’t just sit in it hoping for something to turn up, least of all for the weather to change.
Thanks for the article Bruce, just one quibble in the advanced analogy I would alter the Transient Stability section to say:
“If an engine is decoupled from the train it will accelerate away from it because it does not have any carriages trying to hold it back. If you had a way of rapidly re-coupling it to the train before it got too far ahead, then everything would return to normal, otherwise the engine would escape (and probably crash!). In power systems the coupling is done by magnetic field between the generator rotor and its stator, “escaping” generators is known as pole slipping or transient instability, but if you can rapidly reconnect them before pole slipping occurs; the system is said to be transiently stable.
This should not be confused with circuit breakers suddenly opening causing the generator to unload while the turbine has steam passing through it. The sudden loss of load could cause the turbine and generator to over-speed with catastrophic results due to centrifugal force. Consequently turbines are fitted with over speed trip mechanisms which shut off steam flow to the turbine and release valves which exhaust the steam already in the turbine to atmosphere.”
Unfortunately text colour and cross-outs don’t translate to this comments field so if any of it is unclear I can forward it in a Word doc. P.S. I am a former power station operator so am more familiar with steam than electrical parts of the system.
Regards,
Shane O’Loughlin
Thanks Shane, I generally agree. However I should point out a sustained fault which doesn’t get cleared will also cause the generator to unload with likely catastrophic results.
Fortunately faults are usually cleared within a fraction of a second.
Thanks Shane, I generally agree. However I should point out a sustained fault which doesn’t get cleared will also cause the generator to unload with likely catastrophic results.
Fortunately faults are usually cleared within a fraction of a second.
While some analogies are useful in gaining various insights into the behaviors of a complex system like for example Relativity, this is because what is being describe is beyond our normal experience. Clever people grappling with such concepts can extend their sophisticated understand this way. However, I fail to see why one would want to help those who want to stick their busy beaks into matters which in fact they actuality have nothing to contributed. If someone can’t understand the the system dynamics of a grid using the normal explanations using the appropriate mathematical equations with the correct solutions I think it best that they stay out of the discussion. I am dismayed to read about what I call the democratization of complex technology. I have been lucky enough to spend a working life dealing with various complex technologies that the public has never tried to understand and I would hope it should remain that way. As the clever people of this world do their level best to design, build and run complex systems I fail to understand why they need to taking into account the thoughts and opinions of the what I would like to call the stupid people. There was, once upon a time, nothing more boring than a bunch of power stations supply a grid in the mind of most of the public. All people cared about was that that it should keep working and they did not have to pay too much for the power. While the genie has partly got out of the lamp lets not do any more to let any more out.
You are kind in ONLY suggesting The Duke of Wellington might share my views – there are others that I would not align myself with for other reason who also thought that the democracy can go to far. On a more technical basis I think your analogy would be fine for a DC grid. But with an AC grid there needs to be two degrees of freedom (voltage and power) and the linear motion of a train only allows for one degree. But, trying to shoe-horn your idea a bit further if you had an engine (loco) to pull a number of carriages (cars in US Nomenclature) and that that train set was coupled to a much bigger train set by a rather long and electric “tow rope” (that could take compression as well as tension) then that might serve to explain the stability problems of South Australia as far as the interconnections are concerned. But your analogy works well for the control of governors and the various types of feeds (1 to 5) of the inverters associated with wind turbines. Some loco just set to a fixed throttle (diesel injector stroke) setting, while some locos with their “cruse controls” are set to a certain speed. But some of the very long ore train had delays of more than 30 seconds on the their brake air lines – not sure how that can be worked into the story. And I am not sure how you can work the quadrature current into the train analogy when dealing with the excitation of a particular alternator.
Hi Ken, I generally agree with your observations. Part of the reason I didn’t dwell too much on the voltage side of things was because it doesn’t fit too neatly into the train analogy, and I also wanted to talk about system frequency issues. However, please stay tuned, I do have a voltage analogy in the works which will be published in the next few weeks.
Add in the solar and wind in generation please to the train . What sort of Engine /carriage will that be? a ship’s wind sail on the a special carriage ? and solar would be solar cells on top of an electric driven carriage?
Thanks Paul for a good summary and attempt to explain to often inexplicable. But you do raise two important points.
Firstly there is little doubt that private ownership can improve things. The operation of Hazelwood before and after its sale shows that private ownership improved its availability and extended its life after the SECV had condemned it to the scrap heap in the early 1990s.
Secondly, it is the loss of competition between generators that has the biggest impact on prices. What we are seeing now is the loss of dispatchable generation which is the prime provider of forward contracts. There have been attempts (eg by ACCC over the sale of MacGen) to limit the loss of competition in the NEM but state governments have sold the assets to get the maximum sale price rather than deliver a market where competition is strong.
We still have nearly 90% of mainland generation provided by wind and coal at costs of production in the $40/MWh yet the cost in the NEM reflects much higher levels. This is a function of the loss of competition but I suspect that it is also a function of the market structure which is not designed to address the generation mix we now have
That might be a topic for another Wattclarity article
No I meant wind and coal. These are the low cost sources of generation. Gas is setting prices at the >$90/MWh mark which is where most of the spot prices and futures prices sit.
I think this is a very even handed approach to a knotty question. I know I bring my bias to the table, I explicitly and frequently repudiate the market in energy, and I believe it was and can be seen to be a demonstrable failure. Alternate realities cannot be proven, so I have to fall back on belief, conviction that a public utility function approach with strong regulation would have delivered a better outcome. But then bring liddell to the table.. would the state have been able to decide to stop sweating the asset and re-capitalize? At what price of capital? at what risk to the lending rate which *is* subject to market process and has follow on consequences.
I know I’m biassed, but I’m holding onto my view right now: we needed innovation and we needed state intervention. What we’ve got is innovation but mostly undercapitalized and the state intervention has been a sorry saga of debt reduction and fantasy roles in the pricing war (Queensland government officials should hang their head in shame, having had substantive control over bidding all the time and simply defaulting to not using it)
So for me? its even odder. I want to agree with the News Ltd Journalist, but to different ends: the price rises were unneccessary but retention of Liddell was not, and will never be a good idea, in either state or private hands.
Seems to me you got your favoured State intervention aka directed innovation and capitalisation of a plethora of wind and solar generators via massive subsidy mining and the dearest power prices in the world. If the State was going to intervene like that and put an end to coal it made sense to flog off aging coal stations to private enterprise to squeeze the last drop of revenue out of them before closure.
The fly in the ointment with the cunning plan was always despatchability plus rising demand and concomitant high prices for alternative gas to replace coal’s traditional role. The Weatherill Govt has the simple solution to such oops! Roll out the diesel gennys and a show battery whatever the taxpayer cost and blame private enterprise for the price rises.
Paul, I look forward to reading this article as it evolves. No doubt you’ll get this completed before the villains even realise that root cause analysis is actually the first step they have to take.
Paul you constantly nail it on the head. However there is one big thing that I wish to add. The bill – 80% of bills that I see can be reduced. Yes, 80% of businesses are paying more than they need to pay for the same power. Why? its all too complicated and its not their core business. Well it is my core business and I am helping businesses save tens of thousands of dollars from the grass roots level. Happy to chat to you any day to discuss this mess. What character am I?
The debate over Catastrophic Anthropogenic Global Warming (CAGW) aside, once policy makers were in agreement about that, they had to ask the hard question- What does it take to replace a megawatt of despatchable thermal power with a megawatt of despatchable wind or solar power? That assumes that tidal and geothermal were way up the cost curve and the obvious rider to the dollar cost answer was how much CO2 would be saved anyway?
Instead we simply had massive wind and solar generator rollout with no thought given to their fickle and unreliable nature, but at the same time bankrupting thermal coal power (not so much bankrupting them immediately but drawing down on their sunk cost with no margin for depreciation and reinvestment)
It wasn’t a matter of ‘no thought’-it was with malice and aforethought…that thermal coal power has been killed off.
That was the pre-requisite…that COAL be killed off as a future electricity generator but be forced to hang around as long as needed so the RE industry can use it and gas to pretend that wind and solar can power a modern industrialised society.
By the time the whole world and the general population finds out that wind and solar …even when supported by batteries and pumped hydro…is unsustainable for the continuation of a modern society…the uber-rich generators will have made their squillions and moved on.
Ask yourself why Germany with infinitely more RE …trillions spent…nuclear power still…connectors to French nuclear power…can’t survive without COAL…and is building new brown coal plants even as they claim to be getting out of coal.
And then ask yourself what sort of madness is it that makes Australia…with no nuclear or huge hydro to fall back on…no connectors to other wealthy nuclear-powered countries… prostrate itself before the carpetbaggers and allow them to make us the ONLY first world country that’s offering itself up for sacrifice at the altar of a LW CAGW hoax.
Thanks Paul – I posted a bit of a response and a link to a blog post where I wrote about this, here: https://twitter.com/KetanJ0/status/917926993785536512 also included recently updated Australian data on climate science! Hope that helps clarify my comments 🙂
The public has been voting for the wrong politicians. This is because the global warming debate is circuitous and opinionated. The media indulge in conducting an endless debating club competition. And the reason for this is that we did not have a Prime Minster who himself or herself would have been firm enough in climate science and educate the Parliament and the public about it. A Prime Minister should invite climatologist James Hansen to give us a lesson or two. Lateline’s Tony Jones had an interview with Jim in 2009 but he didn’t mention it in Monday’s Q&A. That’s why we are not proceeding
The real question for the blue line (and conversely the red line perhaps) is- How did you come to believe in catastrophic anthropogenic global warming? The answer lies in the lemming like belief that you can defy a fundamental axiom of engineering, namely that you can build a reliable system from unreliable componentry.
The enquirer can be fooled into thinking that most people made their decisions on a rational scientific examination of the evidence and so we’re now running an electricity grid on e-motion.
Not sure that I agree.
The real problem is that we have had a decade of poor policy decisions by politicians of both sides.
They in turn were poorly advised by Government Bureaucrats who relied on overly optimistic forecasts of gas availabliity.
Just take another look at the Eastern Australian Domestic Gas Market Study undertaken back in 2013 and you can see that there were some heroic assumptions and forecasts about gas availability and price.
Energy Users raised their concerns but they were brushed aside.
Thanks Peter
I appreciate the difference of opinion and that, in all of these sorts of discussions, there is a high degree of subjectivity involved.
I agree that poor policy decisions (over 10 years or more, at state and federal levels) has been one of the main drivers – which is why I nominated our politicians as Villain #1..
In relation to your other points, I’d just note that I haven’t finished my 10 Villains list, yet.
Paul
Paul …I posted a reply laying out why I disagree fundamentally with the explanations for the confusion and train wreck …and with the whole RE takeover of our electricity system and the destruction of Australia for what I believe is demonstrably a CAGW hoax.
Is it because my reply was too long…it WAS long because there’s a great deal more to this issue than is admitted…or was it BECAUSE I fundamentally disagree…not just at the margins.
Could you let me know please for future reference?
“They in turn were poorly advised by Government Bureaucrats who relied on overly optimistic forecasts of gas availabliity.”
But when you believe all fossil fuels need to and can be completely replaced by renewables, why would you care a jot about gas availability, anymore than coal or oil? Caring about fossil fuel availability and the price of evil is akin to heresy for true believers.
“Energy Users raised their concerns but they were brushed aside.”
Ditto. When you’ve seen the light and the way via disciples like James Hansen and know Gaia’s endless bounty of wind and sun will provide the true path to the light on the hill, naturally you have to eschew such impure thought, lest you fall into the sin of skepticism. You’re not continually having these impure thoughts and concerns like we skeptics are you Peter?
You haven’t explained where Howard or that paper were wrong….in the quantities…or just the crime of using a fossil fuel?
If the latter and you are a true believer in CAGW…can you tell me why Australia alone in the world must be the country that commits economic and social suicide because the world warmed a small amount …over a period of 20 years at most…a tiny blip in the scheme of things….starting and finishing with two entirely natural warming events….and followed since then by an IPCC-admitted…recently-reiterated PAUSE??
Can you tell me why that tiny all-natural warming trend set within a steady slight warming over hundreds of years as the earth emerges from the Little Ice Age …warming that wasn’t initiated by a rise in CO2…the CO2 rise having followed the warming …can you tell me why these entirely natural events require Australia to commit hari kari …destroying our children’s futures ?
Can you tell me why the earth WOULDn’t be slightly warming …or did you …IF you’re a warmist…thinks the earth should have retained all that ice from the LIA…and be all iced up now and maybe cooling towards ICE AGE ARMAGEDDON
Exactly right Observa.
The RE industry and other true believers find it very easy to brush aside those little inconvenient facts like fugitive emissions of methane….which is variously described as having 80-100 times more warming capacity than CO2…is ubiquitous and very hard to mitigate…many old wells that have been shut down continuing to release it for decades.
Sierra Club at one point admitted methane was a much worse problem than CO2 but seems to have gone usefully quiet for the sake of the RE industry and the financial and other behemoths with all that investment at stake.
They can always let us know when it’s all too late…our children’s futures destroyed…Australia a 3rd world energy-poor hitherto energy superpower.
The RE industry finds it so easy to put the lid on these pesky truths because they have a completely compliant and protective LW MSM that’s always prepared to destroy a Prime Minister when it’s deemed necessary-to trash the people’s vote…eager to work their guts out to put a stake through the only legitimate PM’s heart in hopes of burying the corpse…anything to keep the heist going…the biggest money shift in history from poor to uber-rich…the millions rolling in and the populace in the dark…literally so before too long.
I’m becoming fed up with the endless prattle that spews out of print and e-media about climate change. I’m becoming more and more convinced that it is all about vested interests and greed. Take away the subsidies and watch the CE industry collapse faster than a house of cards.
Kogan Creek was originally developed in conjunction with US utility Southern Corporation until CS Energy bought them out. Swanbank E was developed by CS Energy without any JV partner.
It is also worth mentioning the Braemar 1 (450MW in 2006), Braemar 2 (450MW in 2009) and Darling Downs (630MW in 2010) gas fired stations which added significant capacity to QLD.
The overall utilisation of Liddell in recent years may seem similar to the early 2000’s but the operations of the units is quite different. Between about 2000 and 2010, Macgen normally ran two units at up to 515MW (following the turbine upgrade) with two units on standby, with more than two units only used during outages at Bayswater. AGL are running more units (when available) but have derated them to 420MW in order to improve their reliability.
Muja AB was retired by Verve (now Synergy) in order to meet a cap placed on Verve’s generation capacity. It was then brought out of retirement as project between Verve and a private company in order to increase non gas fired generation following the Veranus Island gas explosion. The problems were mainly becaue the boilers had corroded significantly when the plant was shutdown because it had not been laid up properly because no one thought it would be returned to service when it was shutdown.
According to the ACCC the breakdown of cost increases is…”made up of network costs (48 per cent), wholesale costs (22 per cent), environmental costs (7 per cent), retail costs (16 per cent) and retail profits (8 per cent).”
But of those ‘gold plating’ network costs and to some extent wholesale costs we have no measure of the added costs to the supply chain in managing voltage and frequency with the addition of a plethora of these disparate unreliables and we need to address that. Admiral Nelson holding a telescope to his blind eye would no doubt approve of the modus operandi here, not least with our Chief Scientist recommending all future unreliables be despatchable, but we’ll overlook the retrospectivity folks.
Nice to read a really high academic quality article – thank you. Only at the end do you mention what I think is the event that will cause the greatest trouble and that is when Adelaide and Melbourne are both extra hot. This is where better demand side management would help. And, by better I mean more carrots and less sticks.
The first paragraph in section 1 “A fundamental concept underlying the NEM’s design was … etc etc” outlines the intended outcome of the NEM and in this paragraph the writer implies that this dream can, in an ideal world, be accomplished. I would challenge this underlying assumption. I believe it is not possible to devise a free market framework where price signals will provide feedback that will deliver a stable and technically desirable out come. I think the free market economy is, for a great many processes, the best of a bad bunch of alternatives, but for making and delivering electrical energy to consumers is one of the silliest ideas I have ever come across. No amount of tuning and fine tuning will ever work – the whole idea stupid. The NEM is a process where a small number of people can make some money. From that point of view it works. But there are no other winners only losers in the NEM landscape.
Well frankly the penny is slowly dropping that you can’t build a reliable system from unreliable componentry. That’s a fundamental axiom of engineering and where in the world did so many bright sparks ever witness a precedent to the breaking of that rule and yet they charged ahead with active promotion of power generation at the mercy of sunlight and weather. If your first year apprentice mechanic hadn’t understood that axiom of engineering by second year you’d have to fire him for the sake of your customers.
Power is an essential service so like police and the fire brigade, if you want to outsource its provision you don’t go round accepting tenders that have fine print saying no one will be on duty at night or when the wind don’t blow or blows too hard, even though the tender price looks good.
Where do we have to go with a communal power grid for a level playing field? Well whatever the supreme guru thinks the price of CO2 should be to get the earth’s temperature back to the temperature only the supreme guru knows it should be is imposed on all the tenderers. Then they are only allowed to tender those electrons to the communal grid they can reasonably guarantee 24/7, all year round.
Yeah I know all those red faces that thought they could ignore that fundamental engineering axiom and have to tell all the mums and dads with rooftop solar that if they can’t guarantee their electrons they can keep them. That’s why even Chief Scientists duck such retrospectivity with the bleeding obvious.
If I read this correctly there is a reasonable chance that Victoria’s peak demand could reach 10-10.5 GW. Total despatchable capacity including VIctoria’s share of the Snowy is 9GW. maximum imports from Tassie, 500 MW which leaves 1 GW to be provided by demand response (150MW likely) local wind and solar (500MW possible) lmports from NSW zero, imports from SA possibly 350MW. reserve margin zero.
I would agree that the author may know very little about engineering. However what is so unusual about that in the current debate about how the NEM works? There seems to me to be a plethora of economists and bankers and spin doctors with virtually no rational professional engineering input. We would not have the mess we now have if engineers had been left alone to deliver cheap, clean reliable power.
Well try a chain is as strong as it’s weakest link. I’m not saying wind and solar power can’t be made reliable to contribute to a reliable grid system but at what cost and that was completely overlooked or deliberately ignored. Why were so many so stupid or culpable in doing that?
Allan
Thanks for the very useful dissertation. What the article does is to highlight that the sketched outline has a number of wrinkles that have to be addressed, and that addressing them might not be as straight forward as implied..
I am concerned that the NEG overlaid by the incipient 5 minute settlement rule change will have an even greater downward pressure on competition and lead to even higher prices.
Is a GT which can provide energy within 10 minutes a “reliable” source under 5 minute settlement? Is a battery that discharges within 5 minutes a reliable source under 5 minute settlement, noting that it then increases demand as it recharges? Is a coal fired generator with the minimum ramp rate a reliable source under 5 minute settlement?
The more I dig into the NEG and 5 minute settlement, the more I realize that the economists involved really don’t understand the limitations of the technologies that our electricity system needs to ensure that electricity supplies are reliable
“What the article does is to highlight that the sketched outline has a number of wrinkles that have to be addressed, and that addressing them might not be as straight forward as implied..”
Now while that wind energy sector can produce an annual average output of installed capacity around 30%, just ask yourself what sort of storage investment would they require to even guarantee 5% or 10% of their capacity 24/7 all year round? They haven’t had to but get to dump those peak outputs at almost zero cost onto the market and leave all the thermals to pick up the insurance tab for all the troughs. Nice work if you can get it particularly when you’re further subsidised to do it via cash transfers from Govt mandated taxes on your competition. I guess you could call it a free market and we’ll no doubt all be free of despatchable thermal power sooner or later.
Very interesting. The energy balance that Peter Farley makes is very interesting. The whole SA issue depends on probabilities of demand and generation events. The first is that Vic and SA demands are highly correlated. AEMO has summarised the outlook in terms of SA demands, plant failures in Vic (not in these terms) and limitations in solar and wind resources. One issue is that this is fine for one day and all going well but if you get a series of very hot days, then it only requires a couple of these and a plant event to make for a potential loss-of-load event. Any flaws in this logic
Good article.
The NEG proposal has a problem – around 85% of forward contracts are now done through the ASX24 futures, and bilateral OTC contracts are only a small part of the market.
The September 2016 system black was not due to a failure of contracting, reliability or dispatchable generation – it was due to system instability and lack of synchronous inertia.
However, the January 2017 problem was a market contracting failure. There was apparently a lot of customer load that had not contracted with generation, and generators were left undispatched because they had not been contracted.
Why would customers choose spot exposure rather than take out a contract?
I suspect that a large reason is because the $300 Administered Price Cap enabled them to go uncontracted without a significant spot exposure.
The Administered Price Cap has not been reset for many years. It needs to be re-examined, and re-priced to take account of much higher current short-run marginal costs.
The contracts for reliability guarantee may not require financial contracts, i.e. the derivatives. Dispatchable generators and load may sell their dispatchable capacity into the market in 1 MW packets for each 5 minutes dispatch intervals. This can be standard contracts tradable in secondary market. The challenge would be for AEMO and AER to enforce compliance. It would be a serious challenge to verify that a dispatchable generator or load have the dispatchable capacity available for each dispatch intervals and they have not sold more capacity than physically available. This verification would involve checking the technical and operational data records, not merely a reconciliation of contract volumes against physical demand.
It is possible, but I think it would be cost prohibitive.
Ian,
I think you are missing a few things e.g.
1) SA Spot prices compared to Qld for the last 2 Calendar years and 2017 so far yield the following:
SA_SMP QLD_SMP
2017 $108.99 $109.13
2016 $69.67 $67.44
2015 $49.58 $51.97
Which makes me doubt those who spout the economic benefits of coal.
2) . It appears highly variable and unpredictable because it is variable but predictability is improving according to AEMO at least. In any case there are numerous extended periods when the SE Interconnector flow is SA to Vic but it works as intended generally so we get the benefit of lower prices in both regions and the highly flexible gas fired plant in SA is more than capable of operating counter cyclically with the wind generation.
3) Reliance on the Interconnector has always been high but not nearly so much since the wind generators were installed and Victoria benefits considerably from SA exports at times of high wind generation. Far from SA not thriving on wind it does, and so does Victoria.
4) True up to a point but not nearly so much as in the past. The interconnector was initially installed to allow SA to benefit from “opportunity” power purchases from Victoria whenever their price made it economical. Since the NEM start quite a variety of scenarios (draught, plant failure, Basslink OOS etc.) have led to a wide variety of inter-regional power & FCAS provision outcomes.
5) The Heywood interconnector is subject to a lot of constraints for all sorts of reasons but only some are related to wind generation. Apparently Basslink’s availability and its import to export deadband can have significant impacts. Even deep-in-the-system Victorian transmission line, transformer and SVC availability (I believe) can have significant effects. It is even affected by outages in NSW & Qld but so are other interconnectors, though I suspect that SA’s small size and unique location at the far end of a much larger system has an impact as well.
6) I would be surprised if Tasmania hadn’t learned the lesson of the Basslink outage which occurred just after they ran their reserves very low. Remember all those diesels they had to import! I think many of them are now in SA for a couple of years but I doubt they will run now the ride-through settings on the windfarms have been corrected.
Ian,
I think you are missing a few things e.g.
1) SA Spot prices compared to Qld for the last 2 Calendar years and 2017 so far yield the following:
SA_SMP QLD_SMP
2017 $108.99 $109.13
2016 $69.67 $67.44
2015 $49.58 $51.97
Which makes me doubt those who spout the economic benefits of coal.
2) . It appears highly variable and unpredictable because it is variable but predictability is improving according to AEMO at least. In any case there are numerous extended periods when the SE Interconnector flow is SA to Vic but it works as intended generally so we get the benefit of lower prices in both regions and the highly flexible gas fired plant in SA is more than capable of operating counter cyclically with the wind generation.
3) Reliance on the Interconnector has always been high but not nearly so much since the wind generators were installed and Victoria benefits considerably from SA exports at times of high wind generation. Far from SA not thriving on wind it does, and so does Victoria.
4) True up to a point but not nearly so much as in the past. The interconnector was initially installed to allow SA to benefit from “opportunity” power purchases from Victoria whenever their price made it economical. Since the NEM start quite a variety of scenarios (draught, plant failure, Basslink OOS etc.) have led to a wide variety of inter-regional power & FCAS provision outcomes.
5) The Heywood interconnector is subject to a lot of constraints for all sorts of reasons but only some are related to wind generation. Apparently Basslink’s availability and its import to export deadband can have significant impacts. Even deep-in-the-system Victorian transmission line, transformer and SVC availability (I believe) can have significant effects. It is even affected by outages in NSW & Qld but so are other interconnectors, though I suspect that SA’s small size and unique location at the far end of a much larger system has an impact as well.
6) I would be surprised if Tasmania hadn’t learned the lesson of the Basslink outage which occurred just after they ran their reserves very low. Remember all those diesels they had to import! I think many of them are now in SA for a couple of years but I doubt they will run now the ride-through settings on the windfarms have been corrected. I probably just missed it but I don’t remember AEMO fessing up to their lack of diligence in that instance, maybe their lawyers wouldn’t let them.
Does anyone know when the other 7 (or is that 4) units will ne commissioned?
Somewhere the goalposts seem to have shifted from 9 units down to 6 units.
At whose instigation?
From Adelaide Advertiser 2 August 2017
Mr Weatherill has announced plans to lease, and possibly later buy, nine new turbines which can run off either diesel or gas, and provide the state up to 276MW of energy in times of emergency.
[..] Yogi Berra who quipped “you can observe a lot by watching” (he even wrote a book with the title). [..] and I made a little blog post on his passing in Sep 2015 ..
So many great ‘observations’ in this post. Will take a while to read, understand and absorb. As always, thank you for your generosity in sharing your experiences and learnings.
Seems like a lot of complexity to tell you what you must have already known really. Solar works best around the middle of the day, less well when its cloudy, sometimes it will supply more than i need , sometimes less. Its not clear to me how i benefit or what i would change as a result of converting basic reality to data and graphs. I must be mising something.
I agree to an extent, but have observed (over many years of user interface design for process plant operators, and frequent reading of other types of data presented in analogue, numeric tabular, or graphical X Vs Y or X vs T forms) that different people have widely different abilities to gain meaning or understanding from that data.
Thus while I might be like you and fairly adept at quickly interpreting data either by comparison with what we expect from our excellent understanding, or looking at a column of numbers and intuitively calculating average and variance, min, max etc, it remains true that there is incremental value in presenting it graphically.
Sometimes you will see learned patterns or changes to patterns that alert you to something significant. As the number of variables and data points increases, the ability of virtually zero cost additional flexibility in display the data in a variety of ways becomes nearly essential.
The trend graphs in Excel can offer even further insights especially if you resist the temptation with the polynomials to go to too many levels of complexity.
Paul,
By way of further analysis; the remark at: http://australianenergycouncil.cmail19.com/t/r-l-jlbjyll-hyltdykhjy-i/
that “Port Augusta was the ideal location for a hybrid renewables power station because wind strength increased throughout the day and it had a high number of days of sunshine…mainly driven by the temperature differences between land and sea…” got me thinking that with the advent of large solar inputs at the transmission level about to arrive on the scene, including National Grid Metering, perhaps some accurate analysis could be done of the level of cover provided by the intermittency of each e.g. if the wind strength grows at Pt. Augusta as the day progresses it could alleviate the duck curve considerably.
Cheers,
Interested to see your thoughts on this. I regularly see another bias at work – optimism bias – which is where people think that things that have happened in the past to others wont happen in future to them, without any genuine reasons for thinking this.
Last summer was quite eventful in SA, but Oz lived up to its name as the lucky country when a very hot spell, culminating in a total wind power around only 50 MW at peak demand, fell on Christmas day rather than on a full working day:
at first I thought that reduced residential PV installation data after the end of the RET would be a huge problem. I am a data geek too. but then I thought, PV is really a negative load (at least until it gets to a very large portion of the daytime load).
similarly we don’t have good data every time someone installs an airconditioner, but in aggregate we see their effect with the electricity demand increase on the very hot days. won’t cloudy days impact on PV be just the same as aircon on hot days but in reverse?
I fool around with numbers in the same way to get a handle on an issue. I contend there are two sorts of minds word-smiths and numerate minds. The world is mostly run now by and controlled now by the word-smiths because (1) there many of them who don’t have a worthwhile engagement in life and (2) their thoughts resonate with the pubic. But I can’t thing without numbers being a Professional Engineer and I believe no one else can who wants to make a worthwhile contribution to the current debate about electricity.
But to your topic. Most calculations suggest that cycling electricity in and out of a battery adds between 20 cents and 40 cents per kWh ($200 to $400 / MWh). Those figures assume the battery runs for its expected life at about 1 cycle per day. These costs need to be paid by someone and they are far from insignificant. Whatever the reasons are to have a battery these costs are real and to my mind dominate the other considerations.
Hi Paul,
Olivier here. Thanks for relaying our messages.
We definitely agree that the marginal should be used.
It’s actually already available at pro.electricitymap.org.
Will write another article about that pretty soon.
a couple of observations that I think will make a big difference to your analysis.
1. Most 100% renewable studies accept that if you have a week of really good solar (and/or wind), then you will likely be better off spilling some of the excess rather than trying to store every last kWh. Spilling 10% of your annual resource during the good times will only add 10% to the LCOE of the wind or solar, which is not an outrageous imposition. If you redo your calcs assuming a 10% overbuild, and assume that 10% gets wasted during the good days, then I think you’ll find that storage gets reduced considerably.
2. Our existing hydro is quite flexible, sometimes it puts out less than 500 MW, sometimes it puts out more than 4,000 MW. If you run existing hydro on average 1 GW harder during a week of poor wind and solar weather than you do during a week of good weather, then that reduces your storage requirements by 168 GWh, ie an amount far greater than the 68 GWh of storage that you came up with.
Having said that, your assumption to ignore seasonal differences in generation was clearly very generous. Put that back in and storage requirements increase massively. The AEMO and UNSW 100% renewable studies found that the NEM would need around 100 to 200 GWh of storage, though they both had essentially full back-up from dispatchable gas generation (converted to biofuels) that would run during the periods of poor wind or solar.
The recent ANU (Blakers) and the older BZE 100% renewable studies didn’t rely so heavily on this dispatcable back-up, and as a result they required much more storage, 400 to 700 GWh.
The Summer Operations 2017-2018 document released by the AEMO in November provides details about a number of actions that provides the 2,000 MW of additional resources: increasing capacity and maximising availability of generation and transmission system in the NEM, additional emergency non-market generation, ensuring sufficient fuel supply for generators and encouraging demand-side resources.
Please simply explain how the purpose of Big-Battery in ultra-short-term load demand, frequency and voltage-stability.
How even if you were designing a co-gen i.e gas-coal super-critical plant that you would add battery load-equalisation
“I will take the liberty of saying what I think the poster is getting at.
1. That there are solutions available to those Tesla battery design issues mentioned in the article, but the known solutions will kill Tesla’s range claim.
2. That there are three considerations when talking about batteries for E.V. (electric vehicle) designs. Range, charge time and battery life. The most effective solution for one of these considerations negatively impacts at least one of the other two.
3. That there are solutions readily available that solve any one of the three. Any potential method of solving two of them at once is very difficult and definitely not readily available. There is no solution available to solve all three considerations at once that have even reached the stage of a successful test of the theory in a laboratory. Which needless to say leaves a multitude of real world conditions that aren’t even subject to conjecture yet.
4. Unstated in his post is the oncoming calamity which is the underlying goal of some countries to have every single lawn mower, tractor, car, truck, emergency response vehicle, industrial equipment etc replaced by e.v. in the next twenty years. A goal enforced by law in the hope that a solution for some fundamental battery problems will be solved in the next couple of years when the rollout of the conversion will have to start.”
Good analysis.
But what amazes me how quickly good engineering management has been forgotten and replaced by market place substitutes. You write how no one in SA had bid for frequency control. This remark is akin to suggesting that on flight XYZ no one bit to put wings on the plane. These are not market place discussions – these are not things that come from “commercial’ decisions – common sense suggest certain things should happen or be done. If this ludicrous idea of leaving critical decisions to the market place is to continue then there has to be consequences of such omissions – like for example what happened in SA should have costed someone something like $50m. There are minimum technical requirements for reliable and cost effective outcomes. If all of this is handed over to the market place then there has to be the service guarantee with heavy consequences for not meeting objectively stated outcomes. What we have now in Australia is not more than 50% of the processes that are required in a completely commercialized market.
Telling one of my engineering friends that migrated to California 50 years ago about the SA event last year and his first reaction was who will pay the $50m that someone in California would have had if that had happened there.
The main problem Australia faces is the rules of engagement were written buy the guys that hate rules. It is as if we had asked the criminals to write the laws. Of course they would have stopped people going to jail and all the nasty consequences.
Our problems will continue for ever no matter how many inquires or investigations until (1) we stop asking people that have vested interest or
(2) we ask someone without vested interest but who know what they are are doing.
Because up to now all they people that know what they are doing are too close to be impartial and impartial people don’t know what is what having never run anything let alone a grid.
“There are minimum technical requirements for reliable and cost effective outcomes.”
Such as no generator can tender any more electrons than they can reasonably guarantee the supply of 24/7, all year round? Furthermore should every generator also be in a position to contribute to frequency control if and as required? Are some specific types of generators being given a free pass with those basic requirements now and if so why so?
OTOH don’t those 2 requirements really matter for overall grid performance and end consumer satisfaction?
Hi observa, the current FCAS rules allow all generators to get a “free ride” as in they don’t have to contribute to frequency control if they so choose. The market is meant to sort that out by price signals encouraging re-entry. However my observation is that the ancillary services market doesn’t work very well, and in my view it was a flawed idea in the first place. It is like setting up a market to determine which side of the road you should drive on, penalise drivers who drive on the wrong side and reward drivers who drive on the right side. Do this on the basis that the most efficient side will be chosen in all possible circumstances. If we organised our road system like this we would have more efficient transport ? or would we have chaos?
Hi Ken, I agree, good engineering philosophies are being overridden by dogmatic attitudes in many cases. In part that is because the engineering profession is not communicating very well but communication is a two way street and it is also the case that regulators and rule makers etc. are not listening. Maybe a few more failures need to occur for the points to sink in.
Hi Bruce,
There are many lessons to be taken from the story of the railways in Britain between about 1830 and 1870. In many ways what we call “capitalism” was born at that time with people buying shares in railways And then in the latter part of that period what was happening in the US. But the more important lessons come from Britain. It was at that time the ICE, the Institution of Civil Engineers moved to near Parliament because the story of the railways is a story of equal measure of Legislation and private investment. The railways wanted land in a continuous strip that could only be purchased by compulsory acquisition by Legislation and many conditions were placed on the railways in exchange for Parliamentary assistance in getting the land for them. And so capitalism in many ways was not born where there were no rules and money bought everything but in a partnership between government and private investment. And the monopolistic conditions of the grid are very similar to the monopoly of the a railway. Our government be it Labor or Liberal have no idea about how to write appropriate Legislation, so they have turned to people that do know what they are doing but they have serious vested interests and endless conflicts of interest. We Engineers need to get off our bikes and advise the various Ministers. I, acting alone, have done a lot of this in the past few years and in my long life I have never met situations where those I have tried to talk to are not even slightly interested in what I have say
Ken.
A nice increase from the low end constraint for wind from 1200 to 1295MW. I noticed the doc mentions Heywood still only modelled to 600MW VIC>SA. Not sure if we will ever see it opened to 650MW! SA>VIC still also not approved over 500MW. Hornsdale battery mentioned but only the market part of it (30MW). Perhaps with the January update the 70MW component (which bids into FCAS markets) will be able to contribute to system strength?
Not sure you can deduce the duration of the discharge from the 5-minute NEM data. Is it defined what the MW figures are exactly, average over the 5-minute period, or peak, or something else?
Well AEMO’s demand forecasts were more than 2,000 MW out on the high side (over 1,000 in Victoria alone), although capital city maximum temperatures weren’t too far from the BoM forecasts.
Similarly last Wednesday which was hot in both SA and Vic saw demand come in well below levels forecast in predispatch.
Either AEMO is being very conservative this year or demand behaviour has changed – possibly both.
Forecasting difficulties the other way for Victoria on Tuesday – at 2:30pm local time Vic max half hourly demand was forecast to be 7751 MW (although actual 5 minute demand was already something over 7600 MW at the time of the forecast). 90 minutes later the max demand forecast had increased to 8400 MW!
Interesting also that it is only the second December that we have seen values over 13GW (currently 13,010). Given that historically Jan and Feb have been higher, it could be an interesting summer. Off to fire up my BBQ and watch the summer with interest.
If there were more points of input, would it get harder or easier? If all wind sources from now on are required in (hypothetical) law to have associated storage would that help? I suspect that more points would help but complicate the models and mandated storage would only add hysteresis at the cost of requiring oversupply built into the model at almost all times.
SA may be an outlier in terms of wind power predictability because much of its capacity lies in the hills near Adelaide, and wind there tends to have a very large diurnal variability, which means that it cannot be predicted solely from the spacings between isobars, which generally don’t change much over 24 hours.
Theory Alert (I’m not a meteorologist): The diurnal variability arises from air rising and falling as it warms/cools, a phenomenon known locally as gully winds. I’m guessing that localised vertical air motion is harder to predict than wide area horizontal motion caused by pressure gradients.
The total wind power must be some sort of sum of the local effects and the synoptic effects, sometimes with almost exact cancellation, so even if both parts can be predicted to within (say) 10%, when they are in opposition the error in their sum can be relatively large.
We have done quite a bit of investigation in this area with some promising results. The difficulties lie in the non-linearity of the problem.
The power output of wind turbines is related to the cube of the wind speed, so small errors get amplified. Above a certain wind speed the turbine output becomes constant, and for very high speeds the turbines may trip.
These characteristics vary from turbine model to turbine model.
Local topography also affects wind farm output.
For these reasons, meteorological schemes may be an insufficient approach because they necessarily sample wind speed measurements at discrete locations and extrapolate to areas.
We have been experimenting with approaches which approach the problem from a different angle.
Very interesting Paul, it’s also an illustration of the relatively high correlation of windfarm output in SA, with so much capacity concentrated in the mid-North of the State (the Hallett group 350 MW, Snowtown 370 MW, Hornsdale 315 MW plus Clements Gap 58 MW and Waterloo 130 MW also close by), obviously similar localised meteorological factors can affect a large proportion of the wind capacity in the state near-simultaneously. That clearly puts a premium on gaining a better understanding of what these localised factors are and if possible incorporating them into AEMO’s forecasting models – or perhaps just adapting the very short term forecasting models to better track actual rates of change seen in aggregate wind output, which they currently seem oblivious to.
That said I think that your comparison of the reduction in WF output to a large thermal unit trip – which can happen near instantaneously – could be misinterpreted. The latter has system security implications in terms of the fast-response contingency reserve required on the system, whereas the reduction in WF output happens much much more slowly than that from the perspective of real time operations (the reduction in the half hour up to 13:00 yesterday looks precipitous but was actually only about 20 MW/minute). Of course any extended change like this typically has impacts on spot price, just as do the ramps in demand associated with morning, evening, and (unnecessarily) “off-peak” water heating peaks.
Thanks for the comment, Allan
And thanks for highlighting the confusion I inadvertently might have caused some readers. The comparison with a big coal unit was to provide a comparison of scale – though each event has its own challenges. A trip of that magnitude is unexpected and quick, leading to short-term system security implications as you have noted. The unforecast oscillations (increases and decreases) in wind output is somewhat slower and so represents different challenges in terms of gearing up something(s) dispatchable to wind up and wind back at the same rate.
Well personally I don’t think this macro problem can be solved until all generators are restricted to only tendering power they can reasonably guarantee 24/7 all year round. However we wouldn’t want to start that process from here.
In the interim should wind farms be compelled to install perimeter wind monitoring and advanced warning anemometers to aid short term generation predictability and if so at what sort of distance (radius) would that be optimal for thermal generators to benefit from?
Excellent graphics Paul and an interesting discourse. A couple of comments if I may:
George in the shorter (5 minute) time frame storage can usefully be used to meet enhanced (by technologies such as lidar, skycam and cloud motion vectoring) forecasts. AEMO is exploring this with ARENA.
Climan: yup the best organisations to do short term wind forecasts are the wind farm owners and operators because they know the characteristics of their site best.
Bruce: AEMO produce non-linear generation power curves for every wind farm but due to the effects mentioned by Ciman these are not simple – especially for some farms (refer last comment).
Allan – in this discussion we’ve had wind generation congregated around the Adelaide hills and the mid-north. In reality it’s a mix. We do see pretty fast ramping events (up and then straight back down again) but they’re as fast as asset trips – but often larger. Solar ramping events are hellish fast though. The problem is combining a weather ramping event with another, or a demand ramping event or an asset contingency. These combinations have drastically changed the risk environment for system operators.
Yep, this sort of event would certainly give AEMO much bigger headaches and have higher cost implications if SA were not connected to the rest of the NEM.
As things are, I imagine a lot of the ramping capability needed to handle these large wind output swings would be supplied from outside SA over the interconnectors – one of the benefits of interconnection and its diversification of supply/demand.
Even so I would think there is enough ramping capacity and fast-start plant in SA itself to physically handle this sort of event. But as Mike points out, the risk environment for market participants and system operators is heightened because they would need to factor in the chance of this sort of unforecast event in order for enough of that capability to be available.
Are you referring to the discussion around the response of the Hornsdale battery to the unit trip at Loy Yang A?
The battery is contracted to provide Frequency Control Ancillary Services (FCAS) to the NEM. When the Loy YangA unit tripped the battery responded to help maintain system frequency until other FCAS providers could also respond.
Am I on the right track? Can you provide more context to your question?
What you are doing is essentially applying logical thought and reason to a situation. The way things are done since AEMO stated has never been logical or even according to so called market forces. It is mish-mash undergraduate 101 economics, that is fundamentally flawed. Some things make some sense if you think of the workings of vested interest but not much.
Sorry to spoil a good conspiracy theory, Ken, but there are valid reasons why the prices separate. Too much to go into here, but to do with co-optimisation of 9 commodities (Energy + 8 x FCAS) across 5 regions simultaneously.
Thanks Allen,I,ve only found & been following Watt Clarity for the last 5 weeks or so and commend you on your contribution to enlighten us mer uneducated public on the workings of the national grid system and your unbiased accessments of the vaigrencies with in of its very intricate workings.Please keep your observation coming so some of us can glean some more enlightenment.
What conclusions do you draw from the trip at Loy Yang B? Do you think this reflects a broader reliability challenge or just an unfortunate malfunction?
It’s a great question Dean but one for which I haven’t assembled the evidence to judge either way.
What is certain is that large-scale thermal generation units – particularly coal fired ones – are very complicated pieces of machinery with many intricate subsystems. There are plenty of points of failure so it’s no surprise that across the whole coal fleet we do see unit trips and unplanned outages or capacity reductions relatively frequently. And as with (much) simpler machines like cars, it can be harder to keep them running optimally as they age.
But I haven’t seen clear evidence on trends, although I’m sure owners and system operators must be monitoring these closely.
Wouldn’t the reason for the drop in GT capacity be purely the temperature function? As the air gets warmer, you get less mass flow through a unit so the MW go down? For CCGTs, the MW would drop as the circulating water temperature for the steam turbine condenser went up
Hi Paul, great article. I work for Reposit Power and I would like to mention that Reposit takes no cut of the GridCredits earned during a market event like this. We charge retail partners fixed fees so there are no perverse incentives.
The $1/kWh is set by retailers and we are actively working with Retailers to introduce additional benefits. In a trial with one retailer we already offer the customer the full spot price which is how some customers were able to earn up to $37 in a single day just by trading excess solar. We are also looking at other incentives such as reducing the daily access cost to add benefit year-round.
It’s also worth noting in this arrangement that the customer takes no risk – the $1/kWh is paid to the customer regardless how the 30 minute period settles. For example, if the 30 minute period settled at less that $1000/MWh then the customer would still earn their $1/kWh if they were called upon during this time. Due to this risk I imagine we will always see some difference in what the consumer earns in these events vs the spot price.
Reposit is working to empower consumers on the grid and compensate them accordingly. Whilst is not full spot price it’s better than they can currently earn otherwise and it’s better than what they would otherwise buy for. Watch this space – increased incentives will be coming, it just takes some time.
For the demand revision downwards on Friday, you’ll see that there was a “cool” change in VIC from 2:30pm local. Further AEMO also called ~140MW of RERT which had the impact of reducing scheduled demand (compare the demand_and_nonschedgen and dispatchableload in the intervention/what-if cases – there is no change to scheduled demand).
Just to clarify further for anyone puzzled like I was:
– data published in the “Intervention” case represents how the market was physically dispatched, including the effect of demand reductions in Victoria called by AEMO under its RERT arrangements
– in the “What-if?” case AEMO estimates how the market would have been dispatched without intervention. Crucially, this “what-if?” case sets the market prices.
– you might expect that in the “what-if?” case scheduled demands would be adjusted upwards by the amount of load reduction called under RERT – representing what demand would have been without intervention. In fact the way AEMO deals with this is to leave scheduled demand unchanged and represent the RERT reduction amounts as additional “dispatchable loads” – equivalent to demand from pumped hydro for example. This causes the amount of generation + imports required in the “what-if?” case to increase, just as it would if scheduled demand increased by the same amount.
– this approach also makes the RERT load reduction amounts more explicit in the dispatch data. As HR says, you can see reductions of up to 140-odd MW, averaging 69 MW across the period of intervention.
Not sure if I agree with you about scheduled demand though. There is no what-if scheduled demand – it seems to me that for the duration of intervention AEMO considers the curtailed load of reserve traders is shifted from scheduled demand to dispatchable load – in the what-if case the reserve trader would have continued to consume (and thus there would be a dispatchable load for that period), and the intervention constrains them off.
I never was the greatest at the seemingly infinite definitions of demand… But you’ll see that there is a what-if demand_and_nonsched (which I consider to be the demand required to be served by transmission connected generation).
Yep we’re in furious agreement about this. Demand_and_nonschedgen includes the demand of dispatchable loads, whereas most AEMO definitions of “demand” exclude the dispatchable load component.
Pony Club?? – You’re confusing me with another Allan O’Neil I think.
Dave,
I think the purpose of the generators is to prevent blackouts, not to reduce prices and they are probably prevented from generating until just before an LOR3 event i.e. load shedding. if they were dispatched before this they would be interfering with the market and would upset the big boys making their killings on the hot days. You could say the heat would cause blackouts which would cause political heat and that is what they are designed to prevent; price is another matter. Sorry for the cynicism.
Hey Paul. Thanks as always. Any chance you could make it shorter – i would love to read this stuff but time is tight? Maybe take as read your frustration with the superficiality of popular commentary, and jump to the gist of your analysis.
Hi Allen,
You are a great communicator yourself – great writings, but I don’t agree with your criticism of the uninformed twitterites. If the information is kept to just the Experts then they can use all the jargon they like if that helps. But if the public is offered the same information in the name of openness then the responsibility for accurate communications rests wholly with the speaker not the listener. If the public don’t appreciate the differences between LOR1 and LOR3 that is the fault of those that use such dreadful jargon. It reminds me of NASA’s jargon with the moon project. They used that jargon for the same reason that scholars once spoke to one another in Latin. It was to impress people not to communicate. If the this industry must speak they way they do then the fault in communications is 100% with industry. Courts are open so that justice is seen to be done. Judges don’t talk to the Jury in long words they speak so they can be understood. This industry if they speak so the public can hear what is being said needs to gets their communications in better shape. I don’t think it adds to the debate to poke fun at how the public does not understand what is being said. The industry needs more folk like you, Allen.
Fair comment Ken, my dig at “instant experts” on twitter could have been better directed at people there – including those who know the subject matter reasonably well – presenting highly partial views and data.
On something like the LOR issue I certainly don’t expect the public to know the jargon or appreciate its nuances but the journalists reporting to the public could generally do a better job of informing themselves IMHO.
“But AGL’s director of operations Doug Jackson said the issues at Loy Yang A were unrelated to the summer heat but were driven by “not uncommon” problems that arose while restarting a unit after a major overhaul to ready it for peak summer demand.”
Thanks Rob
… but that was for Loy Yang A. it was Loy Yang B that tripped on Thursday last week – in which case the quote you wanted from the AFR article was:
“Alinta, which earlier this month completed the acquisition of Loy Yang B for about $1.2 billion, said an unexpected trip of one unit on January 18 was rectified and the plant supplied critical energy into the market during the hot spell last weekend.”
Surely the extra trips of coal plants are due to the fact that the deeper the penetration of intermittents into the NEM…the more coal plants are relegated …by the artificial favoring of intermittents in the merit order for dispatch..coal plants .made to operate outside the baseload mode they’re designed for…as MIT and others have reported.
The extra cycling and ramp-ups damage coal plants and make them less efficient…causing them to need more frequent maintenance…which generators are no doubt reluctant to pay for with their plants on death row anyway.
AEMC has also said that intermittents damage the grid..increasingly as they penetrate further into the market.
The wind and solar enthusiasts are deliberately trying to make Australians believe that all these years COAL has been an unreliable power source when that’s not true.
They MUST KNOW that the system that gives an artificial advantage to intermittents causes coal plants and the grid to be damaged .
Why are they allowed to deliberately deceive the Australian people?
Just curious … Is this forecast is using the 50POE curve from AEMO? Just looking at the numerical weather forecasts for reference weather stations, I think we might be heading for a 10POE day on Wednesday. If – and that is a big if – the weather forecasts holds and the forecast thunderstorms stay away, we might start nudging the 10,000 figure. (AMEO operational demand forecast for Wednesday is 9954 @ 17:00hrs on 14/02).
There are some scary-impressive numbers being generated by AEMO’s POE10 forecast for tomorrow (Wed. 14 Feb). There is a value of 10,115MW @ 17:00 and 10043MW @ 16:30. In the Energex DAPR, they provide a 50 POE reference temperate at Amberley of 30.65C ([max+min]/2).
The reason I keep harping on the 10 POE is that private numerical meteorological forecasts for Wednesday has a forecast average temperate at Amberley of 31C.
A comment out of the Energex 2016 DAPR is interesting. “Importantly, while air-conditioning load growth is beginning to flatten, the sensitivity of the significant >>latent<< (my emphasis) air-conditioning load already on the network remains a challenge to manage.” Out of the 2017 DAPR this comment “Air-conditioner sales have been higher this summer indicating customers are more likely to update or add to the air-conditioning load in their homes in response to the hotter summer.”
And that is the key to what we are seeing. There was a gap in hot summers from 2010/11 to 2016/17 that coincided with the classic peak-demand time of three hot days in a row, in the middle of the working week, in February. The heat has returned with vengeance, and so too has the load.
Perhaps those customers who did run their air conditioners last summer were pleasantly surprised at their electricity bill, or have become re-addicted to the comfort of air conditioning.
Evidence suggests that customers have a “break point” with the heat. When it gets too hot, they will say “stuff the cost” and turn on their units.
Further, the more affluent who also have both solar PV and A/C, also appear to have a new behaviour to run their A/C hard, offsetting late afternoon and evening kWh consumption against export income during the day.
The peak demand is definitely an interesting challenge to manage. If the weather forecast holds, Wednesday will be interesting to watch as well.
I find it very interesting to read remarks about “managing peak demand”. Remarks like this reminds me of Soviet type think “managing” meaning “controlling”. Supply and consumption of electricity since Edison’s time has been on the basis that the consumer sets the load, rather like a customer enters a shop because THEY decide – not the shop. I think we need to stop talking about managing. Think about the structural steel in a building to resist a storm-wind that might blow at the design speed for 10 minutes every 100 years. At all other times there is a capital investment in steel that provides no tangible return on investment. Those involved in power designs need to get over worrying about peak to average ratios. Many industries have capital investment that is not needed for most of the time. Just spend the money !!!
Managing peak demand has been around in the Queensland electricity industry for a very long time. The Brisbane City Council Department of Electricity introduced hot water load control using the Zellwegger system at Holland Park substation in 1957. Yes, over 60 years ago. (I believe the Sydney County Council introduced the same system a few years earlier). At present, there is approximately 1000MW of customer’s load in Queensland connected to Zellwegger demand management schemes. Over the decades, using the scheme has saved system operators from having to do load shedding during generation deficiencies. AEMO’s forbear (NEMMCO) once had the hot water systems in Queensland “instructed off” for an extended period during a significant generation shortfall.
Back in the 80’s, I did the field install for the under-frequency fast hot water shed. (A system that is still in operation today.) I doubt a single person notices when their hot water gets load shed. But everyone certainly does when the lights go out. Ironically, because this demand management system has worked – that is, it has prevented load shedding for standard lights & power – the concept does not attract media attention, and people assume that it does not exist.
I see little difference between centralised demand management and other flexible purchasing arrangements. For example, if I want to purchase a Qantas ticket from Brisbane to Sydney tomorrow. I can purchase a ticket for a 5AM flight out of Brisbane for $177, or I can purchase a 6:30AM flight for $490. If I can management my demand, I can save 64% of the price. Ditto cheap movie tickets on Tuesdays, and so forth. This is free market economics in action, straight out of the textbook for first year Uni students. There are options available for both buyers and sells, and both adjust their behaviours (ticketing prices and willingness to purchase) based on supply and demand. It does not bear even a passing resemblance to centralised control “Soviet” economies.
The thing that I find damn irritating is the collective social amnesia – most annoyingly from senior leaders inside the electricity industry – that parts of Australian electricity grid has a world class demand management system that works. Instead, the industry is accused of being behind the times and that we should do demand management by TCP/IP, or something like that …
Looks like QLD schedule demand peaked at a new record of 9928 MW today.
But returning to Feb 12, it’s worth noting that while scheduled demand peaked at 18:50 (9550MW), once you include the effect of rooftop PV, you can see demand (scheduled + rooftop PV) actually peaked at approx 9835 MW a couple of hours earlier around 17:00. So rooftop PV reduced the peak by about 300 MW, and moved it back a couple of hours.
Well if you keep adding quarter of a million new consumers into Australia each year and you don’t build anymore base load coal fired power stations there’s a certain inevitability about all this. Well given the low hanging fruit of power efficiency and economising has largely been plucked due to hefty price rises.
Fear Not!!!!. South Australia is showing what a Battery can do to keep things in check.
Yes a 100MW one, that does not produce power obviously but stores it. It actually takes like 10% more power to fill it and then it also has 10% less than finally got into it come out. Thats if its not hot, if its hot, well then it depends on how hot, as to how much can actually come out!!!!!.
Anyway, if you start to think about a battery being used in a peak period of high demand for a long time, its, well actually useless.
In fact the whole setup will actually need power, so its kind of good most the time but in peak times it adds to the demand. Bit of a over sight. However in the real world thats reality.
there is plenty of capacity between 8pm and 6am for EV charging. I have been charging my EV by timer each evening sometime after 8pm for 5.5 years. It’s trivial load shifting and works for everyone as long as their car is topped up by when they need it the next day.
Trivial, until all of brisbane wants to charge their EV at the same time.
Then I doubt the wires that have been run and the transformers that are in place will be enough.
But Thats for the future.
Got to get there first.
Maybe users of EVs can assign control over the timing of their charging to their retailer, the retailer can stagger the charging of EVs through the night so that they don’t get every single EV in Brisbane being switched on at 8pm.
I daresay most people won’t need a full charge every single night.
Fair comment, but there is a LOOOOONG way to go, as we also need to look at heavy EV’s too.
Meaning trucks and buses.
They will need to also be charged.
To replace fossil fuels is going to be a tough gig.
Mind you, Tesla was a leader in wireless energy transfer, so maybe we will not need to charge anything ever, just get charged as we drive.
Curious that Russell is able to come to a conclusion as to what is the most “cost effective” option for emissions reduction and dispatchable power without mention of the costs involved in the solutions he proposes and no comparison of costs of the alternatives.
Thank you Russell for an interesting and thought provoking article. Your exploration of the possibilities for improving the performance of the existing coal fleet is great. Don’t we need to estimate the cost of these improvements and compare to other options before concluding that it would be the most cost-effective, dispatchable, low-carbon solution? Internationally new gas fired plants are the common dispatchable companions to high penetrations of intermittant renewables. Gas emissions are around half that of even a finely tuned coal plant but still not low enough to meet CO2 targets in time. Going even further, a genIII or genIV nuclear plant would cost more but produce zero emissions and greater output. New gas or nuclear plant would have service life far in excess of reworked coal plant. An analysis of total lifetime cost, emissions reduction, dispatchability and other key parameters would determine the most cost-effective way forward. Investing large sums in coal plant mods Australia-wide could be akin to rearranging the deck chairs on the Titanic both economically and environmentally.
Or we could burn wood in coal plants, like germany is currently doing.
Or as South Australia has just done, buy big diesel gensets!!!!!!!!(Yes I know they also got the biggest battery, but really, does it actually do anything?, except use power)
Interesting article, but it seems to assume that all power generation is equal. As long as we can balance generation with demand, then all will be well.
Hugh doesn’t take into account frequency stability. A key factor in the SA system black event was that SA wind generation couldn’t stay in sync with VIC, once SA lost a large amount of wind generation.
The Heywood interconnector tripped because the frequency of SA generators fell below that of VIC, not because VIC couldn’t supply the load.
What followed was a rapid sequence of under frequency load shedding events, as transmission systems tripped off the remaining renewable generation that was unable to remain synchronised.
His conclusion that “the value to the system of large, inflexible coal-fired generators will shrink” seems to miss this point.
There have been some interesting articles on WattClarity about how AEMO now manages the situation when the total SA generation becomes mostly wind. Following the system black event, AEMO can now direct gas generators to come on line, so there is a minimum of synchronous generation to ensure reliability (for example: 2 x Pelican Point generators, or 1 Torrens Island and 1 x Pelican Point generator), even though there is enough wind generation to supply the load.
The value of large thermal generators is that they stay in sync, even when subject to a large changes in load, unlike wind generators.
A great article all the same! I enjoyed reading it.
Thanks Hugh for a nicely written and factual article. I wonder why you omitted the inconvenient truth behind the system black event in SA. In this case AEMO expected a strong contribution from wind turbines, but they shut down unexpectedly (tripped) because of multiple short circuits to ground (aka lightning).
The operators of traditional generators and distribution systems have a social obligation to increase output, often under duress, to supply their customers loads. On the other hand operators of variable renewable energy generators are only obliged to maximise the revenue available to their owners and place protection of their assets ahead of supplying the load. Consequently SA is now a circus ring for spruikers like Elon Musk.
I have great difficulty getting my head around much of the commentary on the electricity system in Australia. I could follow much of the sentiment if it was 1925 and the grid and all the vagaries associated with the “new” electricity were not yet understood. But the system is now a very mature discipline in 2018. Mention was made about the high temperatures for example, with the implications that there were no materials capable of working at 50 degrees C or above. We (humans) might find 50 degrees hot and we do turn on our air conditioners more. But electrical systems which over heat under these conditions are themselves wrongly designed or maintained. Class H insulation work well above class A insulation of cotton-covered-wire. It is not 1925 !
The public can reasonably expect to have electricity for a good deal of the year. They can also reasonable expect to be the sole determinant of the load – just as a customers chooses to go in to a shop to buy something – solely when the customer decides – not the shop.
As the country transitions into a period when more and more electricity is provide from renewable sources they also a right to expect that the transitions will be done by professional engineers who typically get it right first time and hopefully on budget. Sadly this is not what is happening. We have policy written by people who are self styled experts who often have no idea of what they are doing. We have commentary offering derision of the public that are asking for nothing more than it works. Of course when it falls short of reasonable expectations the public offer their advice which often is wrong. But the public are not the cause of the problem because they offer the wrong advice – it is the silly people that have got into running the show who are behaving like what happens when the asylum inmates take over the asylum that are the problem not the public.
Paul C writes that when SA was blacked out that the SA frequency fell below that of Vic. This is not correct. The Haywood inter-connector is not a DC link – it is synchronous connection albeit a weak connection. The two grids must at all times share exactly the same frequency. It is true that in times of instability that there will develop an unacceptable phase difference between SA and the rest of Australia but the frequency remain the same. If frequency is defined the reciprocal of a single period then may be the problem could be seen as a frequency difference but that is not the normal meaning of frequency. The same thoughts apply when thinking about frequency modulation and phase modulation. Over a very short period of sampling then it is hard to tell the difference between the two but they are not the same. The wind turbines took themselves off line because someone had written software that asked for this to happen. But the owners of the turbines and the AEMO had no idea this was the case. After multiple lightening strikes like what happens on most days in Far Nporth Queensland in summer the wind turbines threw the towel in. The system was never deigned to loose so much capacity so quickly and that triggered a text-book cascade fault condition, which involved the Haywood inter connector also tripping. After the inter-connector tripped with the systems now separated the death-throws of the SA grid obviously involved frequency collapse but not until after the separation.
It makes perfect sense that whatever power generation is currently in place at a certain location should continue, unless there is a VERY good reason to change. The fuel supply chain will already be there, as will the skilled workforce, and the power distribution infrastructure.
Paul, re blown LEDs, could I suggest that you check your mains voltage. I my case the mains voltage was sitting at 253 volts on winter days – but it showed a weak correlation with PV output. I wrote to my electricity distributor with a spread sheet of 5 minute average voltage readings during a week, and to my surprise they responded immediately. I was also surprised when they said they had to setup their own monitoring (I expected that in suburban Melbourne it might already have been monitored). Within a few days of seeing with their own eyes they changed a distribution transformer tap. The voltage is now right on specification and with very little variation over 24 hours (except for the notable air conditioners causing ‘fuses’ to ‘blow’.
A suggestion: try looking at cumulated data. A first cut would be simply to subtract the overall average from the daily production, and cumulate the daily differences. The difference between the highest and lowest points on the chart show how much storage is required if you assume that demand is constant and you are depending only on solar+storage. The next step in sophistication is to subtract instead daily demand, scaling solar to the same annual total, and once again cumulate the differences. You can then look at various percentages of overbuild. You should also account for the round trip loss in any energy stored – probably 25% is a sensible starting point. Of course, you need to add in the storage required to meet overnight demand on the biggest demand night of the year – that has to be in store at dusk.
I think you may find that the storage equation starts looking more formidable – particularly if you look to cover worst cases. The real killer is that seasonal storage only gets used once a year – not 365 times as in storage to cover overnight demand – which makes it 365 times as expensive. Worse still, if you want to cover a 1 in 20 bad year, the added storage you need to do that only gets employed 1 year in 20 on average – which makes its economics another 20 times worse. You can reduce the storage requirement by also including either other renewables such as wind, or varying amounts of dispatchable generation. You may be surprised how much you have to add back to deal with those pesky worst cases.
An interesting article. I think our national electricity market and renewable energy certificates go close to getting the mix right. I’d like to see the target higher but we should also consider efficiency improvements at existing generators, both fossil fueled and renewable. Continual improvement can only be a good thing.
Right exactly. That”s why it”s *NOT* Internet of Things. It”s just Connected Things On A Private Network. We used to do that with leased lines and AX.25. Big wup. Thanks for re-inventing SCADA. You are still tied to a single wireless technology, to a single MVNO, to using SIM cards, and your devices only talk to the cloud, not to each other.
Is this poor frequency control a reason why we get such poor performance from our inter-connectors ? For example, between NSW and Vic only about a quarter of the line rating is made available to the market, whereas between SA and Vic about half the line rating is made available. The market constraints are contained in equations that are derived by regression from records of frequency. I expect very few market participants understand what these constraints are. The business case for the SA/Vic upgrade was for it to release 650 MW of capacity in both directions, but this has still not been delivered to the market. How many other transmission upgrades might we make on flawed business cases, when the least-cost path to increased inter-connector capacity is simply to set the governors properly ?
To a small extent only-
I think you are referring to stability limits that would apply even if governors were well tuned.
The lack of frequency control does impact a bit on inter-regional constraints but these may not be fully taken into account in the existing constraint equations, which puts the system at risk.
part of the justification for interconnectors is the ability to share reserves between states thereby making the system more robust and cheaper to operate. The interconnector constraints are more related to contingency reserve sharing where a contingency in a state can be managed by sending additional generation from interstate hence interconnector capacity for use in the energy market is set at a level below the maximum interconnector capacity. This would not be the case if the interconnector was set at maximum capacity for the energy market. The interconnectors are basically working as per the business case.
Ron R
I agree, however if we have sloppy frequency control that also manifests itself as sloppy control of interconnector power flows. This has a significant negative economic impact on the delivery of power.
Great article Bruce. As a long term proponent on renewable energy and the benefits it can and will bring to the Grid I find it amazing that here we are in 2018, with all the technology advances that allow us blast a tank to nothingness using remote controlled drones, we still have major issues with power and the grid. At “watt” point will all these competing disinterests come together and devise a sustainable solution…that provides security at competitive prices.
This is one of the main issues holding back the introduction of renewable generation. If we had a more rational approach to frequency control I am confident we could successfully negotiate the technical obstacles currently holding back the industry.
Great article Paul unfortunately some of these companies have been disrespecting customers for some time.
It is amazing to see some of the assumptions made in their marketing. This is also the case for large commercial solar so companies are given incorrect information to make the business case look more attractive.
Try http://www.smappee.com for a real power monitor. Hasn’t missed a beat for me in 4 years. Class 1 energy meter that actually measures the voltage (unlike most).
Unfortunately energy pricing is complex and only understood by a relatively small proportion of the population. If an advertisement promises $10,000 per year return it is easy to see the attraction, and harder to understand that it may not be realistic.
Hey Paul, you are turning into a grumpy old man. Of course its complex, in electricity more so than in others. But the incentive to mislead is huge. Here is an idea, why don’t you keep a register of the most egregious claims. I am sure regulators and others would be willing to pay for that. Cheers, Bruce.
In terms of savings of greenhouse gasses emissions it is correct to discuss the total quantity of energy generated by the different methods. It is also enlightening to look at the coincidence of generation with load. But discussion like this always chooses to ignores what happens when there is a long period when the renewable energy production falls short of demand. The larger the amount of energy produced from renewable sources the less often alternative generation will be called into service, but what seems to always get ignored is to note it will be called on every so often and it is essential to retain this backup capacity or even to deploy new backup capacity when the old generators become US. What needs much more discussion are the economics of this infrequently called upon backup standby generation plant. Plant with a very low duty cycle is always expensive to operate. It is wrong to assign these high costs of non-renewable production to the bad old guys. Backup capacity is an essential component of renewable generation and as such the high costs of the low duty cycle production needs to be added to the renewable cost and not sidelined as the cost of old legacy plant – out of sight out of mind.
I guess that Solar thermal with storage will replace most of the coal generators over the next 10 years – solving the economic problem of uncompetitive coal and the need for legacy plant backup capacity.
Justin, if coal is noncompetitive into what market are you referring? In a market without RET then coal is the best solution. I refer to the collapse of the solar industry in Germany and the falling % of renewables being used there.
So lets delete the subsidies immediately and see how this market goes.
BTW Solar and Wind technology was developed in the late 1800’s along with the first coal fired generators. They were dismissed due to intermittency.
They are all old technology, I am amazed that we cannot do better than this after more than 100 years!
thanks Alan. I am not sure I understand your point here. So thermal generation on AGC responds. Yes, we know that. Surely the interesting here, as pointed out in Renew is the relatively superior responsiveness of batteries. The fact that the volume of their response is small misses the point.
Bruce, in line with WattClarity’s theme of “making Australia’s energy market understandable”, the purpose of the post is primarily to present a clearer overall picture of how the system responds to a trip, especially for readers not as familiar as you might be with how the electricity grid operates and the characteristics and behaviour of different types of generator. I don’t think anyone could say that the original Reneweconomy article had that aim in mind!
In this particular incident, as Andy points out below, the response of the battery is neither remarkable (a battery can quickly switch on from zero output – who knew?!) nor critical, and of course nor is the response of any other individual generator. There are definitely all sorts of valuable characteristics and demonstration effects that HPR and other new technologies bring to the energy transition – but this event doesn’t really illustrate them. In fact it’s my understanding that the battery’s capabilities have had to be “hobbled” somewhat to fit in with operating in an old-style predominantly thermal system.
Allan
(PS Update 7/4 – see the just released report by AEMO on initial operating experience of HPR at http://www.aemo.com.au/Media-Centre/AEMO-Hornsdale-report – gives a much fuller overview of the battery’s capabilities and services)
Why dont we just have a heap of batterys as the back up. Money is not of concern, because we have much cheaper power with renewables and its also saving on co2.
Yes we might need a lot more capacity for batterys because so much is wasted getting it in and out of them. Lets also not forget they add to demand because they do not make any power so need power for fans and lights and the compound they are in and the control room.
So just to have a battery you need more power.
But they are fantastic because they save co2.
Germany and South Australia are proof it all works.
The fact that several generators started to increase load before NEM frequency decayed thru 50Hz implies their loading instructions were being received via AGC as participants in FCAS (AGC usually also receives digital signals from stations such as trips to pre-empt frequency decay). The rate of these load increase will most likely be driven by the control algorithms in the AGC rather than the governor actions of the active units. Little natural governor action appears to be present. (And many generators in the NEM not participating in FCAS market will actually have their governors switched off) Hornsdale response seems to have commenced after frequency fell below 50 Hz and looks to be proportional to the (0.2Hz-deadband)/2 x Capacity – about 9MW based on 4% governor droop. All looks about right. Not sure there is a story here for the battery guys (or anyone else really). Very good, practical analysis however Allan.
Great writeup Alan, your stacked bar chart of unit responses is excellent. Have to point out something this explainer overlooked though: there’s two ways to rebalance supply and demand, and recover to nominal system frequency following a generation contingency: you can increase generation, or reduce demand (or both). Indeed, in this trip, both types of responses would have occurred. In the dispatch interval where this trip occurred NEMDE had enabled 19/50/58 MW of R6/R60/R5 FCAS from demand side loads (across three dispatchable units)… so the generation responses in your chart are only part of the story. (Of course, you’re forgiven for being unable to represent the demand side contribution in your chart, as the data isn’t immediately published in AEMO’s 4s SCADA logs)
Really good point Matt – I should have mentioned this, along with the intrinsic “load relief” that accompanies a frequency dip. I did actually check APD (Portland Smelter) which does show up in the 4s data and was enabled for 17/11/11 contingency raise, but its response was minimal. As you say, the other demand side DUIDs enabled in that interval don’t appear in the 4s data set.
Really interesting and useful analysis Tristan. A couple of thoughts:
1) Until there is a lot more interconnector capacity, regional demand peaks are also important reliability drivers, so perhaps looking at some regional peak days would be instructive.
2) In the medium term reliability may become less a function of demand peaks and more a function of “supply lulls” – eg extended periods of low wind / sunshine. In other words reliability is driven both by the energy and capacity characteristics of the “residual” demand to be served by storage and other forms of generation (I don’t think demand shifting addresses the extended lull issue). I know the Blakers / ANU study dealt with this, but it made some pretty heroic assumptions about diversity of RE and supporting grid investment, as well as relying on demand reduction for the last piece of reliability.
Wow that graph puts the whole article’s message in a different perspective.
It’s unfortunate that this article was put together without black coal on the graphs, to the innocent viewer it would appear that renewables generation were higher than coal, when coal generation by far exceeded renewable generation. Even though the author mentions that black coal is left out, it still appears to me that the article is still intended to create a positive bias on renewables.
I knew this was going to be the case, thanks for the graph. There is so much BS in this market when it is a critical part of Australia’s economic survival. We need a brand new approach to cheap and reliable electricity.
Don’t have an answer but the problem is there for some smart person to come up with something. Nuclear will be buried by red tape and coal needs a rethink at best. Wind and Solar offer little for the future growing demands without subsidies.
Another point worthwhile doing a post on is that some of the new solar farms will be producing at near-maximal rates until nearly the end of the summer evening peak in summer. Data from the Moree solar farm, which has tracking, shows that it in summer it produces at 80% or more of its rating until 18:00 NEM time (except when affected by cloud). Because of being further west, a new solar farm with tracking at Balranald will produce to 80% of its rating until 18:24 NEM time, while at Port Augusta until 18:48 NEM time. These times are equivalent to 19:24 and 19:48 Eastern Daylight Saving Time, and are about the end of the evening peak. Spot prices during these periods would tend to be set by open cycle gas, unless undercut by hydro. There would also be pass-through of Eastern States peak price timings into South Australia, except on the rare occasions when export from SA to Vic during the evening peak exceeds inter-connector limits.
This has two implications. Firstly, the solar farms will be getting peak prices for about a quarter of their output (~3pm to 6pm NEM time), which would improve the economic viability of solar farms with tracking in the western part of the NEM. Secondly, the amount of dispatchable backup for these solar farms would be much less, because of reliable production during the evening peak.
The world’s best engineers and technologists have an inborn aptitude for the disciplines in which they have studied and worked.
Until such people are given their due respect and tasked with planning the future for Australia’s electricity supply, we are going to slip further into the fiasco now unfolding.
It’s also clear from the many comments posted to its website, that ignorance and naivete are prerequisites for believing Reneweconomy’s propagandist messaging.
With the design and development of the nation’s electricity supply being taken over by politicians and activists, is it any wonder that the horrendously expensive and fragile grid that has resulted comes about?
It appears that collapse and chaos is the only way to bring the nation to its collective senses.
That would be great Rob if some of the ultra conservative energy professionals of yesterdays-year hadn’t been so resistant to the need to replace fossil fuel generation with renewables and storage. Like it or not means big changes in the operation of the grid.
Fortunately the tide seems to be turning and even the older engineers resistant for a great time to renewables are finally seeing that back to the future is not gonna fly, New solutions to old problems and new problems will need to be found. That’s surely what engineering is about, sticking with what they already know (i.e. fossil generation only) is not engineering so much as a basic fear of change, and it’s no longer an option for engineers, politicians or finance people.
It’s a great shame it took a rag-tag crew of activists to smoke out the blockers within the gentaliers, network owners and operators and regulators. I work with a brilliant group of engineers, coders and finance people who model what energy grids of the future may look like and assess where the risks and opportunities lie. It’s great when engineers can solve problems in this way and show creative initiative in contradistinction to modelling ostriches with heads in the Nullarbor sand.
Yet even with these great solutions, engineers, modellers and financial people are still needing to play activist to get the recognition and reform at the level of government policy and mechanisms put in place to take advantage of the now cheaper renewable generation and storage possibilities for better economics. If RenewEconomy becomes cheer squad at times, you need to recognise the mammoth task in converting old thinking about energy in not just the energy industry but also finance, heavy industry, resources and politics.
Thanks Allan for the great analysis. Very clear. I echo with you on the problem with the time tags, which makes difficult to see how much increased power of generators is from the governor control and how much is from 6sec and 60sec services. The 6 sec and 60 sec services should kick in only when the frequency deteriorates to a level below 49.85Hz. Is that right? Although many generators choose to quit their governor control because of wear-and-tear cost and not being properly awarded in the market, I believe there are still some royal generators fulfill their responsibilities.
Thanks for the feedback Jin. You’re correct that in most cases the 6s and 60s raise services would not kick in until frequency drops to below 49.85 Hz, although I think in the case of providers with Switching Controllers (smelter potlines might be an example), there might be lower frequency setpoints before raise services are triggered.
While I didn’t go into this in the article, looking at which generators were enabled for fast and slow raise services vs the responses actually seen was interesting – there were some significant responses from units not enabled for 6s/60s raise – HPR fell into this category – while other units enabled for 6s/60s raise appeared to provide smaller amounts than expected, based on their enablement level. However this last is difficult to be certain of from the 4s data, and accurate verification of contingency responses requires higher resolution data as set out in the MASS.
As the AEMO paper makes clear, the battery does provide very fast and high quality response for both regulation and contingency FCAS, and in very different circumstances from the LYA3 trip referred to in my post, that fast response (harnessed via the newly developed SIPS scheme, see p4 of AEMO’s report) could be very important in maintaining South Australia’s connection to the rest of the NEM and to keeping the lights on in that state.
The point of my post was not to knock the battery at all, simply to explain in full context what happens when there’s a large generation trip in the NEM, and to correct the possibly misleading impressions people might have gained from the original Reneweconcomy article.
A huge problem in the energy ‘debate’ at the moment is people’s tendencies to take sides and hold to their positions regardless of facts – something applying across the spectrum of what Paul has termed the Emotion-O-Meter (https://wattclarity.com.au//2016/07/the-emotion-o-meter/ and covered in many subsequent posts).
Thanks Allan. I fully agree with you that people should not take sides when looking at the FCAS. Although battery storages could kick in very quickly, but it is unfair to say the conventional generators are too slow to respond. Although governor control is slower than the battery, but think about the big inertia that kicks in immediately after the big disturbance. The kinetic energy stored in the conventional generators go nowhere, but convert to the electric energy trying to fill in the gap. Compared to the battery contribution to capture the frequency drop, the contribution of conventional generators should also be applauded. Saying they are slow is a little bit unfair to them.
That second graph with the same scale really shows how much the battery has been overrated in the media. In the first graph, it appears that the battery ‘prevented a blackout’ by meeting the shortfall when in fact it was only a small part of many FCAS responders. The capability of the battery to respond so fast to stabilise the frequency should not be ignored however.
This shows the bias towards renewables on the RenewEconomy website, they disregard selected facts to make renewables look good.
Interesting analysis, but I disagree with the implication that maximum demand coincides with maximum rooftop solar generation. This is factually incorrect, with demand increasing around 6pm when rooftop solar is well down from maximum output.
Rooftop solar actually exacerbates the evening peak i.e. the baseload generator ramp up curve is steeper because it starts from a lower value.
Large scale solar with tracking is a different story (as mentioned by Malcolm M), and I predict that solar tracking can also be utilised in reverse to reduce solar output if required for grid stability, to allow room for synchronous generation.
Very interesting. I live in Adelaide. I fitted many years ago an over voltage and under voltage relay to my a/c. It has never tripped with over voltage set to 265V in 25 years. A few hours after that event at night with no neighbourhood PV feed the relay tripped. I measured the voltage with my multi meter and it was 268 V and stable. I went back to be bed and can add more more info. More than just Hz went feral that night.
Thanks Paul. This is all excellent but even exceeds my nerd quotient. I do generally find the water analogy helpful. But one nit picky thing is that Image #2 refers to “electrons delivered down the pipe”. It is an interesting fact that in a 50 Hz AC circuit electrons don’t actually travel very far. Energy is transferred at the speed of light despite that fact that individual electrons only jiggle back and forth a short distance along the conductor.
In G) the first sentence isn’t finished.
Three paras up there is another unfinished sentence “We just need to keep in mind that Scheduled Demand is”
PS this article says “The truth is way more spectacular: the energy doesn’t travel through the wires at all — it shoots through the space around them, at the speed of light.” http://www.abc.net.au/science/articles/2014/02/05/3937083.htm
Not sure I can get my head around that!
Thanks for your time in putting together this insightful article.
I would love to know more about how the different States are using electricity, for example, why does QLD sometimes use more electricity than NSW, when they have a much lower level of population? Is it due to Aircon usage, or industry?
Thanks Paul, very useful article in explaining some of the more abstract concepts.
Just picked up on one thing; in section B, the units of the last two lines below should be watt and megawatt?
3,600MJ/h
= 3,600,000,000 J/h
= 1,000,000w
= 1MW
Indeed, as stated above, “We have many dashboards … “.
One bit of transparency we no longer have is the comprehensive Hydro Tasmania weekly report on dam levels. This report showed the installed generation capacity and whether dams were rising or falling.
With Basslink still out of action and Tasmania totally reliant on its own resources and unavailable to assist with load balancing on the mainland, this report would be making very interesting reading.
Why was this regular report “disappeared”?
Thanks Rob (and for explaining more when you called yesterday).
For our other readers:
1) it’s Hydro Tasmania’s spreadsheet (which used to be on their website) which you’re missing
2) it seems to have been removed when the website was upgraded
3) to understand why, we’d need to hear from Hydro Tasmania.
ARENA has done a wonderful job with its seed capital in establishing the RE sector.
Proponents of new projects in that sector now need much better meteorological tools to assist in planning the location of their wind and solar farms and managing the output.
They also need storage so that they don’t have to spill energy because of constrictions in the network. Even better, they need much cheaper connections to the main grid from the locations where the sun shines and the wind blows (all pretty obvious).
Perhaps a step by ARENA might be to stimulate research into much cheaper methods of construction of spurs from the main grid to these locations.
1. Storage will need to become an increasing part of the energy scene. Windy periods in southern Australia are separated by lulls of about 3 days, so there needs to be enough storage to cover these periods.
2. Solar would be partially counter-cyclical to wind. Solar definitely needs storage, and this storage will be helpful more wind into the energy mix.
3. While the wind profile in various parts of Australia may be decoupled, showing no correlation, there are probably few negative correlations, which would indicate complementary wind patterns.
4. Bureau of Meteorology wind data would be of limited usefulness in conducting studies of wind complementarity, because the measurements are at 10m, whereas the newer wind farms are harvesting some energy at 200m. The relationship between wind speed at different heights is a “bent stick”, with the lower height becoming becalmed when there is still significant wind higher up.
5. Since most new wind farms will have a height of 200m (hub + blade), wind output data from older wind farms will be of limited value in conducting simulations of a grid with a higher renewable component. Wind farms with the greatest relevance are Kiata and Sapphire, which are both nearly 200m high, and I believe are attaining capacity factors of close to 50%, which should reduce the periods when the wind farm is becalmed.
6. New wind farms under development with the higher hub height include Silverton (NSW), Coopers Gap (Qld), Lincoln Gap (SA) and Stockyard Hill (Vic).
7. Professor Blakers of ANU recommends storage close to major load centres. Yet the largest storage development underway is Snowy 2, a long way from load and other generation. The Murray power station on the Vic system is close to the feed-in point for Snowy 2, and has a marginal loss factor of 0.92. So round trip transmission losses for Snowy 2 could be 8% + 8% = 16%, in addition to the round trip losses of about 30% within the pumped storage system when operated at its full 2000MW (but lower losses at partial capacity). There would need to be some substantial transmission investments to provide for both the capacity of Snowy 2, and reduce the round trip transmission losses.
“Storage will need to become an increasing part of the energy scene.”
Blind Freddy can see that but at what cost? SA now joins the elite having the highest power prices in the world and hasn’t paid for the storage side of the equation yet.
Who is going to break the bad news consumers aint seen nothing yet with their power bills? The obvious reaction will be- Whatever happened to all the cheap power we were implicitly promised from Mother Nature’s limitless free bounty?
Storage for 3days Wowa!!!!!!! I am not sure many people understand these two facts about the world saving storage.
1. They add to the overall demand.
2. They produce nothing but to put energy it into them and then get it out of them the process wastes a variable amount which is depented on a number of factors. But allow 10% minimum, that’s bare minimum like all the stars have aligned. The worst thing is on hot days when the demand is its highest the storage is worst. Not only will nothing be going into them or coming out, the storage project needs power!!!!! So adds to demand.
I am sure however all the pro RE people will have a different view, but unfortunately that’s the reality.
History is history but the additional interesting fact concerning the remaining 32 year old Muja C/D (854MW) and 17 year old Collie (340MW) is although these units are nowhere close in age to Liddell and Muja A/B (52 years as indicated in the article), they are now in the crosshairs of the looming “duck curve”. As consumers desperate to avoid increasing power cost are installing solar PV at unprecedented rates, midday load has fallen off to the extent that on many weekends, the minimum load of the day is now 12-12.30pm and not as historically, 3-3.30am. The point is rooftop PV is robbing these baseload clunkers of their needed demand and within 3 years, the ramping load on the remaining coal units in WA will see them in terminal decline. The daily load swings simply cannot be accommodated by coal fired generation. The government must now act to actively plan the transition to a dispatchable power system philosophy and bid farewell to the baseload one that has served us well in the coal era of the past.
The decisions are not going to be easy but they are mandatory for the security and cost of future power on the SWIS.
It is good to read about some of the cost/price implications of significant PV and wind feeds. I have contended for a long time that generation cost comparisons of renewable and non-renewable energy have been very misleading. As the sales opportunities for traditional generation become impacted by renewable feeds the consequences of these changes should applied to the renewable side of the ledger. The cost of generation of what ever flavor is still largely the repayment of the capital costs. These capital cost need to be met even if there is no product sold. A cut of 50% of the product almost exactly double the unit cost. Even if a renewable feed was zero, for example, because there still remains the need for standby generation then about the same number of dollars need to flow to run the standby plant for it to remain viable. Cost comparisons seems to imply that when renewable feeds are less than traditional feeds the total cost of generation will drop. It will not. Even if renewable feeds were zero the cost of product will be almost the same because of the need to pay back capital on the standby plant. It will never be that cheap. Good to see some discussion about this very badly presented subject.
Nice analysis. When we (Gamma Energy Technology and Red Vector) looked at this issue using historic wind data matched to existing and planned wind farms we found a very significant correlation between wind droughts in each state. In fact there was a period in May 2010 when the output from a diverse portfolio of wind farms in each state would’ve averaged just 5% across the entire NEM for 8 days.
That is not something that storage can help with (the volumes are far too large), other forms of generation will be needed during these periods if wind becomes significant, and if the purpose of wind is to decarbonise the system then these will need to be low carbon too.
Wivenhoe pumped storage (500 MW) could be used much more for balancing, and questions need to be asked as to why it is used so sparingly. So far this month one of its generators has been used 6 times at between 60 and 100% of capacity, and the other has not been used at all. For comparison the Shoalhaven pumped storage has generated 23 times. Wivenhoe is operated by State-owned CS Energy, which also operates several coal-fired power stations. It has a conflict of interest in despatching Wivenhoe because this would reduce its income from the coal-fired stations within its portfolio. But as an asset owned by a State government with a target of increasing renewable energy, its management by CS Energy seems to be detrimental to implementing government policy.
The graphs show a typical power price of $70/MWh, apart from price spikes and daytime troughs when there was an excess of solar. With current steaming coal prices an operator competing for potentially seaborne coal needs to recoup at least $47/MWh just to cover the cost of coal. Furthermore the power stations near Rockhampton and Gladstone get paid a 5% discount on the State price because of line losses. So there’s little margin to cover the other costs, show both a return on capital, as well as profit. Hence their reliance on price spikes.
Paul, interesting, thanks. Not sure I like the title though – its pejorative and ideological. There is nothing terrible special here: production increases when the sun shines. We have seen the same thing repeated whereever solar becomes significant – Italy’s main trasmission has been recording peak demands at night, not day, since 2011. This provides incentives to switch load into the day time period.
I would love to know where the $70 per mwh is needed to break even for power stations comes from.
Somehow over the last 15 years they have operated for up to 4 consetive years where the average spot was around $30!!!!!!!!!!.
To anyone who wants to debate the cost of coal fired energy, I would first like to level the field, and allow to discussion to be around, coal fired plants being able to be built as easily as wind and solar and the bidding for power being filled in order of lowest bid, as it used to, not renewables first.
I am quite sure in Queensland if that were the case the $70 per Mwh, would quickly become $30 again.
Remembering that coal generation has not sat still and has advanced a huge amount over the last 15 years.
The constraint to coal is simply the belief that co2 is a problem, there is no cost problem where renewables are cheaper.
I did use NEMReview to get a bit more granular and looked at hourly average data from 1st April. It gives another great perspective. It does show that the Rule has effectively put a floor under the spot prices during periods of high wind generation and low demand. It would be interesting to model how much this has increased average wholesale prices in SA, that ultimately also flow into retail prices, and the shift in wealth from end users to generators.
With the cost of solar components continuing their global descent coupled with the move towards storage and the increasing awareness/consciousness of the “average” Australian, these trends will continue and magnify over time. Will they in fact help with reduction in grid electricity prices? Time will tell. Interesting article, thank you!
As prices rose to pay for decentralised renewables and their delivery the low hanging fruit of consumer demand was plucked with energy saving appliances, lighting, judicious RC aircon use, etc and now the hard yards of increased population and concomitant inelastic demand begins in earnest. With our thermal generators wearing out due to freeriding renewables we not only have to pay for more wind farms and solar panels but the elephant in the room- storage. We live in interesting times now.
When doing this sort of thing it is valuable to normalise at least for customer number. Having done so we see steep declines throughout the NEM except in Qld, where CSG compression has swapped demand contraction.
Normalising (e.g. by customer number) would provide useful additional details. That would be food for another post at some time.
The numbers in aggregate (as in this article) provide a useful measure in their own right – such as, in my case, helping to identify that in aggregate average demand per region had stopped declining in NSW.
This will have implications for the future in a number of respects.
Paul
When you do the sums on a financial year basis, the growth in demand in NSW has increased by only 100 MW over the past four years supporting the view that Bruce M makes
The only real growth that has occurred is the addition of the CSG compression in Qld
Thanks David
I did try to make it clear in the article that the growth rate in NSW was very modest. Sorry if that was not as clear as I meant it to be.
What my main point of surprise was, and is the thrust of the article, was that the decline in demand had stopped in NSW.
As you know, I have a particular focus on QLD and have noted many times on Wattclarity about the impact of CSG compression. My assumption had been that the decline was continuing in VIC and SA (true still), along with NSW (not true for the past 4 years, which is embarrassing for me).
Paul
Might be interesting to put these numbers against economic activity in the respective states and also compare total energy consumption by state. Given the recent national statistics, there may well be a simple explanation – that we should all be happy about. People forget that at a high level, domestic energy consumption is only about 35% of total energy but causes over 50% of maximum demand. The NEM was well protected during the last economic cycle by a large over supply of generation. Retirements in the last 5 years have removed that. This next cycle might be more interesting on days when the wind isn’t blowing. First sign will be forward pricing on industrial bilaterals. Might be interesting to watch from a distance in WA !
the only measure from an electrical system point of view is the “native demand” which is the underlying demand and energy consumption or the total of the demand and energy consumed by customers. In this case, for the NEM connected customers. It must by definition be a sum of the NEM generation plus embedded generation plus behind the meter generation. On this basis, the energy consumption of the NEM is increasing. To measure it in any other manner for system security and planning purposes is to ignore a large growing segment and an action that will lead quickly to reduced reliability and higher pricing.
A key thing that I believe is worth mentioning is that Tuesday night LOR was down to the Bannaby-Gullen Range transmission outage (they were cutting in a new substation), which reduced flows into Sydney by ~1000MW. This outage was supposed to finish at 17:00 but kept being extended, finally finishing at 19:15 or so.
I don’t believe that RERT was dispatched – there would have been constraints #RT_NSW1_OE/PE reflecting this. The decrease in demand following the MPC spikes was most likely Tomago pulling potlines, and the 2nd spikes due to increased demand which was probably them putting them back in.
Gas availability coincident with peak was Colongra bidding unavailable to avoid uneconomic start. Not sure I agree with how Snowy managed this in light of LOR2 – someone probably had a word with them given they ran on Friday night.
The SA LOR was due to some planned maintenance by ElectraNet to take a key 275kV line out which would reduce Heywood capacity to ~100MW.
“That’s not getting us anywhere closer to a properly managed energy transition.” While talking about your emotion meter and others stongly held beliefs you expose your own.
I would say the Author will be nailed to a cross for saying that wind and solar and not reliable energy.
Quite a lot of people believe it’s the other way around. Coal and Gas are intermittent and wind and solar are predictable. Due to the weather being able to be forecast so accurately. Yes I am a fond boater and the forecast for boating weather is so broad and covers all most every possible out come, yet my experience is that they still have a very poor accuracy.
I find this article really fascinating given all the talk and evidence of rooftop solar booming nationally. How is that we get so much generation from solar yet NEM demand increases. Where would we be without solar and other renewables…probably all living with constant blackouts, shortages etc. So much for the current government line…
Paul, is there experience in other jurisdictions with prices going negative, or is it happening in Oz for the first time round the world?
My crystal ball has a blurry image for about 2020 where the price bounces around at a much lower level than now, especially when the interconnectors are in heavy use and the renewable generation is high. I do hope that the crystal ball the investors pouring into the Oz market have, shows a different and sharper image.
Let’s always entertain that there is a bust around about with every boom and to plan for that too, whether it be Dutch tulips, mining construction or renewables.
Your prognosis is valuable to the industry as well as to your own business. Thanks for sharing.
With gas prices at $9.34/GJ the fuel cost for gas generators alone will be $9.34/GJ * 3.6MWh/GJ @ 30% efficiency = $112/MWh. Which makes it even more interesting as to why TIPS would be running 7/8 units at less than 35% CF as shown in the figure at that price. But the real issue is the interconnector outage – it would be much harder to get these outcomes without it.
The interconnector is no issue. There was always available capacity all day, physically at least.
(Only one of the two circuits was out for maintenance).
Available capacity is not indicative of market outcomes. And the interconnector was limited to less than 100 MW throughout the day (currently limited to 37 MW).
TIPS would be running units at part load so they can quickly ramp up and meet AGL’s own peak demand in the evening. They are probably not running at higher loads during the day because they are long on capacity during the day and want to retain line pack in the pipeline to meet the peaks. Just because the pool price and gas price is high doesn’t mean that this reflects a gentainers real position.
But if they could sell more power at close to max wholesale price now, and potentially fall short of their retail demand later (assuming interconnector/wind issues ease by then and prices are more typical), wouldn’t that still present a good arbitrage opportunity?
Maybe they didn’t want to assume the interconnector/wind issues would be overcome by the evening peak? AGL are probably a bit cautious following the issues in NSW a few weeks ago when they were very short on capacity due to multiple unit outages.
Allan
Thanks for the very good analysis and I agree that the primary issue is lack of competition. There is a view that 5 minute settlement will also depress competition so perhaps that is not the answer.
The work you did also highlights
that gate closure (say 1 or 0.5 hours before dispatch) would also address the problem
Another solution could be having the lower SE a separate region,or even part of Vic region
1) Re Gate Closure
You’re not the only one talking about some kind of gate closure mechanism (e.g. the recent Grattan review) – and these calls tend to do the rounds periodically (e.g. when the SA Govt sponsored review at the AEMC into bidding a few years ago). However each time the call is made, no-one seems to be able to help me understand how no change would avoid the fact that there’d always be a “last bid” (hence the motivation to try to maximise profitability by changing bid). Doesn’t moving gate closure further out from “now” only end up making AEMO procure more Regulation FCAS services, given increased uncertainty about what demand will actually be?
2) Re Zonal and Nodal
My sense is that you’re onto a solution that would have a chance at being more effective, in terms of creating a separate Region for South-East of South Australia (and will need same solution for North Queensland in 18-24 months as well). However the politics of this would prove very tricky – which is why (despite the fact that it was allowed for in the rules) States have never been disaggregated into Regions.
To the layman, this post is incomprehensible, which is not in the least the fault of the author, but was almost certainly intended by the crafty folk who created this thicket of legislation, pseudo-market prestidigitation and regulatory gobbledygook. A mere two or three decades ago, the average voter could understand how the electricity system operated – a handful of large, often state-owned generators produced reliable electricity at low cost, which was distributed to everyone via an equally simple distribution network and sold directly to consumers as a regulated, essential monopoly product. State governments each operated their networks separately, and were thus responsible for the outcome. After two decades of massive subsidy, interference, distortion through hundreds of pieces of legislation and the creation of a dozen or so unaccountable regulatory bodies, all this simplicity has been swept away. By whom, and for what purpose? Can sanity in this essential industry ever by regained?
Thanks for the article Paul.
I will add a couple of points increasing the complexity further:
1. Climate change and potential for an increased number (and duration) of extreme weather events e.g. heat waves.
2. Increased temperature responsive load, driven by renewed housing stock which, by being generally well insulated and installed with airconditioning, is space heating/cooling load efficient in mild temperatures, but significantly increase demand when temperatures are outside the mild range.
I’m constantly amazed at the blurring of Energy (MWh) and Peak Demand (MW) in the discussion.
50% renewables (ie MWh) does not mean 50% installed capacity (MW).
There is very little understanding about diversity, and capacity factors, especially for wind generation.
Queensland’s 50% renewables target is a 50% energy (MWh) target. In essence it’s also a 50% non-renewables target, and the policy relies on the contribution of coal generation to stabilise the network and provide reliability, but no-one talks about that
You’re not wrong when you have a play around with wind outputs for Australia and compare say July13 with July17 overall. That’s bad enough to cope with but then knock off all the State’s outputs except South Australia for July13- http://anero.id/energy/wind-energy/2018/july/13
How on earth is a national grid (ie a rational system) supposed to cope with that because MLFs and gobsmacked doesn’t begin to do the problems justice and where is the omniscient benevolent dictator to solve it all or wind back the clock?
Is villan no. 5 the sheer rapidity of change, to the extent that there are very few people that really understand some key aspects of the power system? The best example of this is in “grid security”, which is often used in South Australia to curtail wind output, but also limits most inter connectors to well below their thermal capacity. How many wind farm developers in South Australia included this as part of their financial assessment? Even the NEM rules were until recently were for an energy-only market. As an example of this, on 13 November 2016, just a few weeks after the system black in South Australia, the SA market operated with only a single gas generator, and all other energy from wind. This would now be considered “not in a secure state”. Even AEMO don’t seem to have many people skilled in this area.
WattClarity does a great job of helping to reduce the evilness of Villlan #4 (including me). I would be wary of the strength of the emotion-o-meter villain #3. I have witnessed how it has ruined the gas extraction and GM foods industries and no calm reasoning of fact based website has found a way to bring them back from the margins to the sensible centre. It’s a little like a story about a horse, a gate, and a bolt left unlocked.
Paul,
Under Q2 you say “In the chart above I have assembled monthly aggregate production from generators for which AEMO has data”; this doesn’t seem to hang together for the only chart in the article. Am I missing something or is there meant to be another chart
great post. I didn’t realise how the 5/30 split was leading to that incredible volatility. Looks like AGL is doing an ‘Enron’, and they’re being allowed to get away with it.
A few more novice questions, which I hope are addressed in future issues.
1. Why are interconnectors between regions constrained to half or less of their thermal capacities, whereas within regions there don’t seem to be problems using transmission lines up to their thermal capacities? In other words, does physics work differently between States than within States?
2. Why are interconnectors often constrained to well below half of their thermal capacity?
3. Why are there now system strength requirements requiring a minimum number of gas generators to be on-line at all times in South Australia, whereas in the recent past it seemed to operate without problems without these system strength requirements? For example, for several hours on 13 November 2016 there was only a single gas generator operating in South Australia, with all electricity generated from wind, and there didn’t seem to be any problems reported.
4. In determining the system strength requirements in South Australia, why have none of the simulations to date have included the 70MW of battery capacity from Hornsdale Power Reserve to include it within the minimum system requirements.
5. In the Victorian market, why would the brown coal bids be so low that night-time spot prices drop as low as $11/MWh, and at this price none of the generators are at minimum operating load? Surely it would be in their interests to put in higher bids down to the minimum operating load.
It may be less a case of the windfarms _actively responding_ to negative prices, than simply having capacity offered (bid) into the market at prices well above the -$1,000/MWh price floor. In cases like today where there is low demand, a strong constraint on exports – due to yet another 275kV transmission line outage in SA – plenty of wind, and the system strength constraint requiring fixed levels of thermal generation to stay online, economic market dispatch will necessarily reduce generation from the “higher priced” windfarms to keep supply and demand balanced.
Effectively the windfarms in question would have set up their offers so that their output is automatically reduced by AEMO if the spot price is lower than some threshold, which may be related to the value of LGCs, or the bundled price of energy & LGCs. Not that different to how other generators bid except that for contractual reasons these price thresholds tend to be negative for windfarms.
It is a pity the SA/Vic boundary wasn’t moved from Heywood to Tailem Bend for the duration of the works. What is the point of curtailing the SE wind farms when the transmission restriction is between Tailem Bend and Tungkillo, and there would be more than enough capacity on the Heywood interconnector to export all their output to Victoria.
Good question Malcolm. I haven’t looked but if the SE windfarms are being curtailed by the Tungkillo-Tailem Bend outage (rather than just dispatched down on price), then that will be due to a transient stability constraint not a thermal limitation on exports to Vic. The issue is that if the remaining 275kV line in SA trips, the SE generation remains connected to SA on the much weaker 132kV network, and if that generation is running too hard at the time the sudden change in impedance leads to transient issues impacting those generators. The constraint’s generation limitation effectively controls what is flowing back into SA from the South East rather than what is exported to Vic.
Won’t there be even greater problems of managing the generation/load balance within the SE SA region if it is part of the SA market than moved to the Vic market for the duration of the works? For example, if SA prices turn negative and the SE wind farms turn off, AEMO needs to quickly find perhaps 200MW to support the SE load, while the capacity from north of Tailem Bend is limited to 100MW. So it would have to find the balance from the Heywood interconnector. According to an AEMO diagram, there are 3 lines from Tailem Bend to Adelaide
– a 275 kV line to Tungkillo (which is being worked on)
– a 275 kV line to Mt Barker
– a 132 kV line to Mt Barker, which has a likely thermal rating of 100 MW
The latter line is no doubt the limitation, so if the Mt Barker 275 kV line trips, there is still 100 MW of capacity via the lower voltage link.
As Ben says below, location of the regional boundary is really just a settlement issue. The dispatch algorithm loads generators according to least-cost principles consistent with all constraints, regardless of which region they are in.
In the scenario you paint, the “missing” 200 MW would simply be provided by increasing loading on other generators (including possibly in Vic / NSW / Qld) and network flows (including across Heywood interconnector) would adjust accordingly.
Also note that it’s not only the thermal capacity of the 132kV network that is the limiting factor on flows to / from SE SA, it’s the much shorter term transient impacts of a trip on the remaining 275kV line that have to be protected against.
The dispatch system should still dispatch the generators downstream of the constraint correctly assuming they are bidding below the marginal cost of supply in Victoria.
The constraint orientation should recognise the generators’ locations OK.
The regional boundary affects settlement, but should not directly affect the efficiency of the dispatch process. Sometimes generators will rebid to avoid being run and settled below running cost, but probably wouldn’t occur in this case.
Why were they bid in at $-1000 in the first place if they are not covered in some other way like a PPA? Is there some technical reason that wind turbines cannot increase or decrease their output on demand, such that the owner would be prepared to take such a hit?
Wind farms will bid in at -$1000 for two reasons:
1. PPA covered all risk, as you mentioned.
2. When there are too much wind in the region, AEMO issues a constrain to all wind farms, and WF with the most expensive bids would be curtailed out of the market. WF would typically bid low to stay in the market because prices are still likely to be profitable for them – that is until the -$1000 hits.
With 130km/hr winds and severe chill forecast across the southern States we may be about to live in interesting times again, especially with the NEG up for debate.
So does this mean that there will be a inverse effect on the value of dispatchable storage capacity as the opportunity value of zero/sub zero PV energy represents an even greater earning potential for storage projects such as Kidston?
If I were a wind generator, I would bid in at -$1000 to ensure you get dispatched for most of the time and take it on the chin when prices a negative for a number of negative dispatch intervals. Almost all of the time wind gets dispatch at positive prices and unless they have a ppa, I would prefer to take pool risk.
Did the two interconnectors actually trip at exactly the same time or did one trip fractionally earlier? Rapidly falling load on the Heywood interconnector could cause it to open to protect the SA network with consequent drop of frequency in VIC & NSW causing the QNI to trip.
So we didn’t lose a single generator on the system but had to trip more than 1GW of load and on a low load Saturday afternoon.
Mick, as noted to Liam (comment below) I don’t know the exact sequence.
All we know is:
1) Two trips occurred within a 5-minute window. We don’t know how close (but suspect within seconds).
2) Which one came first, we don’t know.
3) What caused the first to trip, we don’t know. There were storms in northern NSW so perhaps lightning was a factor, but that is purely speculation (and was not implied in Market Notices).
4) It seems likely that the 2nd trip was a result of the 1st (perhaps due to rate of change of frequency, as Liam suggests in his comment).
Per AEMO’s market notices, simultaneous tripping of 330kV transmission lines in QLD 8L and 8M Dumaresq to Bulli are now considered credible contingency, synchronous export from QLD to NSW have been constrained. The cause is unknown and AEMO not satisfied this event will not reoccur.
Perhaps the chain of causality is opposite to your suggestion, ROCOF in VIC was too fast, overloading Heywood? I will certainly await the report into this one.
Sorry, Liam. Yesterday when I posted, did not intend to suggest either sequence of interconnector trips – i.e.:
Scenario 1 = whether QNI was first and then Heywood, or
Scenario 2 = whether Heywood was first and then QNI.
The Market Notice you refer to was published at 15:44 (more than an hour after this post).
My reading of it still does not confirm which trip came first (i.e. either Scenario could hold). Your hypothesis is one possibility in Scenario 1.
Like you, we’ll be awaiting the report. A sensible thing for any reader to do, to reduce the risk of forming wrong conclusions….
A small update. One part of the mystery is solved.
“64012 26 Aug 2018 17:52 RECLASSIFY CONTINGENCY
Cancellation of a Non-Credible Contingency Event: Simultaneous trip of 8L and 8M Dumaresq – Bulli 330 kV Transmission lines as a credible contingency lines in NSW due to protection mal-operation.”
At the time I just happened to check my solar inverter on a sunny day in southern Victoria, expecting to see high output. But instead its screen was going through what seemed like a boot-up procedure, then after finishing this there was the “clunk” of the relay and it began showing its power output. This is worrying from a system point of view, that a new inverter (Fronius) tripped out with a rapid change of frequency. At that time Opennem shows 802 MW of roof-top solar in Victoria. If much of this were to trip because of bad fault ride-through settings, a lot more load-shedding would be required.
Perhaps AEMO needs to commission some physical tests of inverters under simulated rapid changes in frequency, rather than trusting manufacturer specifications. Inverters that fail are then banned from the Australian market.
WRT the question in observation #4, it doesn’t look like there is a significant time stamping issue because the small frequency deviations seem to match quite well between the measuring devices.
The drop then rise you are talking about in observation 10 (and others) looks like inertial response to me. That will look like a response proportional to the rate of change of frequency. It is governed by the physics of the machines. So not really a controlled response.
So I would say for Loy Yang, Bayswater, and Gladstone there was no governor response or there was but it was too slow to see in that time domain.
Thanks Kate, I have been following your conversation on this topic since I saw a presentation of yours in 2017.
Obviously the deadbands need to be set and prioritised correctly and I hope that your advice is acted on in record time.
I have two very minor critiques, pleased accept them in good spirit!
1. You referred to renewable intermittency a couple of times when referring to frequency, and seemed to imply that intermittency applies to poor frequency control. In my experience the term intermittency applies more often to weather dependent generation capacity, not so much frequency control capability.
2. You mentioned FCAS contingency market and spinning reserve, and obviously you are correct that a generator has to have some spare capacity in order to increase active power output to manage frequency, but my pedantic nature forces me to point out that there is actually no market for spinning reserve for generation capacity.
Ok, thanks for listening, looking forward to some proper governor control soon!
Regarding Observation #15, I believe the decrease of the output of the 6 renewable generators is consistent with the frequency control they must implement to meet the requirements of clause S5.2.5.11 of the National Electricity Rules. Given the capability of power electronics based generation, these generators are probably registered with an Automatic access standard to clause S5.2.5.11, i.e. they must reduce their active power output by at least 20% of their maximum output for each 1% frequency increase above the upper limit of the normal operating band (being 50.15 Hz on the mainland).
This frequency control is a security measure to prevent the power system to collapse. All generators implement such a control regardless their participation to the FCAS market.
Solar farms won’t be able to propose an Automatic Access Standard under clause S5.2.5.11 since they generally cannot increase active power output in respond to a reduction in system frequency (S5.2.5.11 (b) (3)). It is more likely to be a Negotiated Access Standard which can meet S5.2.5.11 (b) (1) and (2). And if they are not going to participate in the low FCAS market, they do not need to activate this function. They simply just need to demonstrate that they are capable of doing so. But I know that some plants enable the function without fully understanding the consequence.
As you wrote, solar farms only need to demonstrate they are “capable of”. If they are not dispatched at their MPP, they can increase their output in response to a frequency fall. So they can propose an Automatic Access.
I’ve carried out lots of plant failure analyses even back to the days of sprecher & schuh relays and you’re on a hiding to nothing trying to use 4 second data especially when you don’t have access to the source of the scans.
There will always be timing issues when different systems are involved. And even though a time is assigned to a scanned value you have no idea of the exact time the data was actually collected.
Modern failure analysis relies on high speed monitoring systems runnng independently of the process control system. While all the above comments are interesting, I suspect none of the data AEMO is currently providing has the required level of integrity to allow for the sort of analysis required.
Thanks Watt Clarity for your excellent blog. This may interest evertyone as I stumbled onto this website last night. Looks like this idea has been a long time coming….. they talk about a ‘reverse ebay style’ auction to help drive down energy prices. Looks interesting. http://www.givemeyourbestprice.com.au
Perhaps SA should consider high-speed inter-tripping to shed SA LOAD when Heywood is overloaded in the import direction and to trip WIND GENERATION when OVERLOADED in the export direction.
Stated power supply from QLD to NSW prior to the incident: 866MW
The QNI capacity of both[1:] circuits is 1078W[2:]. Both circuits must have been in operation to carry 866MW.
Q1: If you want a single failure proof QNI interconnector, then what is the maximum permissible load on both circuits of the interconnector?
A1: 539MW, preferably shared half and half.
Q2: If you do not operate in a single failure proof manner, then what are the consequences of a single circuit failure for any reason?
A2: A cascade trip of the second circuit, that leads to a systemwide transient that is only halted by a AUFLS trip of both Tomago and Portland smelters. The QNI trip shed 866MW of supply and the smelter load shed 850MW of demand. A neat fit that made the transient survivable without a total system black.
Tomago drops 600MW
Portland drops 250MW
Q3: Why would AEMO expect any different result?
A3: Because they have not learnt the lessons of the SA statewide blackout. Is AEMO going to make a habit of doing this[3:] ?
[1:] https://en.wikipedia.org/wiki/Queensland_E2%80%93_New_South_Wales_Interconnector
The Queensland – New South Wales Interconnector (QNI) is a 330 kV AC interconnection between New South Wales and Queensland, Australia. The link was commissioned in 2001. It consisted of double-circuit 330 kV lines between Armidale, Dumaresq, Bulli Creek and Braemar, and a double-circuit 275 kV line between Braemar and Tarong.[1
[2:] Stated power supply from QLD to NSW prior to the incident: 866MW
The QNI capacity of both[1:] circuits is 1078W[2:]. Both circuits must have been in operation to carry 866MW.
Q1: If you want a single failure proof QNI interconnector, then what is the maximum permissible load on both circuits of the interconnector?
A1: 539MW, preferably shared half and half.
Q2: If you do not operate in a single failure proof manner, then what are the consequences of a single circuit failure for any reason?
A2: A cascade trip of the second circuit, that leads to a systemwide transient that is only halted by a AUFLS trip of both Tomago and Portland smelters. The QNI trip shed 866MW of supply and the smelter load shed 850MW of demand. A neat fit that made the transient survivable without a total system black.
Tomago drops 600MW
Portland drops 250MW
Q3: Why would AEMO expect any different result?
A3: Because they have not learnt the lessons of the SA statewide blackout. Is AEMO going to make a habit of doing this[3:] ?
[1:] https://en.wikipedia.org/wiki/Queensland_E2%80%93_New_South_Wales_Interconnector
The Queensland – New South Wales Interconnector (QNI) is a 330 kV AC interconnection between New South Wales and Queensland, Australia. The link was commissioned in 2001. It consisted of double-circuit 330 kV lines between Armidale, Dumaresq, Bulli Creek and Braemar, and a double-circuit 275 kV line between Braemar and Tarong.[1
AEMO’s 4s data has been referenced a few times, which contains two measurements of frequency – one each for NEM NORTH and NEM SOUTH. Do you know where these are measured from?
Just after the event I happened to look at the display of my inverter (a new Fronius), and was surprised to see it rebooting. Then after the reboot there was a “clunk” of the relay, and it gradually began to show the amount of power from DC to the home and grid. It was a sunny day here in southern Victoria, so there would have been a reasonable contribution of rooftop PV to demand. The inverter specifications show it tolerates frequencies down to 49 Hz, and there is no setting for the rate of change of frequency.The 4-second data show it didn’t quite do down to 49.0. Their data are presumably the average frequency of the 4-second interval. However what AEMO should be logging is the max and min frequency in each 4-second period, because it is the max or min frequency that will turn off the small-scale inverters. As inverters become an increasingly larger proportion of daytime power supply, should we be requiring a wider range of fault ride-through settings?
Can we get some analysis on what impacts the Hornsdale Power Reserve had on the spot price in SA during hte islanding event.
It was meant to be the saviour during events like this but it seems they were low on charge even though Hornsdale Wind Farm had been operating above 75% capacity factor for most of the 2 days before the event.
Hi Paul,
although I can’t contribute to clearing up these obscurities, I’d like to thank you for yet another great article that helps anybody trying to get a better understanding of the NEM.
I especially liked the “quick notes” in the squares as a kind of glossary.
Many thanks, keep up the great work!
As an aside tThis might interest the number crunchers at Wattclarity as they’ve released the cost of the Tesla big battery and Hornsdale wind farm facility- https://www.msn.com/en-au/news/australia/tesla-battery-cost-revealed-two-years-after-sa-blackout/ar-BBNBBaD
That’s an investment of $900/KW but if wind can only produce 30% of that on average that jumps to $3000 and then the question of what sort of investment cost per guaranteed kilowatt 24/7 all year round? Marginal costs are low but what’s the bottom line for a level playing field kilowatt of battery backed wind power here?
Actually, I think the most glaring example of villain #5 is not focusing on mitigating the impacts of rising temperatures. As stated above, villains 1 thru 4 ensure that, if you believe CO2 is the primary cause, that prevention is not possible. Mitigation is therefore the only viable option and it also addresses the issue that, if CO2 is not the problem, rising temperatures are going to occur despite all of our efforts.
Paul, you mentioned carbon pricing, therefore I take that as an opportunity to mention the logical inconsistency in the IPCC position. I explain it thus:
– CAUSE: IPCC position is that carbon emissions are currently the greatest threat to humanity, a threat created by developed economies and to be solved by developed economies
– ACTION: IPCC advocates reducing carbon emissions, mostly by the following mechanisms regardless of the effect on developed economies: increase renewables, reduce meat consumption, increase electric vehicles, reduce poverty and increase equity
– LOGICAL INCONSISTENCY: if reducing emissions is the most important action for humanity, and the energy system of the western world is the largest source of emissions, the logical solution is rapid transition to nuclear power generation, globally.
Since the IPCC does not recommend rapid transition to nuclear energy (which incidentally would provide electricity to more people than renewables thus reducing poverty and increasing equity), I conclude that carbon emissions are NOT the greatest threat to humanity, and therefore carbon pricing is a purely political strategy designed to restrict developed economies.
But I agree that if our chosen politicians insist on carbon pricing then it should be like the GST but with less exceptions.
Wind output was quite low at the same time, so SA exports may not have tested the limits there, but AEMO stability requirements will still require some synchronous generation at the SA end (to mitigate risk during unplanned interconnector trip).
The AEMO Q2 dynamics report showed some significant curtailment of semi-scheduled (wind) generation, so it would be interesting what the system would look like during a scenario of the same low demand but high wind.
Thanks for putting this together, Paul–quite interesting. I will preface by saying I’m working from knowledge of other systems and have no specific familiarity with AEMO, so I hope you’ll correct misinterpretations.
If I understand correctly, it looks like they are abusing the Target, which seems to penalize injections above the desired dispatch point but not below. Since this underproduction has to be balanced by providers of FCAS, I would expect there to be a penalty of some kind–does that exist in AEMO? If not, given how quickly they seem to be able to ramp down, the -$1,000 bid is strategically correct (although maybe not good faith).
I am a bit surprised by the speed with which the rebid is incorporated. Based on the timeline, a rebid at 14:33:12 gets incorporated into the SCED in time for the 14:35-14:40 interval. It’s also confusing that dispatch instructions for that interval only arrive at 14:35:20. Am I misunderstanding this timeline?
A semi-related question: does AEMO use a multi-period dispatch model?
Hi Tristan, regarding your comments about the rebate. I quote you in two lines below.
Firstly the word rebate. The rebate is actually a subsidy, which is actually a levy. The levy is not applied exclusively to those who install solar panels, the levy is applied to all electricity users, but not referenced on their electricity bills (no transparency), increasing electricity bills. The levy then has GST applied to it (no transparency). Those with solar installed then have some of that levy returned to them, which reduces their electricity bills. Typically the people and businesses with solar panels are not in the lowest economic group. The net result is your “rebate” increases cost of living for the lowest income groups and reduces cost of living for the higher income groups – that is effectively transfer of wealth from poor to rich (in simple terms).
Secondly your implicit assumption that subsidised solar panels on roofs in Australia can have any effect on global warming is far removed from reality and deserves to be called out as fanciful ideology. In other words, “you’re dreaming mate”. I encourage you to provide some evidence to support your position on this topic.
“So are we faced with a serious problem of there being too much solar and does that mean we should scrap the rebate?”
“Then there’s also an issue about do we care about the dangers to our children and others around the world from global warming, and is this solar rebate a worthwhile expenditure to address these dangers?”
Hi Ben,
I wonder if you could clarify your second argument a little, as I’m not sure what the angle is.
It seems intuitive to me that energy produced from rooftop solar would otherwise have come from fossil-fuel power sources and so has reduced power-related carbon emissions from a business-as-usual scenario. The subsidies would seem to have accelerated the uptake of rooftop solar (this makes economic sense) and helped get the cost of solar down.
If your argument is more around this phenomenon only representing a fraction of carbon-intensity of the Australian power industry, which is a fraction of Australian emissions overall, which is a fraction of the world’s emissions, then I sort of see what you’re saying, but I would offer two rebuttals to that:
1. There’s an important difference between something having a small effect and having no effect. A lot of small effects added together make a big effect. Arguing that this is too small to make any impact is like arguing that there’s no point voting in elections because one vote won’t sway an election.
2. Global warming is a global problem and demands that human society works together across borders, continents, political divides, etc. to solve it. Australia is a developed country with excellent renewable resources, and so is well-placed to lead the global economy through a transition to sustainable, fossil-free energy systems. We may be a small player in terms of things like GDP and population, but small players can punch above their weight particularly in kick-starting things (indeed as Australia did with solar PV R&D). It’s a depressing thought to believe that we might as well not bother trying to do anything meaningful or important because we’re too small to make a difference.
Feel free to correct me if I’ve misinterpreted what you were trying to say, or shoot down any of the points I’ve tried to make!
Tristan, loved the article, there’s some graphs and data analysis here that I haven’t seen anywhere else in this debate and help cut through the spin and misinformation.
Hi Jonathan, the renewable incentive schemes have certainly absorbed some of the market share of coal and gas generation, and hence emissions, there’s no doubt about that.
Your point 1 (the voting analogy) only works if each solar panel has an equal value in addressing emissions (incorrect because of global and regional solar resource variability), if each individual has equal access to a solar panel (incorrect because of differences in subsidies / cost / income / infrastructure), and other implicit differences e.g. are human emissions the root cause of global warming in the first place… plenty of debate there! This analogy also only works if each individual (or even country) has an equal say in emissions – plainly incorrect when you consider the EU, India and China who have no emissions reduction agenda.
To your point 2: if emissions are directly “the greatest threat to humanity” then the IPCC should recommend the logical action that will have the quickest and greatest emissions reduction – rapid deployment of nuclear power generation. The fact that the IPCC does not recommend nuclear, and instead recommends variable renewables is telling. If emissions are the greatest threat, surely the most logical response is to reduce emissions rapidly on a large scale, therefore surely nuclear energy would be the recommendation?
Alan Finkel stated that Australia can make no measurable difference to global temperature. This is an incontrovertible fact.
There is no “global community” working together to reduce emissions.
The IPCC does not recommend the logical obvious step to reduce emissions.
Nothing I’ve said above is controversial, hopefully that helps you to understand my objection to subsidies(or levies, these are not rebates).
They are bidding in good faith because they intended to honour the bid before market conditions changed.
Their rebid describes the market change that required a change of bid. Maybe Torrens should have rebid a little earlier eh?
Insofar as the constraint is a submarket within the NEM they have to sell at -1000 to remain generating. When the submarket begins to impact the actual dispatch price they have to change bid.
Paul, I trust you will be highlighting all of the gas generator rebids for “avoid uneconomic start” where they fail to ramp up 100 MW when prices spike?
I find it laughable that you expect Infigen to cop massive losses due to extremely complicated market conditions. Keep in mind that during these periods of Intervention pricing that predispatch price is totally misleading. Traders are flying blind and NEMDE is trying to minimise their revenue. Good luck out there!
Hi Tristan, regarding a couple of statements in your blog post, I feel obliged to highlight what I believe could be some factual errors.
“Solar – shown in yellow – is simply knocking off the top of the peaks in demand that would otherwise be serviced by what has become quite expensive, and hard to secure gas.”
The afternoon load peaks in the NEM occur after combined solar output has begun to reduce. I agree there is some overlap between combined solar output and afternoon load peaks, but the amount of overlap differs across the regions and the time of year. I believe it is misleading to state in simplified fashion that, “solar reduces the load peaks”. The duck curve created by the day time solar output forces dispatchable sources to increase their output at a faster rate than without the solar.
I agree with your analysis that in terms of bulk quantity of electricity generated, solar is nowhere near the amount of electricity demanded. No argument there!
But what you might have given is a summary of the various recommendations and concerns from AEMO about inertia, fault level and other grid stability matters. For example in SA there is a minimum combination of synchronous machines required to remain on the grid, with AEMO regularly and increasingly being forced to intervene to ‘direct’ renewable asynchronous generators to reduce output and synchronous generators to start or remain online. There are plenty of reports on this in the public domain.
I suggest that the intent of the conversation being directed by the various stakeholders in the NEM is more around the quality and stability of the grid, rather than the absolute quantity of solar power being generated.
Regarding the voltage, the allowable limits for voltage are 230 volts +10/-6%, which is the range set out in Australian Standard (AS) 60038 (Standard voltages).
230 x 1.1 = 253 V
Another note on this is the transformers closest to the loads are only manually adjustable off-load, so they have to be set to cater for voltage swings between day (solar inverters pushing up voltage) and night (no solar inverters). Transformers at higher voltages and substations can be adjusted online from the control room.
This could be why there are limits on both the capacity of individual household solar panel installations and the number of solar panel installations on a particular distribution transformer.
If you look at the first chart illustrated patterns on December 2017 and in particular the days with the highest peaks it shows quite clearly that the absolute peak occurs where there is a whole lot of yellow. The peak which occurs with the red colour is lower than the peak that has yellow. You see similar results in January and February which tend to be the other months where peaks in NEM-wide demand occur.
Secondly, if you look thoroughly at the voltage distribution charts you’ll see they provide two lines – one for night-time voltage and one for daytime voltage. In both cases the voltage distribution is very similar and reflects the fact the voltage is clearly not remotely aimed at achieving nominal voltage of 230V. The point of this is that in most cases the networks could adjust the transformer taps (yes I know this is done manually) to deliver lower voltage and not be at much risk of falling to voltages that were too low, while allowing significant room for solar to push voltages up without much risk of reaching excessive voltage levels.
On your point about maintaining frequency/fault levels etc. yes these are legitimate issues in the case of SA’s very high levels of renewables penetration but not a major issue in other parts of the NEM and were not the subject of media reports which I am addressing in this article which mentioned issues such as overloading transformers and causing blackouts and producing far more solar generation than could be absorbed by demand. I would note that many of SA’s issues (exception of fault level) will be largely resolved with SA-NSW Riverlink interconnector that has been recommended in AEMO’s ISP. I find it quite irritating that the media uncritically report complaints by some sections of industry to support a nation-wide change in policy which when you scratch the surface are based on solar reaching very high penetration on boxing day in a single state representing a small portion of NEM demand and which will be resolved by undertaking actions that AEMO has already recommended and are in the process of being implemented.
I won’t bother to address your argument about emissions because you make a series of assertions that reflect someone who I suspect has made up their mind a long time ago that they find the conclusions of Academies of Science across all major countries so unpalatable that you have become impervious to logic, evidence or just simple ethical regard for your fellow human being, particularly the poorest of the world who are most vulnerable to the impacts of global warming.
Tristan,
With regard to the street level transformers averaging above nom 230v; would you agree that firstly, current is lower, therefore transmission more efficient. Secondly, domestic PV presents as a sudden load when a cloud covers the sun momentarily, at the same time the line voltage drops, so the line loses are momentarily compounded. Voltage drops because loss of PV and sudden increased load on the grid causing any inductive load to draw even more current. The upshot is the swinging loads caused by domestic PV at a local level are problematic. Power factor can also swing due to local conditions. Batteries will help but they are an extra cost.
Thanks Tristan, I’m happy to be corrected on the peak load, you are right that the ‘behind the meter’ solar output from rooftop panels is greatest during the day. I had been looking at the AEMO dashboard which of course only shows the solar output as a reduction in demand during the day. The unintended consequences I have mentioned (dispatchable generation ramp rates) still apply.
I’m pretty happy though with the logic I have presented on subsidies, nuclear generation and emissions. This logic is, after all, not open to attack because it is based on facts.
Switching the argument in an attempt to prescribe guilt at the theoretical unproven sometime in the future (maybe) plight of the world’s poorest is difficult to align with the argument that for zero effect on global emissions, Australia has some of the highest power prices in the world, which disproportionately affect our own poorest citizens. Choosing this style of argument does nothing to help the world’s poorest, even if they are prone to some consequences of ‘climate change’ – refer my argument above on the IPCC recommendations – since actions in Australia to subsidise renewable energy can have no effect on those people.
If you think it will have an impact, and that I am “impervious to logic”, then please take the opportunity to educate me and other readers.
Ben
Thanks for methodically taking this claptrap to task. Tristan traditionally takes a one dimensional view when developing his pro renewable propaganda. That the ESB, AEMO and other responsible authorities need a lesson on the virtue of solar panels is a preposterous assertion.
Tristan, I think you need to understand how distribution low voltage feeders are necessarily designed. Bottom line is domestic demand peaks always occur after rooftop PV is essentially gone for the day. So the design criteria for the feeder is around the peak load. Now a transformer is actually a voltage source – built to a price with fixed tap changing. The lengths of low voltage feeder runs are determined by the allowable voltage drop along its length. It is a compromise. A customer connected near the transformer will see high voltage (at the top of the allowable range) whereas a customer near the end of the feeder run will see voltage at peak demand near the bottom of the allowable range. The transformer off load tap changers (normally 2.5 % steps) allows a small degree of tuning to this profile over time. This is simply good engineering design – function at least cost. The overnight voltage difference will be small simply because domestic overnight load is small – volt drop goes away as an impact. The problem with distributed generation is suddenly the volt drop issue runs in reverse with power flows back towards the transformer in mid morning to early afternoon – very little domestic load (empty houses) and peak solar irradiance. The design criteria is basically now completely wrong. The solution is the installation of on load tap changers to distribution transformers (usually a transformer change out) – not normally very attractive as distribution transformers are generally set and forget – tap changers are not – or you install more transformers and shorter LV cable runs. Both solutions are going to cost money that will end up in the capital base of network service providers (what happened to “causer pays”?). This stuff is not free.
Paul, can I suggest that in the case of blown LEDs, you attempt to get the temperature in the suspect downlight fittings, and compare that to lights that work fine? Voltage will affect all lights equally (and any other electrical appliances), while temperature and moisture can affect individual fittings separately.
I have had combinations of flouros and LEDs for years with no issues, and my power sockets read 245V pretty regularly (I have 3kW solar panels and am about 50m from a pad mount substation).
I recently changed 8 flouro donwlights to LED (Edison thread) and they are all fine, except for one that lasted 1hr and went kaput. I put this down to a manufacturing fault, since any other bulb works fine in that socket.
I have found that temperature, moisture, vermin, dirt and build quality have played a greater role in bulb life at my place than voltage.
Your general quest seems to be to understand more about power quality in general, in which case there are some terms that you might need to understand in more detail:
– flicker, unbalance, harmonics, overvoltage, undervoltage – there is more than just voltage, and any number of IEEE articles discussing the power quality effects of solar inverters and similar non-linear power electronic devices.
Q1. Does solar output drive voltage higher – yes! I visualise this by thinking about the situation with no solar – all the power is coming from a fleet of generators far away, with associated losses and voltage drop. With solar being exported locally you are effectively reducing the load and the losses, so the voltage increases.
Q2. Voltage is supposed to be 230V +10% -6%, but its a big deal to change all the off-load tap changers in distributed transformers in all the suburbs… takes time and coordination and priorities
Q3. Implications for equipment – see above discussion on power quality and local conditions separate to voltage
Q4. You can buy separate power metering devices that plug into a power socket, gathering information might be the first step.
A1) Voltage rise during load is a characteristic of Line Drop Compensation for customers at the front end of the 11kV feeder. Wikipedia has a good article on LDC and associated problems due to distribution generation. https://en.wikipedia.org/wiki/Voltage_regulation I strongly suggest reading this.
A2) While some claim that the 230V standard was introduced to “allow more solar” this is a furphy. As a matter of fact, the 230V standard was introduced back in the year 2000 (well before widespread adoption of PV). It was actually introduced to facilitate more voltage drop along the feeder, to squeeze more out of existing electrical assets. The old 240V standard (only recently superseded) was 240V +/-6%. This makes the voltage range from 254.4V down to 226.4V. The 230V standard has a range of +10% to -6%. This makes the voltage range from 253V down to 217V. As you can see, the upper limit hardly changes (1.4V), yet the lower limit drops considerably (9.4V). To be more specific, the system voltage has to be below the upper limit 99% of the time, and above the lower limit 99% of the time. If it is outside these limits, then complain to Energex. They are generally very responsive.
A3) Maybe. But there are a multitude of factors. As Ben has pointed out, temperature, moisture, vermin, dirt, and build quality all play a part.
A4) I would start with ensure adequate ventilation around the LED bulb and buy good quality components.
Just some other points;
Energex has been installing power quality metering equipment on the 11kV/415V transformers for quite a while now. I would suggest reading their public domain document titled, “Power Quality Augex Forecast 2015/20”. It is very detailed and also addresses some of your specific concerns in the article.
The old BC/ES fittings have significant safety issues. A child or adult can readily put their finger onto energised terminals. The newer GU10 and/or 12V screw fittings are much safer.
Hi Paul, I’m suspecting that your down lights aren’t on when the sun is out, so a higher voltage due to solar output is unlikely to be the issue. Heat build up during operation on the other hand would be, as Ben suggests. I’ve had some flush fitting down lights now for a few years and not a single one has failed to date. I suspect it’s because the switch mode power supplies are separate to the lights themselves and live on top of the ceiling insulation. They were about $100 for a box of ten from ebay, great value! The previous LED’s they replaced were bulb type lamps installed in can type down lights and they got as hot as Hades and failed regularly. LED bulbs fitted to open holders on the other hand have been fine.
Great use of Solar Analytics data, Paul. I couldn’t help but dig a bit further. I’ll send you some more info.
For the moment, on question 1:
Yes, PV generation will correlate with grid voltage to some extent. Your inverter needs to raise the local voltage in order to push current back upstream towards the transformer. How much depends on what else is happening on your local feeder. So if you and your neighbours are all generating lots of energy, voltage will go up but if you are all consuming lots, it will go down.
You’re seeing some pretty massive swings throughout the day, so I’m guessing you are towards the end of a feeder with plenty of solar on it.
Possible solutions:
1. Balance the generation. You’re on three phase, so a three-phase inverter might have helped reduce the voltage rise on any particular phase. I also see that the voltage on one of the other phases is much lower, so putting the inverter on that phase would even things out a bit. In these cases, some installers put Solar Analytics on before they install the system so they can find the phase with the lowest voltage.
2. Reactive Power Control. Qld now requires power factor of 0.9 lagging to “consume” reactive power, which helps to limit voltage rise. This doesn’t affect your real power generation unless your inverter is maxed out. Perhaps check if your installer can enable a reactive power setting on your inverter.
3. Load shifting. Balancing generation and demand at local level helps to balance voltage and increase the amount of solar that the network can handle. We’re working on a virtual peer-to-peer product to help non-solar customers participate in that load-shifting. Most networks also have demand response trials but they need to get a move on.
4. Tap change. As noted by others, the standard is now 230V +10%/-6%. It looks like they could easily go down 5-10V. It is a tough gig as a DNSP though as they have so little visibility and these changes cost money. We’re trying to help as our database grows with loads of useful data like you’ve demonstrated.
I would like to know from someone who writes here, can solar from a house go into the grid, and back through a transformer? to say the next area and then to another house. Or does it have to stay on the transformer supply side?
In short, yes. The transformer will change the voltage (and therefore current) but the power will flow throughout the grid to wherever it’s required. Generation and Consumption need to be balanced across the grid as a whole (e.g. the NEM connecting Qld-NSW-Vic-SA-Tas) otherwise the frequency will rise or fall accordingly.
So solar power from a roof can go all the way back to say a main generation point, being stepped up?. Lets say the Local Wind farm that is for this example say 100K away?
Yes. Electricity will always flow from a point of higher voltage to lower voltage. Solar inverters push power into the network by injecting it at a voltage slightly higher than what it’s connected to. Hence the topic of the article.
Transformers don’t care which way power flows. Either direction is fine, although because they were designed for one direction, they may be a little more efficient in that direction.
There are a number of feeders (11kV lines from sub-stations) that already go to net positive areas. Currimundi for example, as a suburb has periods of good solar that generates more power than it consumes and that power goes back to the sub-station and out to another feeder.
Technically, there is nothing wrong with all of a sub-stations feeders being net positive and the whole sub-station would push power further up the transmission network. So long as voltage levels throughout are within acceptable range, all is fine.
Thanks for sharing, Paul.
Your plot shows a very large difference between day-time voltages and night-time voltages. It made me wonder how many sites have a similar range so I ran some numbers and found that you are indeed at the upper end. I’ll send you some more data.
The bias against nuclear technology in the IPCC is changing, albeit too slowly.
Using it to legitimise the idea that mitigating greenhouse gas emissions is not a very important thing to do seems just as fallacious as the thinking you seek to identify.
I certainly share your frustration. I think the inclusion of those technologies, rather than causing conflict, could prove a remarkable circuit breaker in Australian energy policy.
I would like to present an idea which makes a convincing argument for extending Australia’s existing National Grid as soon as possible.
Extending Australia’s National Grid should be our highest priority.
Technology today includes UHVDC, which allows for transmission across great distances (3,000km), with very little transmission loss.
All renewable energy projects by private enterprise will automatically follow.
“Build it and they shall come”
Please check our website for detailed information on our proposal.
As far as you emotion-O-meter goes, I’m aware of at least 5 organisations that have made some kind of detailed study of 100% RE and one of these organisations keeps the lights on 99.998% of the time for the NEM, the AEMO. Are you suggesting BZE, UNSW academics, AEMO, SEN, UTS for GetUp! are all “loopy left” organisations and not competent at engineering analysis and modelling?
Trying to claim credibility by asserting one is part of a “sensible centre” is nothing more than rhetorical spin. I could deconstruct the “sensible centre” concept if you wish but my point is more about you positioning 100% RE (or say 95% RE for that matter) as some kind of unrealisable ideal only subscribed to by hippies who don’t know what they are talking about.
Alastair Leith
Outreach Coordinator
Sustainable Energy Now
In amongst the questions that form the premise of the article could be another question:
if we are providing subsidies to large scale generation of RE including utility solar plants (i.e. to corporations) in the form of LGCs (soon to be quite modest if RET target is not extended past 2020) then should small businesses and householders be able to participate in the same kind of benefits?
So far it’s only 1239 MW of wind, which would be negatively correlated with solar, so there should be enough local demand and transfer capacity for it not to be constrained. However, there are some other proposals not on the list such as Gennex and Kennedy.
Sun Metals have plans to increase the size of their zinc refinery, and securing cheaper electricity is part of their business case.
The North Queensland region is similar to the South Australian region in terms of demand and transfer capacity to the adjacent region, but there are no system strength requirements for local synchronous plant to be spinning. Is this a historical quirk of it being a single Queensland market, or does physics work differently in Queensland to South Australia?
Interesting point Malcolm M. North Queensland as a region would be a similar size to South Australia. It is only because SA is a separate region with the possibility of islanding from the rest of the NEM does it have an arbitrary requirement to have x amount of synchronous generation for system strength.
North Queensland has no such requirement at the moment (despite operating just fine with little synchronous generation), yet one could argue that it is just as prone to islanding as SA.
So if it became a new region, would such a requirement be imposed on it? (to be able to continue islanded in case of interconnection failure).
Regardless of becoming a new region or not, it would be wise to establish the capability to be able to operate islanded. Traditionally this capability would only be done with sufficient synchronous generation, but that is very unlikely to happen at all. So other methods must be explored.
There are alternative solutions. But it isn’t likely to happen with slow incremental evolution of the network. Planners have to step back and consider much more revolutionary changes to the way the whole network operates, and then start planning the changes required to achieve that. Thinking only one or two steps ahead as is the case now will not work.
At some point, the train has to jump rails. The current track is on course to a disastrous dead end. Time to design and build a new track that keeps going far into the future.
Strange comments.
In reply to yarpos:
The purpose of derivatives is to make solar power more reliable and dispatchable. Who are the rent seekers? Perhaps you are also against lending money by banks.
In reply to Shilo:
Solar and wind power is predictably intermittent, but not controllable. Coal and gas are controllable. The proposed derivatives will bring more private funding into solar and wind forecasting. This should also improve your boating experience.
It will be interesting to see if this forecast holds. It is the Friday before the Christmas break, and historically, a lot of industry (both heavy and light) closes up shop at midday and opens the beer fridge. I am not sure if AEMO models beer fridge loadings correctly. Further, the numerical weather forecasts from BoM ACCESS-C and ECMWF have the sea breeze front pushing in somewhat deeper across metro areas of Brisbane, Sunshine, and Gold Coasts. By 1600hr (time of forecast peak demand) the sea breeze is starting to push past Brisbane’s western suburbs. Finally, there is a vigorous south east change moving through, and by 1700 it is forecast to be sitting in a line from Southport through to Boonah. Shrug. If AEMO forecasts do hold, then we could see some amazing system loads if there is a hot spell in February. (Isn’t the electricity industry supposed to be in a death spiral?)
At least somebody is discussing this risk. Not likely to be discussed much over at the toxic bubble that is RenewEconomy…
I’d also draw attention to the AEMO SA Electricity Report 2018, which amongst several other worrying statements, has this to say on page 5 within the executive summary:
“AEMO forecasts risks to supply in the next five years in South Australia, due to the region’s interconnectedness with Victoria. To avoid unfairly penalising one region for a supply deficit spread through several interconnected regions, an equitable load shedding principle applies in the NEM. This principle states that load shedding should be spread pro rata throughout interconnected regions when this would not increase total load shedding. Therefore, while the reliability standard is not expected to be exceeded in South Australia, high risks of load shedding projected in Victoria are likely to result in any supply deficit being spread across both Victoria and South Australia.”
Equitable load shedding principle?
Load shedding should be spread pro rata throughout interconnected regions?
Wow, look at that, 9602MW @ 16:25. And on a Friday afternoon that is part of the Christmas shutdown. I wonder how much the cloud band coming across Brisbane contribute to this peak. For example, the western facing bank of my PV system rolled back from 2462W @ 16:13 to 445W @ 16:18, all in the space of five minutes. (No way I was turning down my air conditioners.)
From the last panel it would seem that an extra 2 GW of low marginal cost wind and solar generation would push the bid stack far enough to the right for prices to be determined by black coal rather than gas. There would be at least 2 GW of wind under construction at present, and with designs that produce well under moderate wind conditions. So it would only take moderate wind conditions in the evening peak to squeeze out gas and peak prices would be determined by NSW coal generators.
I expect there would then be some repricing. The coal units that cycle their output regularly (such as Eraring) would put in low bids for perhaps the first 60% of their capacity, and bids just below the cost of gas for their remaining capacity (used in peak). This would be necessary to protect revenues. Meanwhile the Snowy hydro stations would also need to reprice their output in order to maintain downstream flows, again to just below the price of gas.
It’s going to take a mountain of those Tesla 2170 lithium lego bricks to smooth that lot out and keep the lights on but if you can get a sizeable heap of them up and running early you can clearly make a motzah creaming FCAS returns off the top.
For the battery, you’d be looking at FCAS prices, which have been reasonable so far. That said, forecast price spikes to $10000 per MWh did not eventuate yesterday.
What I’d be interested to know is what “negative settlement residues” are and why they can lead to constraints being set on inter-regional interconnections? (for example Market Notice 66430 issued by AEMO this afternoon).
Thanks for the article Jonathan, found it very informative. I’m wondering if you might have come across a company called Faraday Grid, and what you might think about their products / claims if you have read anything about them? They claim to have developed power flow control devices called “Faraday Exchangers” (I believe they replace existing transformers) that autonomously maintain voltage, frequency and power factor efficiently in distribution and transmission networks. The Exchanger is also said to generate its own inertia providing greater security in a grid that is increasingly reliant on VRE. I wonder if you think the “Faraday Exchanger” could revolutionise NEM infrastructure, as they seem to claim, and could displace ancillary service providers? Keen to hear your thoughts!
A mountain of Lego Bricks (Tesla) will only increase the overall demand, they are like a water tank with a hole in the side of them. 20% minimum, leaks out to onto the ground. (And no one that i know of has worked out, how far the power coming out of them can actually go. Meaning what really are the longer distance losses from operating them. (Yes I know they can be put near the load, however you must look at the average and when longer distances require power they must deliver it)
Another fun one today, at dispatch interval 16:40 to 16:45 on 2019/01/22 the price SA, Vic and Tas hit the ceiling of $14500, it then plunged to -$1000, and oscillated between positive and negative price for a few intervals. There appears to be no market notice associated with this event, except for “prices subject to review”.
As the ‘gap’ between power demand on the one hand, and the some 20GW of power still available from coal, plus solar plus wind power on the other, increases above 5,000 MW, on the east coast, large peaks of gas and hydro output are called on for the difference. Given gas/ hydro have little competition in responding to these peak demand situations, these sources will continue to charge very high prices at such times. Whether the market actually goes crazy or not seems more a function of market characteristics.
New solar power capacity, presently being rolled out, should reduce such high price occasions, during daylight, if the network is up to it. But the naturally high incidence of 5% to 25% wind load factor occasions, will ensure high price periods will continue to occur often, when power demand is high. These occasions will increase too, as coal power capacity is driven out of the market and the ‘gap’ to be filled by gas/hydro, is thereby increased at low wind times.
Storage may become a solution if the increasing cost of grid power is seen as of little consequence.
The situation was extreme, amd also imports from NSW seem to have been constrained, possibly due to a bushfire.
LOR3 declared in Victoria, per market notice 66670号
“The maximum load to be interrupted is 266MW at 1810 hrs 24/01/2019”
“Unfortunately we also see a significant loss of supply from aggregate wind generation that comes before both the sustained price spikes (800MW lost before spikes) and the peak in demand (>1,000MW lost prior to peak demand)”
That is the nature of renewables and why we are in this mess to begin with.
I rarely if ever see the mention of capacity factors for renewables with wind averaging 20-25% of nameplate capacity. so regardless of the fact we have over 5600MW of wind installed, we on average get 1000ish MW and sometimes much less.
Thank you for including the Hornsdale battery information.
It is important for people to see that even large scale battery storage cannot fix these problems.
Is that correct? During peak demand we had 2000+ MW of wind and solar and AEMO can always constrain wind or solar if there is going to be a forecast shortfall…
Australia wide wind power oputput had been declining all day, hitting the low at 3:35pm (16% output or 939MW) just when solar was coming off its peak.
SA wind power began its decline from 7am. It dropped from almost 70% (1300MW)in the early morinng to 50%+ (100ishMW) output at 9am and then plumeted to around 15% (300MW) from 2pm, falling as low as 4% (80MW)later in the evening.
Victorian wind power output was relatively stable at between 30% (500MW) and 40% (700MW) for the morning but then suddenly declined after 3pm and stablised at the low point of 20% (360MW) between 4:30pm and 9pm which was when demand was rising and solar output was declining.
The fact is that you cannot rely on solar and wind to provide reliable supply on hot days with warm nights as these usually occur when there is low wind.
Wind capacity factor in SA is typically 28-32%, long term average. Go to the AEMO website and download the SA Electricity Reports, that tells the story very well.
Yesterday there was more diesel generation (400 MW) in SA than wind generation (270 MW) around 7pm after the solar generation went to 0 MW.
Constraints during peak demand today were for net flows toward NSW, even while load was being shed in Victoria. That one might take some explanation from AEMO.
Rereading the post, it is likely the same constraint was in operation as yesterday N^^V_NIL_1, preventing voltage collapse on the NSW end should basslink or a Victorian generating plant trip off. Perhaps investment can be made to improve this situation (for example STATCOM equipment at Darlington Point).
Being retired and from the “old school” where things were centrally controlled I am endlessly amazed at the discussion about the complexities that go with this market system. If the control were under one roof and basically up to one person supplied with all they need to be told about the weather how simple thing could be. A list of serviceable plant. Forward planning not much different to a steam engine driver about to go up a hill and getting as much pressure ready as needed. Not rocket science. Now with forward trading – derivatives – no useful integration – AEMO only now seeming to realize that they need to take some form of control there are complexities that are simply created by the design which is not fit for purpose. Thank goodness I don’t have to work with something that I could never believe made sense. You guys have my total sympathy keeping a straight face as you seem to do, or may be being too young to know that it need not be like this. Ever done some work on a washing machine or the car at the weekend and wanting to meet the idiot that made something hard to do. Who was the …. that made it like this?
The Battery is being used like a water tank, with a hole in it. Why on earth could it be cycling on a day like yesterday. WHERE was the power coming from to charge it. WHY was there power allowed to change it??????.
That has to be looked at in detail, because which ever way you look at it, on a day like yesterday, the battery is only making the situation worse. It takes more power to charge it than it can put into the grid, ITS SHEER MADNESS, to allow it to run yesterday!!!!!!!!!
Agreed, just like Snowy Hydro 2.0 which will require huge amounts of energy to pump water uphill and then only be able to provide 70-80% of that energy back when it is needed.
These storage solutions aren’t fixing the problem of renewables and prices
It’s true that Snowy 2.0 is a net load, with no additional generation. What they are hoping is that wind power can be used to replenish the pumped storage, instead of curtailing the wind at nights. Not much being said about the additional transmission line CAPEX and OPEX necessary to ensure wind farms can send their power to the grid without being curtailed, and what happens when the wind disappears for days…
I did not see it “cycling” during the high prices. It discharged at 30MW until it was spent. No surprise there. There is simply a shortage of dispatchable generation. Wind and solar is somewhat predictable although not necessarily dispatchable when required. Fossil fuel generators are generally dispatchable but their reliability is unpredictable. We see coal generation having issues and other generators severely limited in output due to heat. If there was a more storage available in pumped hydro (for example) the system would have been much more stable. High output – but only needed for a couple of hours. Blaming wind,solar, coal or any other form of generation is a bit like blaming who lost the football game. Was it in the last 2 minutes when the team missed that goal and hit the post, or was it the 1st quarter when the losing side didn’t kick any goals??
@ Ian: regarding your post your words are largely correct but the context matters, in this case the context is blackout occurred because of a lack of power generation. Power shortages are a logical result of weather dependent power generation.
Lack of maintenance and zero investment in dispatchable generation are a logical result of subsidised renewables.
The whole scenario is entirely predictable.
But we have commentators like Nicky Ison from The Guardian, with zero technical knowledge in the subject, touting an old AEMO report discussing what it would take to achieve 100% renewables. The report says the idea is rubbish, but the article goes on to say it’s a great idea and perfectly achievable.
Hi Ben, yes I agree with you pretty much. Paul regularly talks about people at each end of the “Emotion-o-meter”. Some people blame wind and solar (whose output was low which unfortunately does happen at these peak times). Others blame coal/FF generation, eg X units were out of action due to failure, or were limited to X amount of output due to heat (all true and correct). What will it take to change the situation, while the small amount of storage introduced recently in terms of batteries I think has worked well for it’s intended purpose, the amount of stored energy is very small when this type of event occurs. Do you think we will see a change of market rules in the future introducing capacity payments perhaps? We certainly need something to incentivise storage, although if we have many more days like this I think the monetary incentive might be already here!
The 600 transmission constraint equations seem like a black box that are probably only understood by a handful of people within the industry. When a constraint equation is imposed how much analysis is there of risk vs consequence? For example on Friday 25 January Victoria had to shed 200 MW of load, but there was spare generation in South Australia that could have been used if there were a temporary lifting of the 500 MW constraint on South Australian exports to Victoria via Heywood. The physical works on this link were completed over 2 years ago to allow 650 MW capacity in both directions, and an export of 600 MW from Vic to SA is now allowed. But are the risks that the 500 MW constraint is trying to avoid sufficiently great to start load shedding in Victoria? Or would it have been better to try increasing the constraint to 600 MW and monitor its effects. If there is problem, the worst that can happen is it is reimposed to 500 MW and load is shed.
THe low output of wind during Victoria’s price peak needs to be contrasted with the high output from solar farms, which until 5:30pm were producing 200-250 MW. By this time next year this should increase by 500 MW through new solar farms under construction, and completion of the current solar farms. If there is an equivalent amount of roof-top solar, we would have quite a different situation of perhaps another 1000 MW of solar into the Vic market, and a much shorter period of price peaks.
Inefficient infrastructure attracts capital
There is astonishment in the industry about the timing of EnergyAustralia’s maintenance program for one of its Yallourn units, and concern about the ability of both Yallourn and Loy Yang A to provide “firm” power when needed, particularly given their owners intend to keep them operating for another 15 and 30 years respectively.
But there is also anger at the bidding practices of some of the big generators in the lead-up to the heat-wave and over much of summer, something that has been building up for the last six months following AEMO’s warnings, the spike in futures prices and the scarcity of contracts available on the market. https://reneweconomy.com.au/fury-over-generator-bidding-this-time-they-have-gone-too-far-39712/
The coal generators had been running at 100% output for the last 2 months due to the closure of Hazelwood.
Blaming coal plants for not being able to run at 100% when wind and solar are only able to produce on average 20-25% output and aren’t dispachable is disingenuous.
reneweconomy is a garbage fire of a website and should be ignored.
According to an article in the Herald Sun, AGL had just spent $136m on preventative maintenance on the generator unit at Loy Yang A that had just failed. There must surely be some questions within the company about the returns on such life extension investments in coal-fired stations, when similar investments in solar or pumped-hydro would be much more reliable. The Loy Yang maintenance is $0.25m per MW, compared to about $1m per MW for a pumped hydro that would probably last 30 years.
Hi I read yesterday on http://joannenova.com.au/ that the cost for electricity in Vic and SA was nearly a billion dollars FOR ONE DAY! Amazing and one poster there claimed ““Three days and you could buy a HELE plant with the money wasted.” That’s a power plant that could last 70 years, and provide electricity at under $50/MW”.
Renewable Wind and Solar was developed around the time of the first coal-fired electricity generator circa 1895. It was very limited engineering then and it’s the same now!
Paul, is there any room in your work to look forwards and consider this scenario occurring in 5yrs time? There is another 2100 MW of wind committed for VIC in the AEMO generator register, which will be in addition to the existing 1050 MW of wind. Back of the envelope calcs tell me that if existing wind capacity was delivering 400 MW (Thu) and 650 MW (Fri) during peak demand, then a total of 3000 MW wind would have delivered around 1200 MW and 1950 MW for the same wind conditions.
This additional wind capacity is already being planned, and if > 42 degC is about a one-in-three year event we can expect similar conditions to happen again.
You have to consider that the wind farms in place now are at the best locations.
The more wind farms we install the worse the actual output will be because they are going to be located in even more subprime locations.
You also have to consider that while the average output (20-25%) is useful indicator, the system is only as strong as its weakest link which means the numerous periods where we have 24 hours or more at less than 10% output need to be the focus point.
Just look at Victoria wind output for 21, 22, 26, 28 and 29 January. It is way lower than output on the 24th.
Bear in mind that Victoria has 1740MW of wind farm nameplate capacity:
The period from 8:15am-12:15pm on 21st the maximum wind output in Victoria was just 40MW
The period from 6am-10:40am on 22nd the maximum wind output in Victoria was just 23.8MW
The period from 9am-12pm on 26th the maximum wind output in victoria was just 20MW.
Batterys, and heaps of solar and heaps more wind will fix it. How any one makes money from it, does not matter, because the bigger picture is that the world will be saved from catching fire, even though the sea will be so high everything will be covered with water.
An observation. The price for the NEM electricity grid is very volatile, for a number of well documented reasons.
It may be comparing apples to oranges in some ways, but the Western Australian stand alone grid is much much more price stable.
Paul do you have information and data about the WA electricity grid ?.
I understand you are very busy, but the differences may be of interest.
I noticed that on those hot days where Victoria was struggling, there were a few occurrences of NEGRES CONSTRAINT on the interconnector between NSW and VIC. I don’t quite understand what that means, but from the little bit of reading I did at the AEMO web site I get the impression that flow from one region to the other is intentionally reduced because of some sort of accounting consideration even if a region really really needs the electricity to be flowing on the interconnector. Did this cause VIC to be under-supplied on the 25th of January?
I guess another reason for the low flow from NSW to VIC on that day might have been the fact that MURRAY power station was going hell for leather, and that meant that there wasn’t enough capacity on the NSW to VIC interconnector(s) for NSW to export to VIC?
Perhaps you can write a brief article that helps us to understand why NSW wasn’t exporting to VIC at VIC’s time of need?
The big change over the next year will be an increase in solar into the Victorian market. Over the last few days large-scale solar has produced 200-250 MW until 5:30 pm, and there is another 500 MW under construction in addition to rooftop solar. Add to this another ~200 MW nearly complete in South Australia (Bungala 2 and Tailem Bend), which would be able to help out Victoria if there is capacity on the interconnector (which there usually is around 5:30 pm).
Utility solar cracked 250 MW once in the last 7 days, typical is about 170 MW peak. You are correct though, that more solar will help the grid for 6hrs between 9am and 3pm. And of course rooftop solar will continue to increase, with around 900 MW peak now. And solar is relatively predictable, much more than wind.
And wind is the big problem for the grid, in my opinion. Another 2000 MW committed for Victoria, eroding competition when its windy and rarely available for peak demand.
I wish people would stop talking about batteries when discussing bulk power delivery. Batteries are good for small power islands, short duration low capacity backup and frequency stability services.
And even 10,000 homes with a 5 kW battery is only 50 MW total, and only short duration, so even time shifting loads or peak shaving is a drop in the ocean compared to Vic peak demand of 8,000 MW.
The loss of coal generator units is going to happen more and more as we run them at 100% to make up for the closure of other coal plants (hazelwood) And the lack of reliable supply from renewables.
Of course we will get people like Di Natale blaming coal but the reality is that when we had hazelwood in operation there was spare capacity so you could take a unit offline for repairs and maintenance and they only had to be run at 80-90% output so they wasn’t breaking down as often.
Look, Wind/Solar and Batterys, do not have any issues, they work, when the sun shines, or the wind blows, it depends on the amount of wind, and direction, also the amount of sunlight and temprature, as well as how clean the pannels are, the battery’s work when ever they are needed, not as good when its hot, and as they age, but they still go. Now all of this stuff has to go through inverters and they work better when its cool, not so good when its hot, and the inverters can maybe at times not work, but there will be so many, you will not miss a few here and there not working and they will not be down very long because they will just be fixed quickly. So all in all, RE, just keep working all the time, they never stop. Vs Coal plants that stop. Cannot anyone understand that.
John, no rational person despises wind farms and solar panels, but a power supply network has to be able to supply electricity to match demand. When demand increases, either the power generation increases to match, or a storage option is called on, or the power is not delivered (blackouts, load shedding, demand management).
There is a place for demand management (as long as it is voluntary and doesn’t cost more than generation) and there is a place for storage (as long as it doesn’t cost more than other forms of generation and does not reduce the overall capability of the power supply network e.g. harmonics, fault level, inertia, reactive power).
Wind and solar cannot increase output on demand.
Power electronics (inverters) cannot deliver fault current, and only the latest technology can support frequency.
A business cannot operate when its power source depends on favourable weather.
Trying to compare a 40yr old coal plant suffering tube leaks with vagaries of the wind is not a suitable comparison.
Ben those numbers are quite large. But if the will is there sometimes a way is found. Germany is also still going forward with a lot of will.
next 10 years its going to get interesting.
But I will guess this, if the will falls short, it will only be, becuase of this or that, and it can still all work, it just need this and that!!!!!!.
John, investment in generation technology and assets has to be performed at the lowest cost that meets the system requirements. Government subsidies have effectively removed the function of the market.
Despite the massive subsidies and investment in renewables:
Electricity cost to consumers has increased
Electricity reliability has reduced
Electricity sector emissions have reduced
I put it to you John, what in your opinion is the priority for a country’s electricity system? I suggest that “reducing emissions” is not the core business of an electricity system.
Grid
Hi Ben, I think number one is, that we need power, I think number two we need as close to the lowest cost we can get in the world. We with coal had number one sorted, but failed with number 2, even though we could have had the cheapest in the world.
The path we are going down now, I just hope gets number 1 covered, because useing history as my guide for number 2. We are not going to have cheap power.
Hi John, I agree completely. Priority 1 = delivery of power when it’s needed. Priority 2 = lowest possible cost.
These priorities are closely linked through cost-benefit, with the largest deciding factor being the cost of not delivering power when needed. If the cost of blackout is less than cost of power, then power cost is too high. The reverse is where we normally sit – cost of power is less than cost of blackout, hence the NEM 99.998% reliability target, normally most at risk by network issues (much more wires than there are generators).
The “renewables at any cost” advocates and lobbyists want greater reliance on interconnectors and try to dismiss reliability concerns with “distributed generation”, except then we are at the mercy of a greater percentage of power in transmission lines PLUS the vagaries of the weather.
Hi Ben,
Yes plus just how much needs to be maintained its simply mind blowing.
I have been trying to work out Rooftop Solar, and am wondering if you might know about how far it can travel within the grid.
So by that i mean, if say all of one area at midday is awash with solar, and the solar has to travel backwards through a transformer to the go where ever it has to go, to then go through a transformer again, and to another person house.
What are the losses going backwards through a transformer, and also what distance can it go, once it has gone that way.
So lets say the transformer it goes through is 400 meters away, to start with, what would that loss be?, then being stepped up, what is that loss, then say another 400 meters to another area, in HV, then back down through another transformer, what is that loss?, then to someones house, say another 300Meters.
And for fun lets say that houses demand is 80% to charge a Tesla Powerwall battery?
What amount of solar power is going to make it into the battery?
Then what amount will make it out of the battery.
So all up the solar power only has to travel 1100meters?.
(I am going to guess the % is going to be lower than going just one way from a power station that is designed to deliver a long distance to a lot of people)
Good info. Could any updates please include a quick look at:
– interconnector activity, including any generation change in NSW that would have affected QLD loads
– coal and gas performance during elevated temperatures i.e. any derating occurring
– which gas peakers stepped in and any abnormal activity there i.e. fail to start
The system seemed to work well and coped with both high ambient temperature and high demand, with technical and economic components operating as required.
To answer your question at C1, the loss could easily be due to temperature derating. If the effective cell temperature of the solar panel is 60ºC the derating factor could be around 0.82 so your 5kW output is derated to 4.1kW On really hot days the cell temperature could even be up to 70ºC which would derate even further.
Looks like how someone would explain economics or something that is really is too complex to work properly.
Just my opinion.
It would seem that the system has lost its way on price. Things used to be far more stable and the price much lower.
Day by day its getting more and more all over the shop and the price just keep creeping up, yet i read how well everything is being run, and AEMO is doing such a wonderful job.
Agree on the complexity, not sure why there are separate prices in each when all states are connected.
I think criticism of AEMO is a tough call, since they just have to deal with the policy, without being able to influence it or influence any investment decisions. Although I believe the CEO panders to certain interest groups unnecessarily and against the intent of the AEMO reports that are escalating the language around reliability and cost.
There is no indication that cost of generation will come down at any stage.
John,
On what basis are you saying AEMO is controlling the price? Have you read this article? AEMO dispatch generators according to their bids and (generally) in line with the market rules. What you are saying is like blaming the ASX for the stock market for being too high/low/unstable.
Just because something is complex does not mean it cannot work properly.
Interesting that the LOR was cancelled yesterday but is back up again today with a 5 hour peak price period forecast in Vic and SA, similar to the 24 and 25 January 2019 event.
SA and especially VIC are sitting on All time high for a year of energy. With SA, the question has to be why?. What has gone wrong with the planning for renewables? They have known about them and must have planned for them, and have had a long time to plan for them, but the price is SA just keeps going up.
But, renewables are cheaper?. In every single study done from CSIRO, to banks, to industry.
It is more than just coal being shut down. It is a lack of control.
I hear you about the market complexity vs physics argument (twitter chat) but those extreme prices don’t seem to be triggering investment decisions. I suppose the RET has somewhat superseded price signals…
‘I flagged a propensity (by us all) to reach for some Magic Wand (or Silver Bullet) as the answer to the tricky problems inherent in this energy transition’….
‘I’d like to just reiterate that the prime objective of the Energy User in the electricity grid is not to “keep the lights on”, but rather to focus on the result of whatever useful work they want the electricity to do in the first place.’
Paying a business not to use electricity seems completely counter-intuitive. I understand that if generation cost goes high, that a business might have a threshold beyond which it is cheaper to stop producing. But surely this is cave dweller stuff? How can we justify developing an electricity supply system that deliberately restricts business output?
“Hi, sorry, we can’t make that coffee for you right now, we are practising demand management, since the cost of power is greater than what we can sell the coffee for…” That’s not a helpful conversation.
(And I know it’s a facetious analogy, but I’m practising humour demand management… 🙂 )
I enjoyed that write up, good job thinking outside the box. It is correct to point out that recent planning reports have escalated their tone and language around reliability.
So, a separate market for demand response sounds ok to me. What about a separate market for semi-scheduled generators?
Hi Paul, we already understand the AEMO bid stack where the power supply tenders are accumulated using the lowest price first, then the next lowest option until supply meets demand.
I was visualising two bid stacks, one each for dispatchable sources and another for non-dispatchable sources. The dispatchable demand would be total demand minus the non-dispatchable supply.
I believe this would calibrate each market correctly based on their true value. It would also remove what I believe is the masking effect of the RET on dispatchable investment signals.
I haven’t bottomed this idea out in my thinking yet, so I thought I would drop it here where somebody might give me some deeper insights.
I think the cost of an underinvested network should include the cost of blackouts due to capacity shortage. This is logical because an overinvested network shouldn’t have a capacity shortage.
How you’d measure or estimate the cost of a blackout I’m not sure.
I’d like to hear perspectives on the cost of network not related to demand, but attributed directly to reducing congestion from intermittent generation or facilitating new intermittent source connections. The demand is not increasing (rooftop PV) but the networks are…
Absolutely. The cost of blackouts is not just limited to the cost of fixing the infrastructure and getting it back up to code, but also the serious flow-on effects of power outages to the broader economy which will not occur with over-invested networks.
On the second point, the impact of increased intermittent generation is a major concern for regulators. As well as the impact on dispatch forecasting which I mentioned above, the AEMO is concerned about:
-the effect of DER generation in excess of regional demand during periods of minimum demand. There is no current mechanism to constrain down generation from DER;
-the progressive displacement of synchronous plant by inverter-connected DER plant removing inertia from the system. This challenge for system security and reliability was a major concern for the Energy Security Board in developing the National Energy Guarantee as well;
-the challenge of maintaining voltage within technical limits without visibility of DER.
Dealing with DER is going to be costly. Whether that cost is best passed on to retailers through a National Energy Guarantee is another question
Thanks Drew (for the original article and the response)
The AEMO report has some particularly interesting language around conservative safety margins and larger reactive reserve thresholds – all reducing efficiency of the system.
It seems the current market and regulatory environment are not set up to pursue the most efficient system of generation, transmission and distribution.
I find the “gold plated” argument frankly annoying. For starters, those who perpetuate the myth clearly have no knowledge of the external boom, bust, boom cycle that caused these gyrations in network investment.
Unlike Drew, I was there. I doubt Drew has ever stood beside a 50 year old creaking 110kV switchyard running above full load, hoping like heck the nearby circuit breaker wasn’t about to explode into a fireball, and wondering if he would be alive to see his wife and kids at the end of the shift. I did. More often than I cared to. (So, please excuse the passion.)
It is not a problem these days, but it sure was back in the summer of 2004. Every night, the local TV news would broadcast pictures of power transformers in Brisbane suburbs being sprayed with fire hoses in a desperate attempt to keep cool. How badly overloaded? Well Albany Creek had 30MVA load on two by 12.5MVA (25MVA firm), and these transformers were originally 10MVA uprate to 12.5MV.A with fans (no oil pumps).
Or how about 260MVA daily summer peaks at Victoria Park substation on two by 120MVA (240MVA firm). Note this is well above N-1. Even better, one of these 120MVA transformers had an electrical age of 50 years and wet windings. But under the severe capital rationing that Energex was under at the time, there was no funding to replace it. At the time Victoria Park supplied critical load like the Royal Brisbane Hospital (arguably Queensland premier hospital), St Andrews Hospital, and Brisbane Central Hospital, and a quarter of the Brisbane CBD. I could go on, but you get the idea of what the network was like in 2004.
Which is a nice segue into the AEMO report that Drew mentions. Why does the AEMO report select 2006 as the start year? My view is because it slants the story in the direction its author’s want it to. Why not start the comparison in network expenditure from, say, 1996? Because it tells a completely different story, and one that runs counter to their argument, and deflates the “gold plated” investment story.
But it goes back even further than that. The electricity industry in Queensland had a significant investment boom in the late 1970’s and early 1980’s for an anticipated mineral boom that never really eventuated. (It did, 25 years later.) This lead to severe capital rationing in the mid-1980’s and coupled with a strong push on cost savings via use of contractors, lead to the SEQEB dispute of 1985. After the government won the SEQEB dispute, there was even more rationing of capital in the late 1980’s across the entire Queensland electricity industry. Buoyed by their success, the economists kept up their severe capital and operational expenditure rationing during the 1990’s. And why shouldn’t they? After all the engineers were just gold plating the network.
In Energex’s instance, along came Greg Maddock in 2000. Greg was an economist. The short story is that he too believed in the “engineers are gold plating argument” and decided to cut network investment even further.
But something else was happening at the same time. Yes, electricity tariffs were low (around 14c/kwh), but the Queensland economy was moving along at a great clip. Price of electrical appliances, in particular air conditioners, was decreasing. Net migration into Queensland was running at 1000 people per week. Mining investment was starting to increase as well. But investment in the electricity grid did not follow.
Something had to break. And it sure did in the summer of 2004. Yes, the EDSE report provided some details into how bad things were. But it is what happened behind the scenes that explains what happened next. Economists and regulators all over Australia saw clearly for the first time how their “regulation” had resulted in significant under investment and this had taken the electricity network to the edge of the abyss. They peered over the edge into the black(out) below and knew what they had been doing was clearly wrong, and if it continued they were about to be publicly flogged.
So, the economists threw the problem back into the lap of the engineers and said; fix it, money is no object. So, not only did the expenditure in 2005 to 2015 have to match the phenomenal grow of the day, it also had to fix up the under investment of the last twenty years as well. Hence, the investment megacycle.
This is the true reason why the investment in the grid – yes “investment” not excessive gold plating expenditure – in the (very suspect) AEMO chart looks the way it does.
Now, if you think this is a fairy tale told by a grumpy old electrician, I point you to two excellent papers.
Simshauser, P. and Catt, A. (2012), “Dividend policy, energy utilities and the investment megacycle”, The Electricity Journal, 25(3):4, 63-87.
And the classic “death spiral paper”
Simshauser, Paul & Nelson, Tim. (2012). The Energy Market Death Spiral – Rethinking Customer Hardship.
The death spiral paper is more accessible outside of paywalls. Check out figure 3 in this paper (which has CAPEX investment from 1955 to 2011). Note the investment peak in the early 1980’s. Note the investment blackout in the 1990’s. And note the investment mega cycle (Simshuaser & Catt’s term) from 2005.
But, there is more.
This investment mega cycle in the Queensland electricity grid (hopefully the gold plating argument is now looking exceptionally weak) occurred in the midst of a once in a lifetime mineral boom. Not only was coal seam gas (and LNG trains) coming on stream, there was rampant mining activity in Queensland. Mining companies too were chasing electricians, electrical contractors, and things that contain copper and spin. The prior under investment had also result in under investment in apprenticeships. Guess what happened next? Wages boom for sparkies.
Take the price of copper (the stuff transformers and power lines are made of). In late 2003 the price of copper was around $1500 per metric tonne. By 2007 it was near $8000 per metric tonne, peaking just short of $10,000 per metric tonne in 2012. The raw material – copper – that is integral to the electricity network has risen by over 600% !!!!!!
The icing on the cake was the regulator’s decision on how to calculate WACC. Despite being warned by electricity utilities in their regulatory submissions, the AER decided to change the WACC calculations. Then along came the Global Financial Crisis of 2008, and the cost of capital went through the roof. This too added to customer’s electricity bill.
So I see a completely different story. Where some see “engineers gold plating” the network, I see inept and clueless economists causing bust and boom cycles in investment. Where some see over investment in the decade from 2005, I see chronic under investment and missed opportunities (when costs were at record lows during the late 1990’s) to make sensible investment decisions for the future.
As for the future, well, that is another story.
Dr Anthony Spierings is a registered electrician. He holds an Associated Diploma in Electrical Engineering. A degree in Economics, Banking and Finance. Master of Project Management. PhD in Information Systems. He previously worked at Energex for 35 years; including extensive stints in substations, project management, program management, and network operations.
What is often overlooked is that end users see massive swings in their MLFs and have done so since the NEM began. What is even harder for consumers is that the MLFs are driven not only by generator locations but by their bidding pattern as well.
What we have also seen is that MLFs consistently over recover the actual losses incurred but the reallocation of the residual back to consumers does not reflect the amounts each consumer paid.
Generators have never been interested in minimizing consumer MLFs so I find it a bit hard to be all that sympathetic to the current out-pourings of being hard done by
Anthony, I was ‘there’ too. And I can tell you, at that time in history, gold plating is an apt description. The other difference in our experience was my employer took a little more care with WH&S. Why an earth did they put you in such situations?
Thanks very much for this, Anthony. There is a lot there that I didn’t know about pre-2006 investment and thanks for the links to other articles. It is a good reminder for me (and maybe all of us) that regulators and policy-makers are often partial to a certain narrative and we always need to look at their reports and reviews with a critical eye.
Interesting discussion Paul. For my part the way i conceptualise a 2-way market is for demand side (retailers, large users) to bid in their demand in tranches of decreasing prices (an old-style downward sloping supply curve) with non-curtailable load bidding in at the MPC. Retailers’ bid stack would reflect the sum of their agreements with customers. One further advantage is that there is less onus on AEMO to forecast demand accurately, that responsibility is devolved to market customers. This sort of approach would require a lot of retailers, of course, but at least initially, if they wanted to be pure price takers they could.
I’m not sure your idea of mandatory bill info would in itself do much to jumpstart greater customer interest in DSP. I think the ACCC-led process for 3rd party data access is more likely to help – with customers’ permission other parties can take their consumption data and use whatever analytical/presentational tools they like to present the DSP opportunity to that customer.
Paul-Can you or anyone here tell me what the role of the existing and augmented grid will be once almost all or perhaps ALL present grid customers…households and some business…are largely independent of the grid …ie with rooftop solar and batteries formed into VPPS orchestrated by aggregators…as AEMO urges?
Who will be paying for the even more expensive grid then…with no regular customers?
Solar and wind farms will be in the AEMO-designated Renewable Energy Zones presumably with their own microgrids.
Who but big industry will take and therefore PAY for power from the grid,,,and will they be enough to sustain it.
The whole ongoing costs of the grid would surely be a huge impost on industry and ultimately on all of us….and especially on Australia’s competitiveness.
Can you or someone here give us an overview of who will take power from the grid and who will be the regular customers of it in the DER Australia that AEMO hopes for please?
That’s an interesting question, because governments are actively promoting policies that reduce grid demand – rooftop solar, batteries, demand management and efficiency measures – while at the same time incentivising large scale intermittent sources that require massive investment in networks.
I’d agree with HR about the fundamental reason for this spike being Snowy moving generation from Upper Tumut to Lower Tumut. Essentially when V>>N-NIL_HG is binding, shifting 220 MW of generation from UT (coeff -0.3) to LT (coeff +0.2) forces a reduction in Vic exports to NSW of 110 MW. Snowy rebid this way effective from 16:00.
Part of the reason for the export forecasts & price forecasts being out is that the representation of the constraint for predispatch is only approximate, and was overstating the export limit from Vic by about 100-120 MW. My guess is that this systematic gap was as much a factor in the “surprise” element of the spike (and in plenty of others like this one) as the other possible causes identified by Paul.
Just to clarify, the fundamental reason for price volatility on the day was the level of demand and overall set of supply offers for NSW; Snowy’s impact on exports from Victoria was then the proximate cause for the spike occurring at 16:15.
I’ve been struggling for a while with the concept that all generators get paid the marginal price. Why don’t all generators receive the price they bid?
Thanks Paul, I agree that “pay as bid” may tend to increase some bids (and therefore wholesale prices), but if that’s the case, then would you say currently we do not have a market that prices cost of generation accurately?
A nodal model such as in New Zealand could lead to a better dispatch outcome for power stations near regional boundaries. For example the Murray Power station is in the Vic region but is penalised if sending power into NSW by a Victorian marginal loss factor of 0.9 (based on sending power to Melbourne), multiplied by the dynamic loss factor for export from Vic to NSW (which would be calculated from Melbourne to Sydney). Likewise the solar farms near Mildura are penalised by a Victorian marginal loss factor of 0.8, based on sending the power to Melbourne, multiplied by the dynamic loss factor of sending the power from Vic to SA (because Adelaide is the nearest major load centre). What are the downsides of a nodal model, and why wasn’t it used in the NEM?
The diagrams you show of the marginal loss factors are close to the physical layout of most States in the NEM as it was 20 years ago, before the near-border power stations were built, and Snowy had its own region. But has it come time to update it to a system that more closely approximates the physics?
There is a trade-off between engineering/physical specificity in locational pricing, and the ability of participants to effectively hedge risks that arise in the market (the current concerns about Marginal Loss Factors being just one manifestation of that trade-off).
I will say again, the elephant in the room is roof top solar, and paying 100% for the export of that. As every day goes on and more and more is installed and more is exported and all still getting 100% of the export at the silly prices. The higher the cost to the system.
Someone needs to work out some kind of average loss of this stuff coming out of 240 (230volt) inverters and going back into the grid to be used, where???, and if its being used how much of it is able to be used. Keeping in mind that as more and more of it is going into the grid, the average distance its having to travel is further, on average.
Well unless its every second house in every street thats getting solar and on average its only haveing to go next door.
Meanwhile on 20 March (just 5 days before this topic was posted) wind output was:
Below 800Mw for the period 00:00 until 15:20.
Below 400MW for the period 06:30 until 12:40.
Below 200MW for the period 9:00 until 10:15. https://anero.id/energy/wind-energy/2019/march/20
Celebrating a peak output is an odd choice given the system is only as reliable as it’s lowest output.
Paul, thanks for your balanced view. Facts are necessary! The complexity of our power system proves that time after time.
I can only see forces imposing upwards pressure on consumer electricity prices. Removing a big chunk of the bottom of the ‘bid stack’ can only force more generation in at the top of the stack. Seems obvious.
The only way to reduce generation cost will be:
– massive over-capacity in low cost intermittent sources (but with massive increased cost of the grid)
– gas market intervention (or huge over-supply of gas)
– large coal units adjacent to their own mine, not exposed to international coal prices, using existing transmission eg Liddell and Hazelwood replacements
The grid is expanding (to accommodate intermittent sources) while the customer base is shrinking (solar PV and batteries). The logical conclusion is that customers who can afford it will get off the grid, the rest will see spiralling costs and companies/governments will face grid asset write-downs and customer subsidies.
Government policies are effectively paying people to leave the grid while paying to build more grid.
Isn’t the NEM a “free” market?
Private business exists to make as much profit as it can and the Vic governments decision to privatise the network made sure of it.
How much cheaper would the average Vic bill be if it were still government owned and run!
Paul
Thanks for the assessment – I found it very useful.
It is important to note that in its June 2018 report, the ACCC considers the NEM to be highly concentrated and this leads to higher prices than a competitive market might deliver. The Major Energy Users attempted to get the ability to exercise market power that comes from a highly concentrated market addressed but all it got was market monitoring. The issue of gaming the market in Qld was addressed by the owner directing its generators to cease the practice and sudddel]nly it stopped.
We all forget that as demand increases, so does the number of generators making offers reduce and we have seen in other markets that there is a recognition that competition reduces as demand increases and that actions are needed to limit the potential to exercise this market power – something we don’t see in the NEM.
While we haven’t yet seen what the outcome of the 5 minute settlement process might be, there is a real fear that it will result in further market concentration in each dispatch period.
The ACCC in its report came up with a number of sound recommendations to address the issue of market structure and market concentration but there doesn’t appear to be a strong commitment at a Federal level to these recommendations other than “the big stick”, brow beating of some gentailers and getting more coal fired generation in north Qld.
Probably most of all, we need to have a clear policy on carbon emissions and how this will be transitioned to as a nation, rather than the various state governments doing there own thing. Providing policy in this was a recommendation of the ACCC too. See page 101
“The ACCC considers it critical for there to be a stable energy policy in Australia which incorporates the need for Australia to meet its climate policy obligations but at the same time does not distort the way in which new generation capacity enters the market”
Hi Paul, can I add my name to the list of people thanking you for writings. I can’t agree more as a Professional Engineer with your remark about using non emotive wording. The facts are the facts and facts have no feelings. If the rules are to be reviewed then I would drill down to the very bottom. Bids are noted, they are stacked in rank order of price and that part of the stack that matches the load is activated. I am using my words deliberately. Then all the bidders are paid the same price. That is the most illogical process I can image. If I were to buy a string of contiguous properties for a property development I would negotiate each purchase in a confidential manner and try to stop the different property owners from talking to one another. I would expect to pay a range of different prices for what might be identical properties because of the different positions different owners would have taken. If I can and am willing to supply power for $30 per MWh then I think that is what the transaction should take place at. If someone else wants $100 per MWh because they have different cost structures and might even be making a smaller profit but that is their bottom line then that is what they should be paid. I know the model of stack pricing is based on the share trading model where some shares are traded as a result of an offer and acceptance ( the big guys ) while others are traded at the current spot price ( small traders ) and that in this model there is only ONE price at each moment of time but that is shares or could be Brussels sprouts in a vegetable market but it is not electricity. To follow the vegetable model a retailer might have purchased two bundles of Brussels sprouts from two different market to fill his purchase order. The two prices he paid need not be the same. Paul, if I might be permitted to use one emotive word – I think the present model is STUPID. Change that and the price of electricity will be harder to “Game”.
Dr Ken, I have also come to a similar conclusion. I accept that if a generator were to receive income relative to their own cost, then the bids may look different to how they look now. I suggest the overall generation cost would be lower than it is now.
However, is it generation cost or network cost driving consumer prices? Bit of both maybe?
So network cost has to be dealt with too.
Current policy is rewarding consumers for moving away from the grid while at the same time forcing a larger grid to accommodate intermittent sources.
This is divergent policy – paying customers to reduce grid demand while expanding the grid. This has to reduce efficiency, ruin market signals and increase consumer cost.
Complimentary policy would be rewarding those customers far from generators for leaving the grid, while also rewarding customers close to generators (or existing permanent large load centres eg cities) to remain on the grid.
At a macro level complimentary policy is logical. Divergent policy is not.
Paying all the clearing price (as the NEM does) is a standard auction mechanism. Pay-as-bid, which is the alternative you are suggesting, changes bidding strategies. It therefore does not necessarily result in lower costs to consumers. Additionally any evaluation of the wholesale cost of electricity has to include the contract market. The spot and contract markets influence each other – you can’t look at one in isolation to fully understand the dynamics. Given you started by agreeing with Paul about “using non emotive wording”, it’s unfortunate that you ended by calling a model you clearly don’t understand “STUPID” (sic). I’m not saying there is no case for pay-as-bid, I’m saying you haven’t made one.
Paul, I’m not sure what you missed in your statement, “I did not the broader recommendations in the ACCC report with respect to rooftop solar”. As I understand it the only two recommendations about solar were for states to take on board the cost of the excessively generous feed-in-tariffs and to terminate the Solar Credit scheme in 2021 rather than wind it down progressively until 2029. ACT electricity electricity consumers still have 10 plus years of paying $15 million annually towards eating the subsidy on these schemes on top of the cost of solar credits.
I think this is a very reasonable way to write about a complex situation. But I think the take-away would tend more to the John Quiggin view of utility function pricing and economics than any other: if we want to constrain market pricing of electricity supply then we should be asking some pretty big questions about why we had a market in the first place. If the market pricing is not within bounds which the consumer is willing to accept (and I think “everybody wants it cheaper” is a bit flippant, yes, true, but we’re also open to realistic pricing:cost relationships and what is not being exposed is the cost:price differential, or the curtailment costs, and we’re certainly not being offered price discounting for supply variances except in the case of the very large industrial consumers)
The Market in electricity wasn’t created to make market value: it was made to try and avoid state capital investment costs by inducing commercial investment for a long tail profit. The price controls were meant to satisfy consumer needs. They patently aren’t working, but the legalisms around varying the model are escaping the ACCC. Regulators that don’t want (to be seen to, let alone actuall) act, is very distressing.
So I too wouldn’t entirely blame the generators: Acting within the rules to maximise profit is not “rorting” but it is “distorting” and I think this kind of price distortion should be controlled, and constrained.
Ben…intermittents as it stands now are paid the same averaged price as other bidders even if they don’t deliver are they not?
So when intermittents fail to deliver they leave a shortfall and the extra bids to meet demand would be higher than those of unfirmed RE so the averaged price would be higher?
Intermittents can’t do without coal backup ….but the government-mandated priority dispatch for intermittents [ that are favored by law in the merit order process despite the fact that they’re not dispatchable]…. relegates coal to standby mode for ramping up to rescue the NEM when RE fails…wherein coal plant is economically and physically damaged …metal fatigue…tube leaks …boiler and chemical problems etc…as well-documented in MIT studies and many others.
Because COAL is indispensable right now…without it Australia would be blacked out…it must get a high enough price so shareholders will agree to its sticking around to be used and abused in these ramping up and down ways it’s not designed for.
Many RE enthusiasts try to make Australians think the increasing maintenance problems coal plants have are due to their being old obsolescent unreliable technology…in order to ensure Australians are gung ho for getting rid of coal.
Nothing could be further from the truth.
But who will pay the operational costs of the grid when household customers are thin on the ground?
No one will answer that question.
Truth: I’ll attempt to paraphrase your first two paragraphs, so we can see if I understand your point.
– the design of the market ensures that all the contributors to the 5min period, including low cost intermittent sources, get paid the same as the highest contributor – I think we both understand that equally
– your point about the effect of an intermittent source “not meeting its bid” is interesting and i’d like to know how often this situation occurs – obviously there are forecasts for weather dependent generation, but your question seems to be aimed at how often these forecasts are wrong
– I think its a fair question, but I also think to be completely fair, then you would have to assume that there are occasions when the weather is both less favourable and more favourable for generation i.e. sometimes there will be less generation than forecast, sometimes there will be more generation than forecast
– AEMO will have the data, I’m not sure if it’s publicly available
– the effect on wholesale price I think would be marginal, since the forecasting error would usually be less than the volume of marginal generation eventually setting the pool price
Your final sentence rings true – that we currently have contradictory energy policies resulting in less customers (rooftop PV and batteries) AND more gird to accommodate widely distributed intermittent sources.
I think the only way to stop network price increases is to freeze network investment, then write down the asset base.
Generation cost reduction (under the current market) needs lower cost marginal generation across all the 5min periods or rule changes (market change) to have generators paid what they bid.
No option seems to be on the table with any political team.
BM: Qld coal plants are government-owned and they WERE gaming the market under government ownership until the QLD government was shamed into putting a stop to it
The NEM is anything BUT a free market…there’s too much government intervention for that…not the least in its mandate that intermittents must be favored in the merit order for dispatch even when they’re unfirmed and therefore not dispatchable.
That changes everything…as everything IS changed…rules…enforcement of safety standard requirements for Li-ion batteries…everything…to prop up RE and damage COAL.
You say….’despite the enormous volume of discussion and debate about the climate change, and the causes of it all over the airwaves and internet (and even a full length movie or two) over the intervening 10 year period. I find that staggering, and quite explanatory.’
I find it staggering that you could think that has happened…..especially the reference to the movies.
Questioning and dissent on this issue was shut down from the moment the dodgy consensus was declared decades ago…when we were told the science was ‘closed’ …before most of the science was even done…before clouds and oscillations were researched….when temperature of 70% of the earth system…oceans…were not even able to be reliably measured…long before the advent of the Argo float system in 2005…at which time they found there was no significant rise in ocean warming….and the earth cannot be warming significantly if the ocean is not.
Other explanations for the slight warming that there is….are infinitely more logical and likely than the CO2 hypothesis…but any scientist …no matter how highly-credentialled who questions the political hypothesis that’s designed by warmists as a means to a political END…any scientist who offers alternative theories ….must risk his job and future to do so.
If the debate has been so open here , why did BoM so fiercely resist an audit when Australia’s PM wanted that done…shortly before that ‘problem’ was solved by COUP.
The universities’ massively politically-charged treatment of dissenting scientists and their views has been a disgrace to Australia
Why are the Australian universities so militantly closed to alternative views on this absolutely seminal matter of science? Nothing could be more vital to get right than this….and yet science is being unprecedentedly corrupted in order to prevent anything but the politically correct dogma …..that just happens to require the suicide of Australia as a 1st world nation ….to drive our public policy .
If a foreign enemy had somehow gotten hold of the energy policy levers in Australia in order to deliberately destroy our country for the foreseeable future, they could not IMO have devised a more existential threat to Australia than this ‘transition’ from simplicity and efficiency and affordability to unfathomable ad hoc…cut and paste complexity.
The problem of growing wind is that the absolute power gap left when the wind drops off on a national scale, also grows. When this occurs at times of high demand and with coal/gas being driven from the market also by increasing wind/solar, what is there to fill such ever larger gap events? Such shortfalls at times of low wind/high demand seem to be in part behind a number of the soaring spot prices in the last six months. Even $50 billion of grid changes and costly storage could only do so much. Disappointingly, it seems more wind = higher electricity prices.
It pays to keep reminding ourselves that electricity is a means of delivering energy and it’s EROEI that rules it all and replacing every ICE vehicle with battery stored energy is only adding to competition for battery production and output. It would appear from Mr Mills’ calculations that as fast as all the Gigafactories we could muster for the DUAL task to get rid of hydrocarbons altogether they’d be also be trying to keep up replacing spent ones. That’s before we even deal with all these spent incendiary items that at the moment are predominantly going into landfill unlike the recycling of lead acid batteries.
I am amazed that wind and solar generation which was invented in the 1890’s, along with the first coal-fired generator is thought to be the solution. I want to go forwards not backwards to intermittent energy. Nuclear energy and hopefully Thorium is the way to go.
I also think electricity should be much cheaper if we are going to power our planet.
Hi Michael, I just came across this article (April 2019…!), thanks for sharing it. The benefits to the end user and the generator from contracting power seem obvious, I guess also the retailer when the difference between spot and contract works in their favour.
Do you have any insights into what effect the contract market has (if any) on the spot market? Does contracting power tend to reduce spot prices or increase them? How much of the market is contracted at any time?
Also I am trying to understand how the gentailer arrangement with contracting power is affected by / affects the sport market.
Hi Ben,
Thank you for your feedback on the article.
The contract market would have an impact on individual generator bidding strategies. Generators would typically bid more competitively up to their contracted volume and then more aggressively (higher prices) for the uncontracted proportion of their generation capacity.
I cover this in more detail in my book Power Profits that you can find here https://www.amazon.com.au/Power-Profits-Comprehensive-Framework-Electricity-ebook/dp/B07C4HB5XL
I’m not sure how much of the market is contracted at any one time but I would expect that it would be a very high percentage. The number of users fully exposed to spot prices is quite small but has been growing over the last few years. Some end users have spot exposure plus some contracted positions (hedges).
In regards to gentailers, my observation has been that the retail arm of a business is not necessarily given more favourable contract prices to any other party that wishes to bilaterally contract. “The price is the price”. That is why retail prices offered by competing retailers are often very close.
Michael Williams
Altus Energy
Thanks Michael, I have your book on my list now 🙂 I read your bio and it’s safe to assume you are aware of the ongoing debate about renewables, cost and reliability.
In terms of the spot market and the bid stack, I can see that more low cost generation at the bottom of the stack should push some higher cost generation off the top of the stack, so the spot price is set by a lower cost generator. This will result in a higher margin for the retailer in the periods when SRMC is low, so retailers like this to happen. On the flipside, when intermittent VRE is not available, the SRMC is expected to be higher.
Since there are market caps for high and low prices, but the high cap is far higher than the low cap, it seems logical to me that retailers would hedge more against the periods of high wholesale prices than against low prices i.e. there is risk to retailers on the high wholesale price side, but rewards on the low side.
If this logic holds true, then intermittent sources result in higher retail prices, and the market cap on the high side can act like a target (not a cap) for SRMC generators.
So two questions (finally):
1. is it appropriate for intermittent sources to bid apples-for-apples against dispatchable sources in the same market?
2. if the market high cap (target) was removed, would extreme pricing be prone to self limiting?
Does all this mean that Queensland,particularly South East Queensland, should switch to daylight saving in the summer,thereby shifting the demand an hour earlier and make more use of the solar generation?
Mr Mills, is but presenting the glass half empty. Lets see just where we can get resources that the group of people that were happy to believe Oil/Coal/water/food and just about everything else was going to run out, yet, are now happy to believe that there are enough resources to build all the stuff they think will give us zero Co2. I find it amazing that all of a sudden all of the minerals that are needed for RE are not even talked about.
Has any of them added up the tonnes?????. or guessed at them. Wind turbines use a huge amount of steel, and there will need to be a bloody lot of them.
(They were so wrong about Oil and coal running out it was a joke, yet steel for wind turbines is a snap. The coke and iron ore)
Hi, wondering about a couple of questions on the combined solar tracking installations vs rooftop solar:
– average capacity factor (all periods all stations)
– capacity factor per hour of the day for all seasons
– contribution to peak demand e.g. the 2017 AEMO SA renewable energy report discusses the 85th percentile of peak demand periods only receives 1.6% of solar nameplate capacity
– stations with highest and lowest capacity factors
Ok so its not wind, which when it is hot normally goes to close to zero output, because there is no wind.
Solar as it goes over 25c gets a little less, so when hot a bit more.
Coal plants do
Gas plants do
Battery’s once hot after discharge, struggle to recharge if hot, so their 24 hour cycle will go down when its hot
Inverters suffer when its hot, both ways, the cooling needed in them goes up a lot (Fans), so to charge a battery must get a double whammy, the battery struggles in the heat, and the inverter to charge it and then to re-send its power back out at the set voltage.
So which one do I pick, hmmmmmm
Steam turbines – constraints due to ambient temperature can manifest in the condenser hotwell (impacting STG condenser vacuum), the cooling tower (also vacuum), the lube oil and jacking oil coolers (heat exchangers). Ambient temperature of 45degC doesn’t have much effect on a boiler operating at hundreds of degrees.
Gas turbines can be directly constrained by high intake air temperature – like a turbo vehicle GTs are sensitive to density and exhaust temperatures.
For a steam turbine plant there are some external factors not directly related to the equipment that can force reduced output. The most obvious being cooling water discharge temperature – if that is tightly monitored, as it is on discharges into rivers and lakes, then reducing load might be the only option to keep those temperatures within their regulated environmental thresholds.
The behaviour of the Mystery DUID looks very much like what happens at the wind farms around Goulburn in NSW. My own analysis of weather data and AEMO data shows that Taralga, Woodlawn, Gunning, Gullen Range, and Cullerin Range all have maximum power versus temperature curves that look a lot like the Mystery DUID.
Solar farms show almost constant maximum a.c. power output as a function of temperature. Probably because they tend to have oversubscribed d.c. power and so the a.c. output can still reach registered capacity even on very hot days.
The Mystery DUID sure ain’t Kogan Creek. It managed to provide about 94% capacity factor at 40C in 2018. Both units at Millmerran have been at about 82% capacity factor when the temperature was 40C in 2018. Keep in mind that Paul’s graph shows MAXIMUM power, and so while some coal DUIDs have been known to suffer from heat stroke, they seem to be able to keep going some of the time when it gets hot 🙂
You will have to give is a clue so we can narrow the field Paul!
Correction to my previous post: For solar farms the “registered capacity” in AEMO’s Registration & Exemptions List appears to be the d.c. photovoltaic module total capacity. “maximum capacity” appears to be the solar farm’s maximum a.c. (inverter) output power. When I said registered capacity in my earlier post I meant the thing AEMO calls “maximum capacity”.
Hi Mike, How about the Big Batt, the cells that have discharged in 40C, how much actual power comes out, vs what was put in when it was cool, plus, how hard is it then for them to cool and take a charge. Meaning lets say the 30MW they use (If they in fact only use the same 30mw of the battery) to cycle in say 20C, what is the charge losses, then discharge losses, and speed in which they can recharge and then discharge again over a 24 hours period. Vs a period of time where its 40C for 12 hours and over 30 for the rest.
The Hornsdale Power Reserve battery has only been called upon to deliver its full output power of 100 MW on a few occasions. Most of the time its output is limited to 30 MW so that it has 70 MW in reserve for emergencies.
You’d expect it to struggle most when it’s at full power because the heat dissipation will max out when the power is at its maximum. It’s possible that they use some of the batteries at 100% when they are delivering 30 MW. Or they might only use 30% of the power capability of all the batteries when they are delivering 30 MW. I don’t know how they manage it.
These are some of the times when it has jumped to 100 MW and the weather has been warm to hot at the time:
* 13 Dec 2017 at about 11:30 and about 37C
* 24 Jan 2018 at about 16:30 and about 32C
It was delivering 100 MW on both occasions.
It hasn’t gone to 100 MW in 2019 so far.
On a typical day its output is mostly zero. It seems to sit there looking for opportunities to make money on the spot market (energy and FCAS). You can often see it jump to 30 MW when wholesale prices are high and then recharge at 30 MW when wholesale prices are low. Sometimes you see it at 30 MW in one 5 minute AEMO SCADA interval and then recharge at 30 MW in the next 5 minute AEMO SCADA interval. It has done a few thousand of these +30/-30 MW cycles since it was commissioned.
Tesla designs electric cars. Cars have to work at quite high temperatures (air temperature plus insolation). Tesla’s batteries have a liquid cooling system. My Tesla battery at home has the same sort of cooling system. I can get the full 5 kW for my air conditioner on days that are 40C.
So I reckon the Tesla battery systems and other properly design grid scale batteries will have no problem on hot days. They will be designed not to overheat when they are charging or discharging. You just need a well designed cooling system.
I haven’t looked at how much energy is lost in the Hornsdale Power Reserve battery but it kinda doesn’t matter if you can buy $50/MWh and sell at $300/MWh.
Hello Mike, Thank you for your messages. Ok firstly. What i am talking about is simply the maximum rated amount of power the battery can deliver in 24 hours. It does not matter if its the whole 100, or just 30MW, as long as its just the same sections of battery’s that are being used. Just like a gas plant or coal plant have a rated output. Kogan Creek is like 750MW. And as you know that is 750 by 24, to equal what it can do in 24 hours. CF (Capacity factors are done over a year) So its 750 by 24 by 365. would be 100%. On a hot day as we know Kogan creek has to wind down, it cannot run at 750 during hot weather it can only roughly get out 680 to 700. (Roughly)
So the battery must have a set maximum it can run at in 24 hours. Like it can be charged and discharged lets say 10 times so in 24 hours it can deliver 1000MW / 24 equals an average of 41MW per hour. (I am guessing at that, what is its rated day capacity? 100mw is how much it can hold, which is diffrent than what it can deliver) However on a hot day, its not going to deliver 41mw per hour, its going to suffer from the heat, its not the batterys will be hot, the inverters will be hot, the power lines will be hot. The battery must also have a reduction in its rated capacity. What is it? That is all I am asking, or alluding to.
Now as for the battery buying low and selling high, that is 100% fine, to make money. Its not the best, if you are short on power because the battery takes in less than 100% to start with, and then lets out less than 100% of whats in it. So there is a loss both ways. The battery produces zero power, it uses power to deliver at a later date, or as you put it to buy low and sell high. But not matter which way you cut it, unless its being charged with excess power that would otherwise be lost, its adding to demand. Which that battery has been doing most the time as it mostly chargers from the grid.
Your tesla will also have a maximum amount of times in one day it can be charged and discharged I would call it the cars 12 hour driving range or 24 hour driving range.
Hi John, I think you might be trying to compare apples with oranges? Capacity factor does not appear to be a relevant figure of merit for a battery.
At the risk of sounding like a Tesla fan boy (and with apologies to everyone for taking up a lot of space) here are a few thoughts regarding your doubts about modern batteries (…ok, I admit it; I am an Elon Musk fan. Anyone who can nail the synchronised landing of two booster rockets deserves some respect if you ask me).
In an electric car the battery discharges whenever you are on the accelerator and charges whenever you are on the brakes thanks to regenerative braking. That means it might switch between discharging and charging hundreds of times a day if you’re driving in city traffic like Tesla hire car drivers do, or like Toyota Camry Hydrid taxi drivers do.
As I have already mentioned, the Hornsdale Power Reserve battery switches between charge and discharge many times each day (and night).
My home battery switches between discharging and charging pretty much instantaneously. On hot afternoons when my solar panels are generating not quite enough to meet the peak power demand of my air conditioner, as the air conditioner cycles between almost no power and peak power, the battery has to switch between charging (using excess solar power that the house is not currently using) and discharging (when the air conditioner needs more power than the solar system is currently producing). This can happen lots of times per day and Tesla guarantees the battery for 10 years. By the way, my battery has carried me through a dozen power interruptions in the nearly two years I’ve had it. The most recent event was 3 hours in January this year in suburban Sydney. Uninterrupted cricket viewing in a nice cool house. That battery is worth its weight in icy cold XXXX Gold. 🙂
In summary, as far as I can tell there is no _practical_ limit to the number of times per day a Tesla battery can switch between charging and discharging.
On the subject of the Hornsdale battery acting as a load, that is not necessarily a bad thing. It can save inflexible generators (e.g., thermal ones) from having to rapidly follow fluctions in demand. That sort of help should save the owners of thermal plants some money and _should_ drive down the cost of electricity (sigh). A big battery could have saved Queensland from some pain when it recently disconnected from the rest of the NEM and the sudden loss of a large lump of load caused the frequency in Queensland to go troppo for a while. Also note that when power is in short supply prices tend to be high and so it is unlikely that a battery would be charging and adding to demand.
I know it’s politically incorrect to say so in Australia but all the data seems to be saying that batteries might just save the world (or at least what remains of the Hunter Valley).
Hi again John, apologies for not addressing one of your important observations. In 2018 the AEMO dispatch SCADA data says that for the Hornsdale battery about 49 GWh went in and about 40 GWh came back out again, so it returned about 82% of the energy that went into it. 82% is comparable to pumped hydro plants (Wivenhoe, Tumut 3, Shoalhaven, Snowy 2.0, etc). See https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricity If you’re worried about the energy losses of a battery they pale into insignificance compared to the energy inefficiency of any thermal generator be it coal, gas, diesel, or whatever. Coal is typically only 33% to 40% efficient even for HELE (ref: https://www.worldcoal.org/reducing-co2-emissions/high-efficiency-low-emission-coal). My home battery returns about 94% if I deep cycle it and about 85% for the shallow cycling that happens most days. You might think I’d be unhappy with 85% but it’s a lot better than the 25% return I’d get if I sold my excess solar energy to the grid at the NSW IPART rate of 8 cents per kWh during the day and had to buy it back again at 32 cents per kWh during the night.
Hi Mike, Thank you for your message. You do like Mr Musk.
As for a Battery not having a maximum capacity, and not having losses, you seem to some how have that part missing. The way you write is as if no matter what time of day, temp, the battery works the same, and so does the inverter thats between it. Not matter what.
Really there will be a difference as there is with everything else.
I am not saying a Battery is not excellent, I am just saying its also suffers in the heat. As do inverters. As do thermal FF plants, as do solar pannels, as do aircons and lastly as do humans.
As stated on a 40c day, a batterys ability to meet its maximum supplied 24 hour total amount of power, will be less than on a 20c day. Due to the increased losses charging it and increased losses discharging it, both from the battery itself but also from the inverter / as well as the speed to recharge it will be reduced by the battery and also by the inverter.
Another candidate for the magic wand category is power electronics. While it can support fault levels (part of AEMO’s termed “system strength”) and provide reactive power control (including distribution and transmission voltage control). What it can’t do is let grid operators escape power engineering design.
On the one hand, the capabilities and limitations have not been considered in VRE location or connection requirements, or protection design. On the other it is just expected that *this or that feature* (synthetic inertia, fast frequency response) will allow wholesale replacement of synchronous plant. The result is many “surprises” in plant response appear at inconvenient times.
Hello Mike, I am sorry I did not see your second reply.
Yes those numbers are correct. The battery lost around 18% of the power that went into it. Thats a average. So as I have said above. On a hot day, its going to be higher than 18% and on a cool day its going to be lower.
Hi Ben; I haven’t looked in any great detail but the quick answer is high wind generation in SA, modest grid demand, plus significant restrictions on exports to Victoria due to transmission works on the Heywood interconnection. SA prices were probably being set by negative offers at one or more windfarms. Cheers, Allan.
Thanks Allan. Apologies if these are dumb questions, but how does a negative bid from a wind farm still sit on top of the bid stack to set the price, and why would a wind farm bid negative anyway?
Are the ‘forced on’ gas turbines exposed to the negative wholesale price?
Ben, good questions, not dumb at all. Answer to the first is that a negative offer can still sit on top of the *dispatched* subset of offers and thus set the spot price if there is enough volume to nearly satisfy demand (including exports) bid at even lower negative prices.
Answer to the second (in short) is that when a generator is directed to run by AEMO, that generator subsequently receives compensation for the difference between its costs and the spot revenue received (or paid out in the case of a negative spot price).
Good job, something that’s needed and will add to the conversation. I think this will be good for the country. Let’s see what the various partisan commentators pull out of it!
Dear Me, some people are going to be upset with your information. Maybe you should have done what the Polling groups did or what humans almost always throughout history have done when information does not fit what is believed to be true, fact or scientifically proven. Changed it, so it fitted in.
The news did lead me to wonder where are all the stop Adani voices, et al calling for said generator unit to be left in mothballs like they all have to be by [fill in last chance to save planet date here].
You forgot to mention the “planned” outage in late 2017 which resulted in the “super load” that Victorians seems to get so excited about when a generator was sent to Germany for repairs in early 2018 and returned earlier this year – this was a generator outage and the planning was done following the unplanned failure.
When prices are suppressed in pursuit of political ends (and compensation denied), necessary investment lags, with resulting, adverse reliability risk/consequences. It ultimately catches up and customers suffer.
Yes – I have seen it happen in various countries/jurisdictions where I worked – and we never seem to learn. In the book I am (not) writing, the antidote to this kind of ‘populist’ government is a properly educated/informed electorate – but governments around the world measure their education programs’ performance by how much they are spending; not the results.
The 20th of May was swaption expiry for FY20. I believe the volume you’re seeing on the exchange is the delivery of underlying for in the money options. (If you look at the ASX trade log, you’ll also see a crazy range of transaction prices in that daily volume, which relates to the different strikes exercising. And the volumes for option exercises come in pairs (of 2, or really strips of 4) of the same volume.)
Ha, good one! I hate being wrong but I like knowing the real answer too…
On reading this I immediately reviewed some old files I have for one of the grid connected wind farms in NSW, turns out there are several temperature sensors in the nacelle for the mechanical systems.
Below are the two high temperature alarms described in the technical documentation from the WTG supplier (in this case RePower). It is possible (and likely) there are others:
gearbox oil sump temperature
generator stator temperature
Page 3 is interesting.
As all of the best sites are used for solar and wind, you will get diminishing rates of capacity factors for future installations being in lower quality areas.
Re those bids dispatched where offer price is higher than spot – it looks like they’re mostly coal units limited by their offered ramp-down rates (typically ~3 MW per minute; note that this is a bid parameter not a fixed technical limit). The dispatch engine would prefer not to dispatch them but can’t move their targets by more than 5 * ramp rate in a 5 minute interval.
And for anyone wondering why those bids don’t set the spot price at their higher offer prices – it’s because those bids aren’t marginal – change demands by 1 MW and their dispatch levels wouldn’t change, constrained as they are by ramp rates.
In QLD there is still over 4GW of coal online during minimum demand and very low spot prices. For the coal generators to bid their capacity so low, is it fair to assume this capacity is already sold as PPA, so they don’t care what the spot market is?
And if that’s correct, or close to correct, then is there a ‘shelf capacity’ of PPA where the spot price changes from ‘care’ to ‘don’t care’?
And what happens when the unscheduled rooftop PV generation eats into the PPA capacity and the coal plants start to get dispatched out of delivering even their PPA volumes?
The same will eventually happen to utility solar and wind – they will be constrained like any other generator – when the rooftop PV eats into their market share. At least a coal plant can make money overnight and on cloudy days and can be held online for FCAS.
None of this gives me any confidence the market is working to price generation correctly.
Whenever generation is running at spot prices below its marginal cost/value, someone is losing money ie someone cares. In a true PPA situation where the generator is paid a fixed price per unit of production, the offtaker is receiving spot price (+ possibly LGCs if it’s renewable generation) for sale of that electricity into the pool, and that offtaker would prefer the generator not to produce once spot price (+ LGC value) is less than the PPA price. The generator is earning PPA price less its marginal cost, so in principle doesn’t care about the spot price. But work through the cashflows and everyone can be better off with no generation once spot price + LGC value is less than marginal cost (the offtaker could compensate the generator for its value of lost output, [PPA price – marginal cost], and still be better off than wearing a loss of [spot price + LGC value – PPA price]).
More typically, traditional coal fired generation is not contracted through PPAs but via swap contracts where they receive fixed price and pay spot price to a swap buyer. Again the incentive for these generators is to not generate once spot price < marginal cost. (In effect they can "buy" electricity from the pool, for financial delivery to the swap buyer, more cheaply than by generating themselves. These contracts are not for physical delivery and carry no obligation to actually generate – it's a financial decision).
So why do these generators bid $0 or -$1,000/MWh and keep running through zero or negative price periods? Typically if their costs of shutting down then restarting units once prices recover are likely to exceed the losses incurred while remaining online. For coal fired units these startup/shutdown costs can be quite significant and their marginal costs are relatively low so they choose to keep running through short periods of low price. But if you look at gas-fired generators with higher marginal running costs and lower startup costs they regularly decommit units rather than stay online at a loss.
In your great piece on price setting, you explain how MLF doesn’t affect the actual output but the price paid to the generator.
My question is: what is the output measure used in a PPA? Do you know if most PPAs take into account the output at the RRN (if so double counting the MLF on both the output and the spot price) or do they use the generator connection point?
“We hope that the results presented in the Generator Report Card can raise the level of awareness of what’s actually happening..”
Well all I know is I just got my last power bill in suburban Adelaide and they’re charging me 51.2c/kwhr peak in an all electric household (19.05c for off peak HWS). It was 43.2c peak last quarter so the poor folk must be wondering when the cheap renewables are coming down the wires.
On Friday night there was a grand total of 150MW wind output across the entire NEM. Low correlation won’t help with that.
Correlation is a popular argument used to advocate for more transmission lines. But interconnectors expose regions to price volatility and system strength issues as much as they expose generators to more load.
Has SA found their interconnection with VIC reduced their c/kWh? Or gas QLD being linked to NSW helped QLDs c/kWh?
I think a region with low volatility and generation over-supply would be better off without an interconnector.
The inherent variability of wind turbines and its effect on price volatility and system strength would be better off not spread throughout the NEM.
The math is changing, larger wind turbines, can drive water turbines, to pump water uphill, for power storage and do it economically. If we look at the challenges, resource shortages, we can develop ways, of using more plentiful inputs, that’s how economies of scale work.
The cost per kWh, km, gram of protein, compared to a competing method is lower. An electric engine uses 1/5th as much energy, as an internal combustion engine. A solar panel can capture 10X as much energy as a natural plant.
Improve the cost factors, with research and mass production and the inherent physics advantages shine powerfully. Near worthless desert land, producing 10X the energy, of the best agricultural land. Vehicles using 1/5th as much energy, requiring 1/10th as much maintenance.
Moore’s law gave us the tech boom, 1997-2000, until market saturation, gave us the tech crash. Strange as it sounds, in a decade boom, most cars will be self driving electric, most power will be renewable.
But these vehicles will be low maintenance, as will be the solar transistorized power. At market saturation, we’ll need the basic minimum income. Not difficult, with cheap transportation and energy, as well as a doubling of assets, due to Clean Disruption.
A second roaring twenties, taking 3rd world countries, to second world standard of living. Second world countries to first world standard of living.
Solving pollution, noise, transportation, infrastructure problems, in the first world countries. It won’t be paradise, but people, will live to 110 years old.
Perhaps the longer testing periods required to before new generation can feed into the grid at full capacity have affected their decisions. AGL must be smarting from delays in allowing the Silverton wind farm to run at full capacity. Better to keep an old unit in reserve until the new capacity is fully commissioned.
The extension of 3 units of Liddell gives “cover” for decommisioning one unit early in 2022, which would presumably be the least reliable unit. This decomissioned unit can then be cannibalised for spares to keep the other units operating.
AGL’s results announcement on Thursday said re LYA2 “Work under way to return to service, expected mid-December 2019”. As this is material information in an ASX release you would have to assume it remains AGL’s best estimate for return date and would have been carefully vetted, probably ruling out worst case scenarios like manufacture of a new rotor or stator. Of course it doesn’t rule out risks of the return date slipping by weeks or even months. It will be fascinating to see how AEMO deals with delay risk in its ESOO update.
In Example 1, Very Simple, the ROP is $65.44. But then the Trading Price looks like $80.32. What give? I’m already lost. I’ll go back to the simple guide now. But this may be worth explaining here, no?
1. The decision to reduce flows on long distance transmission in the face of storms is common in many jurisdictions and had the practice been followed before the system black it would not have happened. Between 100 and 150 MW of gas generation around Adelaide would have solved the problem.
2. The anti-renewables rhetoric may delay the transition at the margin but economics will win so before the end of this government non hydro renewables will provide more energy than gas and brown coal combined and black coal will fall below 45% of total generation
Nice analysis. We shall have to see what the E(3)C, ESO and Ofgem investigation(s) produce; and how much is published.
Stepping back from the specifics, my own take is that (1) there is inadequate competent strategic oversight of the transition in an industry that tends to predict the future from past experience, and (2) the “energy wars” you describe make this worse. The German experience with frequency-related PV disconnection is another example. I anticipate that the investigation will find that no one body was to blame but they collectively failed to act on a fairly obvious risk.
I am currently looking at the systems implications of the possible transition to EVs in the UK by 2030 (to meet the Fifth Carbon Budget). Lots of opportunity for things to go wrong.
Andrew – you absolutely nailed the problem both here and there. As to why there is insufficient competent strategic oversight, well here at least
a) our participants and institutions are almost exclusively mired in detail,
b) our politicians are almost exclusively preoccupied with picking fights and picking winners.
Please explain what is wrong with my understanding of inertia
In practice inertia is a store of energy. If a generator goes offline then the energy stored in the other rotary machines on the grid, (motors as well as generators)as well as the energy in the windings of motors generators,transformers and the transmission links (inductance) forms the inertia of the grid. Referring only to the mechanical or rotary inertia it is Jw^2/2 which J is the polar moment of the machine w is the angular velocity.
Now because the frequency is only allowed to fall to about 48 Hz before disconnection, the useful inertia of a turbo generator is J x 2 xpi x (1-(48/50))^2. or roughly 4% of the system inertia. So if there is 4,200 MWs in the system then only 168 MWs can be used before cascading failure begins. Assuming a rate of change frequency ROCF of 0.5 Hz/s that is 168 MWs delivered over 4 seconds (2 Hz/0.5).i.e. 42 MW.
If a battery can deliver 42 MW for 4 seconds that effectively provides the same energy as all the available system inertia so one can argue that the Hornsdale battery actually provides 129 x 3600 = 464,000 MWs far more than the available rotary inertia in the system. Even with a ROCF of 1 Hz/s inertia only supplies energy at less than 100 MW and only for 2-3 seconds. After that if governor response has not kicked in, goodnight folks.
Similarly, synthetic inertia can be provided by wind and solar farms via slightly different mechanisms.
a) A typical modern wind turbine has an Inertia constant of 4-5. i.e. a 4 MW wind turbine with an inertia constant of 4 has 16-20 MWs of inertia, but because most wind turbines are asynchronous machines, their speed can be allowed to drop by 10-15% while the power converter still puts out a constant frequency. So if you allow the rotary speed of the turbine to fall 15% then the available inertia per MW is 1 x(1-(42.5/50)^2) = 0.38 MWs/MW. so 1,000 MW of wind turbines can contribute about 380 MWs whereas 1,000 MW of gas turbines with a typical inertia constant of 3 can only contribute about 3 x 1,000 x 0.04 = 120 MWs
b) a solar farm can similarly provide synthetic inertia by running slightly below peak instantaneous capacity. If 1,000 MWs of solar farms are running at a nominal 800 MW but deliver only 760 MW, it can ramp up output almost instantly to 800 MW thus providing 160 MWs over 4 seconds just as the gas plants would.
Thus a properly engineered wind/solar/storage system can provide much faster disturbance response with lower inertia. In fact because the inverters don’t have inertia they can respond to deviations much more quickly.
There is a whole other discussion about what happens next but that is for another time
The UK company National Grid is the AEMO of the UK, and is very heavily involved in “The Transition”, i.e. makes a lot of money from all the poles, wires, controls, analysis, etc that is required.
If you regard these events as match-meets-pile-of-wood, you can be sure that any report from National Grid will blame the match for the fire, and will hide the involvement of the pile of wood.
Wind power earlier in the day was 67% of demand, uncharted territory, so claims of great system resilience are specious.
Great article. In relation to your question about the UK equivalent of FCAS, National Grid ran a tender process for Enhanced Frequency Response services in ~2016 (1s response time, 4 year contracts offered, taken up by various technologies, many BESS projects). There is more info here:
I cannot possibly see how a train wreck can be avoided without mandating that every tenderer of electrons to a communal grid has to be able to reasonably guarantee them 24/7/365 or they can keep them and that includes frequency and voltage. Anything else is delusional State sponsored dumping in my book and normally the ACCC would be all over it for the obvious but alas.
It seems that the initial report on the UK outage has been leaked and it notes the wind turbine failure occured first, which tripped the gas power plant.
“Experts have said giant batteries providing instant back-up energy could help.”…
“The costs are sizeable but if there’s a risk of this happening more frequently, I don’t think businesses or consumers will accept a lower standard of continuity of supply.”
Lots of focus on “unreliability” of fossil fuel plants while ignoring cause which is the market distortion due to government subsidies for renewables.
AEMO’s plan is to spend tens of billions reconfiguring our energy network so it can do what it did fine before renwewables were introduced. And we all know that just means the cost will flow onto consumers, driving our electricity prices even higher.
Ms Zibelman’s AFR contribution states that transmission is needed to replace generation.
Unfortunately it doesn’t work like that. Yes lots of wind power across SA sends power back to VIC. But the power comes from Traralgon when the wind drops.
It is disconcerting that the AEMO CEO did not mention intermittency once in her article. She gives more weight given to factors that will increase costs than factors that will reduce costs.
“Pay more accurately for firming capability” – well you don’t need to pay for it when it’s inherent in the technology.
The move to inverter supplies asynchronous intermittent generation might provide cheap power when the weather is favourable, but it’s also forcing the firming capabilities into separate pieces of equipment – fault level, inertia, voltage and part-time bulk power delivery.
So now we need all that plus bulk power capacity available all the time, plus lots of transmission, plus batteries plus hydro.
Sadly the internet has given a voice to people who have no idea what they are talking about. The energy market is extremely complicated and the information in the reports aren’t readily accessible to the layman so they read headlines from media who either don’t understand the topic or willfully spread misinformation.
The only way to fix this is for AEMO to release a simple/layman version where they explain things in plain english as this post managed to do quite well.
Large-scale solar was modelled as having a 40% capacity factor at the time of maximum summer demand, and wind a 16% capacity factor. The summer peak for unserved energy in Victoria is ~4:30pm for 2020. Yet Angus Taylor wants like-for-like replacement of old dispatchable with new dispatchable based on nameplate capacity, not taking into account that both new solar and wind contribute something to capacity at the time of summer peak. This will eventually lead to an expensive oversupply of dispatchable plant.
Anyone notice that Basslink went out about 11 am this morning. Hope it’s fixed quickly, but if not it’s another factor affecting reliability In Victoria not considered in AEMO’s report.
Hi Malcolm, it’s well buried in the detail (p71 of the ESOO report) but AEMO does consider forced outage risks for key transmission links including Basslink in its reliability modelling. Obviously a special factor with Basslink is that some forms of outage can take a long time to repair and it’s not clear if and how fully AEMO’s modelling captures this low probability / high consequence risk.
Re your other comment on solar and wind contributions to meeting demand peaks, the modelling does not assume fixed percentage factors but instead runs simulations based on historical half hourly traces of demand and renewable output (scaled for changes in installed capacity) in an attempt to better capture their variability and correlation. The figures of 30% for large scale solar and 16% for wind (quoted on p76) are ex-post average contributions (as percentage of generator capacity) extracted from those simulations during periods of unserved energy. Across those simulations there would be a wide range of modelled contribution levels from solar and wind.
Your general point that variable renewables (VRE) effectively contribute some proportion of their capacity is correct. However it’s a smaller contribution percentage than from “firm” dispatchable generation, and the contribution percentages of VRE at times of system stress will tend to reduce as the VRE share of generation increases, because instances of tight supply-demand balance will increasingly be driven by low VRE output in addition to high demand and outages.
Best description of the NEM I have ever seen : “There’s no doubt it’s a sophisticated piece of machinery with a lot of complex science and technology behind it. We then wrap it in a highly regulated market founded in arcane mathematics and towing a boatload of jargon and acronyms, operated and overseen by a plethora of organisations with obscure names and functions, trying to navigate and administer a morass of laws rules regulations procedures codes and other legalese.”
I would have thought there’s more to reliability than simply how many blackouts and for how long. The other side of the coin is how much to pay productive users to shut down and knock off for the day in order to cope with unreliable supply. Greenouts but whose wallet are they coming out of?
Take that to extremes and pay us all a pile to do it and at some stage those of us with the capital can afford our own diesel or town gas generator all year round and dump what we don’t use back on the grid to also amortise the investment. Now where have I heard that before? The same place I’ve heard about a fallacy of composition.
I note the recent UK experience where 1000MW loss of generation capacity resulted in 500MW of behind-the-meter generation, thus necessitating greater response which ultimately included significant blackouts.
AEMO, IMHO, should specifically allow for loss of embedded generation in their ESOO models, yet I have seen no mention of that.
Thanks for the comment. I tend to agree with your analysis of the recent UK event, but note that was a “system security” problem – real-time robustness of the grid against a sudden transmission or generation failure.
The ESOO does not deal primarily with this kind of risk but with “capacity adequacy” – having sufficient dispatchable capacity (generation and potentially demand response) to deal with *forecast* future levels of demand. Unavailability of generation is allowed for in the analysis, based on known outages and failure statistics, but the ESOO modelling itself does not pretend to assess the second-to-second robustness of the system against outages. So things like “cascading” failures don’t need to be explicitly modelled for the ESOO. However the models do incorporate many constraints (for example on interconnector flow levels) that ultimately reflect system security considerations.
This is what happens when industry shuts down because of the worlds highest energy prices,and renewables make so much money in subsidies they just bid in the negatives to ensure they are in the bid stack.
It’s interesting that you note renewables contract structure is such that it’s someone else’s* cost (and risk) when prices are negative – so are happy to sail through with their own revenue covered by some form of PPA independent of spot pricing.
The below article from earlier in the year outlined some of the issues with the distortion of the market due to government subsidies for renewables
Great Article. It’s really confusing to understand why wind and solar bid so far into negative. It really highlights the need for bulk storage. in a fully developed market energy will be extremely cheap in QLD. competition between renewable and coal = cheap power.
Looking forward to seeing what new attacks our polies have on wind and solar. Hard to dispute the benefit here.
– what was the proportion of gov owned generators losing money to stay online compared to private companies, because government backed generators with years of huge profits behind them can afford negative prices for longer than most
– a coal fired generator can turn down to the point the boiler air flow can no longer sustain stable combustion, unless it has a turbine bypass, in which case the boiler can keep its firing above the critical threshold and the excess steam can be bypassed straight to the condenser hotwell – a few tonnes of extra coal on top of the negative prices isn’t much
– I’m pretty sure at least one of the QLD coal fired generators is fitted with turbine bypass systems
– to get below the critical coal firing rate requires support from oil or gas burners which are more costly than coal
– what effect does wholesale price volatility have on hedging contracts – I imagine volatility increases risk for both retailers and generators, therefore both of them cover this risk with increased prices to consumers
– what percentage of the QLD demand is supplied through contract
Seems much more that the CANNIBALISM that’s an inescapable feature of intermittents has begun in QLD.
Solar bids the price all the way down to -$1000/Mwh to compete for each solar plant’s hoped-for share of the favored dispatch they enjoy in the market that’s corrupted by government interference….to prop up intermittents.
It’s all artificial.
When the price gets so low and they either have agreements to switch off at that point as they’re propped up by government anyway…or they just decide it’s in their interests to do so…the solar farms just pull the plug…knowing the QLD GOVERNMENT will order coal plants to bid and generate to fill the gap even though it’s unprofitable despite the hedge..because if they didn’t….the lights would go out and everything would stop in QLD and the SYSTEM WOULD THEN COLLAPSE across the whole NEM…as forecast by the former AEMO boss…the late Matt Zema just before his death.
It would collapse because QLD coal plants…the newest in the country…. are all that’s keeping Australia’s East coast from being in SYSTEM BLACK now.
As Bloomberg[ BNEF] said…
‘Challenge for policymakers is to figure out how to pay for the bulk cheap RE units that run at close to zero marginal cost & ALL GENERATE AT ONCE cannibalizing their own energy prices-& at the same time keep conventional thermal plants online & available’.
How long will coal plants make themselves available in the OTHER states where they’re NOT government-owned and are subject to abuse-economic…material…real and verbal.
Murphy’s Law can pop up too – say the spot market is low and a generator has planned testing that requires it to generate through periods like this.
Braemar 2 was pigging its gas inlet last week, meaning it had to guarantee a constant gas flow for a time period – some of that was through negative prices when they wanted to be offline.
I wonder if some of the actions by the coal plants are aimed at highlighting what is bound to come more often in the future? Not in the realms of conspiracy, but more like, “let’s do this and see what happens, so we may educate ourselves and better prepare for the future”.
Lol if I was into conspiracy theories I would be more interested in the August 2018 islanding of SA and QLD, at almost the exact nadir of SA wind output, which is the only time an event like that could occur with the lowest risk of unintended consequences 🙈.
There is a problem when the WeekEnd Australian which is known as the weak lend of you Australia publishes the opposite type of article.
Perhaps with edited quotes to support the view point.
The None Australia paper says the reason for -$1000 price is because of Solar.
Paul has pointed out Solar turned off.
Missing from the worked example is the cost of fuel. A generator that increases output to crush the spot price will incur higher fuel costs. The rising cost of fuel will offset any gains from the swaps. So the question is the degree of offset.
There could be a system security issue due to system strength, requiring the big generators to be on line even at low spot price. I am not sure whether there is a market mechanism that force the generators to stay on line regardless of spot prices and then get compensated for the loss. The market operator got caught by the huge influx of asynchronous generators and trying to catch up in their plans.
There is a non-market mechanism for that Nadia – direction by AEMO – and we’ve seen negative price outcomes in South Australia when it is invoked there on days of high wind generation. However AEMO has to publicly notify the market when it directs generation to remain online and that certainly wasn’t the case in these Queensland events. As most of the non-synchronous generation in Qld is embedded small-scale solar I don’t think it would be leading to general system strength issues – although these may emerge in sections of the transmission network with concentrations of utility-scale solar and not much synchronous plant. Allan.
Before my time, but… Wasn’t 1999 a time when SA max payment was say $1,000/MWh and the market max was say $10,000/MWh and anytime the price spiked above $1000 SA would have rolling blackouts to cut demand? Maybe this is just an old NEM tall tale…
It’s interesting (or ironic?) that reducing demand is being discussed in the same era as some others discuss the benefits of 200% and even 700% installed capacity of intermittent generation.
It’s a tipsy topsy world these days:
– pay some users to reduce load, to reduce costs for others
– generators pay to generate, while consumers costs go up
I’m not convinced electricity should be this difficult.
If we had our time again, with what we know now, would we allow individual rooftop PV and unconstrained access to the grid by intermittent generators?
Its seems totally inevitable to me that inertia on the grid is going to keep falling and that as with so many other things we need to start moving forwards.
A workshop from the University of Washington earlier this year contained some excellent presentations and youtube lecture videos that I’ve been working through. I also got a working paper from the workshop convenor that was readable to a determined layman. My own view is that Australia could be moving more in the direction that Europe and the USA are already going. I don’t yet think there is a role for markets in this except maybe for fast frequency control, but I might be wrong. To me its about getting the design principles right for the next 20-30 years at a time when technology in the area is moving quickly. Seminar reference attached and if anyone want’s Brians roadmap document let me know. I’d appreciate input from people who unlike me can speak with authority on an important topic https://lowinertiagrids.ece.uw.edu
I remain perplexed that this problem you have highlighted, not only remains unresolved, but seemingly the AEMC still doesn’t get it. The response to the double islanding event surely underlines the desperate need for this to be rectified. AEMO in their report noted that the last Qld separation event that happening in 2008 involved the loss of power transfer from Qld to NSW of 1091 MW vs 879 MW in the recent event. In that event even though the loss of power was greater the frequency deviation was less in both Qld and NSW and no load was shed to rebalance the system.
Surely AEMC should be able to join the dots.
Kate you make a good case for re-empowering the power engineers — the market design has failed us and its about time that good engineering was given a dominant role again.
Paul,
I have just come across this article and noted the remarks about demand in some places being characterised by higher proportions of industrial demand relative to domestic. Many years ago I used to predict daily demand for SA and it was pointed out to me that our demand was relatively lightly affected by industrial load, despite having the car industry and its ancillaries, because gas was such a large contributor to energy usage by industry. Adelaide was the first to take gas from the Moomba gas field and its development was enabled by Torrens Island PS taking 2/3 of the gas from Moomba in the 1960 – 70s; the rest being taken by industry, commercial and domestic.
When predicting demand the major impact was due to air-conditioning so demand was highly correlated with temperature since there was a small industrial base load for electricity and we became quite adept at ‘adjusting’ for number of days of high temperatures preceding, overnight temperatures etc.
It was also necessary to take account of school holidays, Christmas and Easter as domestic, school and commercial air-conditioning all had varying effects.
It was apparently quite a different process with unique considerations compared to the Eastern states.
Fascinating article and great explanation of a complicated process. Apologies. I am little confused about the undershoot references (i.e. “so far under its target?”) above and possible “large-error trigger” non-conformance. In a previous article .. https://wattclarity.com.au//articles/2018/09/a-case-study-in-active-management-at-a-semi-scheduled-plant-tuesday-24th-july-2018-at-lake-bonney-2-wind-farm/ it was indicated that ‘For the Semi-Scheduled plant, the Target effectively acts as a Cap – i.e. AEMO is telling them to “don’t produce any more than Target MW” at the end of that dispatch interval.’ I suspect I am missing something about the semi-schedule dispatch process, particularly when the semi-dispatch cap is set. Any chance of a clarification would be appreciated. Thanks .. Geoff
Nathan
Thanks for this, a really good summary of the traps. One that seems not mentioned but has caused problems to others is the cost for addressing earth fault levels at the nearby substation
Thanks David. Good point on fault levels. Agree this can be an issue and perhaps something that early engagement with the incumbent NSP may help to identify.
Well done and well presented Nathan; you probably regard it as an integral part of the network connection process but one part of that process which seems to escape notice or be regarded as trivial is National Grid Metering (NGM), especially the role of Metering Coordinator. The whole NGM process is complex and riddled with acronyms and contradictions (some more apparent than real) so again early engagement with the NSP who usually act as the MC is highly advisable. More than one proponent has had a shock on learning how complex and regulated the metering process is and the desire to have total control has over-ruled good sense in making use of those who have done it before and have the contracts with other NER defined roles in place.
Alan,
Re “why call it something as impenetrable as the “Electricity Statement of Opportunities”? I would add why do such documents always get moved to locations in the bowels of AEMO’s web site often requiring hours of searching to locate since such moves always seem to lead to any previous links being broken. it is often quicker to google the document than search on AEMO’s website.
This Guardian report would suggest National Grid has a frequency control problem, with three significant frequency excursions in the three months leading up to the event. https://www.theguardian.com/business/2019/aug/12/three-blackout-near-misses-in-three-months-says-national-grid The watt logic article above points out that UK nuclear does not provide a frequency control response, so the amount of MW and MW.s available for Raise regulation may be quite small.
Thanks Warwick – given the (relatively small) sizing of the battery, how is the shape risk of the flat swap managed by the solar farm? i.e. in periods where generation is insufficient to deliver the floating leg of the swap, how does the solar farm manage the risk that the cost to procure that power at spot is higher than the fixed swap price (particularly if the price is spiking above $300/MWh, and the battery is depleted)?
Hi Nick, your question relates to the valuing of risk as there are often periods where there is insufficient energy in the battery to deliver to match the contract level. What needs to be appreciated is that the battery is providing the best financial solution not providing an engineering solution for an islanded system. So yes, you might not always have enough energy in the battery to “defend” your contracted position, but in measuring the balance between the capital cost of the battery vs a desire to match every possible half-hourly price spike, it seems the best compromise is to make the battery as small as possible. Also, the battery might charge multiple times or make arbitrage over night or on weekends, making additional revenue for the project.
This is similar to generators in the wholesale market (including baseload coal) that sell contracts against their output and face the risk than an outage will mean that they could lose money. They take a calculated risk that in the long term, contracting is best and accept the non-zero risk of an unplanned outage.
I am referring to the ability of the solar farm to defend its contracted position, noting it is not dispatchable. Because the flat peak swaps traded on the market (used in your example) do not follow the shape of standard solar production, there is a mismatch in the shoulder periods (irrespective of outages). Is your solution to effectively self insure, I.e to reserve the surplus revenue above the contracted volume to settle the swap in the shoulder period?
It’s true that the solar farm is not dispatchable but the battery is. You are correct that the there will be times when we expect no solar production (i.e. after sunset and until 10pm) or limited production (early in the morning/late afternoon) yet there is a contract position to defend. You could say we “self-insure” in that we accept the risk that there may be a time when the battery is out of charge, solar is below the contracted level and the spot price is above the contract price but this is a calculated risk, mainly weighed up against the expectation based upon observed history that we would expect to earn close to the contract rate of $87/MWh (before battery cost) vs a PPA price of $40-50/MWh if one is available.
The alternative is to negotiate a swap contract that better matches the expected solar production but this would be difficult as the profile does not usually match a customer load, so we’d expect a discount to the standard contract, which is exactly the case with a PPA which is a perfect match to the profile.
In the end, if you’re willing to prepare a more complex integrated business model, you may be able to achieve a better return for your investment with less uncertainty. Alternatively, you can take spot revenue with absolutely no certainty or you can negotiate a fixed low price PPA with no price uncertainty.
It’s also worth noting that usually wind or solar projects “self-insure” even under a PPA as they do predictions of the expected annual energy yield and do a probabilistic estimate of the potential outcomes. Financiers may also require a P90 or P95 (i.e. an annual number that is satisfied 90 or 95% of years). This means that whilst they have no price risk, they do have volume risk, which can actually be insured against through bespoke weather derivatives or other insurance products.
PPA offtakers effectively pay unders and overs. For example, if the strike price is $70, and the pool price is $80, then the PPA offtaker gets a net revenue of $10 per MWh. If the pool price is $60, the PPA offtaker gets a net cost of $10 per MWh.
If the pool price goes to minus/negative $100, then the PPA offtaker incurs a net cost of minus $170, while the generator continues to get paid $70 per MWh. So PPAs have a few ways to deal with this: one method can be to limit the pool price used in the calculations to $0 per MWh. In this case, if the pool price goes to minus $100, and the generator continues to generate, the PPA offtaker pays $70 per MWh to generator, but the generator is also responsible for AEMO payments below $0, so would see a net result of (70-100) minus $30.
(You may already know all this)
Of course, AEMO payments are settled weekly, and PPAs monthly, so 30 minute pricing is less important rather than overall Dispatch Weight Average Pool Price is what counts.
But I do think this incentivise’s better outcomes. Supply should be geared to meet demand, rather than just run of plant generation. If the price is negative, it is a sign of excess supply, and supply should throttle and reduce output. This was less possible for brown coal generation historically, but theoretically much easier for solar farms.
I am mystified by the AEMO requirement for synchronous condensers at new PV and Wind sites. All of these new sites use inverters to couple to the grid and inverters can generate VARs or absorb them just as effectively as synchronous condensers. In any case the overall power factor of the grid used to be set by the network not the generator
Another weird period in the SA market was on Saturday 16 November. In the 18:20 to 18:25 interval the price was -$387/MWh, then 18:45 to 18:50 it was $5000/MWh. The Port Stanvac generator turned on, and there was no flow on either of the interconnectors for several hours. The Heywood link came back just after midnight.
“if we start with an incorrect understanding of what the actual problem is, then it’s a very low chance we’ll ever come up with a solution that actually works?”
Without level playing field power generation I’d suggest the big picture can never work. That simply means every tenderer of electrons to the communal grid must reasonably guarantee them (ie short of unforeseen mechanical breakdown) along with a minm FCAS standard 24/7/365 or keep them. The fact we don’t demand that is the root cause of all the problems we currently face.
It’s a political problem and not an engineering/ economic one even if you want to levy a level playing field CO2 tax on them all which is further compounded politically with the international cooperation required. Nothing like protracted brownouts to call forth the political will nationally to face the truth at some stage I suppose while the international politics is a second order priority.
– on your point about the broken hill solar farm I would argue that despite the initial MLF being above 1.0 that was not a good investment. The risk of that MLF being downgraded significantly was always high, when it was built there was no plans for upgraded transmission or greater load near the source so any addition near them would undercut them. I don’t think it is unreasonable that they should have assessed this risk. I would liken the Broken Hill solar farm to investing in a mining town in the boom.
– on your point on building where resources are best, I agree that would be ideal. The establishment of REZs and adequate transmission will assist this. But development of projects that are better located to supply that electricity, I would argue, is just as important as locating them to generate the most. No point having a perfect spot if you just waste a heap in losses and get curtailed all the time.
This is what happens when you let accountants out of the box and give them control of technology. They forget that the initial intention was to drain the swamp. The financially most efficient place to build any generator RE or fossil fuel is where the recourse is optimal. This applied when we were building thermal power station, near coal deposits, and the grid was reinforced to cope no one applied loss factors. The obvious place to build RE generators is where there is either a lot of sun or wind. These areas are readily identified and measures should be taken to reinforce the network to handle the increased generating capacity. RE generators in these places have the following economic benefits, more local employment, lower marginal cost of generation and reduced carbon foot print all good outcomes for the economy and the environment. Unfortunately we have a mob of twits in charge who cannot see this
I would like someone to explain how the Big Battery has such a high MLF, can it really deliver its power to a end user at a high MLF. What loads are around the battery that can take instant 30MW off and on day and night?. Or can it and does its load go long distances?.
By looking at its output, its pretty obvious its being run to make money, Mainly, not to deliver power, although at times it does, but thats mainly with the reserve.
How far can its power actually go within the grid is I suppose my question, say compared to a coal power station.
Hi John,
The lines in SA were built to take coal power from the north down to Adelaide with limited losses.
Multiple high voltage, parallel lines makes for a great place to connect generators. The wind resource just happens to be excellent too! Nearly all generators between Adelaide and Port Augusta have good MLFs.
Hi Tom,
I thought there were also other factors such as voltage that played a role?
I wonder if the battery is say being charged at the same time as suppling?
I wonder if maybe some of that is in essence going around in a loop?
Thankyou for a very good analysis but if we take Jan 25th as a the base case where we were short 200 MW and coal was only supplying 3,400 MW instead of a rated 4,700 MW, we have already experienced a situation with 4 of 10 generators being off on a hot day.
One can say that if we had had 300 MW more supply/ less demand we would have been OK. We will have almost 400 MW of behind the meter solar, 600 MW of utility solar and about 900 MW of wind additional capacity. Usually hot days are more like today where minimum wind generation in Victoria between 10 am and 7 pm was about 850 MW whereas 25 Jan was very unusual in that wind varied between 380 and 810 MW
Due to difficulties connecting wind and solar farms the installed capacity growth has been much faster than energy delivered, some wind plants are also being constrained down as well as the solar.
I think it is hard to see that demand will not be reduced by at least 200 MW from 9 till 3 due to rooftop supply and new utility wind and solar will provide a minimum of 1,000 MW and an average of 1,500 MW, vs 450 MW min and 800 MW av on January 25th. Today for example wind in Victoria only averaged a bit over 45% CF so it wasn’t a particularly good wind day and yet wind generation was more than double that of January 25. Further at 1 PM on January 25 imports from NSW were zero. NSW will be generating at least 400 MW of new wind and solar in the west and the Riverina, surely some of that will find its way south to Victoria even if it is only 100 MW.
But it is also a reasonable bet that if there is really a small reserve margin, that ways will be found for at least a temporary increase in allowable power from these constrained northern and western Victorian assets, so I suspect if coal drops by another 500 MW at least 250 MW of wind and solar would be found, adding to the 850 MW swing in the demand supply balance mentioned above.
Farley, VIC wind contributes about 10% of nameplate capacity to peak demand. Solar contributes zero.
You don’t have to exaggerate here like you do on the other blog, it’s not necessary, just the facts are fine. Everybody know that no matter how much nameplate capacity of wind is installed, it may not be there when it’s needed.
Having spent years commissioning power stations and other large rotating equipment, I suggest the risk of the two replacement generators having teething issues is understated.
Additionally there will be performance testing and bedding in periods required by the OEMs. “Online” is a good milestone, but doesn’t necessarily mean “fully available”.
It seems to me that last summer was the harbinger of things to come. Tight supply is the new normal, regardless of how much wind and solar go in.
Thanks for the comments Peter & Ben – they both identify important issues and illustrate the complexities of the reliability puzzle, particularly as your conclusions tend in opposite directions. I’ve tried to put some thought into a response.
First some context on Jan 24 and Jan 25 last summer. On both days there was forced load shedding (smelter on day 1 and Victorian consumers on day 2), as well as AEMO calling on load reductions from its portfolio of contracted RERT customers.
Demand and supply conditions were different on the two days. Jan 24 saw a combined SA/Vic shortfall with extreme SA temperatures (~46 degrees) and very hot weather in Victoria (~41 degrees in Melbourne). Two large Victorian coal units were out. The supply shortfall emerged in the late afternoon / early evening as rooftop PV production fell away, increasing grid demand, and wind production across both states reduced to about 15% of installed capacity. Smelter load cut and RERT calls in Vic reached about 400 MW of demand reduction.
Jan 25 saw a Victoria-only event, with temperatures of 43-46 in the Melbourne area following the previous day’s high temperatures – consecutive days of heat also elicit significantly higher demands than the first day of a heatwave. Cooler weather in SA enabled exports to Victoria up to interconnector secure limits, but offsetting this a third coal unit had failed. RERT load reductions of up to 450 MW were called by AEMO but as demand continued to climb and wind production reduced to around 400 MW (roughly 20-25% of Vic wind capacity), rolling area blackouts commenced from midday local time – these reduced demand by around 200-300 MW. Significantly longer and deeper load reductions would have been necessary had rising underlying demand not been arrested by an early afternoon cool change accompanied by increased winds and higher windfarm output. Victorian rooftop PV and large scale solar output was over 1000 MW during most of the load shedding period.
Both days were good illustrations of low probability but high consequence events that can contribute disproportionately to reliability risks. Very hot weather is not always coincident across SA and Victoria. Based on historical outage data for the ten Victorian coal units, the chances of three units being out on the hottest day of summer, Jan 25, was of the order of 1.6%. (Even with doubled forced outage rates that risk would still be only around 8.5%). Combined with that day being one of extreme temperatures – well above those expected on the hottest day of a “median” summer, this may all look like a very improbable outcome.
But there were also a multitude of favourable factors that don’t get as much notice as the negatives – there were no significant gas, hydro or transmission outages, renewables output was broadly speaking in line with or even above median levels for the time of year, the cool change arrived before the afternoon / evening demand peak on Jan 25 and so on. And even if only two coal units had been out on Jan 25 there still almost certainly would have been shedding, perhaps even the case with only one unit out.
Because of all these moving and uncertain parts it’s basically impossible to draw clear conclusions about reliability by using “single point” estimates – even quite conservative ones – of what levels demand, thermal output, renewables etc etc might reach, or how many units might be off, to conclude that there will or will not be supply shortfalls this summer. I don’t see much alternative to the type of probabilistic modelling that AEMO runs for the ESOO which can test literally hundreds or thousands of combinations of the myriad factors with associated probability distributions.
The result of AEMO’s modelling for this summer is that – assuming LYA2 and Mortlake 2 are back in service – they estimate that the “risk x consequence” of supply shortfalls (USE measure) is a little lower this summer at 0.0013% than it was leading into last summer at 0.002%. Behind that number are many scenarios – probably a majority – where there are no shortfalls (Peter’s point) but still a significant proportion where there are (Ben’s point). In a sense you could both be right.
A lot to digest here. I guess paying people for their solar output at a mlf of 100% (or 1) in the long term will more than likely have to be reduced down, as more and more people get solar and on average the solar from houses has to go further to be used.
This is very useful, just picked up on it. Could we perhaps have a think piece on renewable PPA’s from the point of view of the buyer, particularly a gentailer buyer? Renewables are increasingly achieving low NEM prices as their weight in the bid stack increases. To make this worse, I suspect that correlation is rising as geographical dispersion of assets falls. The PPA buyer is paying a fixed price for variable volume, and will frequently have been better off just taking spot – perhaps one reason for large hedge expenses booked by AGL and Origin. So what exactly is the point of a PPA for the buyer? If PPA buyers start to get cold feet, could it make it harder for new projects to get financing?
Posted this on Allan’s ‘negative prices’ piece, repeating it here. Could we perhaps have a think piece on renewable PPA’s from the point of view of the buyer, particularly a gentailer buyer? Renewables are increasingly achieving low NEM prices as their weight in the bid stack increases. To make this worse, I suspect that correlation is rising as geographical dispersion of assets falls. The PPA buyer is paying a fixed price for variable volume, and will frequently have been better off just taking spot – perhaps one reason for large hedge expenses booked by AGL and Origin. So what exactly is the point of a PPA for the buyer? If PPA buyers start to get cold feet, could it make it harder for new projects to get financing?
I think that low point this year tells a story, my guess is that rooftop solar have started to impact, and have created a lower low than the year prior. So there is a chance, the high will also not meet last years high.
However, weather will play the bigger role this year, but going forward in time, both the lows and the highs should track lower due to rooftop solar/ batterys?.
John, batteries are generally fully charged by 10am and fully discharged by the afternoon. The effect of batteries is therefore to accentuate the duck curve.
This is due to the default charging/discharge curve. An optimised curve would spread the charge over longer to help with the trough and delay the discharge to help with the peak.
I thought there was no need for a MLF because inverter settings are meant to reduce output to zero when the voltage at the inverter reaches 258 Volts. This occurs occasionally in my unit between about 11 am and 3 pm on sunny days when there minimal load in our house. However there is no process to formally check inverter settings. How is this overvoltage curtailment captured in the AEMO and APVi data? Our curtailment data are captured in the Fronius database, which may be made available to these data sources.
Earlier this week there was an interview on the 7:30 Report with a spokesperson for SA Power Networks, where he said there was a time recently when there was so much rooftop solar output that the entire southern half of Adelaide fed back into the transmission network.
Hello Observa, I thought all the cars were going to be charged during the day, so the batteries in them could be used at night to power the grid. Or at least when there was not wind or sun for a few days?
It is interesting that the two interpretations of the same event both contain elements of truth. Part of the deal with Aluminium smelters, paper mills etc is that if you get power at 1.4-2.5 c/kWh, you are expected to drop demand in emergencies. It is common practice throughout the world including in Germany where customer downtime is an average of 12 minutes vs our 2.5 hours per year.
As you say wind output on the day was a bit higher than the previous 10 days but only half that of December the 9th, the first hot day of this summer. The cool change was definitely a help but another coal generator Loy Yang A 2 went offline at about 5 pm, more or less offsetting the reduction of demand.
I am not sure how you can claim that there would still have been load shedding if there were only two coal units out, because Yallourn 3 and 4 which were both offline by 1 am were 375 MW each and Loy Yang A1 of 500 MW was offline. Loy Yang A2 never got above 75% capacity and finally gave up at about 5 pm so during the critical period from 10:30 till 6 PM, coal generation was down by 1550 to 1850 MW. One Loy Yang unit and one Yallourn unit is a total of 880 MW. Even if we had one of each out and Loy Yang A2 struggled along at 75% capacity we still should have had 3,500 MW of coal which would have been enough to avoid blackouts
However on the overall point it is ridiculous to demand that we eliminate all risk of USE in a spindly grid like Australia’s where 95% of outages are T&D related. Back in 2009 we had three load shedding events of about 1,000 MW. Did the world stop.
Increasing use of local resources can in fact reduce the likelihood of cascading widespread blackouts. For example in Adelaide during their storm blackout, when the storm warning was issued if they had increased gas generation by a mere 150-200 MW, thereby increasing the headroom on the interconnector, no more than rotational shedding for half an hour to an hour would have been required. If South Australia’s current battery capacity had been available at the time probably no urban or Southern blackouts would have occurred at all
The most charitable thing to say about Ben’s comments is that he hasn’t looked at renewables output graphs for the last few years
Peter my comment that load shedding would still have been likely on 25 January with only two units out is supported by AEMO’s 2019-20 Summer Readiness report which quantifies general customer load shed that day as up to 538 MW (on top of RERT reductions of 625 MW). An extra Yallourn unit wouldn’t have covered this gap, while an additional Loy Yang unit would have left the balance very marginal.
And I don’t disagree that in the majority of circumstances the additional wind and solar commissioned in Victoria should leave supply-demand balance better placed than last year.
If the collapse of day-time prices as seen in the last 4 months was the cause of the lack of completed projects, then it was obviously well signalled and forseen.
But perhaps there are other factors preventing the 2019 completion of QLD projects, including:
– difficulties in getting network connection arrangements agreed;
– lack of purchasers interested in middle of day power volumes;
– State government electrical works licensing changes that dramatically increased construction costs;
– failure of building contractors;
– delay in QLD government renewable auction scheme;
– reluctance to compete in a market where the QLD government holds 85% of generation assets.
It should be no surprise that wholesale prices will be affected when new generation enters an already oversupplied market. The day-only characteristic of solar determines when it impacts the market.
I’m not convinced a renewed subsidy scheme for wind and solar to reduce global temperatures is a positive step.
I’ll look forward to your assessment of price spikes in Vic and SA on the 20th. They coincided with what appears to be curtailment or poor production from wind farms during the heat. I live near the Macarthur Wind farm and it was blowing a gale at 40 degrees, prices were up to $2500/MWh, but output was only about 20%. Other wind farms were similarly affected, and some also had periods of constrained or nil production during the day including Mt Mercer, Morton’s Lane, Mt Gellibrand, Murra Warra, Oakland, Portland, Waubra, and the SE SA wind farms. If wind and solar are the new baseload we need to make sure their output can reach users during the heat when it’s most needed.
Here are Bureau of Meteorology data on wind speed at Hamilton Airport, which is their nearest observation station for the Macarthur Wind Farm, Morton’s Lane, and Oakland.
Solar, wind, batteries, no doubt great for moving money from customer to industry bank accounts, but the real hero gets mentioned last, the Torrens Island gas power station, rapidly adjusting its massive output as required.
The real problem in SA, and elsewhere in the climate change obsessed world, is that nobody is going to build another Torrens, so the clock is ticking, not to any climate tipping point, but to a breakdown of the electricity system.
Hi Climan,
There is going to be so much cheep power 24/7, we in Australia are going to be making hydrogen, plus exporting heaps of power direct to asia.
Well thats what I am getting from all that i have been reading.
Per market notice 71933, the Heywood M1 500/275 kV transformer (taken offline at 13:55 on Friday 20th) did not return to service at 17:00 when scheduled in notice 71906. This continued to constrain inter-regional power flow from Western Victoria during the period of peak demand.
I live near the Macarthur Wind Farm, and was surprised that it was blowing a hot gale but the wind farm was only producing at <20% of capacity. For example at 17:20 NEM time, when the 5-minute Vic price spiked to $14,000/MWh, BoM's Hamilton Airport weather station recorded a wind speed of of 30 km/hr, gusting to 48 km/hr. Temperature was 40.6 degrees, having declined from a maximum of 44.5 degrees at 16:00 NEM time. The wind farm was only producing 60 MW (14.4% of capacity). It is rarely subject to grid constraints because it feeds into the 500 kV line to Heywood. Several Vic wind farms were off at the time including Mt Mercer, Mt Gellibrand, Murra Warra, and Waubra, while the operating Vic wind farms had low capacity factors despite being exposed to a similar wind resource. The SE SA wind farms, exposed to similar high winds and high temperatures had capacity factors of 60-80% much of the day, but went off at 17:20 NEM time with the loss of 200 MW.
Two hypotheses
1. When developers ordered generators for the Vic wind farms they didn't believe it was worthwhile paying any extra for high-temperature performance, because their contracts are only for the energy output without any premium for production when prices are high, whereas SA developers are more acquainted with such temperatures and have paid extra for better cooling systems in their generators.
2. TNSP's underestimated the degradation of their equipment performance over 40 degrees.
Until the pricing of power is changed, there will be times when, yes even solar and wind producers will, try to help the spot price up, its human nature.
The top price is simply just way too high, its too tempting.
But can the price ever be right without guaranteed dispatchability from every player at the regulator’s beck and call? Also presumably TNSPs would no doubt like to know what’s regularly required of them for that too.
I know Aldi have some cheap hardware from time to time but if it’s a necessity I go to Bunnings. It’s not that hard to work out the difference.
Some thing is wrong, no matter which way its looked at. The price of power has gone up 3 fold, and the fluctuations are simply unreal. The whole system, is a dogs breakfast.
We are now hitting the CRAZY high prices, and pretty Crazy low prices, more and more often, and everywhere in between.
Power is a needed for our system and country, and it needs to be stable in Both supply and price.
It is simply crazy, as it is 2019, almost 2020.
“When purchasing a wind turbine, the power curve is guaranteed up to a certain ambient temperature (often 40°C). Beyond this temperature, the operating temperature of some components monitored by the wind turbine’s control system may exceed safe levels and initiate a temperature-induced constraint on the wind turbine.”
If it is indeed – as it appears at this stage – that a lot of existing wind farms suffer pretty severe temperature related output constraints above roughly 40C ambient that is… not going to help the public image of renewable energy. It will certainly knock dead any supposed advantage that deployed VRE has over thermal generation in terms of temperature de-rating. Can we put this down to equipment designed for e.g. European climates? Deploying gear in Australia that goes into limp mode once temperatures pass 40C seems foolish/near sighted. We all wait with bated breath…
Also worth mentioning that Loy Yang B unit 1 went offline yesterday with a boiler tube leak. It will also be interesting to see if there’s any repeat of the heat derating of the windfarms that we saw on 20 December. A couple (Mt Gellibrand and Yaloak South) are offline as I type – possibly not heat-related – while output at Waubra and Macarthur is declining.
I happened to be holidaying with family at Coobowie/Edithburgh in SA today in view of Wattle Point wind farm with its 55 turbines and for the last couple of days they appeared to be working well. Except for around 11.30-12.00 today when we happened to be returning from Troubridge Point in a dust storm with the car temp outside reading 40-41 degrees and the turbines had all shut down which you can see clearly here leaving only WPWF graphing-https://anero.id/energy/wind-energy
Not the only severe dip with the wind blowing strongly all day as you can tell by the wind farm output over 90% of installed capacity alongside the shut down. What is going on with these wind farms?
Re: (E2) Is it true that “Lines from Snowy to NSW have been destroyed by fire”?
Minister Matthew Kean was on ABC Sydney radio just before 6pm saying that at this stage he was being told that the ability to control the output of snowy was lost but that no expensive equipment had been destroyed so there should not be a prolonged outage.
The interesting point is that contrary to predictions it was NSW which went to $14,700/ MWh and Victorian prices went negative, indicating a strong oversupply in Victoria. At 5:25 pm after transmission was was restored, NSW was importing on all three links at limit values, power was still at $14,700/MWh and Victoria only rose to $34.
My summation is as I predicted a couple of years ago that NSW may actually more vulnerable than Victoria
Transgrid (via Twitter) reports upper and lower Tumut substations were subject to fire damage. Lower tumut restored, but upper awaiting road or air access to site.
When a transmission line is reclassified as a credible contingency due to lighting, is the transmission line removed from service or does AEMO have a plan/constraint equation ready in case it is hit by lightning?
Good question; no the line(s) would not be removed from service but a constraint would be applied to ensure the system was secure against loss of both circuits. This might for example involve limiting total loading on the line to levels below normal capacity. Allan.
“Teams of people from regulators and industry participants beavered away under the aegis of the ESB………… I haven’t looked at the relevant sections but I guess they run to dozens and dozens of pages, and associated guidelines and procedures produced by the AER (Australian Energy Regulator) and AEMO run to many hundred more.”
Oh dear when what we consumers really want is tenderers of electrons to the communal grid to be able to reasonably guarantee them 24/7/365 both with their frequency and voltage or else keep them. A level playing field methinks but perhaps I’m being all too simple minded here in the face of great complexity?
For the evaporative cooled plant, it might be worth plotting the capacity factor against the Wet Bulb temperature, as that limits heat rejection of the cooling tower and minimum condensor temperature/pressure.
Thank you Allan for an excellent summary of the situation on January 4. I feel compelled to acknowledge AEMO for its highly professional management of very difficult conditions over an extended period. And I should add to that the people at TransGrid who worked tirelessly on the ground in the the control room to keep the system up an running as much as possible.
Paul, thanks for the feedback. Those of us watching events could well imagine the extreme pressure TransGrid and AEMO personnel were under on that day, as well as many others over this summer. They certainly did a great job. Allan.
A couple of quick responses to your questions on pricing under RERT dispatch and observations on the southern NSW constraint.
When RERT is dispatched, intervention pricing (also known as “what-if?” pricing) is used to determine spot prices. This attempts to back out the effect of the RERT contracts (so add them back to demand if they were demand reduction agreements) and work out what the price would have been. If enough supply was offered below market price cap to meet demand with RERT added back in this way, the resulting price will be below MPC.
This differs from the situation where AEMO orders *involuntary* load shedding in which case the price *is* automatically set to MPC.
On the southern NSW constraint, its effect as I understand it is to limit aggregate flows from southern NSW into the rest of NSW, due to reductions in transmission capacity caused by the fires on January 4th (specifically damage to the Upper Tumut – Yass 330kV O2 line). Whether those flows come from Snowy generation, Victorian imports, or other southern NSW generators doesn’t really matter. Unless I’m missing something, Snowy’s generators bidding in more volumes at low prices doesn’t *reduce* the overall flow available into the rest of NSW, it just displaces other generation or imports within the same overall flow limit. When we see “forced exports” from NSW to Vic in this situation it isn’t because less power is getting to the rest of NSW, it just output from generators on the south side of the constraint which *exceeds* the available northward capacity finding its way into Victoria instead.
I should add though that Snowy’s generators being paid the NSW price even for the excess generation they force into Victoria isn’t really an economically efficient outcome, it just reflects one of the “compromises” in NEM market design where all generators in a region are paid the regional spot price, regardless of where their energy actually goes.
Thanks for the reminder about the mechanics of pricing under RERT, Allan. I was thinking more philosophically …
Viewpoint #1) I’d been thinking of RERT as an “out-of-market” intervention (i.e. in essence a sign that the market had “failed” to deliver sufficient capacity – and that, as a next-to-last resort, better to pay those who volunteer to turn off rather than proceed to (much less surgical) Under-frequency Load Shedding.
Viewpoint #2) However perhaps today is an example of a more risk-averse & precautionary approach to dispatch (i.e. the prices delivered might might be read to suggest that whatever was triggered in Reserve Trader was not actually have been required, in the wash-up (of course, with hindsight being a wonderful thing that just does not exist ahead of time – especially with it needing to be triggered (at 13:26) 120 minutes (or 90 minutes ?) ahead of dispatch starting 15:30 (or trading period ending 15:30?))).
Interested to know if others perceive things the same way, or differently?
The economically rational approach for Snowy would be to run the Jindabyne pump, which pays the Vic pool price (negative $200/MWh at times today so Snowy gets paid to use power), while also generating at capacity in Tumut 3 and getting paid $14,000/MWh in the NSW pool. These economic absurdities will strengthen the case for Humelink and Keranglink as transmission investments to increase the Vic-NSW capacity.
Several wind farms turned off during the afternoon – perhaps through transmission limitations. Boco Rock, Crookwell, Gullerin Range, in all about 350 MW missing, or perhaps displaced by more strongly negative bids from Tumut 3.
Several wind farms turned off during the afternoon – perhaps through transmission limitations. Boco Rock, Crookwell, Gullerin Range, in all about 350 MW missing, or perhaps displaced in their access to the transmission network by more strongly negative bids from Tumut 3.
There were a lot of different things going on. Extreme wind conditions were affecting some of the wind farms throughout the day, causing them to reduce output to quite low levels to protect equipment. Paul’s article mentions the constraint limiting flows north out of southern NSW – there was also a constraint limiting southward flows into Vic to 500 MW, and when this one bound as well it forced offloading of generators in southern NSW, including the windfarms. On a quick scan most of those generators had offered at -$1,000/MWh so should have shared the offloading proportionately, but Snowy had craftily rebid its downwards ramp rate at Tumut 3 so as to limit how much of this offloading it actually wore every five minutes, likely meaning that windfarms took more than their share of the output reduction. This sort of ramp rate rebidding has occurred in the past and needless to say has caused some controversy.
The abrupt shutting down of Wattle Point wind farm in SA I noticed on Dec 30th travelling to Troubridge Point was clearly a congruence of the high temperature (41-43degC on my car external temp gauge) and an obvious large dust storm that suddenly blew up above and beyond steady wind already turning the turbines. That seems to suggest that it’s not just temperature that automatically constrains these turbines but a range combination of wind speed and temperature which would vary with turbine type. That information for each wind farm should be available to the AEMO for forecasting purposes.
Operational AEMO staff called RERT providers on Wednesday afternoon, anticipating the possibility of breaching the LOR2 threshold.
On Wednesday the PASA LOR forecasting was playing up – putting the LOR2 trigger in the 1100 MW range (due to issues accurately forecasting intermittent generation in the PASA process), however operational staff anticipated that the actual LOR2 trigger was 750MW (an Eraring unit) and that supply/demand balance would be heading into this region.
Ambient temperature has zero effect on a boiler or a steam turbine. Heat exchangers in auxiliary services such as lube oil coolers and the condenser hotwell is where the effects of high ambient temperature come in.
High oil temps can trip a unit quickly, especially if a cooling fan fails, but gradual temperature increase can usually be halted by reducing load.
Gradually losing vacuum in the condenser is easily managed by reducing load.
(B2A) WE’VE SEEN NO EVIDENCE ABOVE THAT AMBIENT TEMPERATURE IS A SIGNIFICANT FACTOR IN FULL UNAVAILABILITY!
I was once an Operator at Vales Point Power Station
There are two things which can limit the load on the Generator
1 Outlet temperature of the cooling water from the condenser and
2 The back pressure in the condenser.
The cooling water outlet temperature on the condenser is limited to about 37 degree C. I retired nine years ago and that is what is in my memory. There is a lower outlet temperature (about 35 degrees C) which was the normal limit. If the grid was in danger ie…load was not matching demand then permission from a higher authority was required to raise the limit (37 degrees C) and I do not think that authority was from an executive of the Power Station. This is to reduce any damage which may occur of any marine life within the water source.
The cooling water outlet temperature of the condenser was directly proportional with load and the cooling water’s inlet temperature at the condenser’s inlet. During continuous high daily temperatures (as in summer) the source of the cooling water entering the condenser increases and as a consequence, the cooling water outlet temperature will also rise.
To maintain the condenser’s cooling water outlet temperature within limits will require the load on the Generator to be reduced. So, the volume of the steam entering the condenser, lowers the heat energy contained in the steam, matching the capacity of the cooling water. Energy in equals energy out.
If, as at Liddell, the water source is confined to a fixed source, then that source will increase in temperature as the cooling water is recycled around the condenser, also, as during the previous drought, over time that water source will diminish as a result evaporation. Another problem, the condenser tubes must remain highly conductive to allow the heat to be transferred from the steam/condensate to the cooling water.
At Vales Point the cooling water is drawn from Lake Macquarie. I always suspected a current was developed, in a stream, so that a proportion of the water flowing into condenser’s inlet had a component of water from its outlet.
Another suspicion I had we were drawing cooling water from Earing Power Station’s cooling water outlet. Eraring Power Station is located on the western side of Lake Macquarie, relative speaking, not too far north from Vales Point.
THE BACK PRESSURE IN THE CONDENSER
The most desirable back pressure in the condenser is 5kPa absolute (on the steam side) at all loads. As the load increases so the back pressure will also increase. The back pressure is maintained by the cooling water passing through the condenser. When the cooling water temperature is comprised and is unable to maintain the back pressure within the limits, then back pressure will increase as the load increases. There is a limit on the backpressure (which I cannot remember but I am pretty sure it is not higher than 10kPa absolute). When the back pressure reaches 15kPA, both the Boiler and the Turbine will trip and the Generator will come off the grid. The 10 and 15kPA pressures are estimates of what I remember those pressures to be.
This back pressure must be maintained to ensure the steam leaving the last blades of the LP Cylinder is in a superheated condition. At low loads, the vacuum in in the condenser reaches up way back into the LP Cylinder and as the load on the Turbine increases, the pressure in the LP Cylinder also increases. If the steam leaving the LP Cylinder has no superheat then the water partials in the steam will damage the blades in LP Cylinder (which are spinning at 3,000 rpm) that are exposed to the partials.
Now, as the cooling water entering the Condenser’s inlet becomes warmer, the capacity of the cooling water to maintain back pressure within a satisfactory limit becomes more difficult. When the back pressure is exceeding these limits then the load on the Generator must be reduced to ensure reliability (some load or none).
A factor in the transfer of heat of the steam/condensate in the condenser to the water cooling the steam is relative of the condenser’s cooling tubes through which the cooling water flows. If the tubes become dirty the efficiency is lower.
At Liddell, this cooling water is sourced from a relatively isolated lake and the cooling of their condensers rotates around the different parts of this lake and doesn’t have the benefit of a free flowing river or the high and low tides of a lake with a connection to the ocean. When considering the server drought in the recent past, these two conditions could have contributed a rise in the lake’s water temperature. Because Liddell Power Station does not seem to reach 400MWs implies to me there is a problem with their condensers
(This is my opinion) At Vales Point the cooling water rotates in a stream from the outlet side to the inlet with some dilution of cooler water in the lake depending whether the tide is flowing in or out also Eraring Power Station is on the western side of Lake Macquarie, not too far north of Vales Point. I always thought we were pulling water in from Earing Power Station’s outlet and this increased difficulties in the summer months for Vales Point.
SOMETHING ELSE
With renewable energy creeping into service and consuming energy demand, the coal fired Power Stations must be satisfied with what is left over. As this renewable energy takes more and more of the share this continues to reduce the Power Stations’ share. During summer months, the days being longer and the renewables performing better will leave less capacity needed to be made up.
Boilers, at coal fired Power Stations, have a low load limit. This is to ensure the safety of the fire in the Boiler.
At Vales Point this was once 300MWs with three mills in service with their feeder running not less than 50%. It was later reduced to 250MWs Oil torches where placed into service and the load reduced from 300MWs to 250MWs and when the fire in the Boiler was considered safe the oil torches were removed from service. This has since been lowered to 200MWs indicating the space load share was becoming difficult.
Because of this low load limit and the share over time diminishing there will come a point when a Unit from a Power Station will need to be removed from service. If it is Liddell Power Station, then 400MWs will be removed from the grid. If it is from another Power Station, it will be at least 660MWs. This can be serious because if this Unit “which has been mothballed or it is too inefficient or is too old” is required for service, it can take two to three days to synchronise and close on to the grid.
(Another opinion) In earlier years of commissioning Power Stations the Operators were sourced from steam ships. These Marine Engineers developed a lot of experience operating steam engines and the older “gingerbeers” passed their knowledge to the newer and younger recruits. All these Marine Engineers had a Marine Engineering certificate (part A and/or Part B). These certificates taught on the properties of steam which is very helpful when considering the condenser and other parts of the Unit to analyse what is going on. There are now no steam ships and no Technical Colleges courses supplying that help. Most of the Operators are now Electricians.
Thanks Allan for this excellent summary of the issues that – to some extent – all generators face in managing their revenue. From a policy perspective the key question seems to be: should generators be facing these headwinds – i.e. are they a fair reflection of the costs and benefits they offer to the grid? If so, then they would seem to be just the cost/risk of intermittency/weather dependency and it’s up to the generator to look for cost effective ways to manage this risk – removing these risks would stimulate more renewable development, but at a significant cost to consumers who would end up bearing those cost/risks instead. That would probably take several further articles to discuss, but would be interesting to read your thoughts on this question. I recognise that there are multiple parts to the question – in the case of MLFs, say, it seems reasonable that generators be given some locational signal so that losses are accounted for, but does the specific method of calculation do a good enough job of doing this? And then are there risk management tools available to the generator – and if not, why not?
After the trip of the 500kV transmission line at 1:45 pm Friday due to 8 towers being blown over just west of Geelong, it is interesting that the Mortlake gas power station now fully supplying the Portland aluminium smelter as a local circuit independent of both the Vic and SA networks. While the transmission structures are intact to connect this circuit to SA, by operating as a local circuit it doesn’t need to meet the NEM standards of requiring alternative energy paths in case of circuit trips, as well as standards of inertia, frequency control, etc.. These protocols of a local circuit from Mortlake were presumably developed in response to a power outage several years ago that was long enough that caused potlines to solidify.
Malcolm – some market view dashboards (eg AEMO’s own website) may give the impression Mortlake and Portland smelter are isolated from both regions, but they remain connected to the SA network through Heywood. The zero target flows shown on the Heywood interconnector are an artefact of the way dispatch in the SA + SW Victoria “island” is being managed. I’m sure Paul will post an update giving more details later this weekend. Allan
I can indeed imagine some fairly rude comments about the Federal Govt owned Snowy rebidding entirely for profit. There is a lot more than could be said. I also think that Friday proved once again the point Wattclarity has made on numberous occasions and is now “internalised” that wind output tends to be low in hot weather (ie summer peaks). However it is worth noting that system wide electricity consumption in mid winter is about the same as January and that wind tends to be much more effective then (but of course solar is down). Every generation fuel has its place in today’s market even if not in a future market.
I was going to provide some comment on how fast start plant is dispatched in the NEM, to partly explain what might have motivated Snowy to rebid this way (not necessarily to justify or approve of it!), but realised that’s probably a subject for a whole WattClarity explainer of its own.
Suffice to say here that when that $14,700/MWh dispatch price arrived (published just after 16:50) for DI 16:55 , Colongra would have received a signal to start up, but its “T1” time to synchronise is 10 whole minutes so it would not have delivered actual output to the grid until some time beyond 17:00, after the end of the 16:30-17:00 Trading Interval. The settlement price in *that* Trading Interval would have included the impact of the DI 16:55 $14,700/MWh dispatch price (six 5 minute DI prices are averaged to produce the 30 minute TI settlement price; the price for the half hour ending 17:00 ended up being $2,559/MWh), but Colongra would have produced no output until the *next* TI (17:00 – 17:30), when much lower prices might have applied.
Apparently Snowy was not willing to take this economic risk and thought the better thing to do was rebid the plant as unavailable in both the 16:30-17:00 and 17:00-17:30 Trading Intervals, allowing it to ignore the start signal received – and just as a by-product of this “optimisation decision”, triggering an LOR2 event as explained in your article.
Thanks Allan – so many layers of complexity to explain!
My focus was more on why the LOR2 arrived, not so much the motivation for the rebid decisions (though I understand people will go on their own tangents).
Thanks for the analysis Paul. I was aware that some generators game the system by withholding capacity to boost prices but letting the spare capacity fall to the rating of one of the larger thermal sets is really playing chicken with Mac trucks. If say Mt. Piper tripped at that time it would have really put the fox in the hen house
1. Will the 5 minute rule make such bidding practices more or less likely
2. SA prices have become much more stable with the installation of their batteries would 500-600 MW of batteries across NSW under different owners have defeated the gaming
The factors that drove Snowy’s rebidding are peculiar to the current 30 minute pricing / 5 minute dispatch setup, along with Colongra’s 10 minute startup time. The first will go away with 5 minute settlement, but how plants like Colongra will adapt to that regime are not clear. Can their startup time be shortened? How would they have to change their bidding and operating strategy to remain viable if it can’t? I’m not sure we have answers yet. And no doubt other issues will arise under 5 minute settlement – for example incentive to ramp output immediately, rather than linearly over 5 minutes, when prices go high or low.
The Smelter was offline from 13:24:56 to 16:09:32, a total of 02h 44m 36s with load gradually restored over the next 6-7 hrs. We can also see the output of two 294 MW Gas Units at Mortlake and the Heywood to South East Line Flows where it is apparent that AEMO moved to the reduce this flow to zero by approximately 16:47 hrs on Sat 01 Feb 2020.
That’s really interesting Geoff and Paul.
From the logs you’ve provided it looks like the V-SA interconnector remained connected through the fault and started supply back into the SW region. AEMO have then re-synchronised Mortlake to SA which is supplying the bulk of SA the FCAS (which explains the request for large amounts of FCAS in SA Friday night; can you check that please Paul?)
Seems that SA is still connected to SW Victoria & Mortlake. AEMO’s dashboard is not showing flows though (they’ve disabled it when the website went down?).
I’m not sure on Mortlake’s black start/grid forming abilities but I suspect they’ve used SA to do it. I’ve also noted the SA batteries got a hell of a workout Friday too.
It does appear that the design of the towers is nowadays inadequate to withstand extreme weather events. I would have thought that a engineering assessment of the transmission towers would have been done after the 2016 blackout. Maybe it was but the results not released? Regardless, the towers’ ability to withstand extreme weather needs to be upgraded.
Thanks for the insights Allan. I presume that there is some requirements as to how generators react to this type of situation, and that these requirements have become more detailed over time as grid management matures?
Chris – there has certainly been a lot of work done on protecting the South Australian system in the case of “islanding” events where AC connection to the rest of the NEM is lost, for example an over-frequency generation shedding scheme (OFGS) that starts to disconnect generation when frequency exceeds 51 Hz (and corresponding schemes dealing with under-frequency). However the circumstances of the 31 Jan event fell well beyond most scenarios explicitly considered – because the loss of both 500kV circuits in Victoria would have normally been viewed as extremely improbable. It will be very interesting to see AEMO’s detailed report on exactly how the SA system survived the event as well as it did. Allan
When you remark “no guesses on that one” regarding the 15:30 NSW demand reduction, do you mean that the demand reduction is obvious or that you’re unsure about what caused the demand reduction?
The main reason the SA system formed a stable island is because SA was exporting at the time. If SA was importing from VIC at the time, things would have been more challenging. The portland load connected to SA following loss of the 500 kV lines would have actually helped limit the over frequency event in SA.
Agree that exporting pre-event likely made things easier Winodh – however the 4s data on the first chart above indicates that Portland tripped off very close to the time of the 500kV fault (as indicated by both the Portland load trace and the 1000 MW reversal in Heywood flow). We’ll need to see AEMO’s detailed report with high resolution data to know the exact sequence and timing.
Hi Paul thanks for the insights. I’m interested in the winners and losers of this islanding noting that the Hornsdale Power Reserve has made over $14 million on FCAS markets since the separation. Looking at the GSD2019 this is more than half of what it made on the FCAS markets in 2019!
There are lots of folks more knowledgeable than me on the fine detail here, but a few thoughts anyway:
1. As you show, the behaviour of PV in SA broadly assisted the situation by reducing aggregate output after a rapid jump in frequency when loss of the Vic 500KV lines disconnected SA and SW Vic from the rest of the NEM, and then during a sustained period of over-frequency (your earlier article on SA frequency showed it exceeding 50.25 Hz for about 20 minutes after the fault). Some of the reduction is due to systems tripping off, and some possibly due to systems reducing then holding down their output while the period of over-frequency persisted.
2. Expected behaviour of PV inverters in the fractions of a second after the fault depends on a lot of factors that we can’t yet see in the data – how rapidly the frequency jumped (RoCoF), the peak frequency reached in SA (almost certainly above the 51 Hz in your 5 second data), voltage disturbances caused by the line fault etc. It wouldn’t be surprising if many of the trips related to these currently non-visible factors.
3. Expected behaviour also depends on age of system – there were important changes in PV inverter standards in late 2015.
4. For inverters in SA complying with the new standard that remained connected through the fault, these ought to have reduced output (relevant to pre-event levels) while frequency remained above 50.25 Hz. At 50.5 Hz, where frequency sat in SA for most of that 20 minute period, this reduction in output should have been about 15% (if I’m interpreting the standard correctly – a big if). It would be interesting to know if you can see anything like this behaviour in the sample data you have.
The missing solar analytics data also probably tells a story. Although the monitoring units have a 3G/4G network connection, they are not battery powered and need grid power to collect and upload stats. if there are holes in the data (no samples) it most likely means the local grid was off. I am not sure how resilient 3G/4G towers are to power outages but they probably have some battery backup in at least some of the towers but towers might get congested if some towers run out of power in an extended power outage which was not the case in this instance.
Apart from lack of cooling for wind farms the density of air declines and as the available power is proportional to mv2 and m declines from 1.2 kg/m3 at 20C to 1.1 @ 45C there is almost 10% reduction in available energy, the high temperature also reduces the efficacy of cooling of the generator as well as the gearbox so a 15% reduction in output at a given wind speed is not unexpected in high temperatures. The change in air density combined with Carnot efficiency losses is why gas turbines can be even worse than wind farms.
Coal plants often have to heat and partially dry the fuel so although the ISO ratings are usually performed at 15 C some coal plants particularly those with high moisture content coal often reach peak efficiency at up to 25 C because the higher boiler efficiency offsets the lower Carnot efficiency of the steam cycle. The brown coal plants probably have a better thermal co-efficient because the boilers are oversized to cope with poor quality coal and in Victoria unlike SE Queensland and even the Hunter, high temperatures are less likely to be accompanied by high humidity so the wet bulb temperature will usually be a bit lower.
If nuclear plants were built they would usually be de-rated more because they operate at 260-320 C vs 550-650 C for modern coal plants so the % temperature difference between Tmax and Tmin in the Carnot Cycle is higher.
Finally IC engines with properly sized cooling systems probably have the least temperature degradation of all because Tmax in a diesel is around 2000 C so a 20-30C change in Tmin is not a big amount although there will also be a reduction in charge density in the cylinders due to reduced air density
So part of a battery system is the inverter.
So where you say the output of the battery in Power is not affected, from your data, a battery and inverter (the inverter is the one making the power ac, from the battery which is only dc.) system does not suffer peak delivery power reduction at all? Just length of time at peak supply.
Thanks John – have added the figure from the GSD2019 showing full-year data for GannBESS discharging above in the article. Thought it was the easiest way to illustrate the answer to your question. In terms of instances of output (and availability) there is not a derating seen in the same way as it is seen for other tech. As noted on the image, the way this analysis has been done it does not highlight worsening round-trip efficiency.
Very interesting article Paul. I just want to point out that it is important to beyond direct correlation between output and weather. In Texas, we’ve seen fossil plants show up at higher than expected rates during periods of extreme weather stress. But, especially in July 2018, these events were fairly well separated. When I talked to ERCOT (AEMO’s equivalent) folks, they were clearly worried about similar summer conditions happening with more clustered extreme heat conditions, as they had some concerns that some of the older fossil generators might not be able to hold up under much more continuous use. So it seems that cumulative use can be a predictor of generator outages. Have you guys looked at that kind of analysis?
What is interesting is that Mortlake Power Station has stayed online and this seems to be the main contributor to the Westerly flows across Heywood post the trip (the 500 kV system was severed east of Mortlake). The Portland load should have been able to be maintained under this scenario so suspect there is something else that has resulted in Portland load tripping. The answer to this might sit in the first plot where the Portland load hasn’t reduced as soon as the Heywood flows reversed, but need the smaller timestep data and sequence of events data to confirm. Looking forward to the AEMO operating incident report
PS: This information and plots are great, no doubt a bit of effort goes into putting them together, but they tell us a good story.
Great article Alan.
The graph for Raise 6 Second FCAS shows the Hornsdale battery providing what looks like around 60 MW continuously for days. I thought it was a roughly 100 MWh battery. So how does it provide a continuous 60 MW FCAS supply? Any ideas? Is it being kept charged by the Hornsdale wind farms?
Fair question Paul. The so-called “contingency” FCAS services, including Raise 6 second, only require a supplier to rapidly deliver (or consume) extra energy when there is a significant frequency disturbance. So 99.x% of the time, a battery can provide this service without actually having to discharge. This is one of the major reasons, along with quick response and excellent control, that batteries have been able to occupy a large share of the market for this service. Allan.
PS So just to clarify, for these services, providers are being paid to provide a #response capability# measured in MW over a period of time measured in hours – hence paid in dollars per megawatt-hour of capability not energy, if that makes sense.
Given renewable projects typically have contracted their output via PPA and thus don’t have spot exposure, wouldn’t it be more informative to show the FCAS cost they have incurred divided by their output?
Near the beginning of this great explanation you stae “… subset of Ancillary Services supporting frequency control – keeping supply and demand in close real-time balance, reflected in system frequency staying within a narrow band around 50 Hz and returning quickly to that band if a disturbance shifts it.
Isn’t more of a phase shift correction that’s required? A frequency shift would have generators being at opposite phases for some number of cycles, thus a current short-circuit that increases and decreases as the generator phases slip through the 180 degree point — or the 120 degree point assuming this is a three-phase grid. Can transformers, generators etc. survive such frequency slippages?
Ike, I think you’re talking about “pole slip” or “loss of synchronism” events which have to protected against – very quickly – for individual generators or across interconnections. When supply and demand get out of balance for some reason the natural response of a synchronous system is to slow down or speed up through inertial effects, manifested in system frequency falling or rising (my original wording about “frequency shifts” may have been slightly rough, it’s really frequency *deviations* that FCAS respond to). Frequency control ancillary services provide controlled injections or withdrawals of power over various timeframes to arrest and correct these frequency changes. But if rates of change of frequency (RoCoF) are too high then yes that can lead to those phase angle problems you refer to which may even lead to system collapse.
Overall I think this is an excellent blog and provides some much needed balance on several issues.
However I was a little disappointed in the analysis of which energy source “saved the day”. In my view the coverage there is a little unhelpful as it merely seems to be a criticism of the various claims from “competing “teams”. But surely in analysis of an event like this it is important to determine which energy source you would like to have more of at your disposal. This for instance might allow energy planners to determine which source of energy might require more incentives than can be achieved by the current operation of the NEM.
As implied by this blog, you probably can’t determine what this energy source is by looking at the actual contributions. But it would be a useful “what if” analysis. For example at any given time, what energy source would have been called on if the load had been 10%/20% higher (or what generators might have been cut if the load had fallen by 10%/20%). It is probably that this answer would vary across the analysis, giving each of the various “camps” a win, but it is also likely that one energy source might be identified as the “preferred” across the majority of the period.
Thanks, Paul, for your excellent review of the SA islanding in February. As I understand, only a single 500 kV circuit has been restored west of Moorabool and as a consequence there are stricter export/import limits on the Heywood interconnector as pointed out by Allan O’Neill in his post yesterday. During the islanding there was a period when the normal dispatch outcomes for gas peakers appeared to be disrupted. Osborne was absent, possibly for maintenance. Barker Inlet and Ladbroke Grove also seemed be reduced in output, whereas diesel generators were contributing to greater extent than usual. Daytime energy spot prices of $300 were not uncommon in this period. This may have been a consequence of the greater FCAS requirements that were imposed in SA.
Whatever saved the day as The Advertiser reported-
“Electricity consumers will be charged a combined $93 million to cover the cost of stabilising SA’s power system while we were cut off from the national grid last month”
At last count SA had a population of 1.743 million but how that converts to the number of electricity meters and bill payers is not easy to find for some reason.
Will the forthcoming synchronous condensers soften AEMO’s islanding risk reduction requirements for SA, or reduce the cost of FCAS when interconnectors are out of service?
If the network owns the sync-cons, how do they charge or bid the services, and do they compete on equal terms with gas/batteries ?
The Electranet network appears to be only 10% Australian owned, with the rest shared between China and Malaysia. 65% of Transgrid is owned by Canada/Kuwait/Abu Dhabi. All the investment in this equipment will be recouped through the Regulated Asset Base (Electranet’s 2018-23 RAB was $2.7b without the new gear).
Past assessments have criticised network ‘gold-plating’ where RAB growth has outstripped demand growth. Demand is still not growing, but transmission is, with investment planned / progressing in several NEM locations, with the blessing of renewable lobbyists.
There is not much incentive for networks to maximise the use of existing infrastructure. The AEMC proposal to have generators reserve network capacity is a positive, ad is their decision to maintain the MLF specific to each generator (rather than averaging the losses). Building transmission purely to connect intermittent generation can only reduce network utilisation. That’s less efficient, that’s more cost than necessary.
It seems to me that we have a government guaranteed return, on privately owned monopolies, paid by consumers without a choice.
The foreign ownership of monopoly networks is worth noting, because of (a) the lack of competition and (b) the guaranteed returns.
The National Electricity Objective seems to have been forgotten.
“Will the forthcoming synchronous condensers soften AEMO’s islanding risk reduction requirements for SA, or reduce the cost of FCAS when interconnectors are out of service?
“If the network owns the sync-cons, how do they charge or bid the services, and do they compete on equal terms with gas/batteries ?”
They’re good questions and I don’t know all the answers. The syncons won’t be supplying any FCAS and therefore not needing to bid into current markets, but will indirectly influence energy and FCAS costs by reducing the incidence and impact of constraints currently used to maintain system strength and/or inertia in SA – both under normal operation and when islanded. Allan.
For once I disagree Paul. If there was an economy wide carbon price I might have more sympathy. I think you have misdiagnosed the problem. Cheap wind and solar is the appropriate source of energy but a system built around that, and that is where we are going, needs other elements, a bit more transmission and as inertia falls, a change in the control technology.
I’m not sure we’re in disagreement here, David.
1) An economy-wide carbon price would clearly assist in a number of different ways – however that’s stuck in political never-never
2) Wind and Solar is an appropriate source of an increasingly large share of energy
3) My comments are more related to the form of the incentives which have, in my view, led to a new wave of entrants who have come into the market without really gaining a real view of the risks (and opportunities) involved – and their obligations for participation (such as in this instance, as just one example).
This, unfortunately, has led to unpleasant surprises – and it is also likely to have led to a deployment pattern of wind and solar that is different than ‘optimal’, and might have actually accelerated these unpleasant surprises.
In simple terms (on the outside) it seems so much easier to understand an LGC price than the Spot Prices for Energy and 8 x FCAS (and the opportunities and risks involved), so that is where a disproportionate amount of attention has been focused.
The various regulatory bodies are also responsible for part of the problem. AEMO insists on calling emissions reduction a core function, and all their planning includes an emissions reduction focus. But AEMO’s role is not to plan emissions reduction.
It’s role is to maintain focus on the national electricity objective, which does not mention emissions. The ISP does not include a ‘do nothing’ scenario for coal plant life extension, not even as a foil for the radical ‘step change’ scenario. Investors will obviously take that as a signal that renewables have a favoured position in the NEM. But it’s clear as day that lowest cost electricity is to be found by extending the life of some coal plants and limiting the capacity of intermittent asynchronous generators, and maximising the utilisation of existing transmission.
However uncontrolled rooftop solar will be the worst market force, impacting all forms of grid generation.
Another part of the puzzle is the never ending funnel of “100% renewables” reports and academics spruiking 100% renewables with near zero engineering experience or input.
The CSIRO GenCost is hopelessly biased against the cost of coal as a fuel source, you have to wonder how that happened.
Nobody seems to mention Lazard LCOE since they made it clear they don’t account for ‘balancing cost’ or extra transmission.
I’d prefer states to take back the decision making process in the power system, I’m not convinced centralised unaccountable decision making bureaucracies are giving us the best outcomes.
Paul
You are clearly right. In my own very small way I have been writing about this on comments on Renew Economy etc. for only two to three years, clearly not as prescient as you.
Worse it won’t be fixed by new transmission. Most of the proposed new transmission projects will add to system costs rather than reduce them. For example Kerang Link has two functions,
a) to maintain a robust connection between Victoria and NSW in the face of fire and storm. On recent performance this would have had some value for about 24 hours over summer and would have saved a few million dollars of high price power
b) to improve connections to the Rhombus of Regret that would allow about 600-800 MW of power or about $1.1 bn of generating assets to connect to the load. The link itself will cost about the same again so in effect doubling the cost of energy
Paul
I suspect you are in Sydney. It was a public holiday in SA and Victoria today. So the low demand, while unusually so, is not unlike a Sunday.
Sun was up, so domestic solar was producing away, at least in Adelaide.
Peter and Ben, who paid for coal fired power plants to connect to major demand loads from the location of sizable coal deposits? Who actually built the majority of these old power plants? The state did. So it’s reasonable, and makes most commercial sense or the system, that the state centrally plan and fund REZ and new transmission for REs. Contrary to assertions it doesn’t “double the cost” of solar power. In WA Sustainable Energy Now has modeled various penetrations of RE using hourly calculations for weather dependent generation and projected demand based on hourly historical demand and the cost of transmissions to go to 85% RE or 100% RE would be around $2B amortized over 50 years or more.
At some point in time, hopefully sooner not later, societies will factor in the high cost of GHG emissions, and the longer we wait the higher that will be. US$200/tCO2 would be reasonable today, if not much higher. With BAU to 2030 there we wont even be able to put a reasonable price on GHG emissions, the Great Barrier Reef and most other coral reefs will be destroyed beyond recognition and irretrievable? People point to the billions of dollars annual income from GBR tourism and (over) fishing, but is that all it’s worth to us? Again it comes down to the difference between measuring value versus costs. When the world is at 1.5 ºC and beyond $200 a dollar for CO2 will seem like a rounding error on the global carnage and disruption to economies.
(we’re already at 1.2-1.3 ºC above pre-industrial temperatures and 1.0-1.1 above the 1850-1900 average temp IPCC most commonly makes reference to — BAU has us heading for 3-5 ºC by 2100).
You think CORVIR-19 is a big deal economically? Think again. The US$5.3 Trillion with a ‘T’ dollars gifted to the Fossil Fuel industries by governments globally (source IMF) would be much better spent on planning and building not only 100% RE grids but fuel switching our entire economy off FFs.
Better still read this report “What Lies Beneath” by Ian Dunlop and David Spratt
Unfortunately, we see some of the same misguided thinking in the comments to this article. I reiterate Ben’s point that none of the numerous 100% RE studies are of any value. None of them that I have seen paid any retail attention to the adequate supply of security services and so none of them contemplated the issues that have arisen at much lower penetration levels. This includes the ill advised AEMO exercise that the greens got them to do a while back (there were a few caveats buried in a footnote on p79 or similar…). Accordingly, these desktop exercises are of no value in assessing the actual physical outcomes.
Further, the comments about how older plant got built are fairly irrelevant – we all know there was a fundamentally different ownership and governance approach then so unless your point is that we should return to that fully centralised publicly owned model (in which case there would be no private developers to worry about) so what? In any case over 10 GW of dispatchable capacity was built post market start ( new coal, coal expansions, new gas etc.) so the market and the rules can’t have been that much of a problem for parties who knew what they were getting into.
I have some sympathy with the idea that renewables developers responding to government targets may have thought that those governments would follow through with a comprehensive plan to make their target effective, but at the end of the day if you are building a multi-million dollar asset, you do your due diligence and base your decision on the physical system and the rules as they are not as you would like them to be. Of course you can seek to change them since anyone can lob in a rule change. But don’t be surprised that if you predicate your rule change request on a plea to just have all your risk taken away and given to someone else that the AEMC will not go along with it.
As for the delay in building out transmission to potential REZs (important to remember that is what they are until all the generation turns up) that is not down to a lack of central planning and funding. The RIT-T essentially is that. But the vacuum in national carbon policy for several years has made it hard for any party to plan with confidence when and where to build billion dollar transmission links. Rightly or wrongly, the ISP has got around that by ‘deciding” when and where they should be built and the energy market bodies have since then largely got on with the job of accelerating the investment decision process while keeping some sort of oversight in place.
Finally, thanks Paul for pointing out the irrelevance of the ACT model to the NEM-scale transition. One so rarely sees this acknowledged (and certainly not by the ACT government)
25C max and sunny and 16C minm at night in Adelaide was picture perfect for solar rooftop with no aircon demand on a public holiday so you’d expect that’s as good as it gets.
With the statistics being updated and extended with the Generator Statistical Digest 2019 we have the opportunity to delve further, and AEMO’s Quarterly Insights reports are another good reference.
There are many interesting developments yet to unfold.
1. It is likely that with Queensland gas prices falling below $3.50 with possibly further to fall that Queensland’s CC gas plants will undercut older coal plants and Pelican Point will export to NSW via Victoria. Combined with overall falling demand and increasing renewables, particularly rooftop solar, coal powered demand will fall faster than anticipated which may result in medium term withdrawal of one or more coal units in NSW.
2. Work continues on new wind and solar plants and methods will be devised to gradually reduce constraints on existing farms while demand stays depressed. Q3 may well see 30% renewables and coal share falling below 60% particularly if a wet winter tops up hydro resources
3. The economics of Snowy II, always dodgy at the best of times look like a disaster now.
4. There are conflicting reports of demand for rooftop solar but a sure way to put a lot of people to work quickly after the lock-down will be to broaden support for new rooftop solar and local storage.
5. Although the stink of the Pink Batts fiasco still lingers, a wide scale home efficiency upgrade program is another quick high employment scheme that might see the light of day, further suppressing overall demand.
Terrific article Marcelle, contains a wealth of information about a set of processes that are pretty opaque from the outside – and it turns out not all that well engineered on the inside. The recommendation that bid max availability for S/S plant be treated as a cap and not totally ignored seems like a very sensible and easily implemented first step in addressing some of the problems we saw over summer with the current processes.
I wondered if you had any insights into the disconnect we saw in AWEFS forecasts feeding 5 minute and 30 minute predispatch / PDPASA processes on Friday 20 December – see https://wattclarity.com.au//articles/2019/12/out-of-the-blue-an-lor2/
the dispatch-level (5 minute) forecasts generally tracked actual output levels (but with some strange exceptions), but the 30 minute forecasts were off in la-la land for much of the critical period. Any thoughts on how this happened?
Thanks for your comments Allan. On your question about Friday 20 December and the 5 minute vs 30 minute forecasts, have a look at section 4.2 of the first document (“Australian Wind Energy Forecasting System” from 2016) at this link on AEMO’s website. https://www.aemo.com.au/energy-systems/electricity/national-electricity-market-nem/system-operations/dispatch-information/policy-and-process-documentation
While the document is a bit out of date (doesn’t include the SCADA Local Limit for example), it explains how the 5 minute and 30 minute forecasts are calculated based on different inputs (real-time SCADA for 5 min, weather forecast + availability portal entries for 30 min). If there isn’t a semi-dispatch cap on a generator the 5-minute forecast inputs include the current MW generation, which would explain why these tracked well.
If I read this correctly could it be said that the battery is detecting the phase error between its internal clock and the grid and that phase error initially is much easier to detect than frequency change. There is still a measurement delay but it is much shorter, it can be a few degrees of rotation rather than a few revolutions. In that case it could be analogous to a servo control position system which has an outer position loop (frequency) and an inner velocity loop (phase). Then there is the question of voltage which must also have risen but I suspect that detection of voltage changes is much slower than phase differences.
Whichever way you activate it it is quite important in the first 20-200 msec and would reduce the chance of overshoot in response quite significantly.
Can we assume the same applies to synthetic inertia on solar farms and wind turbines
Paul, the media need stories and “nothing much has changed really” is not a story. Reputex know this and know that if they can give the media a story they will get some free advertising of their wares. Like you I’m not going to pay for the analysis, so I’ll take it with a huge pinch of salt because I can’t sense check it myself they way I do with publicly available analysis. Before you get too frustrated with this state of affairs, please bear in mind that today’s news is tomorrow’s chip wrapper (if only metaphorically these days with most of us reading online).
Your note in passing that 31 January was probably an underlying demand record for the NEM is worthy of more analysis if you ever have the time and inclination. In any case thanks for another interesting piece.
This article seems to claim that the grid forming inverter deployed at the Dalrymple BESS are actually able to provide an instantaneous “synthetic inertia response” performing like any synchronous generators. Let me just quote two peer-reviewed results on this topic :
1)”Effects of decreasing synchronous inertia on power system dynamics” – 12 June 2019 https://onlinelibrary.wiley.com/doi/pdf/10.1002/2050-7038.12128
“One of the main questions is how efficiently synthetic inertia can substitute direct physical response of the system (synchronous inertia). Published results show that response times are in the range of 50 to 200 ms, which are very fast, but in relation to time constants of dynamic power system events, they are far from being instantaneous.”
“For power electronic connected technologies, delivering synthetic inertia and/or fast frequency control, there is a delay between the event the device’s response. The response time can be represented by a combination of the following four different parameters:
– Measurement time, which is the time needed to detect and measure the RoCoF (or frequency in case of FFC)
– Signal time, which is the time required to get the activation signal from the measurement device to the FAR device.
– Activation time, which is the time required from the FAR device to deliver the initial power response once it received the activation signal.
– Ramping time, which is the time required from the FAR to ramp up to the required active power setpoint.
However, an adequate response time depends on the system dynamics and requirements such as system inertia and RoCoF limits.”
Please can you provide the response time, to any Ro.Co.F event, of the grid forming inverter you are talking about?
Thanks for the comprehensive comment. Your question strikes at my motivation to write the article – to clearly distinguish between virtual inertia and FFR.
Due to a lack of clear definitions – many people use the terms virtual/synthetic inertia and FFR interchangeably or as one in the same. However, I describe two very different mechanisms in the article, each with their own benefits as discussed. Grouping synthetic inertia and FFR as the same service doesn’t allow people to distinguish between them and undermines their use for certain applications. This is echoed in many papers eg:
“Many studies on synthetic inertia do not focus on distinction between inertia and fast frequency response. From a system perspective, important measures of frequency stability are the minimum instantaneous frequency, and RoCoF. A relaxed view on synthetic inertia might include services which contribute to improving the response of the system, such as lifting the minimum instantaneous frequency, and reducing RoCoF. However, a more strict, unified view of this term would lead to a clearer picture of services being offered to improve system stability.”
(https://www.researchgate.net/publication/321104126_Synthetic_inertia_versus_fast_frequency_response_a_definition)
Your Paper/Point 1)
This paper seems to refer to FFR as synthetic inertia. It is clear in the article’s plot from Dalrymple that there is no 50-200msec delay they mentioned for the commencement of the response – the frequency measurement resolution in the plot is 20msec and the frequency rise and the real power response occur within one of these periods. I would expect such a delay with FFR for the reasons I discuss in the article and believe this is what they are capturing in their literature review.
That paper was submitted in 2018 and FFR has been a much more prevalent service in the market to date. I’m not aware of another grid forming inverter with Virtual Synchronous Machine software running in parallel to a large interconnected power system such as the NEM. This technology has typically been used for off-grid or smaller power system applications. Dalrymple is among the first and as such, there is still limited awareness and information on what can be achieved with this approach – hence my article.
Your Paper/Point 2)
Again, they are talking about FFR here. They say “delivering synthetic inertia and/or fast frequency control” which groups them as the same mechanism and then they discuss the same measurement and activation delays only relevant to FFR that I call out in the article. With the virtual inertia delivered at Dalrymple, it doesn’t rely on measurement to respond to RoCoF. There are two voltage sources (Dalrymple and the Grid) and whenever there is a difference in frequency between them, there is a real power transfer. Real power transfer is equal to the product of the two voltage magnitudes of the two sources divided by the coupling impedance multiplied by the sine of the phase angle difference – just like a spinning machine’s inertial contribution. (We have a paper that describes this in detail that I can potentially share with you once made public – appreciate that this comment forum may not clearly resolve all your queries.)
So the papers you reference are correct but discuss FFR – they aren’t applicable to the virtual inertia mechanism I discuss. There needs to be clearer definitions established to help understanding here and hopefully this happens soon.
To your final question, the response time in terms of the commencement of real power response to a RoCoF event is instantaneous as seen in in the article. The amount of real power, rise time and settling time are all tuneable in the Virtual Synchronous Machine software layer and are optimised for the application and point of connection through connection studies – but the start of the response is immediate.
Thanks for taking the time to comment – I hope this is useful.
Grid Forming / Grid Following are not types of inverters, they are modes of control. They can be changed/configured as requested. There are also other characteristics of being in grid forming mode here which are ignored. A hybrid model/approach is not discussed here at all. Some sudo-science wothout factual comparisons.
Dooming sells it seems but I note the negative correlation between MSM articles on global warming and coastal/esplanade RE prices so there does seem to be some considerable immunity among our movers and shakers. But with Covid19 and relevance deprivation syndrome for all other messages I guess you have to try a lot harder to rise above the milieu of bad news just at present.
Thanks for you answer Stephen, the papers I reported missed some details on clarifying differences among virtual inertia / grid forming / virtual synchronous machine, I agree on that.
I’ll try now very quickly. Virtual Inertia emulates the inertia response of a traditional synchronous machine (SM) mathematically via pulse width modulation (PWM).
Virtual Inertia based inverters and their controls strategy can be implemented in various applications such as BESS, statcom, solar plant etc.
The objective of the VSM (Virtual Synchronous Machine) is to provide grid forming employing frequency droop and Virtual Inertia by using swing equation.
The emulation of the synchronizing torque in the transient domain is still subject to a delay because of the inherently ‘time constants’ of the inverter.
Therefore it’s impossible to have the same instantaneous synchronous inertia response of a synchronous generator.
This paper should definitely clarify every doubt on this topic: https://www.mdpi.com/2076-3417/9/24/5300
About the plot and the measurement, using 20ms measurement resolution would never be enough to catch delays of 5ms or less.
Hi Paul, I was on the CSIRO team that produced input reports for their 100% renewables study in 2013. I have several of those still, so let me know if you’d like the main report or any of the input studies. Cheers, Geoff.
1) Law of large numbers? (larger population means smoother statistics)
2) Increasing geographical diversity (fewer wind farms reach high CF at the same time)
3) There are more partially completed wind farms and wind farm size is getting larger so your aggregate installed capacity value is leading the actual installed capacity by larger and larger amounts? Your graph shows aggregate capacity currently at about 7.4 GW but it is actually only about 6.5 GW as of today. There’s nearly 900 MW of capacity still under construction/commissioning: Dundonnell, Yendon, Coopers Gap, Lincoln Gap,
Mt Gellibrand, Cherry Tree, Granville Harbour, Cattle Hill, Elaine. Most of these appear to be just in the usual process of connecting WTGs are they are completed and maybe some are at “hold points”.
4) More occasions where output is constrained especially in SA.
5) More occasions where wind storms are destroying transmission lines? (or am I imagining this)
Re the reduced ACF of wind farms the relatively recent rule changes, especially for SA, requiring larger specified synchronous generators to run to supply “system strength” may also have led to less “headroom” for wind generation, hence constraints. No idea about effects in other regions though.
Good morning Paul
I am a retired operator from Gladstone P.S.
On line forced cooling has nothing to do with FD/ID fans, it is all to do with boiler pressure and steam saturation temperatures.
Fans are used after the boiler is offline and drained.
Regards
Darryl
It looks a large amount of the value here is coming from realized spot prices underperforming traded cap premiums. I wonder if it would be possible to simulate outcomes from a quarter where the opposite occurred, such as Q117?
I am looking to a time when some more old coal plants are gone and we start to depend on the wind and sun for continuous power. Then there is the issue of the low points when there is not so much wind about. The aggregate is one of the metrics and it is exciting to see the penetration, especially with the projects that are in the pipeline and the impetus will build with RE investment for recovery. Still, the average is important as well as the aggregate and the low points are important as well because they have to be bridged with batteries and pumped hydro.
Totally agree Tom – the Q1 financial results were largely driven by relatively low levels of volatility compared to what the market was expecting before the start of the quarter. We will win some and lose some over time in this regard. The more important learning was how effective the battery was at providing ‘coverage’ against the baseline of 100% with a financial cap – approaching that levels of volatility will have an impact on this too from quarter to quarter. We were heartened by these results. Although the headline figures aren’t great (mid-fifties), if we’re able to tune out the issues we had with the control algorithm in the two rouge itnervals, the figures would be not far off what you’re able to achieve with a major generation asset (and cap seller) like Wivenhoe. Understanding what % coverage is plausible will help inform how much we can rely on the ‘virtual’ service in lieu of more traditional hedging techniques going forward when trying to prudently build a risk management strategy for each quarter. Of course 5 minute settlement will be a game charger for the effectiveness of batteries in this regard… although will still be faced with the problem of optimising the charge level!
Good read and made even more interesting to me having just read the Hydro Tas articles by Cameron Potter et alia observing that forecasting charge/discarge errors were less important with longer duration storage. I am still unclear how the battery can earn arbitrage and provide the cap service at the same time. Perhaps I’m reading too late at night. A gas plant can generate for the owner’s benefit when price is above SRMC and below the cap price and still be available (assuming gas supply) to supply the cap product. If a battery does the same thing it will sometimes be discharged during the price run up towards $300 and then be unavailable when called on to defend the cap won’t it?
Thanks David. You’re definitely correct that chasing arbitrage with a battery will sometimes leave the owner exposed when needing to defend cap positions due to the relatively short duration storage capacity available. This is where the trading strategy (either manual or algorithmic) is critical to try and get the coverage % as high as possible (but also not lose what may be otherwise good arbitrage spreads). It was interesting to me that even a long duration, manually traded asset like Wivenhoe only achieved a coverage %s in the 70s. The challenges of price forecasting and flow on impact to co-optimising the different services of the battery is discussed in depth in the full report, with a few case study days used to help illustrate these and the pros and cons of different approaches of trying to address.
Thank you David for sharing this analysis.
The caps contract are a financial edge,the BESS like a peaking plant are a physical edge. A physical edge has a capital cost that I think needs to be factored in the analysis to make the correct comparison between the two edges.
My thoughts, coming from having operated several windfarms, is that you can have quite fast and dramatic wind changes associated with fronts moving through. With a front, you often get a change in wind direction as well, and I wonder if the time for the WTG to react to the changed wind direction is affecting this. The WTG take some time to react and yaw to a new heading, with some time delay to ensure the WTG aren’t constantly yawing for each minor gust. With the WTG out of alignment to the wind direction, this also has a significant negative energy output, not allowed for in the forecasts.
You do also see the alignment of fronts across multiple sites, which might account for the same affect across the many sites, including MacArthur. It is was a large front, Lake Bonney would have reduced some time before the other sites.
I enjoyed this article and the webinar, thank you.
I’m aware of the many reports raising the issues that rooftop solar is bringing to the NEM and the SWIS.
Has there been any consideration on what is the
economic limit for semi-scheduled generation? I mean how can a solar farm, like the new one being built in the Western Down of QLD make money when the spot market is heading to zero during the day?
This is partly a rhetorical question because I’m aware that the QLD government has a 10yr PPA for most of the output. It is likely above market price, if so I call it a subsidy. I also would like to know what price the PPA set for the RECs, and if that price reflects a realistic market value.
But there is no slowdown in rooftop solar, so how do private entities make money out of a solar farm?
I don’t think you need linear control of individual solar systems to get a linear response. A couple of hundred thousand individual systems with different setpoints looks pretty linear at a macro scale.
There is probably enough solar in SA to alter the market price for energy significantly. I’d expect the negative price periods would reduce if you switched off all the rooftop solar.
On the positive side of things, at least wind powered generator output from the two wind farms in QLD was at the higher end of their possible output. This really shows why QLD needs much more in the way of wind farms. Yes there are some big ones under construction, but QLD needs many times what is currently under construction.
Thankfully a bit of basic honesty here – “Home batteries still don’t pay for themselves” – that’s despite the obscenity of taxpayer subsidies, but the fundamental problem that batteries eventually die (probably within 10 years) has been ignored. So has the reality that sophisticated electronic equipment failure is very expensive to rectify. Neither battery replacement or equipment repairs/replacement will be subsidised.
People will be shocked when these realities hit home.
Many other serious flaws in this article e.g. “— then you’ll know that it’s not necessary to have any energy storage at all for renewables to supply the majority of a nation’s electricity” – not even worthy of comment.
Its time the rose-coloured glasses were discarded.
Nuclear is the best option.
I think the author misses the mark. In fact his own website carries a spreadsheet of home battery prices and I believe an examination will show that installed home battery prices haven’t fallen. I would venture a guess that the forecasts made by Morgan Stanley and many others were predicated on an assumption that residential battery prices would fall leading to greater uptake every year. Energia is another, very well respected by me, forecaster who had a similar idea.
A better question that the author might have investigated is given that both auto and utility scale installed lithium battery prices have fallen very significantly in the past four years why is it that home battery prices have been static or in some cases increased?
In my opinion the story is more interesting than the author has made it.I look forward to more articles on the topic here at Wattclarity.
Great to see forecasts being revisited and the reasons for the differences to actuals being analysed. There’s not enough of it, especially done with a sense of humour like Ronald’s. There’s no doubt that most electricity customers are economic rationalists who will only buy a battery system when the financial savings are obvious. But so many of today’s battery owners are looking for supply reliability when the grid goes down, or just the warm, fuzzy feeling of using all or most of their PV generation. The question I ponder is… how many of those people are out there, how quickly will that sub-market grow, and what could make it grow, or shrink?
The case of Callide B2 “off target” +17 MW above dispatch was due to governor frequency response with a drop to 49.64 Hz after Kogan tripped at 08:00 on 9th October 2019. Prior to the frequency disturbance it was spot on its target.
for those interested, a whitepaper on tracker strength during high wind events:
Utility-Scale Solar in Australia: New White Paper compares PV Tracker Technology and their Stability in High Wind Events
German Solar Tracker leader Ideematec presents insights into stow strategies
16 June 2020. Ideematec, a German pioneer in Solar PV trackers has released a new white paper examining the effects of stow strategy on system efficiency and reliability. The main focus was on large scale PV installations and how they hold up in high wind situations. The results showed a clear cost-saving benefit from choosing a tracker with a 0° stow strategy.
The study draws on extensive storm data from Europe’s major windstorm in 2018, giving insight into how changes in wind direction can play a part in the destructive nature of major storms. This inclusion of real-world storm data, and a detailed assessment into the frequency and severity of wind direction changes gives a far better idea of the true effects of wind damage on solar installations.
Most houses already have a built-in remote controlled switching device – the hot water service ‘ripple switch’.
Worked well for decades.
The situation with rooftop PV is worse when you consider a black start situation. Imagine having to wait until night time before restarting the grid. Ouch.
Yes until we have a level playing field with electron tenderers to the communal grid being mandated to reasonably guarantee them along with FCAS 24/7/365 or keep them this farce will continue. With up to a third of electricity used to heat water for the home it makes economic sense to use rooftop solar for heating electric mains pressure storage HWS with solar diverter controllers like so- https://www.paladinsolarcontroller.com.au/
Any surplus can then be used to RC heat and cool the home for comfort after nightfall. The unreliables problem is political not technical but it will become increasingly apparent with the demise of coal and with it off peak power pricing.
From a policy perspective, we would want to see any price or control signals from the system or grid operators being set at the consumer front gate. While solar is a challenge at present, the future challenges will resolve around EVs, storage and other in-home devices. The consumer (or their agent) should be the one who decides which appliances respond to these signals. Unless there is a technical reason, the system and grid operators should not be reaching in a operating individual consumer appliances.
Anthony, yes to your sensible points. Central control of millions of individual inverters? There has to be a better way. Some would argue the ‘better way’ was several large generators close to load centres with minimum transmission.
The dilemma is that we are achieving lower grid utilisation by subsidising rooftop PV, while at the same time building more grid by subsidising wind and solar farms, renewable energy zones and interconnectors.
I can’t see a way out of the inverter problem except retro-fits.
But properly designed tariffs could help, especially if a penalty tariff is applied to bigger systems and those on a premium feed-in-tariff.
But if owners opted for a network optimised battery, they’d avoid the penalty tariff and receive a network support tariff.
That would help all consumers by avoiding distribution upgrades.
Then EV chargers – I think you have to put them on the ripple switch anyway.
This has been a weird month for wind power. There have been some extended periods of low wind on each side of some great wind days on the 13-16th and 19th. There were also some periods of very rapidly rising and falling wind power. This must be causing some problems for grid management. Do you think it will get worse as the installed capacity increases or do you think it will settle down?
Hi Paul, regarding your question about national cabinet, the AEMA and NER vests the authority with energy ministers, rather than first ministers. So if COAG EC were to be supplanted, I presume it would have to be a sub committee of the national cabinet. I doubt first Minister’s would want to be saddled with all the responsibilities of being an energy minister (energy management, derogations etc.). Cheers, Jon
Yes, John. But rather this is yet another lesson (for those willing to learn from it – little sign yet – of how regulatory fiddling (including market-sidestepping subsidies) and trying to pursue conflicting objectives have yielded what we deserve. My late-career interests are in mathematics generally of “complex” systems (cryptically: “where life meets reality”) – this is a really beautiful demonstration.
Paul, Welcome to our world. You may do well to have a look at what is happening in Western Australia, a relatively small isolated system now operating with over 1200MW of installed behind the meter PV and no storage. A read of the recently published AEMO Statement of Opportunities (SOO) for WA delves into this in great detail. Now WA you would recall is a capacity / energy market. However we have now reached the point where WAs largest generating unit (350MW) is regularly shut down over weekends simply because it is too big with insufficient turn down to be managed within the residual demand left after Rooftop PV and the ever growing wind capacity (Driven by the tail end bubble of the RET). WA now only has about 1600MW of coal capacity in 7 units and this will shrink by 400 MW over the next 4 years as older units are rendered uneconomic – simply capital reinvestment cannot be justified with their declining capacity factors. But hey, this is what the RET was all about. So fortunately the balance of conventional generation is mainly gas turbines and open cycle at that. And with a capacity market that will fund more there is some chance of keeping the lights on. Apply the same PV / renewables outcome to the energy only market dominated by relatively inflexible coal fired units in the NEM and you are going to have a serious problem. Zero / negative pricing is simply the first symptoms. Worse still, burn the financial viability of the incumbents and expect them to come back and put capital into new (probably GT based) capacity will be an interesting ask, particularly when profit seems to be a dirty word in the electricity industry generally and the expectations of cheaper electricity in the future still seem to abound. I watch with interest. Suffice to say I have a small generator in the shed….because I don’t drink warm beer in the dark for anyone!
The back end acceleration is hardly surprising. Given the RET scheme had a drop dead date this year, and that the obligation was on retailers “to procure” and generally the big retailers are vertically integrated, bringing on new generation at (say) $60-70 / MWhr LRMC that simply canibalises their existing fleet (able to trade at SRMC) why wouldn’t you kick the can down the road as long as you can? And over the years of the RET scheme the technology got better and cheaper.
Many of the really good sites (best wind resource and easiest network access) were probably snapped up first in the process (read South Australia) so lower capacity factors in latter projects may well be expected. Allowing for some seasonal (and maybe annual) variability in wind and it may explain much of it. Might be worth looking again with a couple of years of clean data given many of the projects need to be delivered by years end for RET purposes.
If you take ideas to their logical conclusion, and ignore the reasons for our current predicament (ie perfect hindsight), then all generators over a minimum capacity (30MW is convenient) should be scheduled.
Nobody starting a grid from scratch would allow such a high proportion of unscheduled generation capacity on the grid.
But it’s not too late. If all generators above 30MW had to operate as scheduled generators – either by contract with another generator, or by adding technology), the market matching supply to demand would function correctly. Also reduce the LGC price by 20% per year for 5yrs, then you return to an equitable market.
Also, if scheduled generators were proportionately penalised for not meeting dispatch targets, reliability would increase.
And if the shortfall penalty was distributed proportionately among the generators that made up the shortfall, the cost of making up generation shortfalls would fall less on consumers.
I can’t help but think that uncontrolled rooftop PV is by far the biggest risk to the NEM. And always increasing risk at that.
I’d love to see more about the wholesale price volatility, particularly around the less and less offers in the 0-300 range, and which price ranges are having the most effect on setting the spot price.
For example, spot price is currently low, but if the proportions of price offers keep changing, will the prices stay low?
Am I right in thinking that volume weighted prices are (a)one more degree of difficulty to compute and (b)more relevant to the business outcomes?
I’m thinking that the generally lower prices for the generally lower load periods (ie night, etc) tend to skew the time weighted prices lower than would volume weighted prices.
Great article. A question about your curtailment calculation, where you estimated that 1.7% of generation might be curtailed under the rule change. Was this calculated using 4s data, or just the 5-minute initial_MW and total_Cleared data? I used 4s data for several sites for a couple of different months and obtained a significantly higher curtailment estimate.
Thanks David. I used the InitialMW data shifted 5 minutes to line up with the corresponding TotalCleared (dispatch level) value. I approximated (as shown in the diagram) the curtailment as a triangle – i.e. (FinalMW – TotalCleared) / 2 * (1/12). It was far too much work to look at the 4 second data!
Thanks for the quick response. Agree it’s a lot of work, hence I’ve only done a handful of generators and a couple of months. But output can vary greatly within a 5 minute period, often ramping above the target. So I’m concerned your curtailment estimate using 5 minute data is considerably below the true value if generation has to remain below the target at any given moment. My estimates are approx 3% for wind and 5% for solar.
The only opportunity for wind and solar to put downward pressure on electricity cost is on wholesale prices during favourable weather.
Wind and solar increase the cost of all other components of the electricity chain.
The correct definition of ‘the problem’, is that there is no market for intermittent electricity. The market matches supply to demand. It does not match demand to weather.
When you phrase ‘the problem’ in these terms, it is clear that all significant generators participating in the market should be able to perform that fundamental function.
Great article Marcelle – especially the analysis refuting the AERs (or COAGs) conclusion of the SA output more than 20 MW during negative prices graph was due to deliberate of generation (it was so dodgy when I read the AER paper that instantly dismissed it but you provided some great data driven reasons), and the reference to the NZ market solution that is a possible pragmatic solution.
Sorry, but I can’t really see the problem – surely the market is simply saying causer pays. You cannot keep putting generation into a power system that cannot be relied upon to “do what it says it is going to do” once it becomes a significant share of the market. And certainly not when ancillary service markets (including LFAS) are simply not reflective of the cost of providing the service. How about instead of looking for free rides all the time that developers fix their limitations by either contracting with others to correct their market position for them, install storage to self regulate their position across trading intervals or even auxiliary generation to clean up after themselves. Yep it is going to cost some money – perhaps the immature conversation around LCOE being the sole comparative metric in the generation space might be about to change – not before time.
Since SMEs are the biggest consumer of NEM supplied electricity is it possible to post the Victorian SME map for June 2020 v 2019 and Apr-June 2020 v 2019.
Although total Victorian consumption is only down by 2 percent think its useful to take into consideration that 4 of the AER’s 10 Expectations announced in late March 2020 will expire on 31st July 2020. In the absence of a new AER Statement of Expectations the expiration of the existing AER expectations is likely to significantly impact on consumption.
eg 4 of the existing AER expectations are:
– Offer all residential and small business customers, including small businesses eligible for the JobKeeper Payment, who indicate they may be in financial stress a payment plan or hardship arrangement, regardless of whether the customer meets the ‘usual’ criteria for that assistance.
– Do not disconnect any residential or small business customers who may be in financial stress (including small businesses eligible for the JobKeeper Payment), without their agreement, before 31 July 2020 and potentially beyond.
– Do not disconnect any large business customer, including businesses eligible for the JobKeeper Payment, without their agreement, before 31 July 2020, and potentially beyond, if that customer is on-selling energy to residential or small business customers (for example, in residential parks or retirement villages).
– Defer referrals of customers to debt collection agencies for recovery actions, or credit default listing until at least 31 July 2020.
The graphs in section A and B showing the increasing volume of $0/MWh bids is very interesting.
Would you consider doing the same trend separately for day-time periods and night-time periods to more directly observe the influence of increasing rooftop PV on bidding behaviour?
Very interesting article Warwick. However, I have to take your “backcasting” of battery revenues with a pinch of salt. Firstly, on a global level the AEMC’s analysis showed that the difference in historical market revenues/costs under 5 or 30 minute settlement was negligible. in other words, 5 minute settlement is only worthwhile if it changes bidding behaviour as compared to 30 minute settlement. in which case you can’t predict future outcomes from historical bidding patterns.
Setting aside the question of the extent to which bidding dynamics change as a direct consequence of 5MS, numerous other factors have changed over the last 8 years such that the past may not be a good guide to the future. Legislation and rule changes have strengthened the regulator’s hand against bidding that is not in “good faith”, which in itself should reduce any gaming. market conditions are also eroding the market power that has been imputed to certain generator portfolios. It’s no surprise that the biggest gains in your analysis are in SA and QLD, given the main allegations of market power abuse have been against AGL in SA, where its TIPS A and B are a significant share of the total capacity and against government owned generators in QLD. In the former case, TIPS A is on its way out (although AGL has a smaller faster plant Barker Inlet that has recently fired up) but more importantly, there are already 3 large batteries with more coming and soon a new interconnector with NSW. In QLD the main change is that the government has instructed its generators to bid differently, which admittedly could easily unwind. For a range of reasons and because 5 minute settlement will remove the incentives for late rebidding in any case, the p1 or p6 spikes you refer to seem less likely to occur.
Even when they do, the arbitrage opportunity will be eroded by new entrants. I don’t know what the equilibrium is, or how fast it will change as the generation mix changes, but there may not be that much new battery capacity needed to make the next battery uneconomic. It didn’t take much new capacity overall to make SA’s FCAS markets competitive again (at least when the state’s not islanded…) after Northern’s exit.
We can also logically expect an increase in DR. how much of that will be due to 5 minute settlement and how much to the new wholesale DR mechanism may never be clear.
Finally, the wholesale market works well when both buyers and sellers have an incentive to hedge out volatility. it may not be a great outcome for the market to have lots of batteries happy to play spot price arbitrage rather than lock in a revenue stream by selling caps. logically this leaves some load on the buy side exposed to those price spikes…
Two of the best traders in the NEM are Harry Hindsight and Felicity Foresight, who can always see the future. So, yes, the backcasting is not a reflection of reality. I don’t believe this is too much of problem in NSW, VIC and QLD where the price spikes likely occurred much of the time at times of peak demand, so whilst you’re uncertain of the pricing in predispatch forecasts, you still likely know when the maximal prices occur. SA is different as price spikes and dips are less predictable, so you are less likely to take full advantage of these. These figures are imperfect but present a rough guide to potential values. You will probably underperform against these, but it’s plausible that you could outperform if you judiciously did some extra charge/discharge cycles.
It’s not surprising that adding the revenues/costs under both 5 and 30-minute settlement makes little difference to the whole market value as the usual change from one dispatch interval to the next is rarely more than 10’s of MW’s. This makes little difference to stable loads or generators where time weighted and volume weighted prices are almost the same. Those however at the margin, who operate for just a short while will be impacted the most such as high power batteries.
I would also agree that much has changed over the last 8 years, including the fact that all the history represents bidding under half hour settlement and for much of that time, no batteries participated in the market. Other factors such as “good faith bidding” seem to have crunched QLD arbitrage values in recent years. Changes in the market cap, value in FCAS and other issues also have an impact.
Predicting the future arbitrage is challenging. It’s difficult enough to forecast future electricity prices. I agree with your suggestion that P1 and P6 spikes will not be as prevalent as we should expect price spikes to be evenly distributed. Arbitrage is not unlike FCAS markets in that in probably does not need more than a few 100MW to begin to erode the market value.
What I’m trying to achieve is to highlight the difference than 5MS can make to batteries and some insight into how different regions and configurations impact the potential return.
Thanks for your reply Warwick. I agree the debate about the scale of the arbitrage opportunity doesn’t undermine your basic premise which I think is largely correct. It will be interesting to see the rate of battery deployment over the next few years.
In the meantime I hope your friends Harry and felicity are working in AEMO’s ISP team now…
Thanks Warwick for the interesting perspective and backcasting calculation.
First I have to say that I am not very neutral in the discussion since I work in Business Development for BESS at Wartsila 🙂
We did as well some backcasting simulations over the year 2019 for a larger BESS (based in NSW) and considering the current 30min settlement, and landed on an optimal battery duration between 2C and 1C. Our simulation was considering a merchant application FCAS+Arbitrage, and the purpose was to check what BESS size would have the highest NPV.
I do agree with you below conclusion that the 5MS will further enhance the value of power vs energy.
“One of the key beneficiaries of 5-minute settlement will be high power, short duration batteries”
One question, not related to BESS ; The market price cap (maximum) has been recently updated to $15,000/MWh. Why has the minimum price cap not been updated ?
I think it is challenging to make the business case for batteries in excess of 2 hours given the pricing I’ve seen unless there are other value streams outside of FCAS and arbitrage requiring a longer duration. On the other hand, whilst your return per MWh is highest for high C-rating batteries, the installed cost per MWh is also higher, so the best NPV is getting the balance right in terms of installed cost and expected return.
I’m not sure why the market floor is still only -$1,000/MWh. I believe a lower price has merit. It’s also interesting to note that whilst some might observe that negative prices help batteries, which is true, the value is still in discharging for a short time for high prices i.e. you’d need 7.5 hours at -$1000/MWh to equal the value of a single half-hour discharging at $15,000 and having a 7 hour battery would have a woeful return compared against a high C-battery.
As usual fantastic analysis. I think the point about seasonality is going to be writ very large indeed across the NEM and will be the next shoe to drop. Solaredge inverters are just starting to make their presence felt in australia. I wish I could restrict our household consumption to ~5 MWh. Sometimes I feel like a small industrial company.
Just solely looking at the electricity savings, don’t you think it would be more realistic to assess the payback separately? i.e. you’ve estimated a payback of 7.2 years for only the solar i.e. $8,000/$1118 = 7.2 years and the total system cost/ savings = $18,000/$1438 = 12.5 years, so then the payback on the battery is really net battery cost/annual savings as compared to solar so…$10,000/ $320 = 31.3 years.
This obviously places no value on the potential for UPS or the ability to use your own energy but these items will tend to have a different value for each individual.
Without knowing the exact layout of your house and the regulations in Queensland
1. Would it not be possible to add another 1-1.5 kW of solar panels preferably oriented in a different direction which would be clipped when total output reaches the 5 kW limit of the inverter but would make a useful contribution in winter. It would seem to me that that would do more for your self sufficiency than more battery capacity
2. It must be possible to develop a smarter system to arbitrate between the battery and the HWS, either by allocating alternating 5-10 minute blocks or putting a PWM controller on the HWS so you can regulate the power flowing to it so that depending on the hot water temperature/battery state of charge you can vary the rate of battery charging vs hot water heating. I understand that the closed nature of the software on the various bits and pieces you have now makes this difficult, but it is probably easier than getting the regulatory changes you desire
@David – thanks for your kind words. I was well aware conceptually about impacts of seasonality, but surprised by the magnitude of it in reality once looking at the data.
@Warwick – agree that’s correct when considering the ROI of the battery in isolation. As someone who didn’t already have PV though, I was more interested to compare the two options for the overall ‘system’ as a whole. The PV element certainly helps subsidise the battery part when looking through that lens, but nonetheless is able to provide an overall system that helps achieve my other goals within a not entirely unreasonable payback. As discussed in the article, quite a few things that need to change to make home storage more attractive to more people. Further, first thing I’d advocate to those who already have solar (or perhaps don’t want to spend as much upfront) is to put a hot water controller in – you’ll end up with the bulk of the benefit of a battery with only a fraction of the cost.
@Peter – agree complete about more DC capacity being the solution to help address the issues in winter. Another reader over on LinkedIn kindly pointed me to a document from SolarEdge about the ability to oversize their HD-Wave inverters by 155%, which would give me up to 7.75kWp. I was previously under the impression the limit was 130%. I’m going to look into this further. And to the second part of your comment, whilst that would help ensure maximum bang for back for each interval (if I had spot exposure, which I currently don’t), it doesn’t address primary limitation, which is to be able to discharge and export energy stored in battery that is surplus to be daily needs – eg. if I need 6 kWh on average overnight each day, I’d like to be able to export say 4 KWh into the grid on top of this.
I did the same thing in 2014, but I purchased a hybrid car. It is interesting to see a Queensland report the results of which are roughly what I would expect. Basically, solar PV saves much more than batteries, because the energy saved by a battery is mostly night-time energy.
Grid savings from the 3 kW solar PV system balanced car charging costs, and in the first year compared with the previous year and car, I avoided purchasing 1000 L of petrol, saving $1,300.
Would you consider optimising the charge cycle to better support the grid? Eg charge during the mid day high solar period, and discharge during the peak periods? There are tariff considerations clearly.
The hot water heating period looks good from the network perspective. Discharging the battery at the same time seems obvious from your economic perspective.
Would network complimentary tariffs be something you would take on if the opportunity was presented?
Great article, thanks Andrew. Fantastic analysis and I couldn’t agree more with your clear and logical final thoughts. A few points:
1. Looks like throttling the hot water system in winter to around 2.5 kW and spreading the peak could give you a small quick win if that’s possible, although maybe better to just figure out how to get the controllers to play nice with each other.
2. I suspect the lower-than-advertised battery efficiency just comes down to the old saying “your mileage may vary” (almost literally). Battery efficiency is complex and depends many factors like temperature, charge and discharge rates, depth of discharge, cell degradation, etc. I’m not surprised it’s a few percent lower in your installation than in Tesla’s marketing material, which is probably based on a standardised, more favourable application.
3. Agree with Warwick’s comment that the financial case for the inclusion of the battery doesn’t come out looking great compared with solar alone, and this might be useful analysis to add to the article. However, this is a great demonstration for energy and sustainability geeks of how solar+battery stacks up for energy self-sufficiency in 2020, for those of us who like the story beyond the hard financial numbers. Not to mention that without any changes to technology/costs, the sensible rule/tarriff changes you have proposed would significantly improve the case for adding a battery (and worsen the case for solar alone).
@Ben – this is controllable via the Tesla app, but relies on timers as opposed to the mode I currently have it in, which dynamically adjusts charge rate to only use surplus solar. On balance I prefer the ‘solar self consumption mode’ ion the absence of any exposure to spot price etc. I’m certainly open to any kind of tariff structure that would be a bit ‘smarter’ than the typical offerings out there, although note in a perfect world that if I was able to achieve a higher self-sufficiency % the $/kWh tariff becomes largely irrelevant, but $/day fixed charges do continue to matter a lot.
@Jonathan – thanks for your kind words. Your suggestion about the 2.5kW set off a lightbulb for me – thanks! I realised that the SolarEdge Immersion Controller lets me easily adjust the kW setpoint for the HW system remotely via the app. I’ve now changed it to 2.5kW but starting an hour earlier in the day. Will play with this over the next couple of weeks and look to adjust dynamically as the season changes.
Re: round trip efficiency – our report into the 1MW/2MWh Powerpack system also found lower than advertised round-trip efficiency (which is measured under STC, which are less onerous than the Australian climate!): https://sustainability.uq.edu.au/files/11868/EPBQtyRptq12020.pdf
I’ve had the feedback from a few people that the way I presented the payback for the solar + BESS option was a little misleading – this wasn’t my intention, as to me (who had nothing to start with) the blended payback for either solar alone or solar + battery was the choice I had. I agree batteries in isolation are substantially less compelling, and hence my suggestions at the end. I’ve updated the article where I first posted it (on LinkedIn) to better represent this.
Snowtown 1 Wind Farm and the Hallet Wind Farms are Suzlon S88’s, which are older technology turbines. It’s likely they are more sensitive regarding high wind cut out. This would explain the difference between Snowtown 1 and Snowtown North/South which are co-located, but different turbine technologies.
I just checked my own assumptions, and looks like AEMO rolled back the “Mandatory” HWCO back to “Optional” in the Final Determination. A shame as it’s a simple signal to implement!
Meanwhile wind production the next day (24 August) is/was abysmal with South Australia producing just 21MW of wind power for the morning peak (6:45am)
This is the problem with renewables, they are unreliable and any claims of peaks and records needs to be considered with the days prior or after the peak which were likely below or well below average.
If you like to have hot coffee in the morning and a hot dinner at night you had better be thankful that there is enough conventional power available because your coffee and your dinner would be cold if we had to depend on the wind and the sun to heat them.
Four factors make it hard for RE to work in Australia. First the frequent and prolonged wind droughts (see June for example) and also WA, SA and Victoria today. Second is the “choke point factor’ that means the sustainability of RE is limited by the lowest points of supply, not the installed capacity, the high points or the average (think about our need for a constant supply of air to our lungs and what happened when we are choking or drowning). Third is the island effect, we have no neighbours to help out if we are short, unlike most places in the developed world. California has rolling blackouts but it will not go completely black because it is not an island. Finally, there is no grid scale storage in sight for the foreseeable future. Please don’t refer to the Tesla batteries as storage at grid scale. And pumped hydro? Believer it when you see it:)
I guess that no matter how much wind and solar are installed, if there is a shortage of dispatchable generation, prices will spike along with risk. That’s the point of the market – to create price signals.
I’m not convinced the price signals are adequate to see dispatchable investment. Is that a reflection on the market? Well… it’s not much of a market that matches supply with weather, instead of supply to demand, by overtly favouring semi-scheduled generators..
What’s the solution?
Not more wind and solar, that’s clear.
Batteries? Handy but expensive, a net load, and can’t do it all.
Snowy 2? For a while maybe, then another big coal plant closes and we get a drought.
Interconnection? That’s been shown on this site to be a flawed hope.
Demand management? I think so. Firstly for demand peaks, eventually for low wind and clouds.
It will probably force wealthier people to invest in home batteries and generators, in order to ride out the rolling blackouts (sorry demand management).
It’s a terrible situation that’s coming, and more people in the industry need to speak up.
It’s almost like all the claims of the ability to have 50% or greater generation from renewables were lies and aren’t possible in the real world.
Meanwhile, no one in their right mind is going to build synchronous generation when the government is literally funding renewables to compete with you.
It’s almost like you didn’t read the AEMO report in question at all, or the AEMO RIS, which spells out exactly how AEMO plans to manage the system above 50% renewables. AEMO’s butt is on the line with system security and reliability and accordingly it manages the system and forecasts the future conservatively – do you really think they’re lying?
And you’re right, no one in their right mind is going to build syncronous generation, they’re going to build batteries to supply FFR, because that’s exactly what AEMO is saying is needed (in the specific situation being discussed here, which is securely managing SA as an island).
Batteries don’t provide inertia. The only way to keep the lights on is to have an oversupply of fossil fuel generation but AEMO won’t say that because they are captured by green zealots.
The way to manage an islanded South Australia is to rebuild the Port Augusta coal plant.
Scheduled demand in SA was also well down, 420MW in the 13:55 interval, which I believe is a new minimum demand (since 1999, excluding 2016 system black). Could be interesting times ahead in the next month or so as rooftop solar generation starts ramping up even further, but before the normal summer higher demand from daytime air conditioning starts taking effect.
Can you expand on your comment:
“That’s not to dismiss the financial pain felt by a broad section of the South Australian grid were burnt by high FCAS costs which added a very unpleasant surprise to (including some providers of FCAS services via Demand Response aggregators, we hear).”
Is this referring to this high FCAS lower costs borne by customers in SA, or something else that occurred?
Hi Paul
Was wondering if this was a minimum record for coal generation (black + brown)? even if not quite since market start, at least minimum for the last 10 years, say?
Minimum NEM Coal since 01 Nov 2010 was 10,665 MW on Sun 22 Sep 2019 at 12:30 hrs. Last Sat 29 Aug 2020 minimum was 11,062 MW at 13:30 hrs. Another notable minimum was 11,098 MW on Sun 25 May 2014 at 04:05 hrs.
So my understanding is for both regulation FCAS and contigency FCAS, we sell the generation ability, not the real generation. Am I right?
And the FCAS dispatch interval is 5 minutes right? For each 5 minutes interval,regulation fcas and 6s/60s/5minutes contigency FCAS just account once, calculate the revenue and sum them up.
If I am right, what confuse me is that the price difference between regulation FCAS and 5 minutes FCAS is large.These value should be close?
Correct:
1) FCAS is paid for ‘Enablement’ rather than for actual use.
2) Payments made for each discrete 5 minute period
With respect to your question about the difference in price between Regulation FCAS and Contingency 5-min, the answer ultimately comes from the fact that they are different commodities, with different suppliers, and different size of requirement.
There is one subtlety to the Regulation and 5 Minute Contingency services that shaojing may have been thinking of:
Regulation capability can be used to deliver 5 Minute contingency response, as it is a slow enough service to be dispatched via AGC (unlike 6 second and 60 second Contingency).
Also note that Regulation and 5 Minute Contingency services are not mutually exclusive – a unit on Regulation duty is also able to provide 5 Minute contingency in the same dispatch interval.
AEMO therefore cooptimises delivery of the two services by including Regulation offers in the constraint formulation for 5 Minute Contingency (but not the other way round, as 5 Minute contingency can be offered by generators or loads which can’t provide Regulation – eg loads which offer only Contingency Raise services, or very fast start generators like hydro units which may be off and thus unable to provide Regulation at the start of the relevant dispatch interval, but can start up rapidly to deliver 5 Minute Raise).
Typically, more 5 Minute contingency capability needs to be enabled than the required volume for the Regulation service. This can sometimes mean that the amount of Regulation capability enabled to satisfy the 5 Minute contingency requirement *exceeds* the total amount required for the Regulation service itself – this depends on the volumes and relative prices of Regulation vs 5 Minute Contingency offers. It doesn’t happen that often – perhaps in a few percent of dispatch intervals.
In these circumstances the prices for Regulation and 5 Minute contingency FCAS will be equal, as it is effectively the 5 Minute contingency market that is determining the value of Regulation offers.
In any case, the Regulation FCAS price will always be higher than or equal to the price for 5 Minute contingency, never lower. You can think of this as reflecting the Regulation service being inherently more valuable than the 5 Minute service because the former can substitute for the latter but not vice versa.
Just notice. really thanks for your help, sorry for bothering you agian.
just a little confused about some other questions.
1.Whether the 3 kinds of contigency raise FCAS are mutually exclusive? For example,I have a 100MW BESS, can I take part in 100mw 6s contigency raise, 100mw 30s contigency raise and 100mw 5min contigency raise?
2.If we can, when emergency happens, we just need supply 100mw, right? Also if we find we can not supply energy?(maybe caused by fault) Can we withdraw the bidding or can we withdraw the FCAS service even though we win the bid?
I think Jonathon has already answered your question 1 below.
Re 2. the response provided to a contingency should be proportional to the frequency deviation. This is explained in AEMO’s Market Ancillary Services Specification documentation. Regarding inability to respond due to a fault, if known about in advance, offers can be withdrawn at any time up to about 90 seconds before commencement of every 5 minute dispatch interval. Once accepted for enablement, offers can’t be withdrawn within a dispatch interval but can be withdrawn or modified for succeeding intervals. “Acceptance” of FCAS and energy offers is not forward-looking beyond the horizon of each 5 minute interval.
I’m interested in the implications for the contingency plans and constraints that might be needed to protect against an interconnector trip during high rooftop PV output.
With only 500 MW demand, and the NSW-SA interconnector rated at 800 MW, that’s a big frequency whack.
Big risk for rooftop solar trips, big risk that existing inertia won’t be enough to catch it.
It is a game of chicken indeed when you look through the other end of the telescope at dinnertime and see the contribution from wind power when there is little or no solar input. The game gets serious during the frequent and prolonged “wind droughts” when the wind farms are producing little or no power, often under 10% of plated capacity and occasionally much lower. Have a look at June and July. It is necessary to take account of those drought periods and also the “choke point” factor, meaning that the grid needs continuous input in the same way that our lungs need a continuous supply of air and we can choke to death very quickly if the supply is arrested by choking or drowning. As long as we have 100% of demand available from conventional sources the grid will not choke but when we lose conventional power capacity then the wind becomes essential and the sustainability of the system will be limited by the lowest level of wind supply, not the average or the plated capacity.
Looking at the NOS the Tarrone 500/132kV transformer has been offline for a week or two – not sure why there is no associated constraint for MacArthur output
WRT responding to ‘responding to big changes in the national electricity market’ I do hope they take this on board as one way to ease the political fallout/pushback with the need to manage the problem of rooftop solar- https://www.solarquotes.com.au/blog/divert-solar-air-conditioning/
Of particular relevance-
‘..the timing is important as just a couple of years ago, feed-in tariffs were high and curtailment unheard of, so there was no real need to divert power to local loads. But so much household PV is now live (and more is coming), the situation has changed.
Smith said until recently, users have focused on “how much we made from export”.
“If you’re exporting, you’ve got it wrong”, he stated. “The conversation should now be how much I can consume of my own power because I’m managing it better.” ‘
An interesting statistic is to look at the capacity factor for the sector. It is one thing to have output reduced by the retirement of a facility (eg Hazelwood) but as players continue to try and compete for an ever decreasing energy “pie” then the financial viability of plants comes into question. The first thing that gets crossed off is reinvestment in reliability and sustaining capital. Nasty little cycle now starts because the plants become less reliable and hence have lower capacity factors and around we go again. The serious issue on the east coast is the lack of mid merit / peaking plants to fill in as coal falls by the wayside and rooftop PV and wind takes the daytime energy. And I’m still looking for the commercial drivers to get investment in same in an energy only market. All is well until a hot and still evening arrives. AEMO are flagging this – it does happen – look at California.
Thanks Paul for answering my question! I note that last weekend got down to a minimum of 10871 based on the dispatch target metric, so still not quite record breaking but we have a few more spring weekends to go…
To Andy’s point, i understand that in Europe (e.g. UK, Germany) some older coal plants have been configured to run more mid-merit (carbon pricing and lower fuel price differentials made some of them less competitive v CCGTs), i.e. better able to start up, ramp up at a reasonable rate and then ramp down. It seems to be an article of faith that this is not feasible in Australia, but it’s not clear why not – different turbines, possibly? Or is it just economics – there would be fewer hours in a typical year in which to earn one’s money back?
Kieran, Coal plant response will never get close to an OCGT. Once synchronised and part loaded they can get close to an CCGT in terms of loading rates (they will usually be also limited by their steam cycle). You are right, much can (and is being) done mainly around I & C to improve start up and shut down times, however at the end of the day big, thick walled steel components have limits of heating and cooling rates and hence consumption of fatigue life. Given many are already older plants with somewhat imperfect history of life already consumed (remember,many did not have control systems that modelled life consumption in the 1980’s and before when they were built) so it can be a bit of a raffle. New I & C lets advanced materials modelling be put in to manage life consumption. It gets very expensive when you find a big crack in something at a subsequent overhaul (that you might have to schedule just so you can look – this in itself is expensive). As I say a viscous little circle and it is all economics if the plant cannot earn the revenue to wash its face in an energy only market. Capacity markets can help but the market purists hate these. Not withstanding Govt policy on carbon, I don’t think there are any of the original 4 x 500MW stations left in the UK now. Simply couldn’t get a run. Even Drax (6 x 600MW) survives only as an imported biomass burner. Now there are CCGT plants having trouble because of increasing gas prices.
Given the current amount of storage that is feasible and affordable the monthly balance may be too coarse-grained to show the full extent of the problem. Looking at daily figures there are extended periods in June and July, up to 20 and 30 hours when the wind fleet is delivering less than 10% of installed capacity, down to 2 and 3% for shorter periods. This would not be such an issue if we could run extension cords to neighbours. For instance California has rolling blackouts but they won’t go completely black. In the Californian situation with the conventional power sources run down before the RE replacement is adequate, we would be in real trouble. This looks like a possibility post-Liddell.
Don’t forget that these are record breaking heatwave conditions in California. If Victoria & SA had record breaking heatwave conditions there would be power shortages as happened in the recent past. Even then the weather conditions that caused the rolling black outs in Melbourne were nothing like record breaking. Victoria hasn’t added enough generation and more particularly storage plus demand management to cope with a heatwave even close to record conditions.
Liddell closure is going to be a challenge. NSW is running too late with their new VRE generation plus storage and transmission. The question is how lucky NSW is going to be with weather in the first few years after the closure of Liddell.
My [rejected] comment was obviously an unacceptable comment on the ‘slow decline of coal’….no doubt the foreign-made juggernaut and those making their fortunes out of it just can’t cope with the milch cow consumers supposedly driving the ‘transition’…being permitted to have any chance to have their perspective on it published.
Er, ‘metro70’, I appreciate your perspective – however let’s not try to ascribe motivation via conspiracy theory (e.g. ‘just can’t cope with…’) when the facts are simpler.
There is an automated spam filter used on this site to make the site usable, and it appears to have deemed your comment (a long line of text with many capitalised words and other elements that increase the risk it is spam) as such.
Things look even worse when there is finally the recognition that Victoria can’t hide its high gas consumption for residential HWS and space heating. When this energy use is shifted to electricity, even mainly based on heat pump technology, it creates further problems in the winter months.
Of course factor in distribution of power generation assets during the periods of lack of supply and storage, then it looks worse again. The “excess” solar available in QLD in winter can’t all fit through the transmission lines going south to Victoria.
Hydrogen doesn’t greatly help with the seasonality as winter deficit is going to be too great to even use hydrogen with storage at an affordable level.
A little confused about a question.
Whether a 100mw/100mwh storage system can take part in all 8 100mw FCAS(100mw reg raise/100mw reg low/100mw 6s contigency raise……). Or for raise/low, the sum of total is 100MW(for example 50MW reg raise/ 30mw 6s/10mw 60s/10mw 300s)?
Hi Shaojing,
It can be a little confusing but most batteries can take part in all services, even all at the same time, depending on the technology being used. You are right that it is the summation of energy targets, regulation enablements and max enablement of all FCAS services must be less than or equal to the max capacity – its the joint ramping constraint.
Paul Mcardle has my details if you want more info.
Curious as to what data you’re basing your inference that QLD wind appears anti-correlated, Brad?
In our analysis of over 11 years of weather history for the Generator Report Card 2018 using wind speed readings across all Renewable Energy Zones, we found that (statistically) the correlation co-efficient was close to 0 … i.e. random, and not true anti-correlation.
Good article. I just had a SA device/gadget installed yesterday, so I guess I’m in “time period 1”. Like you, I have the benefit of a high Net Feed-In Tarrif for another 11 months so my goal or value drivers are diff. to those that don’t have this. Not sure if the SA site deals with this yet. I also have a curiosity for more data, but need to temper this appetite with the understanding that more data doesn’t always lead to more knowledge.
Unfortunately this raw annalysis does not explain the uderlying reasons for lack of generation e.g. low wind or cloud or was it that the full capacity of these generators was not commited and why steps were not taken to avoid the problem.
Alexander, you might be misinterpreting what these case studies are about.
These are not so much focused on lack of generation, but on the fact that there is an underperformance compared to what AEMO was expecting 5 minutes before in the dispatch target.
Whilst some instances might be due to challenges with fast-moving cloud, or fast fluctuations in wind, there are other factors as well – as the case studies are progressively revealing.
Paul, is the discrepancy in the second last image perhaps a result of Wivenhoe pumping demand and it being reflected differently in the different values AEMO publishes? I say this because I have noticed before that the impact of this load does not show up in the demand values shown in the popular ‘Pocket NEM’ app, but does if viewing the AEMO NEM web dashboard (ie. the AEMO dashboard demand can often by ~250MW higher than the values in Pocket NEM’. My guess being that Pocket NEM uses a different data set for demand than AEMO’s dashboard does, which them each treating the Wivenhoe load separately.
Interesting to see at dinnertime with solar off duty the wind was down to about 5% of the power supply in the NEM and less in WA. Surely sustainability of RE depends on the minimum level of supply, rather than the installed capacity, the high points or the average! Does that mean that until the whole of the demand can be satisfied by RE at the lowest level of sun and wind supply we will have to maintain 100% conventional power capacity?
Until the rules are corrected there are going to be problems with the balance in the NEM. The delay to the 5 minute rule and the lack of additional rules to reflect the valuable services that batteries can provide has compounded this problem. If these were in place we would be seeing the installation of GWh scale batteries on the NEM.
As a result we have coal fired units operating in QLD and Victoria, which otherwise would be shut down for weeks at a time during these periods of low grid demand as they would be uneconomic. Of course put a price on the pollution the fossil fuel power plants produce and the transition would be even faster.
Demand management of easy high load items like hot water services would make a difference to the balance of the supply and demand, but HWS are boring so nothing has happened.
I was under the impression that Queensland distributors use ripple control to switch in hot water during such low demand periods. Another low-hanging fruit for demand management would be eutectic storage elements in commercial refrigeration (office towers, shopping centres and the like).
At least this Sunday the QLD – NSW connection was able to operate at close to maximum capacity. Other weekends we have seen up to 800MW of RE being curtailed.
If the QLD Government was serious about green house gas reductions, then they would shut down one of their coal fired units during these months of low demand. Doesn’t make a great press conference, so it doesn’t happen. Might up set their union mates as well.
The QNI Interconnector is undergoing a minor upgrade that will continue through to the end of 2021 (according to recent TransGrid tweet). To facilitate the work, the 88 Liddell to Muswellbrook 330 kV Line has been taken out-of-service on a daily (Mon to Sat) and more recently, a 5/6 days continuous basis. This outage results in a lowering of the Qld to NSW export capability by 700-800 MW. These outages for upgrade work and the loss of export capability from Qld (and increase in RE curtailment) is giving us a glimpse into the future with the unstoppable increase in Variable Renewable Energy Sources in Qld. So much generation with no where to go!!
A couple of questions Craig, given that the sun is gone by dinnertime, certainly in winter, what keeps the lights on when the wind is low? Last night the wind contribution was down to 10% of demand across the NEM. Currently the wind supply is down to 5% of total demand across the NEM, zero in Victoria, 1% in NSW and Queensland and 3% in WA.
Second, how many batteries do you envisage to provide for these gaps in wind supply, given that the original big battery at Hornsdale would have lasted the SA grid for five to ten minutes, It is bigger now so it might stretch to 10-15 minutes.
Basing the amount of wind output at 12:30pm during these months of low demand is going to be very misleading as many wind farms could be curtailed due to the low demand and high solar output.
When you look at the levels of wind output when solar has dropped away, it is much higher. Then you need to scale the wind output based on the projected ratios of roughly 50:50 wind and solar. At the moment States like QLD have very little wind installed. There is a 1,000MW wind farm under construction plus a 800MW wind farm likely to proceed. This is will start to balance out the QLD grid.
In terms of storage there is going to be a dramatic increase in battery storage going in over the next 10 years. Based on Tesla’s projections the cost of cells should drop by 50% over the next five years, while production volumes grow dramatically. Already there is a 6,000MWh and 1,400MWh battery under construction in California. In time we will see similar and larger ones built in Australia.
Batteries are relatively quick to deploy and the costs are falling quickly, so the installed capacity will grow very quickly after 2025 and more so again after 2030.
Then of course there is hydro, pumped hydro and as the grid approaches 90% RE there will be other deep storage such as flow batteries, green hydrogen and compressed air.
The biggest problem largely being ignored by the wind and solar advocates is that rooftop solar is detrimental to both the physics of the grid and the economics of large-scale generators. This article highlights one symptom.
The second biggest problem is CfD that reverses the market signal ie lower wholesale is more costly for CfD purchasers, and makes further investment unattractive without a CfD.
After that we have the requirement for additional equipment and services to incentivise (make more costly) the inertia and system strength services that wind and solar cannot provide.
Then we have the additional network infrastructure required to connect new small distributed generation at low utilisation rates.
Also the reducing capacity factor of the entire system which can only increase costs.
Our electricity system is in dire straits – there are ways out of the mess, but who’s listening?
It seems to me that if a generator is contracted through CfD, this effectively reverses the effect of spot prices on bidding behaviour.
The CfD rewards generators for low spot prices, which incentivises generators to bid low and keep spot prices down.
But that doesn’t necessarily translate into low electricity prices for customers, and certainly isn’t providing an investment signal (unless you can score a government offtake)…
The added benefit of being semi-scheduled means following a dispatch set point is almost always optional, which is certainly an advantage compared to scheduled generators who can’t sell their output when they’re dispatched down to make room.
I am wondering how useful a droop characteristic would be for solar inverters, as well as home batteries and loads such as (future) electric vehicle charging in these situations (in particular, commercial systems with “zero export” constraints)? Would it reduce the need for inertia from synchronized generation?
A quick search of IEEE finds quite a few papers on control strategies for use in low-voltage connected inverters, which aim to improve rotor angle stability of synchronous machines and transient stability generally.
Would it be correct to say that over performance is a function of:
– semi-scheduled generators not forced to follow dispatch
– CfD rewarding generators for low wholesale prices
– completely controllable
While underperformance is a function of:
– outage or derating
– unfavourable weather
– largely uncontrollable
Therefore the new rule mostly applies to the over performance situation…
I understand the ESB made the ISP ‘actionable’ and therefore the ISP’s lost of projects must be built (or at least studies commenced) by the respective transmission companies.
I’m not a fan of the central planning model – customers would be better served if a state requiring more power, or more load, negotiated with the adjacent states.
The changes to solar inverters in SA is really too little too late. There needs to be a rapid shift to making hot water services a dynamic load. This alone won’t be enough, but it solves two problems and is one of the only things that can be done quickly and cheaply.
The term ‘transition’ implies an end point, when the ‘transition’ is complete. I’m not holding my breath.
Meanwhile, in a recent interview, Audrey Zibelman literally said that AEMO is full of people who prioritise decarbonisation over consumer benefit. Perhaps that’s part of the problem.
You’d think the CEO of AEMO would have the cred to propose that states hellbent on forcing more rooftop solar into the grid should back off for a while and start turning the ship, before it runs aground.
You would have to have some reservations about the transition when you look at the difference between the RE contribution in the middle of the day yesterday (black coal below 7GW) and the situation at dinnertime with the RE contribution down to 1GW!
I don’t understand what you mean by variable. If the HWS are mostly controlled centrally, via providers if need be, then on days of forecast very low demand, the HWS are controlled in such a way that bulk of the load is run during the period(s) of lowest demand. Based on historic use and knowledge of the ambient temperature a very good estimate of the total load and duration can be modelled. Controllers could be managed over 4G or WiFi and even monitor the load current.
In case of emergency on the grid, controller can switch on or off hot water systems.
There is approximately 200MW to 300MW of residential HWS that could be controlled.
Hey Paul
I think you’ll find it was due to NQLD system strength constraints – won’t be in the public data until tomorrow given constraints only have 1 DUID on LHS. Seems more severe than normal – combination of a network outage and crossing one of the demand thresholds I expect.
If AEMO’s logic hasn’t yet linked their own constraints to their dispatch engine, it probably should. A bit like that time QLD islanded and AEMO were dispatching QLD generators against a NSW frequency…
The insight on flexible load reacting to high prices and exacerbating a system strength problem is a curly one. Some loads also contribute inertia and fault current.
It would be a shame to be trying to save money by reducing load, and being penalised more under user-pays than was saved.
Interesting choice of title for the post Paul – a backhander at the “PV saved the day” stuff sometimes seen on other sites?
In reality it seems that a less drastic formulation of the relevant system strength constraints could avoid this sort of disruptive dispatch outcome. I doubt that the underlying physics of the network is quantized.
It would seem that more graduated constraint would have been better. This would have resulted in a more gradual transfer of the generation. It would likely have resulted in a small increase in the price, without the massive spike.
This time the overly risk adverse constraints almost caused network instability.
So now that the full picture is available, can we agree that yesterday’s Wattclarity article headline referencing “many Large Solar Farms tripping” was entirely misleading, inaccurate, and unhelpful? Turns out no so solar farms ‘tripped’, rather they were doing exactly what they were supposed to: following their AEMO issued dispatch targets.
With all the commentary about energy transition “villans” on this website… it would seem Wattclarity fell into its own trap yesterday. I would think an apology or correction is in order.
This trend is plaguing our fledgling hydrogen industry. Project proponents, funders and industry advocates are satisfied that LGCs or nebulous renewable PPAs for an electrolyser running 24×7 can be classified as green. By ignoring the time and place of renewable energy supply in our largely coal-dependent grid, you end up producing hydrogen with a much higher carbon-footprint than traditional fossil-fuel sourced hydrogen. This doesn’t mean its the wrong approach, as scaling up grid scale electrolysis is important in delivering a future low carbon hydrogen industry. However, it’s hard not to get fed up with the disingenuous marketing of “green” projects which are really anything but.
Hedging for purchasers has historically meant protection against high prices.
Now the hedged purchaser is in the unfortunate position of being exposed to very low prices, which costs them more, under a CfD.
Is it possible to hedge against both high and low prices at the same time, or has the combination of rampant growth in rooftop PV, uncontrolled semi-scheduled generation, and manipulated CfD purchasers broken the market for good?
And yet bankable forecasts are made all the time and banks accept some forecast when lending money. Noone can produce bankable forecasts of the oil price, or of share prices. And yet there is a market for such forecasts because the paradigm for valuation requires a forecast of future cash flows. This is not news.
Its also always helpful when looking at surveys to know the sample size.
And finally like management people don’t know what they want or what is need until they see it. . No-one knew they needed a smart phone until they saw one.
Equally there is quite the challenge in designing rules when the technology is changing very quickly. My own view is that the distinction between market facing and regulated business may be more harmful than helpful going forward and that least cost is not always the same as best. Many of us drive cars that are not least cost. How do you incentivise technology lead investment when volumes are declining and still give consumers least cost? It can’t be done. Anyhow enough rabbiting on on Friday afternoon.
At least the solar farms weren’t off target, but this seems like a perverse market outcome where bidding high can lock your competitor out of the market (and would become worse with widespread price-sensitive demand response).
It can be done:
– by having market rules that do not favour any particular technology
– by having market rules that ensure all participants provide the same product
– getting market distorting subsidies out of the picture
– getting non-market government investment out of the picture eg offtakes forcing RE into a saturated market
– by removing the spectre of a carbon price
– by allowing all possible technology to enter the market
– by limiting hedging to a portion of the generator’s capacity
– by enforcing the intent of the National Electricity Obligation at all levels of bureaucracy
– by eliminating central planning for transmission ie remove AEMO from its central planning role
– removing the ESB
Very nice analysis. The questions are heading in the right direction.
I’d ask a couple of them more directly:
– which VRE CfD purchasers are benefiting from very low wholesale prices
– if the market system is in crisis, then why are governments still forcing in new unnecessary supply using VRE offtakes – surely the government has some sort of obligation not to wreck the joint
– same goes for rooftop solar
The questions are good but as you point out everything depends on your reference point.
Mine is the need to decarbonise. If you start with that as the priority then policy will make the money be there. Same as for police or some other essential service.
The second observation is that there is committed and or under construction new investment sufficient to provide more than a further 10% of NEM energy. And I will offer 3:1 odds that there will be more commitments over the next few years. Yet without a policy to incentivise demand the total pie will not be big enough.
When I first saw the graph, I realised it was the source of the “75% VRE” statement and I thought, “fair enough, the odd half-hour here and there, but that’s going to need a lot of transmission and support services”.
A bit later I realised the ISP scenarios rely on GenCost, which has a built in carbon price, somewhat separating the scenarios from reality, being much more favourable for VRE and batteries than in practise.
Not long after, the NSW and QLD gov’s announced REZ plans a decade in front of the ISP, and now the second major ISP project – EnergyConnect – is going to cost so much more than originally planned it could be canned.
As we speak the NEM investment environment is completely tanked (or not far off), with governments the biggest investors via long term offtakes with VRE.
Before long AEMO will be opening substation feeders to isolate sections with too much rooftop PV, to prevent wider blackouts.
Back to the graph, now I see an increasing number of daytime periods with zero or negative wholesale prices, still no solar output at night or during evening peaks, and coal plant continuing to make room for VRE (except when there isn’t any).
You can probably tell, I’m skeptical The Transition is providing much benefit.
This post touches on so many issues its worth coming back to.
Whether people realise or not there is a big discussion about risk going on.
If revenue is guaranteed it is undeniable that the risk in financing wind and solar is very low, that’s because there is no exposure to fuel costs. Essentially your only risk is construction and breakdown. So then the cost of capital is low and the required price is low. But the risk has been shifted.
So the most fundamental principle of risk is it should be borne by those most able to price it appropriately. In many markets the private sector is the best placed to judge return and risk, that is a fundamental principle of capitalism.
Equally there are society level risks, such as say lung cancer from cigarettes, defence, a sufficiently educated population, that society has decided are best borne by the Government
The fundamental concern about risk is the need to judge the carbon priority. If there was a carbon price, a credible emissions trajectory, the risk of new investment could easily be judged by the private sector. Without that policy, the risk falls back to the Government(s) who signal their support for decarbonisation by providing PPAs without necessarily having to disclose their carbon policy. In this case the Government has better information than the private sector and may therefore be better placed to bear the risk.
There are other factors in that individual wind and solar farms are very small relative to the size of movement and also quite fast to build so that the risk of misjudging the total new supply required is smaller. Still it would be better done by setting a credible target and then letting the private sector work out how to do it. Its that lack of govt policy that is at the root of many of the problems.
Hi Paul, thanks for another insightful case study. I’m looking forward to you unpicking the the ‘Q-BCNE_821’ constraint set. I understand why such a System Strength constraint exists, but I don’t understand why it would trigger an immediate dial down to 0MW for all solar farms. It seems to me that the constraint itself leads to the very concern that it is trying to avoid, that is the rapid loss of a significant portion of variable supply in the region. Why would the constraint not just dial back the solar output to ensure it remains below ‘reasonable’ levels?
I note your comment “Unfortunately, because of the hard ramp on the constraints at the time* this meant hitting the park brake when coasting down the hill” – but I couldn’t find whether the * meant you had some further notes about the reason for the unfortunate hard ramp?
Cheers
Mark
Well written Ben, thanks.
Many people, even within the energy sector, still don’t realise that the states hold legislative power over energy rather than the Federal Government.
How do you see the network cost debate moving, given that much of the upgrade is prompted by a conscious pursuit of renewable penetration instead of re-use of existing sites/regions for generation?
Unlike past eras, this time it’s not growing demand that is driving the investment need.
Latrobe Valley and Central Qld are good examples where those states have better renewable solutions elsewhere geographically but that prompts greater transmission investment to enable the better low-emission outcome (rather than for eg. replacing coal fired capacity with gas at lower capacity factors in the same area of the network.
How does the different emission reduction targets/timelines by different states’ governments affect this?
Lots to unpack perhaps, but keen to get your thoughts.
“…Victoria recently passed legislation to empower the state minister to direct the construction of transmission assets outside of the NEM’s cost-benefit based planning framework5. It would clearly be unreasonable, and a moral hazard, for customers in other states to pay for such follies. …”
*Follies* -Excuse me, but, simply because something falls outside a cost-benefit model does not mean there is only cost and no benefit: it can mean the benefits are less than other choices in the ranking at that time, or the benefit is not sufficiently equalised. If Victoria decides to implement something with asymmetric benefit, you are arguing that the other beneficiary path nothing? How about if we ask the questions: are there other models of paying, and are there other cost-benefit models?
Good post. Confirms what we have all been saying. For this sort of price spike batteries can’t be beat by any other currently available technology. That said I wonder if any demand response was able to take advantage of the short intervals.
Great to have Andrew contribute this ‘Part 4B’ (in a way) in the evolving Case Study.
Readers should also keep in mind that UQ’s St Lucia battery is also not scheduled (i.e. not even registered with AEMO, so able to operate completely independently of the AEMO’s dispatch process – because it is under the AEMO’s 5MW registration threshold for BESS).
These two differences (i.e. size (re: size of market need), and type of participation) are other difference to consider in the considering what happened on the day … especially if thinking through the ‘if I also had a battery, it’s performance would have been X’ extrapolation.
It is interesting that the battery benefits in 2 ways from slower response times of traditional peaking units: from the obvious ability to capture high prices during supply shortfall; and also from “artificially” lower prices due to the slower ramp down response of the traditional units after the VoLL event. Perhaps it should have to pay a “corrected” price for recharging in the latter case.
As you point out, the bidding dynamics and dispatch outcomes that will occur under 5 minute settlement (5MS) could be quite different because the financial incentives for participants will be driven by the 5 minute dispatch price not the 30 minute average trading price. I think one of the current uncertainties with 5MS will be how operators of traditional peaking generation decide to offer and run their capacity – for example they may have to bid “more cautiously” as they won’t be able to rely on “catching up” to 5 minute spike early in a Trading Interval by starting up in the later DIs of that TI.
It’s also worth noting that although limited ramp rates and start up times for peaking (and online thermal) plant probably were a factor in driving the dispatch price to $15,000/MWh when the northern Queensland solar plants were constrained down, it was presumably the very high downward ramp rates of those solar plants that required very sudden upward dispatch of the rest of the generation fleet. If the solar plants had offered slower ramp rates (downwards) to the dispatch engine or if AEMO’s constraints were less binary in formulation, then the price spike probably wouldn’t have occurred. Sometimes very fast fast ramping can be too much of a good thing :).
Interested readers may also be interested in this working paper: We document the increasing frequency of system strength and transmission constraints binding in the NEM, coincident with increasing VRE penetration in SA and QLD. We also document some evolution to the form and frequency of the the SA system strength constraints binding. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3501336
Scrutiny on how the constraints in NEMDE are decided upon should be healthy – there are real trade-offs to consider when setting these constraints – while a more risk averse approach should of course improve system resilience, these tighter constraints can only increase system-wide generating costs. Consequently, innovations that allow the loosening constraints without impacting reliability probabilities will provide tangible economic benefits.
Wasn’t this caused by 800 MW of Victorian wind turning off at 1pm to avoid negative prices? This turned Victoria from an exporter to NSW to an importer. The current dispatch rules are leading to perverse outcomes.
I don’t expect wind was doing anything other than reacting to dispatch instructions.
There was an almost 2000MW reversal of flows on VIC1-NSW1 required by constraints. Given the sudden excess of generation, VRE was best placed to respond by backing off. http://nemlog.com.au/nem/ic/VIC1-NSW1/20201116/20201116/
Was looking at NEMLOG to try and understand why Victoria was importing against the price gradient. It continued for for a few dispatch intervals, as NSW hydro filled in the gap.
I think market notice #80022 may explain what was going on.
「An un-planned outage of Upper Tumut – Stockdill 1 330 kV Line in the NSW region occurred 1252 hrs 16/11/2020
Constraint sets invoked: N-SDUT_1 and N-CNSD_3C
This constraint sets contains equations with the following interconnectors on the LHS.
Negative load should not be a showstopper. Inverters will simply need to operate as though they are on an AC coupled system. Provide primary frequency response and accept AGC requests to adjust output.
Domestic, commercial and grid-scale batteries, combined with synchronous condensors can provide sufficient inertia and any curtailed VRE are effectively ‘spinning’ reserve.
This problem sounds like a local lack of reactive power. I wonder if it can be improved by?
a) installing a static synchronous compensator (STATCOM).
b) Requiring Voltage Source Converter (VSC) based inverters remain synchronized and provide voltage control, even when not generating active power (currently on trial in the United Kingdom).
c) series compensation capacitors to reduce AC line impedance.
d) additional transmission lines or synchronous condensors to increase local system strength.
e) follow b) but additionally investigate voltage support from (Doubly-Fed Induction Generator) wind farms.
Unfortunately I suspect economic rules like the RIT-T will slow down technical fixes.
For anyone curious about N^^N_NIL3 the constraint is published here http://nemlog.com.au/show/constraint/?k1=N^^N_NIL_3 (an extensive public dataset, no connection to it)
So Paul, Allan – (maybe even Ben B or Alex W at AEMO) – any thoughts on whether we will get into a round of debate about the merits or otherwise of having an interconnector term with such a small co-efficient, but wide range (much bigger than any generator) on the LHS? It’s above the threshold, but at 0.09287 you don’t get the best relief – and we have already seen similar equations swing the interconnector 1800MW in a TI recently. economics vs physics anyone? Should interconenctors have a different threshold given their range? do these swings cause any other power system issues, now, will they in the future?
Thanks to Allan for a very clear and concise description of despatch constraints. In a well-ordered system the proposed interconnector to SA would hopefully alleviate this constraint.
Time will tell.
P.S. I propose we refer to the area west of Wagga as the ‘Triangle of Tribulation’.
I never understood why this ever needed to be a “polygon of pain”. From what I can see there is no reason why AEMO could not have operated the Balranald-Buronga line normally-open at Buronga (with a backup auto-close scheme). This would force the power flow away from X5 up through the high-capacity Dederang interconnector instead, potentially avoiding any curtailment. Maybe I am missing something?
Let me guess – retailers are making plenty out of negative prices, so the tariff reform that should follow the duck curve, is lagging… and lagging… until it becomes embarrassing for the retailers and they concede.
In QLD, the government owned DNSPs set peak tariffs between 6am and 9pm. They charge peak rates during periods of negative wholesale prices.
Of course some is hedged, but CFDs incentivise those generators to bid low, which results in very low wholesale prices.
Ben
I don’t think many retailers are enthusiastic about tariff reform – not least because they don’t see any evidence their customers want it. but it’s ridiculous to claim that they are the barrier here. Tariff reform is an orphan policy. Governments, policymakers, consumer reps – no-one has picked up this ball and run with it, and many have opposed it. For example, Victoria banned widespread tariff reform a few years ago. Some DNSPs have taken some steps as you note, but they don’t care about wholesale prices – they are doing it to manage network peaks, which may often overlap with wholesale price spikes but are not exactly coincident. Even they have limited interest – after all they’re on revenue caps so they get the same amount of money either way.
Despite the rhetoric there’s enough competition and innovation in the retail market that business and household customers can access wholesale price signals if they want to (eg via Flow Power or Amber respectively, there may be others).
Hi – some issues – forces a semi scheduled generator to behave like a scheduled generator when in a semi dispatch interval -a lot of the time the cap will be below the available resource energy capability to limit output. In this scenario an SSG can’t go above or below semi dispatch target due to variable resource
As the amount of VRE increases so will semi dispatch – it’s getting rid of semi scheduled operation through subterfuge All they would need to do is semi dispatch VRE
system security would be the excuse
I’ve just dialled back in after 14 months of retirement and it’s great to know that the neurones haven’t completely turned to mush in that time. Thanks Allan for a great explanation. I’m thinking of using it to dazzle some mates when someone asks why their bills are so high or how the NEM works. I am very happy to now be just a spectator though I do miss the access I used to have to NEM data.
Negative prices are endured by generators for only two reasons: because they have sold PPAs (or similar contracts) to naive buyers who don’t mind paying the strike price in their contract plus the negative price(!) for every MWh generated, or because a generating unit is not flexible enough to turn off and back on again later when prices are attractive. Surely, an increasing prevalence of negative prices is therefore both the symptom and the cure for this problem?
Hi Paul
With regards to the forecast LORs, in 2018 there was a change to incorporate forecast uncertainty in the process for declaring forecast LORs. Refer to the Reserve Level Declaration Guidelines on AEMOs website for details.
It looks like common factors in all the NSW volatility this week are:
– significant numbers of coal-fired units on outages, I’m unsure to what extent these are planned maintenance vs breakdowns
– transmission constraints between southern NSW and the rest of the state, due to outages of some important 330kV lines – again I’m not sure if these are planned or forced outages.
While these price spikes were occurring, the transmission constraints were forcing flows OUT of southern NSW into Victoria, until “Negative Residue Management” constraints were invoked to wind down generation in the south of NSW and prevent further counterprice flows.
In essence, despite the reduced availability of coal, wind and solar generation mentioned in your posts, there was still sufficient available generation in NSW + imports from Qld to comfortably meet demand, but the transmission constraints from southern NSW meant that not all of this generation could be dispatched.
Now You Can Create beautiful, natural sounding Voice-overs – Add Pauses, Inflections and Tone – Make Listeners believe it’s a real human talking. Don’t take our word for it, hear a free demo now. Just send us an email here: nancyfadodd@gmail.com
I scanned the explainer and could not see the difference between Total Demand and Operational Demand. I had assumed Operational demand was the demand from the grid seen in Victoria. This includes net imports but not behind the meter generation.
If Victoria was exporting as it probably was, then Operational demand would be grid generation less net exports, battery charging and pumps while total demand would be Operational Demand + rooftop solar. So the question is how can Total demand be less than Operational demand
For those that have the data, skills, systems, resources, etc to do it – an excellent approach. Well done.
I think all retailers/embedded networks (virtual or otherwise) started with these ideals, the challenge is to make sure it remains efficient, and it’s a challenge to traditional retailers to leverage their strengths to compete across their portfolios.
Is there a public source for the terms of PPAs?
I’m especially interested about the connection point: is it usually the node or do some use the generator connection point as in the ACT reverse auctions?
Hi Dan, good question. There isn’t a public source for the terms of most PPAs as they tend to be closely-held commercial deals. I assume by the connection point you mean the generation figure used for the fixed price payment of the PPA? I’ve seen PPAs that pay the fixed price on generation at the node (the regional reference node RRN) and others at the generator connection point (sometimes referred to as “at the gate”, which isn’t always physically where it is). The difference between the two is on who takes the risk of the MLF (marginal loss factor) changing in subsequent years, which can be a key point of negotiation between the parties.
Thanks a lot for your reply, I do have a follow-up. I have understood from M O’Neil’s article that MLFs scale the spot price that the generators receive, not the output they sell. But what about the actual loss of output between the gate and the node? In a PPA, is the MLF-scaled price applied to the output at the gate or the output minus the physical transmission losses at the node? If it is the latter then it would amount to double counting the impact of transmission losses.
Hi Dan, the settlement generators receive from AEMO is calculated as RRP * MLF * metered generation (at the gate). This is typically referred to as the “floating” leg of a PPA (i.e. the money the generator is swapping with the off-taker). The “fixed” leg (what the off-taker pays to the generator) could be calculated in two ways (typically, but every PPA is different):
– “at the gate” – the off-taker takes the risk of variability in the (RRP * MLF) part, and settles at a fixed price on the metered generation.
– “at the node” – the off-taker takes only the risk of variability in the RRP but not the MLF, so settles at a fixed price on the metered generation * MLF. This isn’t the generation adjusted for physical losses to the node but an allocation of who’s taking the risk on the financial impact of loss factors between the gate and node.
In theory (the theory of marginal pricing) the MLF scales the spot price – but in practice the AEMO settlement calculation is to multiply the three terms together (RRP * MLF * metered generation), and PPAs often split this (when “at the node”) into RRP and (MLF * metered generation). The actual physical loss to the node is irrelevant – all that matters is who’s taking the risk on which components of the AEMO settlement.
Thank you.
One thing I don’t understand is how one can reconcile the relatively good position of the Coleambally plant with regards to the X5 constraint with the many headwinds you mentionned about it in your other post, especially the constraint part? https://wattclarity.com.au//articles/2020/01/gsd2019-allan-howgoodissolarfarming/
The fact that Coleambally isn’t caught “behind” the X5 constraint doesn’t undo any of the other factors leading to the headwinds described in my earlier article – they remain just as relevant. The X5 coefficient is just another potential headwind that Coleambally avoided on this occasion.
Another point about constraints is that while generators with (larger) negative coefficients are at risk of losing volume when the constraint binds, there’s no corresponding volume or price uplift for generators with positive coefficients(*) – those positive coefficients typically just mean the constraint may be less likely to bind when the relevant generators are running.
Allan
(*) Except where the same generator sits in two different binding constraints but with a negative coefficient in one and a positive coefficient in the other – there it’s theoretically possible for the effect of the positive-coefficient constraint to offset the impact that the negative-coefficient constraint would otherwise have. But this is not guaranteed and very dependent on the specifics of the situation.
Hi Paul and thank you.
Why would the developper that has signed a PPA with an off-taker care about negative prices? I understand that the newer ACT reverse auctions have a -$20/MWh threshold, but are you implying that this is also the case in many other PPAs?
For the earliest wind farms built in the NEM, the initial PPAs (i.e. primarily signed with retailers back then) might not have had negative price clauses as negative prices were not common at the time, so not forseen as a risk.
Most, if not all, PPAs signed in recent years have negative price carve-out clauses where the strike price does not apply, but instead the generator wears the risk.
Sorry to double post but could you please explain :”If you are the project proponent that has some form of off-take arrangement at the RRN […] you are still wearing the risk.”.
My flawed understanding is that MLFs scale the spot price the generator gets, not the output it sells. If I already have an offtaker at a fixed price and if MLFs don’t impact the volume I sell, why do I care about the MLFs?
Danton,
You note ‘My flawed understanding’, and it is indeed flawed – the output is scaled by the MLF to calculate the MWh notionally delivered at the RRN, and it is this number that is used in the vast majority of commercial transactions – such as:
1) For all wholesale energy purchased by AEMO in the spot market; and
2) In most of the PPAs contracted between generator and offtaker (there may be some offtakers who, for whatever reason, have agreed to take the MLF risk … but this would be rare); and
3) Even in terms of the LGCs registered each year for renewable generation.
All of this makes sense on the principle that the MLF reflects losses in getting the product to market.
Paul, do you have any insight into the logic AEMO uses to determine when to send out a ‘power system events’ market notice, or not? Large units trip with some regularity, but do not always attract a market notice flagging the event. For sure, an ~830 MW trip is a “very large” trip and significantly unusual, but a frequency drop to ~49.78 or so (as your screenshot suggests) is not particularly unusual or dramatic. Smaller unit trips have caused larger frequency drops in recent times, without prompting AEMO to send a market notice. Any insight into how AEMO decides to trigger notices, or not?
Double-checked with a couple more learned NEM junkies than me before responding …
Yes, the reason this event was market notice worthy was because it was a “non-credible contingency”, not because of the level of the frequency nadir.
Stations are designed to not have common points of failure across units, hence this contingency is “non-credible”. This event will trigger an investigation to ensure its cause is understood and not repeated.
—–
As an aside, even after a credible contingency, a frequency nadir below the frequency operating standard, i.e. 49.5Hz, will trigger a market notice.
These type of demand reductions are unusual enough that you are writing about them. This makes me wonder what was so different this time? One of the many problems is that the information that should be available from every point of presence via communicating meters is not actually available because there aren’t enough communicating meters, not enough software to exploit their information, and maybe not enough sensors. So if it was cloud cover or rain or a substations or two dropping the centralised data system really should have awareness. Cloud cover for instance generally has a pattern and and can be predicted. Its interesting but (i) what was different this time and (ii) we need to exploit the information each point of presence provides.
There was indeed a cloudbank with associated rainfall that passed over parts of SA on Wednesday morning. It was mentioned on local radio as being over the Eyre Penisula about 9 am and arrived over Adelaide in late morning. It was clearly visible on the BoM radar proceeding in an ENE direction. By mid afternoon the sky over Adelaide was complete clear.
By the way, above what percentage of yearly curtailment (defined as (AVAILABILITY – TOTALCLEARED)/AVAILABILITY taken over a year) would you consider an asset to be problematic?
Hi Dan, it depends! It becomes problematic when the bank and the investors are unhappy with their returns, so this is different for each asset depending on its financing structure and the assumptions made in the business case. Also, what really matters is revenue – so if it’s a simple fixed-price PPA then curtailment is a fair proxy, but if the plant is merchant (spot-exposed) or has negative-price exposure, then the spot price at the time of curtailment matters as well as how much curtailment there is.
If anyone can confirm or otherwise that it was a big cloud that caused the loss of 500 MW of rooftop solar in South Australia last week please post your thoughts. This is a big deal as it represented a sudden 45% increase in forecast demand for the entire state of South Australia.
Looks like NEM frequency went rather out of band (~50.2Hz!) during the scada outage, this might have tripped some solar. I was looking for public data on DUID responses to frequency excursion, but this is also SCADA based.
Hi again Marcelle,
I have tested out what you said: “The calculation can be simplified by aggregating AVAILABILITY and TOTALCLEARED over a time period (such as months), and ignoring the SEMIDISPATCHCAP, as any time the TOTALCLEARED is less than AVAILABILITY (and therefore making a non-zero contribution to the curtailment calculation), the SEMIDISPATCHCAP will be 1.”
Although what you wrote is the case for most generators, the difference between AVAILABILITY and TOTALCLEARED will change if we include the SDC flag or not for some generators (COLEASF1 over 2019 for instance).
Do you have any idea why that might be?
Sincerely
Hi Dan, good question! There’s another complexity with the dispatch target (TOTALCLEARED) – if the ramp-down rate is involved. Each generator specifies a ramp-down rate in its bid (in the case of COLEASF1 in 2019 this appears to be 3 MW/min or 15 MW in 5 mins). The NEMDE dispatch calculates the dispatch target from the AVAILABILITY and constraints and bids, but then allows the step from the INITIALMW (the snapshot of output just before the dispatch calculation is done) down to the TOTALCLEARED to be no more than the ramp rate. In that case there won’t be a SEMIDISPATCHCAP=1 as the TOTALCLEARED is greater than AVAILABILITY not less than it. For most farms this is infrequent enough to be immaterial in a curtailment calculation but it appears for COLEASF1 that on partly-cloudy days where the generation and forecast moves around rapidly that this can happen more often, and so it is necessary to filter for only SEMIDISPATCHCAP=1, to avoid under-estimating the curtailment.
Fascinating figure! Why do we see such a big increase in total capacity bid from 2000 to 2005? Surely renewables aren’t an explanation, since they mostly came on board from 2010.
The tax refund offsets nearly half of the (projected) PPA losses, so it could be said that taxpayers are somewhat subsidising the wind industry, through AGL.
There could also be some element of bluff from AGL in an attempt to warn off other market entrants, otherwise known as competition. Most people are fully expecting NSW prices to jump when Liddell closes.
Others in the hole with overpriced PPAs include the ACT, VIC and QLD governments and Snowy because of their exposure to renewable PPAs.
Expect the RE lobby to spin this as a government failure somehow.
SA’s grid is getting even more complicated. And that’s saying something.
The system strength shortfall declared in 2017 was supposedly addressed by the synchronous condensers, but there was a further inertia shortfall declared in 2020.
Separate but related to the shortage of synchronous generation in SA, is the imminent closure of Torrens Island A which is prominent in the combinations of minimum synchronous generator. Then there is the likely cancellation, or at least delay, of the EnergyConnect transmission link to Wagga, and the inability to shut off excess rooftop solar. Tough place to be in energy politics.
I wonder which heavy industry, reliant on large amounts of electricity, is considering South Australia as a potential investment location? I guess there will always be government subsidised, unprofitable, low capacity factor, hydrogen to clap about.
How much does the cost of SA’s upgrades count towards the rest of the NEM’s bills?
Any idea of the effects of the current constraints on the Heywood interconnecter due to be repaired late June?
With no where to go, negative prices are inevitable
Disappointed these Syncons are taking so long. Originally instal was late Jan and now we are out to late April
Ben, very interesting set of facts and figures. On a point of clarification: In the Outlook section you talk about a forecast pipeline of ” between 500 MW and 1 GW per month” but in the charts just below, it looks like those are quarterly figures not monthly? incidentally, you might want to swap the titles of those charts around…
thanks for that Gordon. I am also an Amber customer (I wonder what %age of their customers are energy wonks…?) primarily for interest rather than expecting massive savings, as i know i don’t have a lot of load-shifting tools available to me at the moment. I got on the Amber train a little earlier than you about March and have since congratulated myself on my impeccable timing, as that’s pretty much when the wholesale price started to soften with slightly lower demand and falling gas prices. So you and i have front run the wholesale price falls that most other small customers will only see flow through with the change in VDO from 1 Jan (I believe). My energy wonkiness extends to keeping track of my bills for the last ten years. Thanks to Amber’s model and the lack of price shocks in Victoria since I joined I am paying pretty much the same in nominal terms for both the daily charge and the per KWh rate as when we moved into our home in 2010! see here for more detail: https://boardroomenergy.com.au/2021/01/08/the-mystery-of-the-missing-bill-shock/
Thanks Kieran, nice post. You really nailed the timing. With no exit fees (or Amber wait-lists), in usual years with a more challenging summer I guess the least cost retail customer strategy (in terms of dollars, not hassle) would be to go on to fixed-rate for summer and real-time for March-November…
Gordon, that’s what i thought too and wondered about switching out in december. Glad i didn’t as my January bill had the lowest average wholesale price so far – 2.47c/KWh! my total wholesale costs were less than $5. Of course, i agree that we can’t rely on such benign prices every January!
I must be a bit behind but haven’t come across the acronym VRE before and my search (Ctrl+F) only finds one mention in this article with no real explanation “”VRE curtailment” – reductions in large scale wind and solar output across the NEM due to various factors, including system strength constraints” doesn’t really explain what it is nor where the acronym comes from. I would appreciate an explainer greater than is given under Figure 40 please.
“VRE” = Variable Renewable Energy. Basically shorthand for wind and solar generators whose output is variable with incident wind or sunlight. Large scale VRE generators in the NEM generally require capability to control output downwards – curtail – when instructed by AEMO.
Gordon, the environmental certificate cost would include (or be entirely made up of, not sure) the Victorian Energy Efficiency Certificate cost, which is borne by the retailer and passed through to customers.
Thanks Kat, I had forgotten about the state-based programs when I wrote that footnote up. I’m also not across the cost-impacts of the scheme relative to the RECs.
Nice article and well articulated thoughts Gordon. I’m also a recent Amber convert (and work with C&I spot pass through). I love the Amber model, but I’ve got a couple of musings around their long term prospects in such a vertically-integrated heavy market (we have a big 3, better than Coles/Woolies or QANTAS/Virgin[/Rex?] right!)
1. Like Kieran I can’t help but feel Amber’s customer base is mostly energy nerds dogfooding our own products or at least indulging our interests. I was somewhat surprised when a non-energy industry friend of mine recently independently raised Amber over a beer saying that he was on the waitlist and keen to start with them. That being said PowerPal has been seeing great engagement from the Victorian community at large, so there’s definitely some appetite for more novel approaches to managing residential consumption and costs.
2. The back end hedging required to meet both the solar minimum FiT and VDO (at least here in Victoria) must have some significant challenges in keeping the fixed costs low.
3. I’ve joked with others that Energy Locals model of outsourcing their retail licence might be the smartest play in the retail sector, but it imposes some significant risks to the fixed costs – already Amber has had to bump the monthly charge to $15, presumably due to points #2 and #3.
4. The ESB seem intent on legislating in favour of the vertically-integrated incumbents (staring you right in the face RRO). Some of these changes could be seriously damaging to the competitiveness of small, spot-exposed and non-generation owning retailers.
The most serious negative figure is Net cash provided by operating activities down 46% to $614 M. That is close to a 50% reduction. This loss of cash profits is reflected in Underlying EBITDA down by 13& to $926M.
This large change in the net cash receipts is the most concerning as it shows the underlying profitability has been significantly reduced.
If AGL’s write downs were simply due to overvalued assets it would not have impacted the cash flow.
As time progresses the cumulative financial impact on the small business sector from sudden Victorian lockdowns will make it harder for AEMO to accurately forecast electricity consumption. Small business is all about “confidence” and small business working capital has been severely eroded with lockdowns. Is it possible to provide an update to this article with a graph of the actual consumption compared to STPASA forecast? AEMO’s 2020 Electricity Statement of Opportunities Figure 57 states that the largest consumer of NEM supplied electricity in ‘Victoria’ is Small and Medium Enterprises. However, in AEMO’s 2020-21 Budget the largest consumer of NEM supplied electricity in the ‘NEM’ is large business at 57%, followed by residential at 28% and small business at 15%. The 2020-21 AEMO budget also states that “consumption of less than 10 MWh per annum is a household”. Identifying your largest customer and understanding their consumption intentions is critical to AEMO’s forecasts and AEMO’s Integrated System Plan.
Shows the need for those sync cons to come online as soon as possible. Also shows the need for the SA Government to move residential hot water to smart controlled loads.
This load shedding is caused by the US policy that requires Gas in times of severe cold to be diverted for home heating depriving gas generators of fuel. Gas generators do not store gas on sight. This exact scenario has been predicted by Meredith Joan Angwin in her Book Shorting the Grid.
It is a system problem in that generators are not paid anything for storing alternative fuels like oil on sight. As much as I would like to blame Wind for this Grid failure it is not a problem with renewables as such. It is indicative of poor grid management. Note Nuclear Plants do not suffer from this problem. https://www.amazon.com/Shorting-Grid-Hidden-Fragility-Electric-ebook/dp/B08KZ51SDP
“It is a system problem in that generators are not paid anything for storing alternative fuels like oil on sight.”
So the FF generators work on just in time like wind and solar but somehow they’re to be scolded?
My take is unless tenderers of electrons to the communal grid can reasonably guarantee them 24/7/365 along with FCAS then they can keep them and use them themselves. But then I’m just a naive consumer of an essential service.
Correct. Meredith Angwin describes this problem. The gas burning generators are capable of burning oil instead. But no one wants to pay them for storing the oil on site. What is worse is that no one seems to be actually responsible for guaranteeing supply. The market system which is run by the Federal Energy Regulatory Commission (FERC) is a failure. https://www.ferc.gov/
The outcomes for 2020 have been compounded across the board by Covid impacts on demand.
The prices in QLD were negative much of the time due to QLD > NSW interconnector being restricted during day light hours during low demand periods. Typically there is some work on the lines, but 2020 saw an increase due to the upgrades on the interconnector and related equipment. Some of the activities of the FF plant in QLD needs to be questioned as there are times when more units should be taken off line, even if it means a bit more use of gas. It will be interesting to see the impact of 1GW of wind generation capacity is online.
NSW won’t see many negative prices until it has more VRE. I doubt this threshold will be reached in the next two years.
The ACCC should be investigating the predatory pricing of some of the coal generators. Just because it is difficult or costly for them to ramp their plant, doesn’t mean it is legal to force the prices negative due to excess supply. This is particularly the case in Victoria and QLD where the coal generators have significant market power to fall under this section of the Australian Competition Law.
Craig your last para does not make any sense. Why is it not legal for coal generators to bid negative? Why would it be in their interests to force prices negative? if it’s not legal for them, why is it legal for wind and solar generators to do it? the point of negative prices is to signal excess supply, so sounds like the market working as it’s supposed to. also why would QLD generating companies run less coal and more gas when gas is more expensive?
Anyway good luck getting the ACCC to do anything about the state owned generators in QLD…i mean they’d probably quite like to but ACCC can’t always do a lot about the actions of publicly owned businesses.
Predatory pricing has to do with market power. There aren’t any wind farm operators with enough market share to be able to dominate the market. This is not the case with AGL in Victoria and the QLD government owned generators in QLD. If you do a search for predatory pricing and Woolworths you will see there was a successful case run by the former Trade Practices Commission under the former Trade Practices Act. The current Australian Competition Law has been strengthen in regard to predatory pricing.
Generally the ACCC will only take action if it is in the media or has a large impact on consumers. These cases certainly fit the latter as the dominant generators can make it unprofitable for small VRE to produce power. This discourages further investment in VRE and would extend the time that the coal plants are operated for.
Why run gas? If the QLD Government owned generators shut down a coal fired unit in the low demand and high VRE output months, then most of the time there would be no or very little need to operate the fossil gas plants. Now it might be cheaper for the QLD Government to run excess coal plant and push the VRE generators out of the market, but it isn’t fair, legal, in the long term interests of Queenslanders both in terms of the wholesale electricity rates or the climate.
I fully understand that predatory pricing cases are complex and difficult. The Woolworths one run through the courts, including all the way to the High Court, over a ten year period.
Well, the Queensland government who own the coal power stations there have set a renewable energy target of 50% and set up a new company CleanCo to invest in VRE, and write PPAs with VRE to help them get there. So i’m still not sure why you think they are trying to discourage VRE.
AGL have a single coal plant in Victoria, so they are in no way “dominant”. perhaps you meant to type NSW where they have two…although they are closing one in the next couple of years which is odd behaviour for a “dominant” player.
There are multiple layers of ongoing monitoring of behaviour in the NEM. the AER and the ACCC both have roles in this regard. So it’s not like no-one is looking.
When there’s a lot of VRE, i.e. when the prices are negative, no-one has market power and anyone who bids below 0 is only doing so because they are desperate to be dispatched. VRE do it because they still earn via their PPA, coal plant do it because they can’t afford to go offline and have to restart. That being “costly or difficult” is a perfectly good reason.
I don’t imagine i’m going to change your mind, but just in case anyone else is reading…
some useful facts and figures here Paul. First question should be: is the planning standard designed to deliver reliable power at these extremes? If so, then yes, something has gone wrong, and Texas needs to find out what it is. but if not, and it does sound pretty extreme, then the question is does Texas want to pay for a higher standard of reliability?
on the subject of whether a capacity market would have saved them – well maybe, but a) at a cost and b) only if capacity was certified for operation at these extreme low temps and/or there was sufficient penalty for non-performance.
As for greater interconnection, well maybe but that didn’t help CA in the summer. its neighbours were also experiencing demand peaks and didn’t have enough to spare. i believer the cold snap extends beyond TX…
what about better insulation of buildings (weatherization in the US)? this would certainly help reduce the discomfort of rolling blackouts, but it wouldn’t prevent them alone. Logically it would just shrink the demand curve and the supply side would rebalance accordingly. you still fundamentally have the load duration curve issue to solve – are there sufficient incentives for those last few MW to be available just for those rare extreme weather events?
finally – blame the market. well having a single public utility is no silver bullet – you still have to have the standard, someone has to oversee/enforce it, and the extra kit (more capacity/gas storage/pipeline insulation…) costs money and has to be paid for.
“Wind energy provides only about 25 percent of Texas’ total power throughout the year, according to ERCOT data—natural gas sources account for 35 percent—although the turbines tend to generate less power in the winter.
The Texas Tribune reported that only 7 percent of ERCOT’s forecasted winter capacity was expected to come from wind power sources in the state. Meanwhile, around 80 percent of ERCOT’s total winter capacity is generated by natural gas, coal and nuclear power.”
I’m reminded of the cautionary tale with statistics that the average adult has one large breast and a testicle.
It is now several days after this article was posted. I agree with the observations and analysis, including the forward projections included in it.
Congratulations for a broad, well presented summary of the situation which has now pertained in Texas for 6 days and for the balanced conclusions that (a) this is a system problem and (b) electricity is not optional – it is an essential service that warrants higher standards of reliability than, unfortunately, have been achieved in Texas.
Negative prices are ‘predatory’? First I’ve ever heard it explained that way.
Wind and solar are still supported above market rates by PPAs (just ask AGL) and the RET. these generators almost all bid negative just to get dispatched. With governments underwriting RE the market is oversupplied during the day when the overpaid rooftop solar is in full flight.
Currently there is only investment in generation that runs when there is wind and sun. So the energy-only market indicates oversupply during these periods.
Until governments remove the incentives causing oversupply when there’s wind and sun, there will be no investment in anything else.
Perhaps a capacity market would change things, but I’m doubtful. I think we are destined for blackouts
Adelaide, Melbourne, California, Texas, who’s next – Germany, UK?
Very interesting article. I would just have one question regarding how prices are being calculated in both Case 1 and 2 for 30-min settlement:
In Case 1, payment is calculated as follow ($50 x 0 + $40 x 0 + $60 x 0 + $55 x 0 + $60 x 0 + $14,700 x 1)/6 which is different to Case 2 (-$1,000 x 1 + -$1,000 x 1 + $60 x 1 + $55 x 1 + $60 x 1 + $14,700 x 1)/6). As I understand, regardless of when the battery dispatches, settlement price is calculated as the arithmetic average of the 6 dispatch prices. Would you mind clarifying why 2 different calculation methodologies are used here?
just a coda on this topic following unfortunate events in Texas last week. you may have read about people there on wholesale pass through contracts who are facing bills for $1,000s because the price there (for those lucky enough to have power…) hit the cap of US$9,000/MWh and stayed there for days. Amber customers here in the NEM are not so exposed because the NEM has a rule called the Cumulative Price Threshold (CPT). This is currently $224,600/MWh. If cumulative prices over an extended period (seven days) reach this level, then AEMO effectively overrides the spot market and applies administrative pricing of $300/MWh instead. So while this situation would still be quite a bill shock for Amber customers, sustained exposure of the level seen in texas last week wouldn’t happen in the NEM. PLus if you’re in Victoria, Amber have to cap your total bill over the course of a year at the level of the VDO, so that’s a further protection.
My comments are restricted to apparent parallels between extended weather driven price hikes in Texas compared to those in AUstralia.
Prices going ballistic for sustained period of time driven by extreme weather is less likely in Southern Hemisphere then Northern Hemisphere. Why? The polar vortex of the SH is much more stable than its NH counterpart. Whilst we can get big anomalies stable polar vortex means, they would be expected to be of much shorter duration (intensity?) then our friends up North.
The two calculation methodologies are 5-minute and 30-minute settlement. The key point is that the battery would rationally be dispatched the same for each case (i.e. 1 single high price or both high and low prices in a half hour) yet it results in adverse outcomes especially in case 2. Therefore 5-minute settlement does not penalise a battery, the way half-hour settlement can.
It’s Economics101 that if the ACCC sits on its hands with solar and wind dumping on the communal grid with power at the beck and call of the weather and sunlight then alternative dispatchable power with FCAS will be driven out. Basically we’re giving up reliable large scale hub and spoke generation and distribution for unreliable spaghetti and meatballs. When that fails or looks like it will State Govts step in with the diesel generators to save the day like SA and Tasmania you’ll recall- https://www.themercury.com.au/news/tasmania/basslink-backup-fast-tracked/news-story/f338dfade76e55495ccdffe18d98f247
It’s politically nicer if you pop in a Unicorn Big Battery for the mainstream media to drool over while you’re fast tracking that Govt intervention of course. My local IGA supermarket in Adelaide installed a permanent large backup diesel generator after the Labor Govt did that. Probably at the behest of their property insurers who mightn’t renew their insurance or else charge prohibitive premiums if they didn’t.
There are three or four factors which most analysis understates which may make even this report end up being too optimistic in the case of coal fired generation.
1. More governments and businesses are focussing on energy efficiency. Even though Italy has three times as much industry as a share of GDP and its major cities are hotter in summer and colder in winter than ours, Italy uses 40% less electricity per capita than us. i.e. we have huge scope to continue reducing demand from the grid which has already fallen from a peak of 210 TWh to less than 190 TWh in the last 10 years.
2. New solar technologies such as bifacial panels, tracking systems and coatings which improve offaxis collection and finally the increasing share of tracking solar vs fixed tilt systems will mean that annual solar output will increase faster than capacity.so a doubling of solar capacity over the next 5 year will probably mean a 250%+ increase in output
3. New wind farms are approaching 50% capacity factors vs an average of around 30% for those installed by 2018 so the next 5 GW of wind will probably supply as much energy as the last 7 GW.
4. The increasing geographic divesity of renewable installations will mitigate the effects of wind lulls in SA by allowing more excess production in NSW and Queensland to flow south.
5 Batteries , syncons, synthetic inertia can provide most FCAS services at very low cost further depressing revenue for coal plants.
6. If all the above proves out and grid demand falls to 165 TWh, wind output doubles and utility solar rises by 250%, even if gas output halves, coal output will fall by closer to 35%
Peter, your points 2 and 3 claim dramatic improvements in capacity factors for solar and wind in only a few years. Do you have any credible sources for these claims?
Similarly is there credible modelling you can point to that supports your claim that “diversity” of output is likely to be meaningful and not impacted by transmission congestion?
I suspect that with improving value in heatpumps and higher COP,electricity gas price differentials moving in favour of electricity and a start of a movement to build gas free developments we have reached a tipping point where domestic and commercial gas is not economical. The building of the first all electric hospitals will confirm the possibility of gas free operation fro most commercial and institutional buildings so hopefully the next 5 years will see a significant reduction in gas use for heating while gas use for power generation could fall by 85% from 2017 levels by 2024 unless coal use falls dramatically, it has already fallen 40% since its peak
I just love the justification for intermittent t supplies being based on averages, while in reality, “ 26.54% of total electricity consumed over the year is during periods when there is no wind or solar being provided to CN from its Wind PPA or solar generation assets.”
Unfortunately electricity system design is not based on averages.
Hi Ben
Nearly three quarters of supply being matched without firming or detailed DR is a pretty good start. This is a financial contract though not a power system design. I don’t believe managing grid reliability should be a customer responsibility?
“I don’t believe managing grid reliability should be a customer responsibility?”
It’s a fallacy of composition to believe we can all take that view. As for opportunities with V2G I fail to see why on earth you would want to make any necessary grid firming battery capacity mobile rather than fixed. After all if our vehicle batteries are larger than they need be in order to firm the grid that’s simply more wasted energy lugging them around wouldn’t you think?
“A fallacy of composition” – I’d be interested if you’d like to expand on that. When you fuel up at a petrol station, as long as you have a compliant vehicle, is it your responsibility to ensure there’s fuel supply? What’s the difference with grid supply?
As for V2G (and I’m no expert), no doubt vehicle batteries will be oversized for customer convenience (who really needs to travel 400km without stopping?) but the broader point of v2g I think is vehicle utilisation. Vehicles are usually not required for their primary purpose (driving) 24 hours a day.
When I fill up with diesel or 91-98 octane petrol for that matter I’m guaranteed full measure and a quality standard although the price varies depending on competitive supply and seller costs. I won’t expect the same price at William Creek and Adelaide that’s for sure but you’re free to set up in competition to any of them. What I don’t want is my car to play up down the road and be told that’s the way it is sometimes as the Premier’s nephew has special rights to deliver his special Green brew as much and as often as he likes but the quality is dependent on the weather.
As for being forced to buy an expensive BEV only I won’t be thinking about my duty to save the unreliable grid with its battery sizing anymore than servo sustainability with my ICE choice now. After all it’s my hard earned and I don’t have the luxury of sticking up everyone’s rates to finance my particular preferences be they judicious or profligate. Naturally if there’s a cushy Council job going with a long range Tesla thrown in I’m all ears but then we come back to that fallacy of composition problem.
Whilst local Councils or we as individuals can fiddle around the edges with the communal grid the really big picture problem is State mandated dumping- https://en.wikipedia.org/wiki/Dumping_(pricing_policy)
Normally the ACCC would be all over such practice by wind and solar but instead we have the silence of lambs.
Thanks Adam – really interesting stuff. What would be even more interesting would be a net load duration curve for the year: i.e. for each half hour (soon to be 5 minutes…) load – own PV output – the scaled output from the wind farm. this would capture your exposure to spot outcomes and address Ben’s point about averages.
Thanks Kieran, I’ll put that together. Also interested when I have time to see the actual firming size required and also to evaluate average pricing during those periods of unserved load – ie from a financial perspective wouldn’t necessarily need to cover spot-exposed periods of low average prices. That’s where looking and covering the early evening exposure will be the first step.
1) How did you define the BAU scenario?
2) “It is important to note that this analysis relates to the delivery of electricity rather than the associated emission reduction which comes from Large-scale Generation Certificates (LGCs).” – is this saying you haven’t factored in the cost of the LGCs in this comparison?
Hi Sean
1 – BAU is based on fixed price contract pricing supplied at time of tender when the PPA was entered into (I.e the contract we otherwise would have entered into).
2. There’s further commentary on the original post, but yes this is a like for comparison on electricity supply, not emission reduction schemes.
Nice article Adam. I did a similar analysis for the ACT’s 100% renewable target that you may be interested in. I contrast to NSW, the ACT is using a mixture of wind and solar to meet its needs. Perhaps the main headline from this was that 75% of the ACT’s needs could theoretically have been covered by the wind and solar generation without any storage. Kieran, it includes some generation duration curves!
Don’t need a battery, just V2G instead 😉
The right data to make decisions is hard. Converting the data to action is even harder. Pretty hard to do with only 30 minutely data from the connection point. Often only for yesterday too.
I can’t see any comparable EV to the ICE paying for itself at double the price when you can go all electric with solar panels and with a solar diverter controller use the power to heat a storage HWS and any left over to aircondition the home for the homecoming. Solar panels are relatively cheap and so are electric element HWSs so lithium batteries aren’t in the race for energy storage.
As for V2G you go right ahead and pay for larger longer range batteries than you really need for transport if you feel the charitable urge to help stabilize the grid. As a BMW exec was asked when will EV prices come down he responded ‘never never never’ as they can make an ICE car for the price of a battery pack with its BMS. As a consequence any fuel cost savings with an EV is eaten up by the interest on borrowing as nearly 9 out of 10 do with car purchases.
I am in SA and have a solar system. My solar system was curtailed between about 2:45 tp 4 pm NEM time on the 14th March 2021. Mine is not the latest system with agent control, so I assume I was part of the 40MW reduction directed from substations. They must have raised the voltage above 453 volts. Anyone know how much they raised it to?
Interesting article Adam
Two questions;
1) How would the BAU comparison look if referenced to todays wholesale price – it looks like you have set a BAU baseline based on a market highpoint versus the collapsed wholesale prices of today
2) what controls do you have to ensure Flowpower / Sapphire are not selling the same Mwh of generation to multiple end customers
Funny how all these issues keep happening in states where they are closing coal plants at record pace and replacing them with renewables which unbeknown to many people, don’t actually produce reliable supply of energy.
The answer from those in the know is to just install more renewables because they seem to be unaware that installed capacity does not equal energy output.
So we then move on to interconnectors at the cost of the end user which are more likely to fail due to the highly variable nature of a more renewable/less coal grid.
All this because the government supports renewables and penalises coal plants so no one builds firm capacity generation any more.
It’s only going to get worse as more coal plants close.
In case reading comprehension isn’t your strong point one of the key contributing factors was a significant shortage of DISPATCHABLE THERMAL GENERATION. But no, you’re right, some more coal plants will solve complex system resilience right away. Just build them and watch all the historical problems of the NEM design (and the state electricity commissions that preceded it) melt away.
Hell maybe we can export your simple and genius strategy to other grids around the world (hi Texas, you need more gas), along with lots of beautiful black Hunter coal.
How come we never had these problems before the expansion of renewables? The cause of the problem is the expansion of renewables which then means no one invests in reliable fossil fuel plant.
The “historical problems” of the NEM design are only problems if you decide to wholesale change the entire generation mix and rely on large scale transmission from unreliable sources.
Congratulations on taking in this type of analysis. It is refreshing.
It must also be possible to assess each solar farm earnings against an assumed PPA strike price, to gauge the ongoing effectiveness of the CfD as a hedging strategy for purchasers.
I presume a similar effort will be made on wind farms?
Why do people insist on using the $0.00c format. It’s confusing. Are we talking dollars or cents here. It’s either $0.00 or 0.00c but it can’t be both $0.00c.
It’s not that hard to write one or the other format. Plus it takes up extra characters.
Great work Adam.
This kind of analysis is crucial as we move the world towards True Zero (as opposed to Nett Zero) carbon. In the past 6 months we’ve seen Google, Microsoft and IBM all commit to this idea – 100% renewable energy procurement without relying on offsets of certificates. In Europe there’s even a as standards push (energytag.org) towards making RECs completely time based, and it’s what Enosi’s Powertracer is designed to do as well.
Interesting article. In 2019 I mentioned to an energy bigwig that gas was being displaced by wind and solar and he heartily disagreed. The data says otherwise.
In the DDPS section, it should be noted that Origin are also gas developers, so their fuel cost is lower than market. Similarly for Arrow (Braemar 2) and QGC (Condamine).
I think a lot of confusion about the role of gas in the NEM arises from failing to distinguish between sources of energy and dispatchable capacity. The data clearly shows that gas-fired generation’s contribution as a significant energy source has fallen and because of its inherent variable cost disadvantage to renewables and coal, that is not going to change.
The more open question is around the role of gas-fired generation providing (relatively) firm capacity for relatively short periods. In this role only small amounts of energy (as a proportion of total production / consumption) are required so fuel costs are much less important and the economic question turns to which technologies – storage, demand response, gas or even liquid fuelled generation, transmission enhancements etc – provide the best combination of fixed capacity costs, flexibility, and supply duration. My view is that it’s premature to be talking about replacing most of the NEM’s current gas fleet in this role in the near term, but it’s much less clear how much additional gas-fired plant will be economically justifiable.
What the AEMO approach does not do is allow end users that have a weather driven demand (eg shopping centres) to provide WDR. These sites are ideally suited to provide WDR when it is most needed, but are excluded from providing WDR
Absolutely agree, David. The 20 per cent cap on adjustments will prevent many temperature sensitive loads from participating, which is unfortunate because these are exactly the kinds of loads you want to be offering WDR when it’s hot and supply/demand is tight. AEMO seems resigned to its current approach for market start, but I’m hopeful they will open the market to a wider range of load types soon after.
I don’t think its so much VRE.
Isn’t it more “semi-scheduled generation” VS scheduled generation ?
Semi-scheduled generation is the low-cost energy,
yet the system needs scheduled generation. It’s the system energy.
(Especially when they get paid the same price)
Isn’t it time we put the onus on the operator to provide the system with system energy.
Shouldn’t the strategy be all DUIDs register capacity as scheduled.
The operators should take cares of the “managing around” bits. If they choose VRE,
then they need to provide the “managing around” bits thru hybrid generation or partnerships with
other providers.
100MW of Generation should be 100MW of generation. Not something less.
That would be completely the wrong approach, if we want to balance supply and demand at least cost (including environmental cost)!
100 per cent of demand is non-scheduled but system operators have found ways to manage its uncertainty and variability for at least as long as we’ve had large scale grids. Non-scheduled or semi-scheduled variable generation does not create a new kind of problem as far as capacity adequacy goes.
Of course any system will always need a large volume of fully dispatchable capacity, but it certainly doesn’t need that type of capacity exclusively, nor must any fixed or minimum proportion of energy over time come from that source. And you’re further incorrect to state that all forms of generation are paid the same price, in the NEM at least – there are dozens of analyses around, some of them on this site, demonstrating that the average spot price earned by different forms of NEM generation over almost any given time period is very different for fully dispatchable vs VRE plant, with “peaking” plant earning the highest averages and VRE typically the lowest.
I’d suggest you read the Grattan Institute report released this week (https://grattan.edu.au/report/go-for-net-zero/) for a balanced discussion and analysis of how the “managing around” task might work in a high VRE system.
Is this just a case of having 20/20 hindsight now, but not having it at the time? i.e. actual charge/discharge will never be perfect to maximise price arbitrage and you will always get caught out simply through not being able to predict future events perfectly (i.e. who’s the marginal generator, who’s next, who’s holding back capacity, credible contingency events which appear out of nowhere, non-credible events more so, demand forecasting errors due to e.g. cloud cover etc., etc.)……. of course post event you can go back and dissect what happened and perhaps build up a library of better responses, but it will never be perfect simply because the future, even the near future, is to always some extent, unknown……
Also, are these battery responses somewhat influenced by what contracts they have in place and what is being played at any given time by the owners/retailers to meet e.g. some contracted obligation? Or are all batteries, as the article suggests, simply working in the spot market maximising arbitrage opportunities (i.e. in the visible part of the NEM)?
Also, how many batteries (in MW and Mwh) can connect before they eat their own lunch? i.e. before they remove the spot price highs and lows so effectively, they remove their main source of income. And how much influence will shifting domestic loads (e.g. pool pumps, air-conditioning, EV charging etc.) have in this space (reducing the requirement for storage) noting this is also happening at the same time as batteries are being rolled out.
It’s interesting to think about lower FCAS as an “opportunity cost” to a generator i.e. getting paid for FCAS means not getting paid for energy. For batteries, things look a little better, given the energy constraint. Another big question is how different bid stacks of larger generators, and thus dispatch price outcomes will be under 5 minute settlement.
We have known since Paul Miskelly’s paper in 2012 that there are always going to be prolonged and widespread wind droughts when high pressure systems sit over SE Australia.
We also know that on windless nights there will be next to no RE generated.
When there is no spare wind anywhere it will not help to have more transmission lines.
We know that the so-called big batteries do not store power at grid scale.
Pumped hydro costs the earth and wastes up to 40% of the power that is fed into the system.
Join the dots:)
Looking at the situation in SA (the wind-leader,) checking at breakfast and dinnertime – the peaks of demand coinciding with little or no solar, we find SA almost always importing from Victoria (with help from the battery of the nation quite often.) When the wind is low, look at the amount of gas used in SA and the coal plus gas in Victoria. Then contemplate the situation when a big chunk of Victorian coal capacity gone, bearing in mind that no amount of extra wind and solar capacity is any use on windless nights, or even windless breakfast and dinnertimes.
Stop celebrating big wind periods and focus on the lows, like the low points of a flood levee where the water will penetrate regardless of the high points of the levee or the average height. It is the low points of the wind supply that kill the grid, regardless of the installed wind and solar capacity. Build as much as you like if you want to displace coal power, RE can displace it but not replace it.
Have a look at SA and Victoria this morning, they are back in wind drought territory (less than 10% of plated capacity).
Actually Paul knows all this and he has been dropping hints for years, but he doesn’t want to frighten the horses and he is waiting to see how long it takes the RenewEconomy people to work it out for themselves:)
That Capacity Factor (CF) should be one of the most important considerations here.
While there is a total Nameplate now of 8703MW, when CF is taken into account, that means that Nameplate equates to a power generation of only 2610MW
Over the last 135 weeks (a bit more than two and a half years now) that long term CF is 29.6%, roughly the same as it was 135 weeks ago.
In that time there has been a number of increases to that Nameplate, adding more and more ‘better technology turbines into the grid.
In the most recent 12 Months, there have been four increases to that Nameplate, from 6960MW up to that total now of 8703MW, so an increase of 1743MW.
However, that rolling CF for the most recent 12 Month to this date is only 29.2% in fact, so with the addition of better technology, the CF has actually gone down.
As to the weather situation, it’s relatively simple to work out when wind generation will be low. 65% of all wind plants (Nameplate (around) 5700MW of that 8700MW) are located in that area of South East South Australia and Central Western Victoria. With the approach of those large High Pressure systems, and then very slowly passing over that area, it’s easy to see when wind generation will be low, sometimes falling by anything up to 3000MW in hours.
Conversely, that in between times between those Highs and the following lows, when those isobars are closer together, hence higher winds, you can see large and sudden losses of power, ranging between 500MW and 1500MW in around the time spacing of ten to 15 minutes. Those sudden losses are due in fact to those newer technology plants, taller towers, larger nacelles with larger generators, and considerably longer blades. In those high wind situations, vast numbers of those newer tech turbines automatically turn off, so the blades, (higher tip speed) are not destroyed.
Those sudden losses of power can be seen by looking at those wind graphs, and you can see numbers of those ‘spikes’ occurring in succession, off, on, off, on, etc.
And no, I am also not a Meteorologist, just an observer with a bacground in the electrical trade who puts two and two together, having been doing this for so long now.
That CF figure should be the most important figure in all of this.
The only thing being ‘ruled out’ is talking about mothballing, or closing down, the Stanwell generators.
QLD total annual demand is about 60 TWh per year. The 50% RET means 30 TWh has to come from wind and solar. Currently about 8 TWh come from wind and solar. The government requires at least another 22 TWh of wind and solar.
In 2019/20, gas generated 5 TWh and coal 46 TWh.
To get wind and solar up to 30 TWh, and keep the 5 TWh of gas, requires 20 TWh of coal to be displaced by wind and solar.
Stanwell operates about 3000 MW of coal generation, delivering 18 TWh in 2019/20.
To make the 50% target, Stanwell has to close all its coal power stations.
But I’d put money on Stanwell building a wind or solar farm, or at least entering more long-term contracts to underwrite wind and solar.
100% of QLD wind farms are underwritten by the state, and 30% of solar farms. That public ownership rate will continue, and likely increase, as the deadline for the 50% RET draws nearer.
This gov has free reign until at least 2024, with no senate to reign it in. We’ll see a big public spending splurge shortly on wind and solar, plus the transmission necessary to connect it all.
Wholesale is negative more often, demand is not increasing. There is no incentive for private investment.
The existing public-owned generators will run at a loss, and the new RE will run on over-priced fixed price contracts. Subsidies both ways, plus ongoing subsidies for rooftop solar, plus public spending on transmission, plus distribution costs rising.
Working from the consumer back, all consumers want is reliable, available & affordable energy.
In the absence of empirical evidence proving the case against CO2, I believe all Australia therefore needs is a simple, easy to understand, sensible Energy Policy that’s;
• Technology neutral (eg fossils fuels, hydro, the ‘Unreliables’ of wind & solar, Bio-mass, batteries including the nuclear option – ALL on the table),
• NO subsidies ie a level playing field.
• Generators choose their own technology in a free, fully transparent, competitive market place where market forces prevail.
• Generators simply contract in advance (auction?) executing a firm commitment to deliver guaranteed power 100% 24h/day to meet forecast demand (by the likes of the AEMO) for the next day/week/month market needs at their BEST competitive price(s).
• Substantial financial penalties to apply for ANY generators failure to meet contractual commitments in accord with clearly defined 100% QOS standards (not withstanding force majeure in the unlikely event of natural disasters eg earth quakes, floods & bushfires etc).
If those T’s & C’s don’t appeal to the power generating industry, tough. Just revert the power generation industry back to whence it came ie respective State Governments & be done with it.
“If those T’s & C’s don’t appeal to the power generating industry, tough.”
But that’s the point. The problem is not technical but political because you’re outlining the level playing field we as consumers needed all along with this ‘energy transition’ as it’s euphemistically described. Instead we have State sponsored dumping by solar and wind driving out dispatchable coal with a Catch 22 situation evolving.
Normally the ACCC would be all over such market behaviour like a rash but they know their politics and hence their stony silence. To implement the level playing you propose is the only logical and rational position. However somebody has to break it to all the voters with rooftop solar they’ve been led up the garden path and their highly correlated and fickle reactive electrons are no longer wanted by their equally generating neighbours. If they don’t like it tough but that’s the tough political message no politician wants to be the bearer of. So we all continue ‘transitioning’.
What exactly are we transitioning to in the absence of telling it like it is? Well that’s continually trying to disprove a fundamental axiom of engineering. Namely that you can’t construct a reliable system from unreliable componentry and count me a skeptic that we’ll manage to disprove it.
a wealth of knowledge but the presentation and embedded logic is very hard to follow, for me at least. In providing the information, please don’t assume the reader knows all the background to your concise points.
I know a bunch of wind farms schedule their annual maintenance shutdowns at this time of year (in preparation for the more lucrative winter months I guess). Could explain some of the low overall output that you’ve seen recently.
Instead of “the super-secret financial support being provided by the Victorian Government to EnergyAustralia”, why not be above board and pay them to close coal fired stations down provided they use the funding to support the development of renewable energy?
It’s not a question of paying coal to close down but a question of timing as they’re going one by one anyway because the deck’s been stacked against any reinvestment returns. They’re simply running on sticky tape and string now to eke the last returns from their stranded assets. As they go it leaves a gaping hole with dispatchability and that’s one thing politicians understand. Or rather they understand the punters won’t be happy chilling in the dark with warming fridges etc.
Suddenly a unicorn Tesla big battery appears pronto to take the spotlight off the diesel gennys guzzling copious tankers full of refined fossil fuels. What that’s really exposing (along with backhanders to coal to stay afloat) is the dire lack of storage to make wind and solar dispatchable here and now. That’s the inherited problem can that’s being kicked down the road politically. No matter the clever technical folk at the AEMO have it all in hand as they can work miracles can’t they?
Hardly surprising. One of the most difficult engineering challenges is to wind down older coal fired plant ahead of retirement. It involves the application of much good engineering judgement to get right as there will be emerging risks known to those trying to operate the plant. Historic performance is not necessarily indicative of future performance where a particular issue may have been recently identified. No one operating with a commercial mindset is going to retire a pristine plant – maintenance spend is sensibly rationed at the expense of reliability as the plant winds down over several years. The trick is to do it safely. There are big bits of high pressure and temperature kit in a power station and some of it heavy bits spinning at high speed. A healthy dose of conservatism is justified.
The real risk here is that the economics of all coal units around Australia is being seriously damaged by renewables causing their rapidly declining capacity factors. That squeezes available maintenance money and that will inevitably play out as reduced reliability. Someone had better start thinking about how the market signals for firming capacity…..and soon.
FFS is not inertia but batteries and grid forming inverters can and do provide synthetic inertia. By far the biggest shortfall in all the ISP and related work is a clear understanding and plan for how to run the grid using inverters as will be required when coal generation goes away. It is disturbing that we haven’t done the appropriate engineering studies to plan for this not merely foreseeable but seemingly inevitable event.
FFS is not inertia but in some ways it is better. If inertia is thought of as stored energy then it can be shown that virtual or synthetic inertia can actually contribute much more energy per MW in two seconds than synchronous inertia can. If there is about 30 GW nameplate of gas hydro and coal turbines spinning they will have about 240,000 MWs of inertia If the minimum safe frequency on the gridis 49 Hz then available energy is 1-(49/50^2) x 240,000 MWs = 9,500 MWs and that is all.
In two seconds even allowing for a 140ms ramp time 5 GW of batteries can supply as much energy as 30GW of rotating machinery. However if the loss of generation takes more than than two seconds to recover the batteries go on supplying energy whereas the inertia is completely exhausted. In fact if the frequency stays low for more than 10 seconds or so then trips will start across the system. Thus some of the initial generator response is required to replace the inertia.
Even if a minute is allowed for frequency to return to say 49.5 Hz that means new supply has to supply 9,500/2/60 MWs = 80 MW for frequency restoration let alone replacement of lost supply to customers.
Wind and solar that are constrained can also rapidly increase output and inductances along the grid also provide a stabilisation. Hydro plants can be run dry to provide inertia and by 2028 Kidston and Snowy II will be providing another 8-10,000 MWs
If over as little as 5 years, half the coal plants departed the grid that would reduce inertia by about 100 MWs. From the above it can be seen that 2.5GW of batteries will do the job even without pumped hydro or virtual inertia on wind and solar farms.
As plans have been announced for some 7GW of grid batteries plus somewhere around 0.5-1GW of VPPs even if 1/3rd of those plans come to fruition, it is unlikely we will be short of “inertia”.whether real or virtual
This was a whole-of-mainland high price event, with interconnectors unconstrained between the four regions.
The $14,950/MWh price in DI 18:00 was being set by EA’s Jeeralang units in Victoria with plenty of spare capacity in other regions but offered at prices that were in effect higher than this level when translated to Victoria after accounting for inter-regional marginal loss factors.
Overall I think your comment that there was a reasonable amount of NEM-wide reserve (5,400 Available capacity over Total Demand) is fairly pertinent – this capacity was simply being offered at five digit prices, showing that bidding dynamics, in addition to physical supply demand factors, was a major driver for the event.
Trying to understand this a little more deeply. So there was capacity at interconnectors, and there was plenty (well enough) supply, but perhaps this supply was from a small number of participants who worked out they could set prices at high levels without being undercut? (I’m being careful here….)?
My laymen thoughts:
#1) Having less actual demand than the 30min prediction could lead to a skew to higher frequency.
#2) Having less demand makes it easier to keep the frequency above 50Hz and keep it close to 50Hz.
#3) The software control of inverters can change output without changing frequency. More generators using inverters (solar&wind) and less using inertia (coal,gas,hydro) could mean the supply minus demand differences are having less effect on frequency. Or overuse of faster FCAS.
#4) NT had a problem with the software & parameters of their grid connected mix of generators where the FCAS capacity was used too quickly, a lack of frequency reduction failed to properly signal correct load shedding, and a few other issues. The frequency increased during a time of demand>supply so some generation was curtailed until the overloaded battery supply could be sustained, this drop from batteries & the grid was unprepared for a sudden loss of battery et. al. generation. They now say they’ve fixed the control software glitches but it could happen again elsewhere in the country if these rare cases are not tested & searched for.
#5) Just because it looks better (tight around 50Hz) while everything is close to “average”, it doesn’t mean it will be better (follow same pattern to be close to 50Hz) during extreme demand & sudden changes compared to 2 years ago.
Thank you for your efforts to analyse the Australian energy market but unfortunately I’m not a big market player (just a little home user). Knowing my kWh from my megawatts, I hear these claims of green energy nirvana with scepticism. If I had the detail from AEMO etc. you (Global-roam/WattClarity) have I could work out modelling of the supply & demand requiring storage & increased supply (roundtrip losses when using storage) to make it work under different conditions/mixes (gas vs battery vs pumped hydro). During each day and between seasons.
Would you say electrical grids are still run within a minute of disaster and always have been? We would need distributed storage with megawatts to meet more than peak demand and with enough GWh storage to meet natural shortfalls & absorb the peaks (not as simple to calculate as it sounds, as Paul McArdle already knows). 100% green is not a problem of just technology and some of it’s cheaper but “whether or when we have the economics, materials & time to do it ALL when competing with the rest of the world for the same limited resources”.
I think you are trying to analyse this to death guys. The % variations you are seeing here is “noise”. The NEM PFR deadband is nominally +-15 mHz – that says the machines governor action does nothing until is falls outside that band. All very nice, but….that is just a crazy setting. Simply too tight and will wear governing systems out and on thermal machines send control system oscillations all the way back through the boiler. Hydro machines are even worse in this small signal area because of the mass of water in the turbine runner that needs to be either accelerated or decelerated. There is a lot more to tuning a governor than just deadband. Gain and reset action will determine how quickly they attempt to correct the error. To add to that not all machines will be in pristine condition all of the time , depending upon where they are in their maintenance cycle. This will mean their will be mechanical hysteresis in their systems also. So all I am saying is a few % here and there is to be expected. This distribution looks pretty good with an effective deadband of about +-50mHz – about as good as I would expect on a system where the dominant frequency regulator is large thermal machines. It might get better if there were lots of big inverter connected batteries around but we aren’t there yet. Generally on falling load I’d expect to see high frequency and rising load, low frequency. Operationally pretty normal and hugely better than before PFR was introduced as you are observing. Simply all machines are now obligated to contribute to correcting frequency(as I might add historically – pre markets – they always did). The other wild card is where an additional generator synchronises (or desynchronises) . Most need a block load or unload their last few MW very quickly when coming off to prevent chilling of metals from excessive throttling. All will represent frequency disturbance. Also, don’t forget that the inertia of the power system is not only the number of synchronous generators on at the time but also the number of induction motors driving loads in the power system at the time. It will therefore change as industry starts up and shuts down each day – the same time generators are loading and unloading in response. All of this modelling is what system modellers spend their lives doing using very powerful modelling software.
WA has an island grid known as the SWIS. SEN modelling for 90% RE on an hourly basis using 2014 weather data (a low year for wind out of the previous decade) shows us that while there are winter wind droughts that can strike for ten days in a fourteen day period, it is manageable and even with these wind droughts, storage and overbuild still can provide 90% affordably, and 100% for a little bit more cost (summer is not an issue due to PV).
I should add the wind correlation in WA wind farm locations on the SWIS is pretty high. Batteries and PHES really helps, but only goes so far (say 2 hours for batteries today, 4 hours in 5 years time and 10-14 hours for PHES in WA)
Why did the H24 Calvale 275Kv Transmission lines trip 22minutes after the fire event?
Significant power system event – 25/05/2021
AEMO ELECTRICITY MARKET NOTICE.
Significant power system event – 25/05/2021
At 1406 hrs 25/05/2021 a significant power system event occurred.
Description of event – Multiple transmission lines tripped
Transmission element(s) tripped – All 275kV transmission lines out of H24 Calvale
Region Qld
Updates will be provided as additional information becomes available.
Paul,
It may not be generally known that most modern generators are hydrogen cooled so it is possible that it was a hydrogen explosion that was reported as preceding the trip (I believe). I think that is unlikely as there were no casualties, though this may just be a matter of luck (everyone at smoko). Another source of explosion are pipes containing steam which hopefully would only have been a small line e.g. a drain or instrumentation line. Loss of an instrumentation line may effectively cause a false signal to the turbine control systems leading to a trip. All speculation of course at this stage but any explosion in a power station is a worry.
Looking at the diagram it must’ve got very close to splitting Queensland into two electrical islands. It will be interesting to find what caused the tripping of all circuits out of Calvale.
Exactly right Philip. Understandably it’s the incident at Callide C that has attracted a lot of attention, but it was the Calvale trip and its apparent impacts on other generators that had the most widespread effects. It’s not at all clear whether the Calvale trip was the direct result of what happened at Callide C or some other cause.
Hi Allan, Nice to see some details. In what looks a related incident, Wivenhoe was pumping at the time of the trip of the Callide units, and from 5 minute resolution it looks to me that that pumping load tripped pretty much around then. Perhaps that helped to stabilise that first trip. Maybe that was the low hanging fruit to arrest a frequency collapse.
I look forward to your further exploration of this event.
Cheers
Controversially one might question whether the repair cost is justified in a 50% and going higher renewable generation world. If, as is usually the case, insurance pays then I guess it’s a non issue, but Callie C unit 4 replacement will confirm the death sentence of some other no doubt less efficient unit.
The problem is that coal loving QLD is no where near 50% RE and won’t be for a long time. Indeed, QLD is struggling to maintain their 20% RE target with all the VRE curtailment that keeps occurring in the State. This failure does allow the QLD Government a bit of a politically safe out in terms of closing a unit that isn’t really required. In terms of insurance, I would be very surprised if the failure would be fully insured as there is typically a significant “excess” to be paid. Given the decision to repair the unit will be a political one, the economics of the situation don’t greatly come into the decision, even though they should.
Those measurements certainly point to loss of synchronization (est. about 3.8 cycles different at the point of QLD nadir, based on pixel counting your chart). 4 second data also record zero flows on QNI. It would be interesting to know what was going on with voltages, in relation to the transmission line trips. Also loss of synchronization between NSW1 and QLD1 may have been preceded by collapsing voltage on the NSW side, triggering UVLS rather than UFLS.
The AEMO report on the “double islanding” event notes changes were made to the protection scheme “The protection equipment on the Dumaresq – Bulli Creek 330 kV lines operated as designed, although the protection design did not cater for the circumstances of this event. The auto-reclose control for the loss of a single phase of both circuits required a successful synchronism check before reclosing the lines after the fault had cleared. Powerlink and TransGrid have since modified the control logic by reducing the reclose time and not requiring the synchronism check prior to re-closing the CBs.”
Earth lasers’ plasma shield CAN prevent a devastating global blackout/all nuclear plants’ blasting by asteroid explosion (as in Tunguska-1908 & Chelyabinsk-2013) or solar storm hit! NINE times near-miss extinction so far: 1972, 1989, 2003, 2006, 2012, 2017, 2019, 2020, 2021 https://GlobalBlackoutPrevention.wordpress.com
I remember the islanding that had the QLD generators being given AGC set points based on NSW frequency, while trying to deal with local frequency going in the opposite direction.
AGC probably wasn’t an issue here as the two systems appear to have synchronised and reconnected within 15-20 seconds, which is a remarkable outcome – fantastic recovery of frequency in Qld after the under-frequency load shedding, and some very clever controls on QNI. The recovery in Qld must have largely been due to primary frequency response from the Queensland units remaining online. While lots of things went wrong in this event, some very important things went right.
This is a very interesting graph. On your chart of reserve margins, is it possible to point to:
a) when actual blackouts have occurred due to supply shortages
b) generator closures
c) droughts
Also, it’s intuitive that all reserve margin is dispatchable generation – is that a correct assumption?
Also also 😂, how does the reserve margin correlate to inertia levels, or AEMO system strength constraints?
I had a look for diagrams and from an old AEMO incident report it seems that Calvale is a breaker and a half substation, so even a bus fault or a CB fail shouldn’t have resulted in the loss of all circuits from Calvale.
If the media buzz is to be believed, and it is a turbine rotor ripping out through the casing, the parts alone to fix it are the better part of a year to source. And it’ll be weeks (at a minimum) before there’s enough failure investigation to understand precisely what needs to be ordered, and whether you can simply do ‘like for like’, or do you have a design failure (and then what happens with the sister units C3 and Tarong North). And even within all of that, the insurance claim is going to be pretty messy.
Two statements which stand out to me as defining the event.
“At 1334 hrs on 25 May 2021, Callide C4 stopped generating but did not disconnect from the power system.”
“At 1340 hrs, CS informed AEMO of a possible turbine hall fire at Callide C.”
The Toshiba technology is simply a GE licensed design. And there are a lot of people with that license. And specifically, this turbine is simply rated for a temperature and pressure, and the forging capability would be the same as any other thermal (like solar thermal, or a garbage burner, or . . .) turbo-generator.
Well said Jeremy. The article actually says they are stopping coal fired power plant sales, not turbine sales. Toshiba used to do boilers also. There is a huge fleet of Toshiba machines out there that will need to be supported for many years to come, so bashing out another one is no issue. And if they won’t Doosan will – another GE licensee. The big issue will be lead time. If they have wound back their order book they wont have an inventory of big forgings to make shafts out of – and they need at least one from the bit lying on the floor. I suspect there will some very sad insurance underwriters just now. There are a lot of people commenting on this that really have no idea the step by step process to sorting this out and that it is going to take quite a long time. Having been directly involved in the rehabilitation of a lesser magnitude, near death experience with a machine many years ago I can but express my sympathies to the people involved and the no doubt sleepless nights and long hours this is causing them.
Paul Paul Paul, when you say ‘the transition has one or two decades to go”, you imply an end to The Transition!
We all know there is no end, because who in their right mind thinks we’ll be powered by intermittent dispersed low density low efficiency generation?
There is no “one or two decades of transition” with wind and solar – it’s a transition in perpetuity. Transition implies an end point. You’ve assumed 20yrs.
But how many wind turbine sites are refurbished with new turbines after 20yrs? How many solar farms? Batteries?
A sane transition would be from one technology to another technology that delivers improvements in the major facets of energy delivery:
– density
– efficiency
– cost
– footprint
– mining (number of holes in the ground)
– waste
– pollution
– manufacturing, refining, transport
– transmission
Wind and solar improve very few of these facets of energy delivery, and actually make some worse.
Ben, Ben, Ben, [kiddo], since you did not care to provide any numbers or analysis to support your assertions about what matters and how technologies compare, perhaps I can help you out. I’ll do one, then you do one.
Let’s take density/footprint, and compare Liddell [1GW] with the Forest Wind project [1.2GW]
Liddell itself, not accounting for the coal mines, etc takes up about 2 km square, which isn’t usable for any other purpose. Forest Wind’s footprint will consist of 226 30-m circular foundations, adding up to 0.16 km square. All access roads already exist. There will be temporary extra space required for construction, but that can go back to forest afterwards. See https://www.forestwind.com.au/project-overview Therefore this wind farm, conservatively, is FIFTEEN TIMES more footprint efficient than a coal fired thermoelectric power station. This is generally the case for wind farms – while they are spread over a large area, their footprint is small, and the ground space in between is generally fully usable for farming or conservation.
Your turn.
Thanks Allan for dissecting the elephant’s toe nails. This consumer advocate found it fascinating since 5 minute settlement is rapidly approaching. Volatility in wholesale prices can flow through to retail power bills. Since QEUN’s April 2021 survey of over 200 businesses in the Cairns region found over 60% had a medium concern about the payment of their power bills (20% had a very large concern), then are consumers already struggling with power bills about to struggle with fluctuating power bills caused by volatile wholesale market?
Fair question Jennifer. As you know wholesale energy costs are a proportion of consumer bills, and that proportion depends quite a bit on location (network costs), customer size, retail on-costs etc. Most customers won’t be directly exposed to short-term spot market volatility but trends in wholesale prices flow through to forward contracts and to retail pricing with a time lag – the length of this lag can be quite appreciable, again depending on customer specifics. The first post in this little series (https://wattclarity.com.au//articles/2021/06/eating-the-nems-q2-elephant-an-appetizer/) has some views of the impact of recent volatility on average and forward contract prices. For Queensland, average and forward prices are now well above where they were six or even twelve months ago, but still well below the levels prevailing over most of 2017 – 2019. The energy price component of consumer bills for many customers, eg those on tariffs set ~12 months ago would still reflect some impact of the high prices prevailing in that period so I don’t think the current volatility by itself should lead to a *jump* in that component, but obviously if it extends into subsequent quarters that could change. What’s probable is that is that signficant falls in that component which seemed on the cards just a few months ago are now less likely.
The photos of the site show a catastrophic failure starting at the HP end with the steam pushing the I
P stage back into the LP stage and alternator shearing the main shaft that was ultimately spat out. These turbines have multiple temperature and vibration sensors to shut them down if things go pear shaped. Do we know why the shutdown failed
Thanks Allan. If volatility increases then this could be priced into retail offers which could be more problematic for small retailers with small generation portfolios, this could impact retail competition. With the COVID19 pandemic continuing and no real idea as to when the international border will open, many businesses who are at present benefiting from lower wholesale/retail prices will have little ability to suddenly pay more if volatility is priced into retail offers in the future. Volatility is also enhanced by Jurisdictional Scheme Charges – a significant charge (eg 3.5%) that can without warning be imposed on consumers’ power bills. The COVID19 recovery of small business – the largest employer in Australia – is fragile eg Reserve Bank of Australia’s May 2021 charts has lending to small business as flat lining and business investment is continuing to fall despite the extension of generous tax write-offs (agriculture is bucking the trend). Therefore downward wholesale/retail prices are most welcome but a sudden spike or a consistent rise in wholesale prices would impede the COVID19 recovery. Since energy costs to over 700,000 business and residential customers in regional Queensland represent 35% of a power bill and 75% of the energy cost is the wholesale price, we are very interested to see if 5 minute settlement will increase volatility and therefore be priced into retail offers.
Hi people,
A BESS provides much faster response to system events then tradition gen sets, as there is more delay in goveners response and engine speed pick up etc.
A BESS will determine a frequency change in 10ms has it clocks the time a a voltage signal crossing the zero point, which occurs ever 10ms for 50Hz, plus a small processing delay, so realistically say around 20ms response. On the system we see response in under 200ms, which stops a blackout event. This seen on small systems, less than 10MW, using an ABB BESS.
Through the grapevine – in other words just a rumour. The destroyed unit did not have a self starting capability and so the generator could also function as a motor by feeding back off the grid. It has been suggested that a lubrication failure one end was fighting a generator that was acting as a motor due to a software/hardware failure allowing the grid to feed back into it.
The Network Outage Schedule from the AEMO website yesterday shows outages of the Callide C- Calvale 275 kV lines 853 and 854 and the associated circuit breakers at Calvale from 25th May until 7th June.
With solar FIT rates dropping continuously with a glut of rooftop solar and no forseeable economic case for home batteries there’s only one viable option for the little people going forward- https://www.solarquotes.com.au/hot-water/pv-diverters/
They might also store solar energy in the home with the right gear- https://www.solarquotes.com.au/blog/divert-solar-air-conditioning/
Nothing surer than the demise of coal coupled with all those prospective EVs charging at night peak grid power rates will be after sundown.
These winter peaks will only rise as more houses, and to a lesser extent commercial buildings, switch to reverse cycle air conditioning for their primary source of heating. This will be most obvious in Victoria.
Equally important is the completion date of the 800 MW EnergyConnect between NSW and SA.
It seems that within a 12M period the following occurs:
– NSW loses 1000 MW of 24/7 baseload from Liddell
– SA loses 600 MW (?) of dispatchable gas generation from Torrens
– NSW sees 800 MW of new load from SA via EnergyConnect
With NSW the most connected pricing region in the NEM, any price increases in NSW will spread to the other states.
The future looks good for NSW coal, gas and hydro in the medium term.
Longer term it’s likely a few more semi-scheduled RE power stations get built as wholesale prices are buoyed by gas and hydro (note the recent government intervention in these technologies). But I’m still doubtful too many will be built without state government contracts.
The growth of rooftop solar will certainly suppress daytime prices until FCAS costs overtake wholesale costs.
That clearly sets the scene for batteries to make money from arbitrage and FFR, hence the interest and hype.
Interestingly the market for batteries and pumped hydro pretty much destroys any market for green hydrogen production using low-cost ‘spilled’ power.
Given the extent of the damage even that is optimistic. Big question why did the unit’s many safety features not shut it down. The only explanation i can offer is that the high pressure inlet valve failed. Has anyone got more information or a better theory
Hard to comprehend why they’d bother fitting a new turbine when coal has to go. Unless of course transitioning gas is really setting power prices and they’re expected to keep on rising.
The value of electricity always lies with demander/consumer. Up to any free entrant supplier to decide if they can compete with other like suppliers to meet that value with whatever costs they face. But the complexity lies in sum of consumer values at different times and likewise the costs of the various suppliers at those same times including frequency and voltage of said power of course.
Too much autocorrelation of values between consumers and costs between suppliers within each sector but they’re way out of sync with each other and hence those volatile prices reflecting that.
Does this address Villian #8 for both scheduled and semi-generators?
4.1.1 Publishing individual unit available capacity and PASA availability
The proposed rule includes a new requirement for AEMO to publish the available capacity and PASA availability of individual scheduled and semi-scheduled generating units20, scheduled loads, scheduled network services and wholesale demand response units.
It clearly opens up opportunities for retiming hot water, demand response and thermal storage if not batteries. Wind developers will have to think about areas that are out of sync with the bulk of existing wind farms rather than maximum annual energy production Gippsland comes to mind, offfhsore or nearshore in southern NSW. Some existing wind farms may be repowered with taller towers and larger rotors without increasing nominal capacity but significantly increasing output in low wind conditions.
The 5 minute rule may also reduce the incentive to game the system at the high price end whether it will help at low prices is another mater.
Both C3 and C4 have some shared and/or linked systems (fire systems among others). Fully segregating those systems so that C4 can have carte blanche for the next 18 months is never easy . . . essentially going back to a state last properly seen in the late 90s, when the two units were being built.
Gullen Range, Collector, Biala, and Crookwell 2 wind farms all dove to zero at the start of the price spike. So did Gullen Solar Farm. A total of about 300 MW whilst there was a NEGRES constraint on the VIC-NSW interconnector. Nearby wind farms Cullerin Range and Gunning did not change their output. The DUIDs that went off line are all on a pair of 330 kV Transgrid transmission lines to/from Yass BSP. Maybe something happened at Yass BSP?
Same thing just happened again at 18:00 on Thursday 16 July, only this time the price spike happened in NSW and also in QLD. Again those same windfarms went to zero watts.
Another dip for the NSW southern highlands wind farms at 19:00 this evening. 500 MW in total. Is Transgrid testing SmartWires as part of the VNI upgrade? Has it something to do with the commissioning of Stockdill BSP?
“ We need a federal package to help rollout this infrastructure because nobody wants to handle the risk. Really this is the only successful way we’ve done nation building infrastructure projects before. All the historical generation zones relied on government funding to build out several redundant lines at once. The model used in the past has been to build it and socialise the costs later once the benefits are established. The states can’t do it alone, and most of the value of new transmission is diffused across the entire NEM not just in local pockets.”
Tom, I enjoyed this write up, it is largely measured and sane. However the paragraph quoted above misses a couple of important points.
You mentioned risk, and later you mentioned ‘white elephants’. But transmission infrastructure receive guaranteed returns based on the value of the asset, regardless of its utilisation. The only risk appears to be changes to the regulated rate of return.
You mentioned that past practise has been for governments to build the transmission and socialise the cost. That was true when it was necessary for governments to build this infrastructure. That was the case as electricity demand was increasing and remote locations were connected to the grid. But in most regions the industry is now privatised, and electricity demand is not increasing.
You mentioned ‘nation building’. It is incorrect to label transmission projects, that do not connect new locations or large projects as ‘nation building’. Expanding the size and cost of infrastructure, for no increased demand, lowers the utilisation rate of that infrastructure – this is a net reduction in efficiency of the overall system. Reliability and availability can reduce utilisation, but that’s far removed from building infrastructure purely to connect a preferred (by some) technology for reasons unrelated to supplying electricity (emissions reduction).
I agree on the direction of your analysis, which argues for an multi-faceted assessment of transmissions n investment. Good job.
Orchestrated criticism of anything that might delay renewable nirvana? No way!
The fact that an organisation titled ‘Energy Security Board’ even exists is a living example of how the market has been manipulated to achieve poor outcomes.
The options paper described how the market redesign would incorporate subsidies and targets for renewables, as if this is now ‘the new normal’… a phrase I’m beginning to dislike!
I cannot see how the NEM can achieve the best outcome for customers when the market design is becoming more and more a central planning behemoth, catering to lobby groups and political whimsy.
I think Ben underestimates the professionalism of the team that Audrey Zibelman recruited before she handed over to another proven performer in clean Energy implementation.
Nothing too surprising here. Metal fires are extremely difficult to extinguish. They can’t be smothered because they produce their own oxygen and hence self sustain. All you can do is either deluge them with enough water to cool them to a point where the temperatures fall below that necessary for auto ignition (a pretty big ask for such a densely packed module) or let it burn out – which is what is happening here. Simple message – real estate is cheap in Australia – put space around each module to prevent spread from the radiant heat. So next time someone suggests putting one of these things inside your house…..show them the photo. If something (anything) goes wrong simply the energy release is rapid and disproportionately high compared with usual combustible materials and your house will catch fire. By all means have your battery, just don’t put it in or near your (or anyone else’s) house.
Regarding the home battery situation, I recall that a draft Australian standard recommended home batteries be installed in a bunker style housing, separate to the house. The RE lobby got that removed.
On these batteries, in general the strategy will be to let the bad one burn out and try to keep the rest cool. Makes you think about the risk of having them scattered throughout the distribution network, in suburbs…
It would appear that fire barriers between the individual batteries may be worth considering. From the photographs it appears that only one battery caught fire and then damaged the adjacent unit. Substation transformers present the same problem, they can occasionally catch fire or even explode. As a result they are separated by blast walls so that the damage cannot spread to adjacent units
There is a bit tension as to the safer form of lithium battery technology- https://www.energy-storage.news/news/lfp-vs-nmc-not-a-silver-bullet-solution-for-safety
Although fixed battery providers like Redflow may have a point that if transport is to be batteryfied it won’t make much sense to use up light weight lithium for fixed battery backup for the grid. OTOH increasing demand for lithium resources should make that clear fairly shortly via the price mechanism with the rush to EVs.
Medupi: If air was admitted whilst there was still hydrogen in the alternator then it was a pressure vessel and thus an internal explosion has occurred. Hence the explosive damage to the alternator.
Callide: The unit was in service, thus it would have contained pure hydrogen. Failure of the seal oil system would have allowed an external explosion when the escaping hydrogen mixed with air. Hence the alternator in the photo is intact but shows signs of burning at both ends.
Medupi: The Damage is confined to the Generator and surroundings. The end covers are blown OFF and Excitation Gear is damaged but since the unit was OFF LINE and stationery there appears to be little damage to the LP Turbines and the structure of the Turbogenerator set is largely intact. The generator is probably fixable because it was not running. There is quite a bit of Building damage, more than at Callide C4, mainly because the Generator is approximately twice the size.
In comparison with Callide C4 where it is clear that the shaft is destroyed and the turbines severely damaged seems to indicate that the Callide unit had a catastrophic Turbine failure followed by a hydrogen explosion after the Generator Casing was ruptured.
The problem was they basically ran out of capacity – everything that was available was on and full load – the only plant left to come on was a cold coal fired unit or a cold CCGT. Can’t start those up in the hour’s warning.
The generation data shows the windpower significantly dropped in the period 1500 to 16:30. That and weed blocking the screens of a hydro just after 18:05 seemed to have tipped it over the edge.
Was there a hydrogen explosion at Callide at all? Or did the GE of the LP turbine destroy itself during the motoring, possibly without lube oil? The damage done is not what one would expect from a hydrogen fire
Probably a lube oil fire and hydrogen explosion due to leaks following turbine damage. From the damage at each end of the LP cylinder and holes in the roof, I would have said an it was an overspeed causing the turbine last stage blades to fail but there don’t seem to be holes in the LP casing. Maybe there was a loss of lube oil while the generator remained synchronized as you have suggested and this caused the turbine to want to stop while the generator wanted to keep it going and this led to the shaft rupturing?
Since the mid 1990’s NZ population has grown from 3.3M to today’s 5.2M. There has been no new generating capacity, while Watermelon politicians press for greater intermittency of supply and demand the populous drive a ‘trick car or walk (except them), and the NIMBYs block the development of new generating resources. 60% of NZ power (once all domestic consumption) derived from renewables, chiefly hyrdo. NZ has UNEP divested itself of gas and coal investment and exploration, and is currently bearing down on one of its main sources of revenue, agriculture. The virtue signalling Watermelons doom NZ to a future of division, destitution, depopulation, deindustrialisation, delinquency, despair … and hunger.
NZ wind generation accounts for an intermittent, unreliable meager 5%.
Hey Paul – thanks for this. Do you know (or can you explain the difference between) “Operational Demand” and “Demand and nonschedgen”. I actually thought they were analogues of each other – but checking out the data, that doesn’t appear to be the case! Cheers, Mick
Mick – there are many wrinkles to demand definitions, even the ones AEMO uses. If you hunt around you might find exactly how AEMO measures Operational Demand. I think the biggest difference between Operational Demand and the MMS DEMAND_AND_NONSCHEDGEN field is that the latter includes supply to Scheduled Loads (pumps and batteries) but Operational Demand excludes them. Allan.
The published information on the motoring shows firstly that the machine did not overspeed and secondly, there was major resistance to the turning at grid speed. It is very unlikely it could have motored if there had been a hydrogen explosion. And it seems to have motored until the line was disconnected at the substation.
There is other information around on primary and secondary causes but nothing formal.
All that has been published so far hints at operation as an induction motor. Unofficial speculation hints at loss of steam, excitation, lubricating oil, and power to protection systems occuring simultaneously.
Windage loss will heat the turbine, circulating currents heat the generator, and loss of oil destroys the bearings.
Some big issues coming, no doubt. I wonder at the priorities of the ‘influential’ people, and what’s more important to them – more RE no matter what, or a working system… because we are fast approaching disaster.
We currently have an energy ‘security’ board – because clearly the market rule maker (AEMC), rule regulator (AER) and operator (AEMO) aren’t capable of maintaining the ‘security’ of the system.
It’s not that difficult.
1. Restrict the semi-scheduled category to plant below 30 MW nameplate capacity.
2. Limit PPAs to 20% of a plant’s annual output.
3. Constrain rooftop solar to consumption only (unless a battery is configured to complement the grid)
One possible reason for the delay in coming up with something clear and meaningful is that the task they have set for themselves, or the task that they are expected to perform, is impossible.
They are supposed to come up with a plan to proceed towards net zero with all possible speed by building more RE capacity plus [think of a number] GW of dispatchable firming capacity from batteries, gas and pumped hydro. Can we get enough gas? believe pumped hydro when you see it and forget about batteries for getting through windless nights (do the arithmetic).
We are supposed to end up with less emissions, cheaper power and a more reliable grid. Tell that to the Germans.
And don’t mention the environmental impact, the impending avalanche of toxic waste and the extra reliable power required for the great leap forward into electric vehicles.
But don’t laugh, this is serious!
Coal has to go, one way or another, although the Government it also has to stay, one way or another, as long as necessary. How long as that?
With the right Energy Policy in place, I don’t care whether Generators elect to use; wind, water wheel, tidal power, solar-PV, Bio-fuels, fossil fuels, batteries, nuclear, or even pedal power, if that’s what turns them on.
Provided they don’t expect to receive a subsidy from taxpayers to underwrite their fantasy.
In brief, I’m of the view that all Australia, and or any other country, needs is a sensible Energy Policy in keeping with market forces that;
• Is technology agnostic;
• Has clearly defined QOS standards eg 24/7 power available, reliable & affordable;
• Facilitates / encourages strong competition between generators to deliver their best price per GWh for guaranteed power output eg 100% (force majeure permitting);
• Provides no subsidies & repeals past legislation that provides for same (eg in Australia the RET, LRET, Safeguard Mechanism); and
• Importantly, imposes substantial financial penalties for any failure to meet contracted QOS obligations for an essential service.
Hi Paul: Two years ago I prepared a report for a State Government and highlighted the issue of what I called strategic energy capacity when %VRE exceeds ~ 90%. Also the fact that wind energy provides greater value than solar energy because it generates after dark – but still there are times of low supply. The 200 page report was about challenges for the power system transitioning to 2050.
The other thing I pointed out (relating to this article) is is that economic transmission capacity will still provide significant constraints in Australia with a large network (km) to be able to share VRE capacity across the distances. Simply the fact that 1000km of DDCT 330kV transmission line costs over $2.5Bn (800MW per circuit). Even if you take $2.5Bn/1600MW the unit cost is about $1.5M per MW. Hence you can build local wind and solar excess capacity at less cost. Strategic energy capacity would also be less than this taking into account it is say 96% reliable compared with remote generation from distant VRE which may also be low hence less reliable – your data will provide me some insight into this level of reliability over the quarter for consideration of this future issue.
Many thanks for such great articles in WattClarity!!
On the wind output record, that might require some pretty high demand to turn up – otherwise output will be limited by local demand + export capacity less required synchronous generation, as well as wind self-curtailing on negative spot prices. So the record may last a few months yet.
will be interesting to see to what extent directions on GPG reduce. I’m interested to note that “no changes to the SA generator combinations will be made at this stage”. This implies that there won’t be a significant reduction at this point, or have i misunderstood?
And if so, is this just because AEMO/electranet are in testing phase and changes will be made in due course when testing is complete?
Because otherwise the syncons don’t seem like great value…
The plan as announced is to allow the network to run with the machines for several weeks to monitor, then make adjustments to constraints (you are here) and monitor performance for a couple of weeks then repeat until fully commissioned. Updating the minimum synchronous generation constraints is the final stage of the plan.
AEMO’s program (here) is to reduce the minimum synchronous requirements (to 2 x TIPS B or equivalent) about 6-12 weeks after commissioning of the first two syncons, which occurred a couple of weeks ago. So all going well in this interim period, we’ll see reduced gas directions in around 4-8 weeks.
An easy-to-miss rider in the AEMO notice: the current limit will apply for Davenport synchronous condenser or 275kV line outages – there’s one such line outage running until the evening of Sep 14 so the old limit effectively remains in place until then.
Great work as usual. Suggest you may want to do a piece soon comparing how well different ageing plants are going at developing more flexibility under these conditions. In particular which ones are successfully getting to the lowest mingens for extended periods.
On 16 December 2020 Watt Clarity with the permission of David Leitch and Ben Willacy reproduced their article “Wind and Solar transactions: what to buy in the new year sales”. The article contained a table which specified the type of revenue (Power Purchase Agreement or merchant). It further stated the remaining lifetime of the contract based on a notional life of 25 years for wind and solar farms. Would be great to see this table updated together with what public companies are reporting as their asset life. One of the major investors in Australia in its Interim 2021 Results on 30 July 2021 reported “Depreciation and amortization expense was almost stable (up €0.2 million), as the mechanical increase reflecting growth in assets in operation was offset by the positive impact of extending the useful life of wind and solar farms from 25 to 30 years. Are other public companies reporting a 30 year asset life? AEMO use an economic life of 25 years (period of time required for the project to recover build costs) and a technical life of 30 years (period of time the project is operational before life extension refurbishment or retirement).In AEMO’s 2020 Inputs and Assumptions Workbook the retirement costs of single axis tracking solar farm was $21,7366/MW and a wind farm $10,883/MW (real 2020 $). AEMO did not specify retirement costs for solar and wind farms in its 2021 Inputs and Assumptions Workbook but the workbook did update retirement costs for coal and gas generators. The first wind farm in Queensland was built in 2020 and AEMO has it retiring in 2025 – its a baby compared to today’s wind farms ie 20 turbines with a capacity of 12 MW.
Why does MT PASA only appear to apply to generation plant and not to transmission? Many of the times of high prices are because of interstate transmission outages, much of which is planned months ahead but doesn’t seem to be visible to market partipants. There are also intrastate transmission limitations that restrict how some dispatchable plant can bid or contribute to a capacity market. Until Humelink is built there are limitations on how Lower Tumut (1800 MW), Uranquinty (664 MW) and the Maoneng battery (200 MW) can access the Sydney market. How would a capacity market allocate the limited transmission capacity to each of these generators?
The Network Outage Schedule is available on the AEMO website covering a period of 13 months in advance. Search on the AEMO website for “Network Outage Schedule” and “Planned Electricity Network Outages”. The MT (and ST) PASA processes use a model of the entire NEM including generation, demand and transmission. AEMO uses information from the Network Outage Schedule as an input to the PASA process.
On the flip-side of this article, what is the forecast peak load for SA this coming summer? Noting peak during a heat wave will happen at about 7pm when solar is gone and it is potentially still and calm weather across SA. Without NPS, Playford and now all of TIPSA (so minus some 1260 MW), does SA have enough conventional generation plus interconnection capacity (assuming there are no problems elsewhere like in Western Vic and something can come through from Murraylink)? Has any shortfall been flagged by AEMO?
Since most wind and solar power stations are contracted, and any uncontracted gets REC, I expect the headroom necessary to increase output on demand would usually cost more than the available FCAS income. The lost income from capacity not earning will add up over time. Just ask any coal-fired power generator…
“Trying to understand how the current rules are legal in a country with strong consumer protection laws does my head in.” In theory only one extra regulations should be required as the existing consumer laws do cover all these aspects. The problem is the need for the laws to be enforced. The ACCC typically won’t get involved in small or minor cases. This is supposed to be handled by the under resourced state consumer affairs bodies. Solar PV systems are a complex system with variable output depending on the conditions. This makes it difficult for an educated customer determine if a system is performing as specified. The vast majority of customers don’t fall into that category.
The one area that current laws don’t cover is the phoenixing of businesses or operators.
Even even 20-storey blocks of home units are turning up with fatal construction faults. It is stretching credibility to say that we live in “a country with strong consumer protection laws”.
This seems to me to be eminently sensible that all power plants including wind and solar should run with a bit of headroom to cope with system disturbances. On a 5 minute dispatch cycle the headroom could be provided by capacitors or flywheels at much lower cost than batteries. That would be probably be more financially acceptable than limiting production. Perhaps Synthetic Inertia or simply overrating generator capacity compared to the connection will also help.
If all generators do have some near instantaneous reserve then the cost will be widely spread and therefore have little impact o any generator
These changes are long overdue and most welcome. However, the predatory signing up practices are as big an issue as the quality of a solar system. Consumers also need to know that solar systems are not set and forget ie all solar systems need regular maintenance in a way that does not void their warranty… this should be explained before a customer is signed up.
And directly related to this article, what happens when SA eventually occasionally/more regularly drops into negative demand territory due to the amount of non-scheduled embedded solar PV and then SA islands? How is SA run by AEMO as a stable island?
It is well overdue that someone publicly identifies that the option that “states go it alone” is a real possibility.
One of the 49 recommendations accepted as part of the 2017 Dr Finkel led Review into the Future Security of the National Electricity Market – Blueprint for the Future was Recommendation 7.3 which stipulated that:
By mid-2018, COAG leaders should agree to a new Australian Energy Market Agreement that recommits all parties to:
• Taking a nationally consistent approach to energy policy that recognises Australia’s
commitment in Paris to reduce emissions and governments’ commitment to align efforts
to meet this target with energy market frameworks.
• Notifying the COAG Energy Council if they propose to take a unilateral action that falls
within the scope of Australian Energy Market Agreement prior to taking the action.
• Within 28 days of notification, the Energy Security Board will provide advice to the
COAG Energy Council on the impacts of the proposed action taking into account the
objectives of Australian Energy Market Agreement.
Realise COAG Energy Council has been rebadged but it is effectively the same entity without the open communication to stakeholders. Maybe COAG members are using the rebadging as an excuse for missing by 3 years the mid-2018 timeframe for agreeing to a new Australian Energy Market Agreement. States going it alone is a real possibility.
This is a very thorough report almost as thorough as the ISP and coming to similar conclusions particularly if one uses the ISP step change scenario, which is actually my base case.
When ITK started formally forecasting electricity prices a few years ago and said our modelling was different to other approaches in that we just assumed there would be enough policy to get to 50% renewable by 2030 it wasn’t taken all that seriously. And our modelling probably still isn’t although I take it seriously myself. But the point is 50% renewable by 2030 (which includes hydro) is now pretty much base case.
The forecasts in these models are done to meet the reliability standard, and yet they produce reasonable electricity prices and none of them suggest you need nuclear or anything similar. That is one reason why I am so supportive of Matt Kean and the NSW approach. It is based on sound modelling and a proper business and execution approach. And it is done in the open with feedback and consultation at every stage.
Of course as every epidemiologist and many others will point out, modelling is a very error prone process and any good policy maker will not rely solely on models . Still it has to be done and its great to see this effort from Transgrid.
The problem with optimistic projections of RE penetration is that we are only just approaching the point where it matters whether the sun shines or the wind blows. When we lose some more conventional capacity the problem of windless nights will become real and we will have to confront the full cost of firming that has been carried by the coal power stations up to date. See the story of the frog and the scorpion!
Is any work being done by anyone to determine how many MW/MW.h of batteries can be connected in e.g. SA before they eat their own arbitrage lunch, eradicating the very pool price volatility that drove their investment decision in the first place i.e. through there being too many connected all trying to draw a load when there is no load (and when pool prices are/were generally lower) and conversely all trying to be generators when generation is scarce (and pool prices are/were generally higher)?
Didn’t notice the power filter function! By the way, using all volumes, effective as-bid availability showed the change in bid stack from solar and wind plants in a clear way too.
Nice that this article helped you discover that one, Adrian. Remember the individualised ‘Widget Guide’ accessible under the ‘?’ button in the top right of the widget might provide other ideas for how to use that widget!
Not so wild in SA it turned out. High prices only due to lack of reserve in QLD in the afternoon. Perhaps some maintenance work was delayed by weather?
Did they start synthesizing 5 minute offers from 30 minutes, or start accepting 5 minute offers around these times? Are the sensitivities part of the same file, or separate? I believe they are incorporating FSIP into the P5min but I’m not sure that has occurred as yet – would require an additional solve per interval.
Any idea how much the QLD tax payers lost on Saturday due to the poor management of the operation of the QLD Government owned generators?
Without knowing how much of that power is under contract it is difficult to determine, but operating at negative $1,000 per MWh for hours at a time must have cost millions of dollars.
It is actually a very clear case of market power abuse, with the near complete removal of all competing generators priced out of the market.
We knew that the intermittent providers can displace coal but they can’t replace it. SA was importing coal power until after breakfast and it is importing now, thanks to the efforts of the much maligned brown coalers that are running over 90% of capacity while the wind is delivering 2% of Victorian generation.
Would it not be more profitable for the State-owned generation companies to shut down one or more coal generation units until demand picks up in mid summer, and sell the coal on the spot market, which is now extremely high?
What is the point of achieving that amount of penetration for a short time on the weekend? When the sun goes down and the wind fails, as it often does for hours at a time, SA depends on local gas and coal power from Victoria.
A chain is only as strong as its weakest link and the effectiveness of a flood protection barrier is limited by the lowest point, regardless of the high parts. Windless nights are the weak link in the chain, like the gap in the levee.
How hard is that to grasp? Don’t mention pumped hydro storage, when Snowy2.0 comes up at immense financial and environmental cost it will only partly replace Liddell. Big batteries to cover windless nights? You have got to be joking!
Rafe, I don’t think the flood protection barrier analogy works. In that case, if the flood exceeds the lowest parts of the levee, everything on the other side is compromised: a literal tipping point. So, as you say, the only thing that matters is how high the lowest points are.
That doesn’t apply here. Occasionally resorting to gas or even coal is still a huge improvement over constantly using it. “What is the point?” Not burning all that gas when the wind *is* blowing or the sun *is* shining. That’s the point.
Your analogy would only make sense if the entire system were to somehow collapse if gas or coal were ever used, which isn’t the case.
More importantly, what has it cost us in higher electricity costs by not implementing 5 minute settlement earlier? I don’t just mean the delayed implementation, I mean that it should have been brought in at least 5 years earlier than that.
We should be looking at what it is costing us by not moving to 1 minute settlement. I don’t think we are there yet, but it needs to be watched. The NEM grid now has the ability to convert almost all of these 5 minute price spikes, to 1 minute price spikes with grid scale batteries.
That only leaves the political bunfight over carbon taxing penalties for fossil fuels but now we have the worst of both worlds the AEMC has to grapple with. What’s more with the switch to EVs for transport we’re seeing scarce light-weight lithium battery resources being squandered on fixed batteries for temporary grid firming as that’s where the cream is. I wish the Regulators well with it all when we really need rational statesmanship.
Fortunately most very low wind events happen during the day when as you say solar is increasingly covering them. There are some occurring at night which leave a 15 GW gap between hydro + wind and net demand so even if wind is quadrupled over the next 10 years and 3-4 GW of pumped hydro comes on line there is still a fair gap to fill. However moving a lot of non time sensitive demand from nighttime offpeak tariffs which were designed to keep coal plants going to daytime solar sponge tariffs will fill a substantial part of that gap
The era of fixed times for non-time sensitive loads should be dead and buried. The cost of smart control systems are low now and we should be taking advantage of this. This means a much better match between production of RE and loads.
The problem of low-wind periods is never going to go away because when the CP is down to 2 or 3% that is practically zero compared with the demand in the grid regardless of installed capacity. This was clear years ago, as soon as we knew that the 2012 study of wind droughts across the NEM published by Paul Miskelly was not going to be superseded by more recent data. Prolonged wind droughts can occur between sunset and sunrise as well and it doesn’t matter if these events are rare, because until we have several more Snowy schemes or ample nuclear power the lights will go out bigtime sooner or later, and how often do we want that to happen?
The crux of the problem is the “gap in the flood protection wall” effect. Build the high parts of the wall higher and increase the average height but the water will still come in through the gaps. https://www.riteon.org.au/netzero-casualties/#217
The AER rule change request posted by AEMC today (22 December 2021) may assist to shine a light on why Queensland is seeing such insane wholesale prices compared to the rest of the NEM eg the year-to-date FY22 wholesale electricity price in Queensland is $95.88/MWh which is nearly 50% more than NSW at $65.13 and 3 times more than Tasmania at $27.68/MWh.
The AER’s usual report on individual high price events is no longer fit for purpose and this has been identified by the AER in its point 2 below “The timing and format of reporting, to enable the AER to look at prices more holistically e.g. grouping prices in where there is an observed behavioural trend.”
Posted by AEMC today:
Dear Ms Collyer
Re: Rule change proposal – AER reporting under clause 3.13.7 of the National
Electricity Rules
Please find attached a rule change request proposing amendments to clause 3.13.7 of the
National Electricity Rules (the rules), which relate to AER reporting obligations on significant
price variations and high price events in the National Electricity Market (NEM).
Currently, clause 3.13.7 of the rules places a number of wholesale market reporting
requirements on the AER:
• Reporting quarterly on ‘significant price variations’.
• Reporting on any market condition the AEMC or ACCC asks the AER to report on, if the
AER agrees.
• Reporting on energy 30-minute prices above $5,000/MWh.
• Reporting on ancillary service 30-minute prices above $5,000/MW.
While it is important that the rules require the AER to analyse and report on significant
wholesale price outcomes in the NEM, the AER considers that the current requirements are
overly prescriptive and no longer fit for purpose.
The AER’s proposal is to replace the current rule with a principles-based reporting
framework, supported by AER guidance. The intent is to replace the current clause 3.13.7
with a new, less prescriptive framework for reporting on significant price outcomes. This
would provide flexibility on:
• The prices the AER will report on, to enable the AER to adjust to changing market
conditions without a further rule change.
• The timing and format of reporting, to enable the AER to look at prices more holistically
e.g. grouping prices in where there is an observed behavioural trend.
• The factors the AER must consider in the reports, to tailor reporting to the particulars of
the price events.
Increasing flexibility in this way would ensure the AER’s reporting provides more insightful
analysis into market outcomes and behaviour, supporting the achievement of the National
Electricity Objective.
The AER requests that the AEMC expedite this proposed rule change. The AER considers
this proposed rule change is non-controversial, as it relates specifically to AER reporting
requirements and it is unlikely to have a significant effect on the wholesale electricity market
or the interconnected national electricity system.
Finally, I note the work done at a staff level between the AER and AEMC including meetings
and guidance on the principles-based approach to drafting the amendments. I would like to
thank the AEMC staff for their assistance in developing this rule change proposal.
For any further information please contact Scott Johnston on 08 8213 3426.
Yours sincerely
Justin Oliver
Board Member
Australian Energy Regulator
Sent by email on: 21.12.2021
The statement that 22,000MW of coal fired power will be closed down by 2050 strikes me as not terribly remarkable given that probably all of these stations would be at least 10, and more likely 20 years, old. The usual life of such plant is 40 or sometimes 50 years so little to remark on there i.e. I don’t see it as a collapse so much as regular replacement. Its just that in the current circumstances they will be replaced by non-coal and probably non-fossil fuelled plant. The task is to ensure that the replacement plant is reliable 24/7.
“AEMO should provide more explanation as to why thermal coal plants are not assumed to operate more flexibly”
Yes the EU and UK certainly got more flexible with their coal generators this winter and nuclear along with gas got a whole greener in outlook as well. Flexibility is certainly in vogue at present.
BNGSF1 came back around 15:00, after a short run at ~14:30 while BNGSF2 was around 17:00. I can not see a market notice for non conformance, so I assume the action was either intentional by the plant operator, or requested by AEMO/TNSP.
Queensland has now cracked $100/MWh as the year-to-date 2021-22 average wholesale electricity price.
To put this in perspective Queensland is around a 50% premium on the NSW wholesale price of $67.94/MWh and nearly 3 times the wholesale price of Victoria and Tasmania ($38.77 and $37.00).
As a consumer advocate we know this is disastrous for small business which are the biggest consumer of grid supplied electricity and the largest employer of Aussies (nearly 5 million Aussies). Affordable electricity is one of the keys to Australia’s Covid19 recovery. Main stream media hasn’t even acknowledged this sad milestone.
I’ll be provocative here Paul and say that your 86% / 14% split is very creative accounting. It’s capacity that counts, not number of units. On that measure something more like 75% / 25% which is not great for the height of summer.
Ok. This pains me, but you need to add the GT26, most thermally efficient major generation station absence generating 325MW over the last two days too. I cannot bring myself to name the station. What is the state government trying to do. Get themselves voted out on a Eastern seaboard blackout?
Can you do a similar explainer about why so much energy from NSW is being exported into Victoria against a price gradient? For example at 16:10 NEM time on Sunday 6 February 624 MW was exported from NSW with a high price to Vic with a low price (according to the AEMO website). This is now occurring quite often, and it’s a much larger quantity than we would expect than just from the far western NSW solar farms.
Essentially that constraint can choose between curtailing generation in SW NSW by 1 MW or swinging flows between Vic and NSW by 10 MW as long as the price difference is less than about $100/MWh. So we see plenty of situations like the one you highlighted.
But I agree it would probably be useful to summarise the impacts it has been having, and why, in a follow-up piece sometime, now that it’s been in operation for a while.
Hi MK,
Unfortunately, to my knowledge, no. The official calculation of the response – actual consumption relative to baseline consumption – is done for settlement after the event so it would be private to the Participant.
I imagine the original uptick was coincident with P5 sensitivities, although I would have thought they would be run separately… On a side note these were paused for a period because the demand scenarios were being applied incorrectly.
A couple of weeks ago there were some quite bad performance impacts – some PD runs were missed entirely! I’m not sure how actively AEMO monitors – it persisted for some time until reported, worryingly. They investigated and things improved.
I did have a look at the # of binding constraints and didn’t appear to see much correlation (indeed there didn’t seem to be that many in the first place).
I know that were going to add FSIP into the P5 formulation, which would involve a commitment and pricing solve… The other thing that comes to mind is BassLink switching term?
I find it extraordinary that just as SA proudly announces work beginning on the NSW interconnector we get news of Eraring closing much earlier as well as NSW announcing big plans to fast track even more local renewables. Something does not compute with the purported benefits-
“Project EnergyConnect will be a freeway for renewable energy produced in South Australia and exported to New South Wales, reducing electricity prices and fuelling economic growth in both states. Importantly, it will strengthen the power grid, so South Australia can securely add more renewable energy in line with the state’s push for net zero emissions by 2050.” https://energymining.sa.gov.au/latest_updates/work_begins_on_sa-nsw_interconnector
I’m not sure what the future of solar and wind holds but it would appear the ethos of Covid State border controls is catching on with renewables in general- https://reneweconomy.com.au/bob-brown-foundation-slams-marinus-link-as-isp-white-elephant/
Hey you lot pay for your own expensive backyard grid firming and no mooching on ours! Seems the NEM is about to live in interesting times as Origin’s announcement on Eraring has concentrated some minds around interconnectivity.
I didn’t think the weather conditions were at all exceptional for this time of the year in SE QLD where most of the demand was. Yes, I am including humidity. I am not saying it wasn’t challenging weather conditions, I am just saying this normal for this time of the year. The poor design and insulation of most of the houses in the SE, combined with climate change is going to make things worse in SE QLD.
Of course the unbalanced and uncompetitive grid in QLD just makes things worse. If QLD had a similar amount of wind based generation as it does solar, then prices would be so much lower.
Has AEMO made any comment?
AEMO don’t seem very interested in fixing problems with their data and data systems that are public facing. I have raised several issues with errors in archive files and received no response, nor has the situation been rectified.
Due to the nature of AEMO, there doesn’t appear to be anyone to make an official complaint to regarding these issues.
“Again this week I am wondering whether it’s possible for us all to collectively cool the ideological barrage and really think through what we’re expecting to happen as we move together through this energy transition….”
It does appear to be a case of ‘in lithium they trust’ with transitioning to weather dependent power generation and yet some obvious storm clouds appear with that- https://premium.goauto.com.au/ev-fires-become-hot-issue/
as the lawyers weigh in too- https://premium.goauto.com.au/ev-fires-what-the-lawyers-say/
Presumably why I keep hearing all this hydrogen hype all the time. All I know is I wouldn’t be much use at all heading up the AEMO with more questions than answers for those at the coalface. Not an easy task with universal interest in the outcome and as politicians know full well- only the impotent are pure.
Thanks for the analysis Linton. I too have been wondering about the high consumption considering flooding impacted thousands of residential and business customers in the Southeast. I put it down to the Queensland demand was high as the rest of Queensland is suffering from heatwave conditions – one Cairns businessman described it as the worst summer for heat and humidity since 2006 and Cyclone Larry. The other is there are about 2.1 million residential customers in Queensland of which about 1.5 million are in the Southeast. Similarly, there are about 230,000 small businesses in Queensland and about 140,000 in the Southeast. Energex has done an amazing job getting power reconnected so not sure the amount without power was too high and anyone with power would have been using it for cleaning and air-conditioning as it was hot in the Southeast too. You would definitely need a good nights sleep after a day of cleaning and getting businesses back working so air-con would be a must. Also the rest of Queensland actually consumes more electricity than the Southeast due to mining and heavy industry and this was unaffected by the floods.
This is very interesting, I am wondering is there a part c where the possible solution to what is obviously a countrywide situation at certain times?
IS the speed of drop off capable of being forecasted and could it allow for an alternate power source?
Russ Christ effect, well at least cloud location compared to their shade, also depends on the height of the cloud and its thickness. I did some modelling and comparison to cloud on the Gold Coast and found it was extremely difficult to match cloud conditions based on satellite pictures to output of my PV system. Part of the problem was the low resolution and frequency of the satellite pictures. This was particularly the case when the cloud was broken into cells as distinct from a solid wall.
I did try using ground based photography of clouds, but the topography defeated me in this regard.
This event needs to be put in context though. The odds of such extreme rainfall in a particular location might be 1 in 500 years, but the odds of 80% of this level occurring some where in this region of QLD and NSW is much higher. We only need to look at 2017 and 2011 for similar events. Sure the rainfall levels might have been different and the locations slightly different, but the results are not that different. Major flood damage in this region.
I think there needs to be a better way of stating the statistical probability to the region as clearly 1 in 500 years or even 1 in 100 years, doesn’t convey the risk.
What happened to the interconnectors during this period?
They didn’t even average 10% of their capacity. Given there was fore warning about these conditions, how was it possible that they were not brought back online? I can only assume it was to do with the floods, but that is a worrying sign as flooding events are going to become more frequent and more extreme.
Unfortunately there was almost no wind based generation during this period. I haven’t checked for constraints, so I assume it was just that the conditions weren’t suitable. This is going to make solving this sort of event very hard in a near 100% RE grid. It is going to take many levers to solve this sort of event, more demand management, much more energy efficiency, a little PHS and huge amounts of battery.
Craig, we’ve had this discussion elsewhere, but at the risk of repeating myself it’s absolutely pointless to talk about the notional maximum capacity of interconnectors – typically that would mean thermal capacity of the lines etc under steady state conditions – when looking at their actual capability to transfer power over long distances in a real power system.
The key limiting factors on what interconnectors can actually transfer can often be nothing to do with that thermal capacity but with *transient* electrical phenomena that occur if there is a fault or trip leading to sudden changes in flow and/or impedance and/or phase angle along the interconnection path. In the case of QNI and Directlink, there is a very long distance between major generators in NSW and load centres in Queensland, which makes this interconnection quite susceptible to voltage collapse if a fault were to occur with flows out of NSW exceeding certain thresholds.
AEMO uses a large number of constraints to manage these risks – which can arise on trips of different generators in Queensland, lines in NSW and so on. The limits imposed by these constraints are quite dynamic and I wouldn’t pretend to understand all the factors involved. The bottom line is that it is not at all unusual to see net exports from NSW to Qld limited to only a few hundred MW at times of high demand in Queensland.
Recent work on upgrading points along the QNI flow path – involving both lines and new equipment at terminal stations – should see some increase in capability later this year. But that capability will almost NEVER approach the underlying thermal capacity of the transmission lines themselves.
On March 8 there was an outage impact on flows (apparently a Static Var Compensator at Lismore or Armidale), but it still only reduced transfer capacity relative to system normal (nil outage) conditions by something like 100 MW.
Origin is planning to run its gas fleet as baseload (Uranquinty is a big OCGT station), which will be needed when Eraring shuts down of course. Higher wholesale power prices will be a nice reward for them.
Uranquinty at full output will cover almost half of Eraring’s usual output,
But when you rely on your peakers to provide baseload, what supplies peak demand?
Thanks Jonathon. Great background on the CFP arrangements to help understand forecasting behaviour.
Perhaps a naive question… but are the deliberate deviations necessarily bad behaviour?
From your description, it sounds like there is systematic error in AEMO forecasting that e.g. underestimates NEM demand on a typical afternoon, hence dipatches insufficient generation, hence under-frequency conditions. If a solar or wind generator anticipates this and responds by intentionally under-estimating generation, leaving some in reserve, then they would be benefitting from addressing that error. Sounds like a good thing??
Of course it would be preferable to fix the forecast errors in the first place or perhaps create a form of frequency regulation that VRE can formally particpate in (beyond contingency). But in the meantime, perhaps this self-serve proxy is useful??
There are massive difference between the 2021 GSOO predictions and the 2022 predictions.
To believe the predictions in the 2022 version, you’d have to believe all AEMO’s past predictions are way off the mark.
Just last week Frank Calabria stated he’d be running Australia’s largest gas turbine fleet for periods of “days and weeks” ie baseload gas, once Eraring closes.
At the same time Liddell is closing, so NSW is down about 3,000 MW of baseload and adding 800 MW of peak demand from SA when EnergyConnect comes online around the same time.
That’s gas for power generation.
To follow the AEMO prediction for domestic and commercial, you’d need to immediately start changing out a lot of perfectly fine gas equipment for electric equipment, all over TAS, VIC, NSW and QLD.
The ‘most likely’ scenario doesn’t seem all that likely.
I have been reading the ESOO with following questions in mind:
(1) Will there be power shortages as aging coal plants are closed?
(2) Where is the modelling for storage of renewable energy over daily, weekly and seasonal cycles?
(3) How many Snowy2 schemes do we need for mass storage of renewable energy?
(4) Will there be enough power for 100s of apartment towers, skyscrapers (without solar panels) and other power hungry infrastructure governments are planning?
(5) How will additional demand from electric vehicles be accommodated?
(6) As we are hitting peak conventional gas on the east coast, how will that impact on gas peaking plants?
(7) How does the speed of oil/gas decline and global warming determine the amount and sequencing of investments in the power sector?
I have a burning question to the more complex cases. Assuming that the Increase field captures the marginal increase in energy for the ENOF and the marginal decrease in energy for the LDOF type entries (for the extra 1 MW), and then there is FCAS (Increase sums to 0 MW), what exactly does the Increase field capture for the tbslack (tie-break) type entries? And for MNSP type ones (Basslink maybe – although Basslink shows up as non-scheduled in AEMO’s registration list (https://www.aemo.com.au/-/media/Files/Electricity/NEM/Participant_Information/NEM-Registration-and-Exemption-List.xls))?
Paul – The lack of the non-mandatory rebid fields is probably a very deliberate decision by at least some participants (not saying this applies to Milmerran). What’s the benefit to the participant of populating those fields? If they put data in and are assessed for compliance or something else by the AER they may have to show that the numbers are accurate, which is probably not straightforward in at least some cases – “awareness” and “decisions” can emerge over time, not necessarily in an instant – and they are also on a hiding to nothing if they made any errors or omissions. So strategically it may make very good sense not to provide that data given it’s not required. So perhaps I differ slightly on what’s ‘Best Practice’.
Imports into SA via Heywood are severely constrained across Thursday. This combined with the low wind forecast is the cause of the forecast LORs. If Heywood imports were at their normal level there likely would not be any forecast LORs (even with the low wind forecast), although it would be close to a forecast LOR1.
Also relevant is that Pelican Point has completely offline for over two weeks. This could be for maintenance or possibly for establishing a gas connection to the neighbouring Snapper Point generator which is being commissioned using diesel fuel.
Many of the SA peakers have been running this morning peak but Barker Inlet has been offline and Torrens Island limited to 80MW instead of up to 400MW in recent days. Pelican is still offline. Currently SA/VIC interconnectors are very severely constrained.
Hi Paul, it would be great to have a look at the generation profiles and bids that have resulted in the prices.
We’ve had a few occasions in the last week where there has been over 150MW liquid fuel at these times. Would not be surprised if the generators who routinely bid at $-999MW/h have probably been impacted by calm & cloudy days and those who can turn on are wanting to make some cash to make up for the all too frequent periods or even whole days of negative regional prices.
Yes calm and cloudy in Adelaide Friday and diesel generating 127MW and I noticed it as high as 157MW around 2:15 in the arvo yesterday Thursday. Is diesel cheaper than gas or just available?
Here’s a puzzle. Friday night tonight and around 3:50pm total SA output is 1704MW with 564 gas 478 wind and the rest solar when diesel kicks in as high as 173MW as solar shrinks to zero around 5pm and both battery storage comes on from SA and Vic in that time. SA diesel continues down as low as 115MW until it drops out around 8pm and by then wind has picked up. At one stage gas reaches 910 with wind the rest for 1717 total and wind is picking up while all batteries have dropped off.
By 9 pm SA price is around $300 Vic $320 and NSW $340 and Vic is feeding NSW suffering a max shortfall around 1000MW maxm and Vic is in turn being fed by SA and Tas. By this stage SA wind is 1436 and gas 338 in SA and at one stage all States batteries are supplying. The main surprise for me was to see SA diesel running for around 4 hours and clearly competitive with gas for the period.
Whereas earlier in the week when most of the diesel generators were in action, on Friday the diesel output was only from Snapper, which was at 120MW. This is the highest I have seen it in the commissioning process. Pelican Point ramped up from outage during the afternoon reaching 450MW before going offline at 9pm. Barker Inlet returned from outage this afternoon.
11:35 am Monday morning and the prices per MW in round figures are showing Qld-$525 NSW-$130 Vic-$120 Tas $112 while SA is NEGATIVE $95 yet SA power consumers pay the highest power prices. Clearly power retailers factor in lack of predictability in their business models and something for easterners to look forward to as they follow SA’s lead with solar and wind penetration.
How AEMO people deal with that lot beats me but it doesn’t seem like an optimal career choice if you like the quiet life.
Since we are low on coal capacity lately I thought that the experience during low-wind periods would be interesting. Gas probably picked up some of the slack but I expected that hydro would be the cheaper option and it seemed to be the case looking at the Aneroid Energy plots. Maybe crude compared with the analysis here but there was clearly a difference in hydro from month to month, expressed as a larger number of spikes over 60% capacity for all hydro. I appreciate the analysis is better without Tasmania.
For some time I have wondered how long hydro would last if it is required to fill a serious gap in supply.
Drawing an arbitrary line at 10% of installed capacity for wind drought, it has been barely more than that all day.
Still the message from the wind records is clear, the retirement of conventional power has gone as far as it can until grid-scale storage is on deck, on time and on budget, as you signalled years ago. Still waiting:)
South Australia has demonstrated the point, they import power whenever the wind is low, never mind full-on droughts like yesterday. Currently they are importing and 70% of local generation is gas.
The NewWatch widget is the clearest indicator, a glance in the morning and evening is enough to see the unreality of the green dream. Not to mention the way things have gone in Britain and Germany.
Apologies that this is somewhat off the exact topic yet still on vital subject of wind generation.
Wind generation plateau for 4 years in South Australia http://www.warwickhughes.com/blog/?p=6921
What has caused wind generation percentage to plateau for 2018, 2019, 2020, 2021 in South Australia?
Aggregate wind production in South Australia has been limited (i.e. ‘constrained down’) due to various forms of the ‘System Strength’ constraint in recent years, Wazz.
Thanks for explaining that Paul – I tried searching the subject at various search engines but got nowhere. I must have been using the wrong words. I wonder what the implications are for this in other States and further down the path to “Net-Zero by 2050”
As an ex coal fired power engineer, I could always justify turning off the unit to fix something. There was always a reason to turn a unit off. So I would pull out my risk assessment, and say that the risk of keeping the unit online was unacceptable, and that repairs are to commence once the unit has cooled sufficiently (48-72 hours), and that owing to concerns about the complexity of the repair, it will be done on day shift only. I could conceivably keep a unit offline for a week or more in that scenario.
That only buys you enough time to get past the immediate challenge, but to get short run marginal costs that high, you’d have to be exposed directly to very high fuel prices (and presumably water prices for water cooled units). Which in the current environment I would see as gas only.
Would be good Paul if you added a colour legend for the various States in the NEM. I make them – NSW Turquoise – Qld pale purple – SA Red – Tas Green – Vic Brown.
Thanks for Wattclarity info over years.
Tas hydro has no competitors –
is minor in size compared to NEM –
Tasmania pop is ~525,000,
TasHydro is Tassie Govt owned like TasNetworks –
is in horrific debt –
must have Departments of staff whose task it is trying to sucker the Feds to invest in their 2 oversold projects –
the new Marinus Link cross Bass St. copy of their BassLink disaster –
plus their Battery of the Nation version of pumped hydro copying Snowy 2.0 –
I hope the Feds keep right away from investing any taxpayer $’s in above.
Let TasHydro find private enterprise to invest in their great projects.
Oh!! almost forgot my opinion of the answer to my question in thread headline –
The outer galaxy $15,100 per MW/Hr prob due to TassieHydro book-keepers having to balance columns in some spreadsheet.
There is a very good blog TasFinTalk who in Feb had an article on Basslink –
I think the overwhelmingly obvious thing the current situatio illustrates is that you have to build new capacity for energy and power before the old stuff goes out of service.
To me it is incredible that Frank Calabria is talking about restoring coal supply and is closing a generatorbut is not contracting or building any new wind or solar even though every model shows that lots more of it is needed. What does he think, someone is going to give hime 5 or 6 GW of wind or solar the day Eraring closes so he can keep his retail customer loads? The same for AGL and EA with Yallourn.
Nobody with any honest insights believes The Transition (to replace coal with wind and solar) is realistic – either in the realm of physics or economics.
The faint technical possibility of a very high % RE system is dwarfed by the massive cost and the impracticalities of storage, transmission, inertia, fault level, over supply, negative wholesale prices, land use – just one of those challenges is a spanner in the works. Combined they are insurmountable.
At what point does the community of energy professionals decide to speak up and tell the truth?
Tristan Edis states that RE output is not down, comparing May 2021 and May 2022. What he has omitted is that there was 3 GW of rooftop solar installed in 2021, and a bunch of large-scale wind and solar too. In the context of much higher RE capacity than 2021, the same MWh output in 2022 is in reality a lower capacity factor in 2022. This is in surprising because the weather has not been favourable. The weather is also partly responsible for some coal plant scarcity because of mine output constrained by water issues.
Tristan acknowledges that gas generator output is not much different to previous years. In fact some gas generators are price takers, being dispatched at bids of zero dollars.
Gas and diesel peakers, while responsible for short duration price spikes, are not the problem either. This is evident by the lengthy periods of wholesale prices sitting around $300/MWh. This is not a high average price caused by 15min afternoon spikes.
Consider this scenario: a coal-fired power station, with fuel constrained by a partly flooded mine, offers a larger portion of their output at high prices ($200-$500/MWh) to avoid being dispatched, to conserve fuel, unless other generating capacity is short. Concurrently, there are several units on planned outages and a couple of unplanned outages. All of a sudden, baseload generation is scarce, and that higher-priced capacity gets dispatched.
Continuously high prices are no mystery.
As for gas, imagine if a state thousands of km from the bountiful QLD gas fields banned their own local gas development and failed to upgrade pipeline capacity… also imagine that with an early onset of winter (damn that global warming), gas demand for heating suddenly increased … and the pipes are already at capacity.
When demand outstrips supply, you get price increases. Again, no mystery.
No shortage either.
Closing coal-fired power stations and expecting gas, wind and solar to fill the gap cheaply is an outrageous fallacy.
The knee jerk reaction to slap new taxes and levys on coal and gas producers is exactly the wrong strategy. At a time when high prices and some high profits can be expected to stimulate new investment, new development and therefore new supply (and jobs and taxes and prosperity), eliminating those profits by imposing arbitrary levys is a sure way to push these companies out if this country while doing nothing to increase supply and therefore doing nothing to lower prices.
I have to say that high energy prices and coal generation scarcity is exactly what the RE lobby wants.
I can’t imagine why anybody is surprised at the results.
I think most energy realists expected the crisis in the power system to become obvious due to rolling blackouts.
In the northern hemisphere the crisis arrived earlier than expected on account of the war and this may be a factor here although Ben Beattie did not need to refer to the war to explain the situation.
The underlying issue is the fact that the green transition is not sustainable due to the combination of wnd droughts (zero RE on windless nights regardless of installed capacity), the need to meet 100% of demand all the time, and the lack of feasible/ affordable grid-scale storage.
There is also the matter of cost – we are stuck with a hybrid system because we can’t allow the coalers to close until we have storage or nuclear power, the transmission lines and interconnections will cost several arms and legs, and everything containing rare earths, especially lithium and copper, will be more expensive. Please, no more talk about the price of batteries plummeting:)
It’s hard to comprehend why an analysis such as this did not even once mention the nuclear alternative.
Our kids are being indoctrinated with the global warming message even in kindergarten, more so in primary and secondary school, and the process finally being completed in university. As grown ups these naive people haven’t seen the logical alternative for it is rarely discussed, banned by legislation, and terrifies the uninformed.
Reversing this deeply rooted belief will entail a herculean effort at re-education.
In the meantime we will all suffer from a failed power grid that must surely open the eyes of government and an enraged community shaken from its torpor.
Emissions free nuclear power – the ultimate means for providing a viable, plentiful, and cheap supply of electricity.
Paul: I think the local contingency raise FCAS constraints, as well as the negative residue management constraint on Qld-NSW flows, were still fundamentally the drivers for these Qld events this morning – they were all binding in this DI, and as noted in Marcelle’s article (https://wattclarity.com.au//articles/2022/06/interconnector-constraints-and-limits-an-explainer-and-a-few-traps/) AEMO data only lists one constraint as setting an interconnector limit even when there are multiple constraints binding. The only reason N^^Q_ARTW_B1 bound was because NEMDE was trying to increase northward flows to satisfy the NRM constraint. That in turn was only binding (violating in fact) because the local FCAS constraints had previously been pushing counterprice flows southwards.
Gas exporters were supposed to source their gas from coal seam gas, and that hasn’t happened because they’ve been allowed to buy fields like Arrow previously domestically focused. 10% of their sales are spot – exporter source, and can readily be stopped. In the UK, they know war profiteering when they see it, hence a windfall tax. The US has also declared war powers as well. Not surprising given the kind of mess we’re seeing here.
Thanks for your thoughts and the info.
That “symphony” reminds me of Schnittke music. Here is a suitable example https://youtu.be/N1U4tN_ZUt0
Other contender is Karlheinz Stockhausen: Helikopter-Streichquartett for a quartet of string players and four helicopters https://youtu.be/7ykQFrL0X74
You get the idea:)
“It also seems rather odd that the gas industry finds it so difficult to supply a level of gas power generation in 2022 that is not much different to what has occurred historically.”
The gas industry has been extremely vocal in the past few years that we’re facing a gas supply shortfall (with specific emphasis on this year and the next). The RE lobby responded to these calls by saying that total gas demand destruction was imminent (not exactly seeing that right now) and no new supplies were needed.
“What appears to be the real reason for high prices is that both gas and coal miners are charging very high prices for their products because that’s what they can sell them for into international markets.”
We were seeing 100 TJ/day of gas sent north from the Moomba Sydney Pipeline in the beginning of the year which is in line with what they’ve done historically so the LNG producers weren’t pulling up more domestic gas than usual. Because domestic prices have been so high recently, we’ve witnessed a negative chill spread (domestic prices higher than export prices) and yet prices still remain high domestically. Shouldn’t the greedy LNG producers completely cut off their exports and pump everything locally according to your logic?
By the way, OCGT gas generators (the bulk of them in the NEM) act as peakers and not as baseload. Therefore, calculating their total GWh and claiming that it isn’t high compared to previous years is completely misleading. The commercial driver for peakers is based on spot prices, their economics are not tied to long baseload runs. Furthermore, it drives me nuts when people do not appreciate how important those gas peakers have been during this crisis. They were absolutely critical in helping tame prices during the morning and evening peaks. Something that “free” wind and solar isn’t too good at.
Where to begin. How about, simply, Western Australia.
Anyone who wants to talk power and price, and supply, please look at Western Australia’s last 10 years.
It simple what they have done, it was not hard, in fact it was easy. People can make up what ever crap they want to explain the eastern sea board. But it was simply planned poorly.
“…one would hope to see availability exceeding 80-85% from a coal generation fleet performing well”. I thought the capacity factor for coal was 50-80%, so 71% is about average, especially for aging infrastructure.
Dave – thanks, but you might be thinking of actual capacity factor which measures energy actually dispatched relative to nominal capability, and will always be lower than available capacity factor because of market offloading – generators don’t always operate at their available maximum and we see plenty of swing on the NEM coal fleet. But available CF is a more appropriate metric to consider generation reliability. Allan.
Great article Allan. One cant help but observe the coincidence of outages and high coal prices at the start of winter although I have no doubt coal availability at the mine mouth and rail availability to move it to a power station are also down. I do wonder whether the bayswater issues will further impact agl profitability.
Sorry for the very belated comment Paul but having done some similar analysis recently on coal units for Q2 2022 I came to this post. I can’t see that there would have been much material value or meaning to inclusion of semi-scheduled generation in these datasets, so I think the rule change was sufficient. While including S/S capacity might have shown some small variations in maximum capability of wind and solar farms (eg known maintenance on a few wind turbines here and there), or commissioning profiles on newer farms, these variations would be dwarfed by uncertainty in future sun and wind levels, so information on maximum output of any wind or solar farm, let alone the whole fleet would be of very limited value. The STPASA forecast availability datasets based on ~1 week’s weather forecast, and also incorporating known capacity restrictions, seems as much as one could or should expect.
A price cap, in any market, tends to be the last step before outright physical shortage emerges.
That basic pattern has been at the heart of physical shortfalls overseas in various countries. Generation shutting down for purely financial reasons with the end result being physical shortfalls.
Same applies to gas since the recent price capping simply thwarts a shift to diesel, thus holding gas consumption higher and depleting storage more rapidly.
I wonder if AEMO’s own data centre has been load shed? Seems impossible to load Market notices or price data from either Telstra or AAPT connections at present.
Hi Paul,
The last 7 days sum of spot prices for all states except for QLD is well below the $1,359,100 CPT. How can the CPT be exceeded when the sum of that last 7 days of spot prices is at or below $1,000,000?
As consumer advocate we are keen to see network tariff reform include more emphasis on load controlled tariffs. Around 900 MW in Queensland but it could be more.
Thanks Paul, great service to the community!
PS I am doing a list of our world class windwatchers and you will have to be on the list with Tony From Oz, Paul Miskelly and Mike O’Ceirin (Spasmodic Energy.)
Looking at strange events, reserve (dispatched generation less load and export) has been negative at times in Vic and NSW, while generators appeared to operate off target (IC flow suggesting denand was met). YWPS unit 4 started up yesterday afternoon, then tripped right near peak demand.
Anyone know why basslink is constrained to zero in the Tas to Vic direction?
So the chair thinks being forced to sell at $300/mwhr when your inputs are >$1,000/mwhr is not “reasonable cause” to stop offering. Is this guy straight out of communist russia?
Paul, it’s interesting that some have commented that generators have withdrawn capacity rather than be dispatched at a loss making price due to the cost of input gas. Whilst there is a mechanism through the AEMC to claw back costs, I’d expect the prospect of losing millions of dollars per week and needing the cashflow to manage this whilst waiting (months?) for a determination on your compensation may not be viewed as particularly attractive…
Paul, congratulations on the work that you do. I was looking for information on why Swanbank E has been out of service and when it’s expected to become available again. Bingo.
A couple of other questions I can’t yet find answers to are:
1. Where can I find information on each of the grid-connected wind turbines that shows their unavailability due to mechanical or electrical failures/faults? Individually, their forced outage doesn’t make much difference, but collectively their outages may be contributing as much to grid integrity as the coal-fired and gas-fired generating units.
2. Where can I find information on each of the grid-connected BESS units to see how they’re performing and when they’re showing early signs of needing replacement? Just knowing how many charge-recharge cycles each installation has done, would be a leading indicator. I’ve noticed Hornsdale on numerous occasions doing more than one cycle per day. On current usage, what is the forecast date for the original batteries to be replaced.
Hello, can you explain what snowy 1 has contributed during the current crises. My understanding is that SH 1 was designed to step-in to supply during such crises.
I have not heard a peep from SH to say it was on duty. Or in response to the white elephant claims by media commentators.
Whilst we (the general population) might have intense interest in what’s happening inside the wholesale market during this crisis, and so might think that the participants in the market should be doing more to talk about what they are doing, it’s important to keep in mind that their first duty is to do whatever they can to help AEMO keep the lights on … which might mean their communications come a distant second.
It’s unfortunate that some take silence (most likely just because of busyness) to leading to … ‘distrust in what snowy 2 will do’.
So…at a time at which in-market participants were only entitled to $300/MWh and/or their direct costs, out-of-market resources could earn around $20,000/MWh. Does this feel right?
Hello Paul,
How about its also been pretty cold for broad areas for long periods, low winds, the snowy hydro not running properly, high gas costs, high coal costs, water sodden coal mines restricting supply, gas supply tight, very run down coal plants, and I am sure gas plants, poor planning, bit of gaming the bidding system?, and i am sure a bunch of others i have missed. All in all a fair amount has gone the wrong way all at the same time.
We may learn some lessons from another energy sector: oil.
Peak oilers have long been thinking what will happen when the global oil peak occurs. There is the theory of a Seneca Cliff after the peak in which the whole system declines because everything depends on a smooth supply of oil.
The first global peak we had in 2005 (as predicted by Deffeyes) which caused the US recession in 2007 and the global oil price shock in 2008. High Chinese oil demand for the Olympic games also played a role. At the same time banks were so imprudent to have invested in car and therefore oil dependent US suburbia just as the peak was happening.
High oil prices give a strong signal to motorists to economize the use of fuel with fuel prices at the pump following oil prices after 2 weeks or so. If demand destruction happens in all countries then oil prices will fall again. If these prices are too low for oil companies to make money then oil production may suddenly drop, potentially below levels needed for the safe functioning of the world economy. This did not happen yet because of US shale oil but this oil has astronomical decline rates and will also peak.
Drawing parallels to the power sector, capacity of dispatchable generation may have peaked (including from coal depletion in mines near Eraring and Liddell/Bayswater) whereby this is a function of the price. We then have a cold snap with higher demand (similar to China’s 2008 Olympic Games). All the while Australian governments were driving up evening peak power demand by increasing immigration over 15 years, requiring hundreds of new power hungry apartment towers (similar to expanding US suburbia). The surge in renewables is similar in function to the new frontier of US shale oil and occasional/unpredictable recoveries of oil production from unreliable suppliers like Libya, Nigeria etc. and new fields.
The big difference between oil and power:
In oil markets, the feedback loop for lower demand via fuel prices at the pump is comparatively short as mentioned above. On the fuel supply side, oil companies are sitting on inventories and have a certain flexibility in case crude production is short. So this gives at least a short to medium term window to balance supply and demand.
In power markets, however, generators must balance supply and demand instantaneously while the feed-back loop to lower demand is up to 3 months when the next power bill arrives. On the supply side, generation depends on variable renewables without sufficient pumped storage to provide dispatchable spinning reserve.
So in the power market, there is a higher chance that the system collapses (=load shedding)
Looking at this rather unfortunate state of affairs from the other side of the country, I suggest the main drivers are fairly obvious. There were a few coal fired units out for maintenance then a few more were out of service from forced outages – either plant related or fuel supply related. Oh and lets not forget that there didn’t seem to be a lot of wind around for a few days. Now there is some gas fired gas turbine plant around but nothing like the proportion of the fleet we have here in WA. But that only works if you have contracted fuel supplies and hence a sensible price. Given most OCGT’s will have a heat rate of between 10 & 12 GJ/MWhr then if you cap your price at $300/MWhr then as soon as your gas price exceeds $25-$30/GJ you are out of the money before you have even paid for your starting costs on the unit. Spot gas prices that I saw exceeded this so clearly there is little contracted gas reserved for these peakers. Liquid fuels are even worse.
This is one of the serious limitations of an energy only market and why Western Australia doesn’t have one. The NEM used to have lots of spare capacity but this has been hollowed out over the last decade or so as older stations have been retired and a market regularly trading at SRMC results in little incentive to invest in sustaining capital. I have heard much nonsense about “aging coal fired power stations” – coal fired power stations do age but are like Grandad’s axe. A new head or handle and it is still Grandad’s axe. Unfortunately when the economics of reinvestment have been so damaged by the hollowing out of daytime energy served the plants become unreliable. Their eventual retirement (whenever that might be) is also a bit of an are as no one retires a pristine plant – the glide slope is always tricky.
This is where a capacity and energy market works. It guarantees a revenue stream, but it has to come with strings attached seeing at the end of the day the punters (via the retailers) are paying. There has to be firm fuel supply arrangements in place – if the generators want to be paid there are no excuses. Seeing the capacity market pays for some of the fixed costs of the plant then the market price caps get pulled right in. In WA they have always been $300/MWhr based on a GT burning liquids – probably about half the number it needs to be right now but the point is numbers like $15,000/MWhr just break the little guys who have little chance of hedging with a generator.
Capacity markets work – they simply need to be carefully designed and calibrated (in terms of $$) to get the correct outcome and prevent gaming by people whose business model is only capacity payments. They can also provide good drivers for preferentially locating (say) a wind farm where it delivers energy best aligned to the daily load profile. Ultimate energy produced might be less but the capacity payment keeps the selection whole.
The NEM is certainly creating some volatility alright- https://www.youtube.com/watch?v=CSkFPA81Uws
That’s not mincing words and calls for a response from the CSIRO and AEMO.
Paul, 2 things I appreciate from you:
1. Trying your best to stay neutral and wholistic
2. Summarizing hours’ and days’ worth of research into a short, concise and structured article
3. And sharing for free. Well with some subtle marketing but who wouldn’t!
Very interesting and important, suggesting that it may be winter wind droughts that will set the amount of storage required, rather than the maybe more widely studied wind lulls in summer, in association with heatwave induced demand spikes, which often in the past didn’t cause problems because they fell on public holidays.
Solar has a daily drought of course, but must play a role during daylight hours in replenishing the storage, and preventing it from becoming fully drained.
“We look forward to being part of the discussion on what a robust, reliable, low emissions, and lowest total system cost NEM might look like.” Good luck with that, there must be a huge tendency to bury the bad news of possible future problems.
This graph is a classic example of everything that is wrong with the uncontrolled operation of intermittent generation. There is something like 500MW/hr being added at the whim of the wind from 6pm on Sat at the very time when system load will be declining. This would force other generators to ramp down and / or decommit. So please don’t complain when the reverse happens and suddenly thermal plant is expected to restart at short notice with no guarantee of recovering start up costs. How is the large gas turbine fleet needed to deal with this going to be incentivised and where is the fuel coming from? (or for that matter the deep storage. ) Forget the big Coalies – they simply won’t be able to play like this as the problem gets worse – exactly what you are seeing with older plant like Liddell & Yallourn that will simply exit.
Hello Andrew,
The whole point of the market is to get coal plants to exit currently. If the greens had gotten into power they had planned to simply start shutting them, not even have them exit.
And not in any order or age either, it seemed rather random.
Wind (excuse the pun) back just a few days to 14th July and during the evening peak wind production was a paltry 600MW. This is less than 10% of rated capacity, supplying a measly 2% of the NEM demand.
The fact that wind output can vary so much over the period of just a few days is a far more news worthy and important issue than reporting a once in year outlier in maximum wind generation.
Michael
There are a number of article (like the following) on WattClarity that also point out the lows: https://wattclarity.com.au//articles/2022/05/27may-short-term-wind-lull/
It’s becoming increasingly the case that both extremes need to be recognised, and the averages become increasingly just that … average. This has also been noted a number of times here.
Paul
As an electrical engineer who repairs electronics for a living, I can tell you that it is mainly the temperature of the device (or rather, inside it) and hence the temperature of the components that negatively affects its life.
Specifically, in Switched-Mode Power Supplies, the Capacitors are rated in hours at a particular temperature, for example, 2000 HRS @85 deg C. If the temperature in the device is 85 degrees, then it can be expected that the capacitor will be past its useful life in 2000 hours. This is under a year if the light is on for 8 hours a day.
However, if the temperature is reduced significantly, then the capacitor will last much longer.
The amount of Switched Mode Power Supplies I have repaired with failed capacitors is countless, and in larger devices it is usually a small 10uF 50v cap that goes, and this is used to filter the power to the SMPS control chip.
Whenever I replace them I try to install 10000 HR @ 105 deg C rated capacitors for longer life, if available in the size and voltage required.
In an LED downlight, some are direct plug in 12v versions, while others have their own 240v SMPS to enable dimming, and the old transformer style power supply is replaced.
However, it’s usually the direct plug in 240v ones (I.e bayonet cap / edison screw) that fail because they have to pack in the power supply, the LED with its heatsink on top, all into a small area with barely if any ventilation!
At home, we have had a few seperate power supply type downlights fail, however they were just pushed into the roof space, and both failed ones happened to be covered in insulation – further evidence that keeping the temperature down is key to long life of any device that relies on electrolytic capacitors.
Most mains rated equipment generally has a safety margin for incoming voltage, and although a higher voltage will also increase aging in the capacitors, the heat is the main cause of rapid aging.
Ameo, is like a drunk, drugged person at the wheel of a car they have been saying they are getting serviced for 2 years, they have neglected paying rego and insurance and are now about to go around a corner and the headlights have failed!!!!!!!
(Seriously can it be possible that they let gas get down to those levels then have to step in at the last second claiming its anyone else fault but their own for not actually having anything planned, its mind boggling)
Unless I’m reading the chart incorrectly it looks like there were liquids in there too which would be setting the price. Thermal plant all off and / or cold so no way to recover quickly.
Unless the residential market is required to use wholesale linked rates, there is little incentive for households to change their use patterns or invest in reducing their consumption through better efficiency such as insulation. Without this change to wholesale linked rates, we are all going to pay more due to greater storage and transmission requirements.
Due to the fact which Paul and others have noted regarding the intractably low production at the bottom end of the wind scale we have probably gone as far as we can go to get rid of coal until there is a radically new storage vehicle available or nuclear power is on deck.
Digging in to this one, looks like generation at DUIDs CHYTWF1, STOCKYD1, MUWAWF1 and MUWAWF2 were suddenly interrupted. Do not know if these were the result of constraints binding (i.e. dispatched off), or there was some kind of fault in the area.
This may also have been a participant response to the extreme FCAS raise prices. Generators pay for raise FCAS and a number of them will bid their plants down or off when FCAS raise prices spike.
Is another major wind drought forecast?
There was an epic wind drought in several states over the weekend, running into Monday and lifting on Tuesday, hard on the heels of record high winds.
That’s intermittent energy!
I watched this as it happened, and then wrote about it the following day at the blog site I contribute to. I have been keeping all the AEMO coverage area wind generation data on a daily basis for the last four years in my daily Posts, and for the year prior to that all the (AEMO coverage area) electrical power generation data from every source for a year and a half, and then for the year prior to that, all the Base Load data, also on a daily basis.
This was a failure of (ONLY) wind plants and Solar Plants. (not rooftop solar, just the solar power plants) The initial fall at 11.35AM saw an immediate loss of 3600MW, and then over the following 45 minute further losses to take the total loss to 4700MW. Keep in mind here that this was ONLY those wind and solar plants. Now, while you might think that 4700MW is not all that much, at the time of failure that amounted to a loss of 17% of overall total power consumption across the AEMO, so this was a significant failure.
At the same time as the failure began, Hydro came ‘to the rescue’, as it always does when there is a failure of any power Unit, because hydro can come on line almost immediately. The plants at Tumut Three (pumped hydro) Upper Tumut (Tumut One and Two) and Murray (Murray One and Two) immediately put 2450MW into the grid to replace the missing wind and solar power. As you are aware these (some of them older than 50/60 years) Units are smaller Units, around 150 to 250MW, so there was around 12 to 15 Units coming on line here. Also at that same time, EVERY black coal fired Unit (15 Units in NSW and 22 Units in Qld) also ever so slightly RAMPED UP (and hey, haven’t we been told ad infinitum that they cant ramp, despite doing exactly that twice every single day by between 4500 and 5500MW) and in the next 10 Minutes, those coal fired Units also added an extra 650MW to the grid to all but cover that initial loss from wind and solar plants.
Note here especially, that the point of failure and the point in time of rescue were exactly the same, 11.35AM.
It was almost 2 hours before wind and solar recovered to their level at the time of failure, but this was the failure of just wind and solar plants.
There is more text, and images showing the failure on load curve graophs at the following link
Anton
I’ve only skimmed what you wrote at the link in your comment – but it does seem that you have misinterpreted ‘the reason why‘ the wind and solar farms collectively dropped in output. It was not a ‘failure’, in that sense – though the underlying software glitch would seem to be a failure of other processes.
Coincidentally I’ll be posting to explain this here on WattClarity in the next day or so, so stay tuned…
Paul
Paul, and thanks for the quick response here.
Using the same word semantics you used, I didn’t ‘skim’ your article. I read it all.
At no point did you even mention the failure of ONLY those two renewables as the end result of what you found.
You mentioned the Queensland situation, with no explanation.
You mentioned the rise in hydro. (without even going to check why)
You mentioned the spike in the cost of generation, also without checking why.
All of that is down to the immediate loss of 17% of power, and the extreme short notice of finding replacement power ….. OUTSIDE of the bidding process, hence enormously expensive, hence the spike in prices.
That 17% of power was 4700MW, and that’s the equivalent of six to nine of those large scale coal fired Units. If that many coal fired Units dropped off line, even for this same reason, then you would have been shouting it from the rooftops. A failure of delivery of power ACTUALLY IS a failure.
You say that I may have misinterpreted the reason why, but it seems you have also misinterpreted things as well, perhaps more than I have.
I look forward toy your coming article, ….. a week after I wrote about it. Feel free to use anything from your skim, but hey, you might credit me if you do, eh!
Anton Lang.
The one big omission in the AEMO report is the absence of a review of the Administered Price which at $300/MWh was set some 20 years ago based on the cost of energy from distillate fired gas turbines. If this price was higher there may have been less intervention needed by AEMO and much simpler financial settlements. Alternatively a dynamically calculated administered price could possibly be calculated each time.
If a school kid was looking at the graphs, thy would say diesel is the go!!!!!.
I really did not think it was still in the system, but it is, and its in SA I am thinking.
Not sure if there’s a (widely accepted) ‘official’ definition … but would think it would need to be over a longer period than what happened on Friday 26th August.
It is rare for all regions outside of FNQ to have very low wind conditions, but it does happen. It really does highlight the poor progress with wind development in QLD compared to other regions. While additional capacity in SE QLD wouldn’t have helped this time, certainly more in FNQ, NQ and to a lesser extent central QLD would have helped.
Paul,
surely you must know why the output was so low on that day, surely!
Having keep all the data now for more than seven years, I saw pretty early on why these low power times happen, due to those large High Pressure weather systems hovering over Southern Australia.
Here’s my own article for that day, posted the morning after it happened. (https://papundits.wordpress.com/2022/08/27/australian-daily-wind-power-generation-data-friday-26-august-2022/)
Here you have almost two thirds of total wind Capacity in the one area, South East South Australia, and Central Western Victoria, and every time those Large High Pressure weather systems come across, the bottom falls out of wind generation as they hover over that area. You only need note the synoptic chart, which I might add is always shown at the Aneroid site, so you can quite easily see it as it happens, and even as they approach,, you can tell when it’s going to be low.
And umm, it’s not just an isolated thing.
TWO YEARS ago now, I researched the matter, for the earlier time frame of 800 days, so a little less than three years, and in those 800 days, there were 270 times when wind generation failed. All of that information is shown at the following link. (https://papundits.wordpress.com/2020/11/30/wind-power-generation-intermittency-its-worse-than-you-think-it-is-introduction/)
Maybe this looks like I’m linking to my own Posts to drum up readers. Nothing could be further from the truth as this is from two years back now.
Perhaps this may be of some assistance to you.
Anton Lang.
Paul Miskelly published a paper in the scientific literature in 2012 describing the prolonged periods of very low wind that occur across the whole NEM when high pressure systems linger. It was hoped that a wider spread of windmills would even out the supply but Morgan found the same thing some years later and Mike O’Ceirin looked at a decade of date up to 2021 and found the same thing.
Complete failure of due diligence on the wind supply.
Disaster looms. https://spectator.com.au/2022/07/energy-policy-where-parallel-universes-are-set-to-collide/
The root of the problem is obvious as we speak, there is a nation-wide wind drought which means in an hour or two there will be next to no RE across the nation and no amount of additional installed capacity will make a scrap of difference.
At present we have no grid-scale storage from pumped hydro or batteries and there is none in sight in the foreseeable future. The absurdly hyped “big batteries” are only useful for FACs, not carrying baseload through wind droughts.
The so-called transition to green energy has gone as far as it can go until nuclear power is up and running. It is time for the public and the politicians to face the facts and commentators in the industry should be explaining this.
Great Article.. “winter evening periods that will be increasingly critical”. The calm cloudy winters which may last for months, like 2010, will outlast any possible storage even the great 350 GWh of Snowy2. It will probably happen again before 2030. I think “Success is by no means guaranteed” is a positive spin.
A number of factors were probably contributing. Murraylink has been offline for several days and Heywood was severely constrained at the time of your snapshot. Pelican Point has had only one GT running for about a week but was completely offline since 0130. Solar was depressed late morning because of a weather front which has led to transmission line warning market notices for SA. Many of the gas peakers were called on during the day.
“there’s also a significant volume available to AEMO, but not bid into the NEMDE price forming process (which might be for various reasons):
i.e. 4,444MW (PASA Avail) – 1,411MW (bid into NEMDE at 00:30) = 3,033MW (available, but not priced)”
Most of this difference (about 2,400 MW) relates to large scale wind and solar, for which PASA availability is usually shown as facility capacity before taking into account available wind and sun. However in most cases the difference between PASA and bid (market) availability for semi-scheduled plant is obviously not available to AEMO or to anyone else, because actual output is limited by wind and sun conditions.
Interesting read but much of it I don’t agree with.
1 – The article makes a tacit admission that long term PPAs are a marvellous way underwrite what would otherwise ne highly rsiky investments in renewable generation, but as a means o reducing emissions are in fact a bit of a scam.. (I agree)
2 – The sentence “Once implemented, liable entities (retailers) will need to invest in the roll out of not just renewables, but the storage and DR to flatten it out across the day, across seasons and come flood or drought. Where they don’t invest themselves…..” pretty much gives the game away. Just an admition that generators will no longer be obliged to supply stable and reliable power, and users will just be supplied with what ever power is available from intermittant renewables, and the huge costs associated with dealing with such intermittancy (such as storage, firming or DR) will need to be borne by the retailers and users themselves.
Well retailers are not obliged to remain in a market or operate at a loss, so all I can see happening is most energy retailers exiting the market and investing their finite resources elsewhere, leaving relevant state governments holding the can in yet another energy market thought bubble…
Back in May 2019 when we released the Generator Report Card 2018, we wrote about what we saw as an emerging schism between:
Service #1) ‘Anytime/Anywhere Energy’, a name we coined to describe the dominant form of support in use for renewable generation – i.e. not focused on the time, or the place, of its production.
Service #2) ‘Keeping the Lights on Services’, a name we coined to describe all of the other services that need to supplied over the short term and the long term to keep the electricity grid operational.
We then expanded on this Schism in GenInsights21 late in 2021.
You spoke about your ideas for a ‘Three Stage Transition’ back at an Australian Institute of Energy event earlier this year, and it sounded like you were talking about ideas that could potentially address some of the limitations of ‘the Schism’ above. Hence they are worth sharing on WattClarity here today.
James. Great article – I like the simplicity. Can you clarrify how you see storage being treated within this framework? Is storage granted a REC time stamped with the hour it is dispatched and obligated to buy a REC time stamped with the hour of charging? Or are you imagining that storage is a way that a customer is able to shift the time profile of its load (and, hence, move its REC liability to a cheaper hour)?
I don’t think this would be at all simple to implement or manage. Spot prices deal with attempting to value energy when it is most needed already, so this would add to complexity. It’s also unclear what hours are we talking about i.e. which of the 8760 per year do we plan to average? Are there summer/winter/weekly/weekend hours that are different? How are they forecast in advance? What if the market fails to meet an hourly REC target, is there another penalty? Can hourly REC’s be produced in another region? This may also dissuade low cost renewables with storage in preference to high cost renewables producing in the right hour…it makes you want to revert to a carbon price where the externality of pollution is priced and has a spot market that values the scarcity of energy supply.
I think its somewhat accepted now that the term for wind and solar combined is variable renewable energy [VRE]
More contentiously given inter state contstraints there is still value to be had in understanding the position of each region individually.This comment applies as much to Tasmania and Victoria as to Queensland.
Even leaving aside the interconnection limits issue it remains the case that each State has its own policy, for instance QLD has of course 50% renewable by 2030 and this will/should/might impact on investment decisions.
Useful summary of the different elements, thanks. One important distinction between the spot price revenue and everything else is that buyers of energy (retailers/large users) can hedge spot prices. So unless they have rashly left themselves exposed, the net cost to them of energy is likely to be well below the $6.52 bn.
By comparison, the everything else figure of $250m (plus successful claims through the AEMC process) cannot be hedged. While some FCAS costs are part of normal market charges, most of the rest is unusual and will be a sting in the tail for whoever ultimately bears these costs. Who that is is not entirely clear. Roughly speaking, i’d expect that many C&I customers would be on contracts that allowed the retailer to pass through such costs, whereas small customers wouldn’t be (and so won’t directly bear these costs). Whether retailers build in some contingency for similar future events in their market offers in the future, i don’t know. Happy to be told i’m wrong about all this if anyone knows better.
would be interested to know how much the $250m works out as a $/MWh figure. I would guess it’s in the region of a $10-$15/MWh uplift on top of energy costs. That is not insignificant.
It’s interesting that if we were to include the demand drawn by dispatchable loads (hydro pumps and battery recharging), which are excluded from measures like “Operational Demand” and “Total Demand”, then yesterday’s event wasn’t the lowest demand we’ve seen NEM wide or in New South Wales – in both cases 17th October 2021 would remain the record low day. Shows that large scale storage is playing more of a role.
A very useful piece of analysis. Obviously, the main contributor was high spot prices, it would be interesting to know what composed the $6.62 Bn price outcome. Higher demand, plant outages and higher gas prices were the dominant effects, the impact of gas prices seems to be less important unless gas producers were forcing gas bids to be higher by a substantial factor.
The question of what happens next year in my mind prices will be lower but gas prices would still remain high. However, all the planned outages would not go ahead and if demand was lower then market total revenue are likely to be a lot lower.
One might also note, although I doubt if many readers here need the heads up, for solar farms the difference between AC (inverter) and DC (panel) capacity. Solar farms generally have an Inverter Load Factor [ILF] greater than 100% because it is better economics.
Perish the thought that journalists or anyone else would want to spread fake news but it is easy to inflate the contribution of humungous batteries by measuring them in MW so we can be impressed by a 500MW battery that is going to substitute for a 2000MW coal power station. Or the total MW of all the big batteries on the ground at present or in the pipeline.
Of course giving the MW number to indicate the amount of storage is like giving the length of an area without the breadth.
We just continue to have a disgraceful level of energy literacy in this country that can be traced back to a couple of generations of people who had little exposure to basic science and physics at high school. It allows much of the evangelical spin put into the public domain about big batteries and how everything can be replaced by windmills and solar panels to be swallowed hook line and sinker by an unknowing populous – oh, and how it will all be cheaper than it is today. If we want to decarbonise then the sooner the facts get out there the better so we can have a sensible conversation about cost and how it will be funded.
Paul,
Great articles. I am on the pessimistic side as well.
I hope there are clever people working on supplying reliable cheap power to NEM.
It is disingenuous to report “new generation and firming capacity” in one number, do we assume that is 12GW of firmed generation during all months, or for 4 hours.
Boston et al (2022)* modeled using 10 x Snowy2 schemes storage and still had significant blackout during Winter 2010.
Interestingly the wind drought June 2016 you mentioned does not show up in their modelling.
Would it be possible to only run Coal power from mid-May to July?
Of course one should be talking in MWh but nut just MWh produced but worst case. So if Loy Yang A produced 16,000 GWh in a normal year we would need enough wind and solar to produce 16,000 GWh in a poor year or about 17,500 GWh at nominal capacity factors. However, capacity factors are increasing due to better controls, better technology etc. So in 2025 an optimally designed solar farm with bifacial panels etc might average 28% and similarly wind farms with taller towers and large rotors should manage 44%. So, to produce 17,500 GWh the system needs 3.2 GW of new wind and 2 GW of new utility solar. In practice behind the meter solar is still the cheapest power so it will continue to expand even if the Victorian government reduces incentives. So by 2035 at least 20% of demand will be provided behind the meter so to replace Loy Yang A you will need about 2.6 GW of onshore wind and 1.8GW of utility solar. At last count between restricted capacity of plants working up and new plants committed Victoria already had 2.8GW of wind scheduled to be online by 2026 and at least 800 MW of large scale solar for an expected addition with new rooftop solar of about 15,000 GWh/y. So that will easily take care of the production from Yallourn 8,500 GWh/y and 40% of LYA by 2026. Therefore, there are nine years to find the balance. If Star of the South is built, it alone should supply 9,000-9,500 GWh per year and ongoing additions of 600MW of btm solar will provide another 6-7,000 GWh. In sum if the only new large scale renewable plant that is built after currently financed plants are completed is Star of the South, Victoria will add 30,000 GWh/y of capacity and close 24,000 GWh. Now it is possible that no offshore will be built, but the idea that there will be little or no new large scale renewables contracted in Victoria for 13 years is completely unrealistic.
This may pose a problem or small delay for callide.
Callide Power Station’s C4 generator stuck at Gladstone’s Port as CS Energy and TMR seek alternate route amid unsuitable weight bearing bridges
Roads and bridges deemed unsuitable to carry heavy loads are being blamed for a delay in transporting a generator to Callide Power Station to replace one that blew up, cutting power to thousands of Qld homes. Here’s what we know.
Nilsson Jones The Courier Mail
October 10, 2022 – 2:06PM
A vital replacement generator destined for the Callide Power Station is stranded at Gladstone’s Port as unsuitable bridges have left authorities “scrambling” to find an alternative route.
A 250 tonne generator set to replace the c4 Unit at Callide Power Station — following its destruction during an explosion in May 2021 that impacted thousands of Queensland customers — remains at the Port of Gladstone as the access bridges have insufficient load capacity to enable transport to Biloela.
Flynn MP, Colin Boyce confirmed to The Observer on Monday morning the generator had been unloaded at the Port but transport issues meant CS Energy, which operates the power station, and the Department of Transport and Main Roads were planning to relocate the generator further along the Port.
Mr Boyce said authorities were planning to load the generator onto a barge and move it further north along the Port to Fisherman’s Landing, about 17km north near Yarwun.
“The problem here in Gladstone is we have a port access issue that has been here for many years,” Mr Boyce said.
“The generator is sitting on the wharf but we can’t move it anywhere because of the bridge and road infrastructure which is not suitable to carry 250 tonnes of generator across it.
“Everyone has known about this for a long time and now they’re all scrambling, wondering how they’re going to get it up to Biloela.”
Mr Boyce said decision would have to be made regarding which route would then get the generator to Biloela which “itself had its own problems” with roads and bridges on the corridors leading to the Callide Power Station.
Very interesting and thorough analysis of contributing factors. I wonder if the prolonged period of low Solar generation prior to the June event, led to a draw-down of low marginal cost storage supply (ie. Big Batteries and Snowy 2.0), which then led to the perverse withdrawal of bidding to ensure a more ‘certain’ compensation regime when ‘directed’. This in turn led to profit maximising bid behaviour.
It seems that if we want to prevent further ‘inevitable’, but relatively unpredictable similar events. Considerable investment needs to be made in storage.
As a pro-sumer, it seems shocking to me that the regulatory agencies largely ignore Rooftop Solar DER in capacity planning, incentives, etc. and seem only to be considering ‘centralised’ options for planning, storage and transmission.
Further, this seems a perverse and unintended consequence of not pricing emissions, nor kWh, but only Watts.
It would seem to me that like the internet, a more distributed and standards based solution is the most economic, scaleable, resilient, and reliable. One might even argue it will be more scaleable and lower cost to build.
This will indeed be a very interesting incident report to read when it eventually appears more for the things that went right. It would appear Tassie had a very real “near death” experience. The simultaneous loss of what really is the backbone of its North / South bulk transmission system must have come very close to causing separation of the island into two parts. The line faults themselves (let me guess, a double circuit tower line) will have cause a voltage depression and hence loss of some load instantly, but clearly a major automatic underfrequency load shed has occurred probably as a result of the loss of the significant Basslink input to Tassie. It appears that the Hydro machines rode through and responded under governor action as intended (primary response). Hydro units by their physics will immediately reduce export power (accelerating the water mass in the runner) for an increase governor response hence perhaps the small decrease in frequency after the UFLS recovery but then got on with it. Nice job! Must have been a bit hairy for the operators I suspect.
Andrew, I agree that the system did well to survive what was reported to be a tower failure (due to landslip?) on the main N-S 220kV double circuit line. The large scale load shedding may have been primarily operation of the FCSPS which trips specific large loads on trip of Basslink when importing (rather than frequency-driven), but may have involved some UFLS as well. Allan.
The reason for the “plug being pulled on curtailment” (double negative!) is that a prior outage on the Heywood interconnector to Victoria was completed ahead of schedule at 1025 (AEMO Market Notice 102382). This outage was the underlying cause for heightened security risk levels in SA that prompted AEMO to consider PV curtailment.
This is ideological nonsense based on neo-classical economic models. The simplest way to price electricity is to to take the average cost-plus a standard rate of return of return and retquire providers to supply on those terms on annual contracts. The facilities should be publicly owned. The idea that these are necessarily inefficient is just based on ideological observations by Hayek. The idea of letting providers opt out of supply is absurd for provision of a necessary social good. The idea of marginal cost determining price is absurd for a good where there are few providers. The “rent” for such a policy is claimed by private providers under NEM and with a high marginal cost for gas it means that provision is made at enormous system wide cost. My belief is that a courageous government needs to drop the ideological nonsense of neoliberalism and r
go back to public provision of socially necessary goods. Whatever the ‘inefficiencies” of this, they are nothing compared with the inefficiencies of having a small number of retailers, with each supporting a superstructure of trading administration supplying complexly different energy plans to the public.
Good luck with building any form of temporary 220kV transmission line in this terrain. There is a reason why even the 110kV is on steel towers. The real big one will be the loss of stable ground to build foundations in. I suggest this will take a while to sort out.
The different frequency for north vs. south Tasmania may reflect an oscillation between the 2 generation blocks especially if the lost circuits reulted in a higher impedance link between the 2 areas as well as different inertias and governor responses.
My comment about a possible frequency oscillation within Tasmania assumes that BassLink was still out of service given that it tripped during the initial fault disturbance. Another possibilty is the timing of any automatic reclose of the 2 faulted transmission lines.
Interesting work… we have also published work on this that might be of interest.
Boston, A., & Bongers, G. (2021). MEGS: Modelling Energy and Grid Services to Explore Decarbonisation of Power Systems at Lowest Total System Cost. Energy Strategy Reviews, 38. https://doi.org/10.1016/j.esr.2021.100709
A pity the Griffith report talks about “average”, electricity is all about the instantaneous. Average wind and solar will not give a stable electricity grid.
So we have over 8GW of installed wind capacity across the entire eastern seaboard but we can see total output drop to just 250MW and stay low for quite some time.
This seems to fly in the face of the assjmptions that adding more wind farms will increase the minimum ouput during a wind drought
Not necessarily, because new wind turbines perform much better in low winds than older designs, however it is true to say that even with five times as much wind capacity and better turbine designs and wider distribution eg offshore and eastern Victoria, minimum wind output at the time would probably have been only about 2 GW, however as is hinted at in the article if the NEM has 4 times as much solar which is pretty much certain, solar would have supplied between 120 and 200% of the demand between 7:30 and 4:30
At best the capability for solar is 12 hours, often much less.
Solar can only work in conjunction with significant storage which ends up being way more expensive than thermal plant generation.
Hello V,
Thermal Plant generation is not allowed out in time, it must be from what is deemed Renewable energy. Solar and Bats or solar, bats, and hydro or Solar,bat,hydro, and hydrogen. Is what will end up as the used power supply. Wind just has too much maintaining. There is also Nuke, but too many don’t like it
This will be celebrated by the RE community but it is very seriously misleading, like the new Renewable Penetration tab on the AEMO data dashboard.
The green transition is just a massive waste of money until there is some way to lift the delivery of RE on windless nights. As long as these gaps in RE supply persist we will need all the coal and gas capacity that we have at present to keep the lights on.
Pity about that, because it means that prematurely connecting intermittent providers to the grid will be seen as the biggest policy blunder in our history.
It is interesting comparing the 2 papers: Boston & Bongers 2022 and Gilmore 2022.
I have used both MERRA and NCEP RR2 in hindcast studies. Recently I tend to use the ERA5 data, it is finer resolution and contains a 100m wind for wind studies, a height more consistent with modern turbines.
There are a few comparisons papers of MERRA, RR2 and ERA5: https://asr.copernicus.org/articles/17/63/2020/
However both these papers seem to produce reasonable calibrations against recent measurements.
For storage, I prefer the Bucket Model approach taken by Boston.
This clearly show the curtailment issue in Spring/Early Summer and blackouts in Winter with their scenario of using 10 x Snowy 2.0 schemes storage.
Boston do not address Ramping which may be a problem with to many 2GW/350GWh Snowy2’s.
The shorter period of modelling by Boston may not be representative. The 42 years of Gilmore is much preferred.
The 15-year used by Boston doesn’t show if how unusual Winter 2010 Dunkelflaute was.
Interesting both papers note that 17 May 2010 was a bad day. Boston shows this continued for the whole winter and the storage never recovered.
Boston model assumes perfect transmission. Tasmania was not included in Gilmore’s work, it’s pumped hydro will be critical storage.
Boston discussion for the use of offshore wind farms with their time-shift of wind generation and increase CF was very useful.
We need more of these long term studies done to ensure the reliability of the NEM.
Looking at the curtailment are we reaching peak Solar?? I think storage has become the main issue.
I am certain clever people in AEMO are looking at all these issues.
I believe you are destined to be eternally disappointed, because the bureaucrats are in charge now, and they won’t let go without a fight. They have targets to meet you know.
“Where has all our engineering gone?” – indeed.
We often hear from credible experts on scientific, medical, legal, economic, and even defence topics. Yet when it comes to a sensible and informed discussion on powering our national electricity grid, the best the media can manage is politically loaded partisan mantras and slogans, overlaid (or is it buried) within the religious imperative of respecting climate change as undeniable. Who must take responsibility for this monumental ignorance of the community at large.
Can confirm that the SA network voltage was raised for approx. 4 and 1/2 hours just after 12.30pm.
Our monitoring systems alarmed with a voltage increase of around 200v on each phase
Re supply authorities ramping up voltage to trip inverters and Jim’s comment of of a very large increase at his place. I’m wondering if that increase in voltage that Jim commented on was natural. Don’t forget that at the household end of the grid, each suburb has a final distribution transformer, stepping down (usually) 11KV reticulation to 240/415volts. If the energy of all the suburban panels at midday can’t be used up by a load, then a higher voltage will naturally occur. 240v rising to 270v is not out of the question (and this will trip inverters), but 200 extra volts seems very large and would fry many household appliances. Also I find it hard to understand how the SA power authority would easily be able to increase voltage. Most final suburban transformers can only increase voltage with a manual change-over of their output tappings. It’s also very dangerous to increase voltage on a grid scale as things will burn out. Lastly voltage levels are set by the AS 3000 Wiring Rules, and must be no greater than around 10% of normal.
Thanks Dan. Good analysis, but I’m not sure I’d agree with the headlines. These frequency traces seems remarkably good compared to previous separations. It would be worth comparing to reports on them in AEMO reports.
Separation circumstances are always different, but I suspect we are seeing the beneficial effect of fast acting inverter equipment.
Hi Paul,
Your comment “the Murraylink interconnector is still operating, but it is a DC interconnector so there’s a limit to the benefit it can supply to South Australia”
Why is a DC interconnector limited benefit?? Any reference would be appreciated.
Thanks
A DC connector can not provide frequency control services. It could provide some MWs, but otherwise SA’s frequency was completely separated from the rest of the NEM, same way Tasmania’s frequency is always separate. Tasmania’s grid is already more unstable than the NEMs.
Don’t have any references, just my guess.
I look forward for your review on the day in particular the constraints and synchronous generation dispatched during the day. It was surprising to see Dry Creek peakers and diesel thrown into the mix.
My pv has been constrained to internal demand only until thursday when it was completely cut off to zero.
This was frustrating as obviously generators had a good day at my and the publics expense.
No market operating and no feel demonstrated by the powers in charge to allow home pv to cover in home use. This was a planned action to limit pv should have been done with more finesse than it was as the circumstances of too much non scheduled will only increase in frequency (no pun intended!)
Imagine supporting the development of an energy sector where you literally have thousands of megawatts of supply just evaporate or appear out of thin air and thinking this is somehow a good solution.
We are going to get more and more frequency events creating islanding like SA saw last week because of the unreliability of renewables.
Wonder how both models performed on the night of Wednesday 7 September 2022 when South Australia “lost” 433 MW of wind generation that didn’t arrive as forecast at 11 pm.
Thanks for covering this Dan. Just noting that the Reliability Panel has completed the 2022 Reliability Standard and Settings Review and recommended an APC of $500/MWh for the period 1 July 2025 to 30 June 2028 (along with recommended changes to the MPC and CPT).
The Reliability Panel will now need to submit a rule change request to the AEMC in order to implement the recommended changes to the settings. The AEMC would consider any proposed changes through the usual rule change process, allowing further opportunities for stakeholder input and consultation.
One thing I don’t understand is how the Hornsdale wind units can provide FCAS services. That seems to be a contradiction in terms. Gas and batteries make sense, but wind units? Can you explain?
Hi Paul, the three Hornsdale wind units are registered to provide six of the eight FCAS services (Lower 6 Sec and Raise 6 Sec contingency the exceptions). The document below describes the trial process that was undertaken in order to allow them to do so:
Any wind or solar farm that has suitable inverters and control systems can register to provide FCAS. Currently my understanding is that they are only providing a service by quickly cutting their output, but having an output like a big resistor to dump load into, could also provide a similar service.
However there is no reason that a wind or solar farm couldn’t de rate their output, thus holding some capacity in reserve to deliver immediately. To date I don’t think AEMO is registering this later form of capacity as it is considered more risky than having a battery operating behind the inverter.
AEMO forecasts a need for “the equivalent of up to 40 large synchronous condensers” provided for by a “range of technologies” to sustain 100% instantaenous penetration of renewable energy. There are just 4 synchrons in SA today. I haven’t seen this mentioned anywhere. What sort of technologies could provide this service? Does this imply massive investment in synchrons?
Paul Hyslop’s article like many others, including unfortunately some of my own, focusses on the 6300 kcal/kg coal price which is (i) extremely high relative to history and (ii) very visible because there is a futures market. However it is not I have read a highly traded contract and possibly not a good reflection of the lower spec 5500 kcal coal that forms the bulk of sales to Asia and which is, in an even less washed form, what is mostly used in the domestic market at least in NSW. The lower spec coal was recently “only” US$136/ or say A$200 and at 23 gj/t is $8.7/gj and at a heat rate of 9.5 gets you to a fuel SRMC of “just” A$82.6 although that is pre transport. Using the “proposed” and no doubt soon to be legislated cap A$125/t and assuming probably incorrectly the same number of GJ/t gets you to say about A$55/MWh + transport and other costs. Not sure what freight rates are at the moment
It’s not hard to see where this is all going. The AER et al are now operating beyond their core function, by interfering to promote wind and solar. Regardless of one’s position on the reasons, the outcome is entirely predictable – increasing market intervention until the government owns it all.
The media release is incredible, with three ‘recommendations’ including:
– contract monitoring
– increased regulation of coal-fired generators
– where governments have choice over their investments or underwriting, consideration should be given to actions to diversify the operation of dispatchable generation
The AER admits that hydro is setting high prices more often. But governments own the hydro! There’s no price caps on hydro being mentioned!
Let’s look at NSW post-Liddell. With 1300 MW needed to be supplied by the remaining coal-fired generators, they need more coal. But if it’s not already contracted, who’s going to sell it to them under the price cap?
The only way that will happen is if Matt Kean intervenes again and takes the power to curtail exports for local use. That’s a coal reservation.
If not, then the power will come from gas. But Bowen stuffed that up too, so the feds will need to implement the DGRM.
In both scenarios – forced export curtailment – the producers will be compensated. By taxpayers. So we pay anyway, but this time in national debt and less opportunity as resource companies say ‘no thanks’ and gradually pull out. Which is exactly what the Greens and co want.
As private companies start making noises about offloading projects, and politicians realise these power stations and mines and gas projects are needed to keep the lights on, politicians will spend public money to nationalise it.
Which is exactly what the AER has recommended ie diversify government investments into retiring coal to keep it online – Yallourn is already there, Eraring will be next. I’ll be surprised if Liddell doesn’t get a lifeline.
High prices have been a function of several things, mostly fuel and generation scarcity. The AER’s price-setter data shows hydro is increasing wholesale prices far more than coal and even gas. Hydro is not affected by Putin as far as I can tell.
Why the focus on coal, apart from ideology and the narrative?
Given the distance of Cressy from Heywood and Portland the SA island resulting from the 31/01/2020 incident should probably be listed as the SA & Western Victoria island which raises the question of whether SA was able to supply the smelters at Portland and what were the possible/ likely consequences if not. Was SA reliant on renewable sources during the event and how did they perform? What fossil fuel sources were available to assist and how did they perform?
SA was, with the assistance of one fossil gas based generator in Victoria, able to cover the needs of SA and the very significant load of the Portland smelter. Due to the more limited options and the conservative approach of AEMO, many wind and solar farms were direct to shut down. At the time of the event the Hornsdale Power Reserve (Tesla Big Battery) was smaller and not capable of operating as a grid forming inverter (synthetic inertia). Nor where any wind or solar farms. The SA grid now has much more grid stability capacity from at least two batteries and one wind farm, so when SA separated recently far more wind and solar farms were able to keep operating. There is another large grid forming inverter based battery under construction in SA and others planned, so in the future there will be little to no need for fossil fuel based generation in SA to provide grid stability services even during a separation event.
Thanks Dan for hunting down all these transmission fails into one reference.
Re Basslink – I think I read at the TasFinTalk blog that Basslink has never paid its way.
The NEM currently operates in an old style grid where the timing signals for the phase angle and frequency is transmitted via AC transmission lines. This makes sense when the grid is dominated by large synchronous generation like coal and gas fired generators. However it is not the most accurate and creates greater risks when separation events occur. For years now it has been more accurate to have phase angle and frequency coordinated across large grids using atomic or similar reference clocks to drive grid forming inverters. Such as system has no issue with regard to frequency control when separation events occur. The cost of these highly accurate clocks and referencing systems have come down significantly to where they are less than a few thousand dollars, so can be installed at all sites in the NEM with grid forming inverters.
My recollection of the Callide C4 debacle is that both dedicated and the Backup shared 120V DC supply busses failed (or were turned Off). All super-critical plant control functions were inoperable. Critically, the turbine could not be disconnected from the Grid. (The critical OCB is in a substation about a kilometre remote from the turbine.) Unlubricated bearings then overheated and seized, progressively increasing the load drawn from grid until the Grid collapsed some 45 minutes later. A truly magnificent balls-up!
ABC news website had article on 26th May 2022 entitled “One year on from Callide Power Station fire, answers are still being sought”. This mentioned the report on the forensic investigation by Dr Sean Brady would be likely be finalised later in 2022.
All very clever and all very well, but what happens when the millions of roof-top solar installations reach “end of life”?
Is anyone suggesting that taxpayers will again be on the hook to subsidise their replacement – as was vital in the original take up phase?
I take issue with the premise that “they might be able to set “dynamic” limits on households for the import and export of energy.”
From this simple idea and assuming that both electricity providers and governments are not passive actors in this scenario. It won’t be long before you get a text message that ‘your electricity use has exceeded the limit for this time of day, you will be fined $XXX, if it continues further action will be taken.’
This is a neat way to ensure that our electricity network will be protected due to huge fluctuations in supply.
On the “uprating” of C4 to 466MW, it is possible to do with some quite noticeable loss of efficiency by bypassing quite a few of the heaters, which is what was done with C3. Effectively running the unit in permanent “overload”. Given the likely remaining (economic) life of the units, pushing them hard for what time remains probably makes sense.
Are there going to be unforeseen ramifications from the triggering of the RRO in South Australia from 12 January to 13 March 2026?? One South Australian MLO generator has about 900 MW of gas-fired generation of which about 40% are Open cycle gas turbines (more expensive electricity than combined cycle gas turbines) and on current advice to AEMO their OCGTs will close in 2030. Of the MLO’s OCGTs about half run on gas and the rest on diesel. How can a South Australian MLO with OCGTs due to close in 2030 offer contracts for March Quarter 2026 with uncertainty over gas supply and gas prices and Project Energy Connect due to be operational in July 2026? Will the $19.5 million deal between the South Australian Government and the other South Australian MLO be enough to keep the lights on in South Australia?
At the risk of adding more complexity, a better way of assessing curtailment or economic offloading is to just consider the difference between forecast availability and target output. If there is no forced curtailment, or offloading due to offer price, then target output will equal availability. Differences between target and actual for whatever reason – poor forecast or something else – shouldn’t go into calculating curtailment or economic offloading.
If we followed Allan O’Neil’s methodology of assessing curtailment as the difference between forecast availability and target output wonder what the result would be for these two solar farms.
Imagine a summer where green hydrogen vessels are lined up at east coast ports and electrolysers need to operate all day every day to supply the export green hydrogen. Politicians are speaking as though this will be Australia’s reality pre-2030. Add to this demand from EVs and desalination plants and what will be the NEM peak demand? How much pumped hydro and batteries will be operating pre-2030? The alerts may be coming in thick and fast and not just in summer months.
Much of the green hydrogen and related products are likely to be off grid. That is even if they are partly connected to the NEM, the bulk of the energy will generated and consumed outside of the NEM. I notice that the QLD owned transmission companies have been trying to at least obtain a small cut of this potentially large pie. I expect that we may see private transmission lines in a few locations to avoid some of the costs and regulation related to the NEM.
Having had experience with a similar structural failure of a mechanical draft cooling tower many years ago the timeline does not surprise me too much. Even though the tower is not that old as these things go, the more they will have looked the more cracked and rotten timber they will have found (yes these things are big timber structures) and trying to remove / replace / reinforce damaged timbers in a very constrained space says it is a very slow process. Particularly when you are coming at it from a standing start. If you knew this was coming at you and you had a complete kit of timber to hand it is easier, cheaper & quicker to pretty much demolish and rebuild the timber bits and install new fill, but as a break down job it’s just hard work.
As noted in the article, Michael, the ‘N>NIL_94T’ constraint equation was one of the most significant ones. Obviously with the ez2view software we can drill in further … but that would be a separate (and longer) article.
Bit ironic that on 3rd Feb this year consumers could be facing yet another very expensive RERT bill. On 1st Feb last year AEMO forecast a Queensland demand over 10,000 MW and it cost Queensland consumers a whopping $50 million for the period 1700 hours to 2130 hours.
I am not sure why the demand is forecast to be so high. Yes it will be hot and humid, but nothing that is even close to a record. It just looks to be on the higher side of average for SE QLD at this time of the year, not very pleasant.
RERT used to cost less than $5 million per event but now it can be more than 10 times that for a single RERT event…. all ultimately paid for by consumers. This is all happening before the first coal generator retires in Queensland and BEFORE desalination becomes Southeast Queensland’s new source of drinking water. Imagine being dependent on desalination (liquid electricity) for your water supply. Probably why Dubai doesn’t run on solar!
The weather conditions didn’t seem anything like a record, so it would have been disappointing if a record Market Demand had occurred. The Market Demand was only 3% or so short of the record. There are so many variables in the mix for these records, weather, changes in work patterns (more people working at home), fewer international students and lower tourist numbers than pre Covid.
There does appear to have been a better level of self consumption of rooftop solar than there was under similar conditions say four years ago. This is a good result for the grid as a whole. Further improvements would be obtained if all residential customers were moved to a wholesale linked tariff.
Maybe also customers are getting smarter and using simple timers and manual interventions to minimise peak demand. Peak grid demand this summer has been 29.8GW well down on previous years
It would be interesting to know if, and by how much, Energex triggered the demand management on residential air conditioners. There would be significant number of air conditioners in SE QLD that Energex can ramp down the compressors on. This program was/is an opt in program with an up front incentive paid.
Hi Craig and Peter. Thanks for the comments. It appears Energex’s PeakSmart program was indeed activated on 3 Feb. Participating AC units were capped to operate at 50%. This would have definitely helped.
Hi Paul, you are right. This is obviously an expanding frequency oscillation although luckily it does not break the +/- 0.15Hz deviation. A few observations: 1) disturbance always exists in the system, but usually well damped. However, with the high load, the damping is often weak. The system does not look like well damped in this case especially around 13 o’clock. It would be interesting to see the voltage profile. Usually they are decoupled, but not in the case of weak damping. 2) The skew to the lower frequency indicates in average the load is increasing, thus although it happens in the mid of the day, the rooftop solar should not be scolded for this expanding oscillation. It is more likely that the disturbance from ramping up the outputs of the generators and even switching in some generators both introduce disturbances and can cause this oscillation. After the transients of ramping up the generation, the system should settle down with much smaller deviation. 3) The oscillation looks like has a period of 10-15minutes (say 4-5 cycles between 11am -12pm). Does that time scale also roughly conform to the time scale of ramping up the generation and the AGC? Highly likely. Anyway, thanks for sharing and it would be interesting to use these observations to validate the models used for system frequency studies, but of course hard as the non-transparency of the control model, which is fully understandable.
Thanks for the analysis Linton. A quick query for you and the other smart folks at WattClarity. It seems that the spot price peaked at $1,297/MWh, well below the market price cap (MPC). And this was only for one dispatch interval (DI), before RERT was activated. During the period of RERT activation, the spot price barely topped $300/MWh. This implies there was in fact more supply available which surely have been a lot cheaper than RERT. I appreciate AEMO has to make the call on RERT in advance and that we don’t know yet how much they paid. But is this the lowest price (outside of market suspension or CPT situations) at which RERT has been activated?
Kieran on a very quick scan it looks like there was a RERT event with lower outturn prices in Vic on 30 November 2017, and another on 30 Dec 2019. Quite possibly there are others as well. As you point out AEMO has to make the call in advance.
Pricing under RERT uses intervention methodology and doesn’t perfectly reflect what might have happened without RERT being called – I’m sure that’s not an issue last Friday, but may have been on those earlier episodes I mentioned.
It would be interesting to know if in the most recent case AEMO had other RERT options ready to go with shorter lead times, which they were able to avoid activating when demand fell short of forecast. The 21 MW dispatched last Friday seems hardly worth bothering with by itself.
In terms of ‘should we be surprised’, there’s several levels to your question.
In our widely read GenInsights21 report we illustrated how there were a number of drivers towards increasing volatility into the future. Hence the answer’s ‘no’.
To a great extent this seems like a truism: the cheapest generation isn’t available at the most expensive times, but that’s precisely why those are the most expensive times.
Pretty interesting outcome last Friday in QLD I thought. I was wondering if some of the demand degradation might be driven by tariff 31 and tariff 33 whereby retailers can turn down residential load via heat-pumps and other energy intensive devices during the peak?
SA and VIC prices are mostly hard to work out. SA has had the diesel gen sets pumping out over 300Mw for well over an hour yet the price is only $500. One would think to get any kind of return on diesel gen sets you would need a bit more than $500.
Also of note, almost everyday SA have gas running, yet the price is often very negative. They must have cheap gas somewhere. Maybe they have found a new kind of Gas turbine and finance system?
SA prices are currently $4000. All fossil have been online.
Gas running when prices negative are because of AEMO system security requirements to have two synchronous generators in SA. The minimum output from gas is 80MW.
Hello Alex,
So does that mean SA does not operate like all other markets and gas because it has to run does not get costed in.
As I understand it, the last bid for the market set the price, this cannot be happening in SA unless gas is not bidding a lot of the time.
The cost of FCAS (grid stability services) is not factored into the wholesale spot prices, but is applied through the additional costs similar to the transmission and distribution costs. The amount of standing fossil gas operating in SA has decreased due to the relatively new synchronous condensers providing that service at a lower cost. In the future there will be no requirement for standing fossil gas generators to provide inertia to the SA grid as this will be provided at a lower cost by big batteries and the EnergyConnect interconnector to NSW.
So the cost of something that is deemed to be needed is not costed into the cost to run the grid in SA, yet we are fed thats its wholesale price is much less than anywhere else. Hmmmmm.
The Big Battery (HPR) is mainly there to provide grid stability services, so if one of the other generators falls over or the interconnector fails, then it can keep the system stable while that generator or transmission line with the problem either comes back online or another generator fills the gap.
When the additional big batteries come online there will be additional capacity in the system to provide more battery supply during evening peaks.
The diesels should achieve about 36-38% efficiency so 3.9 kWh of generation/litre at a tax free price of $1.40 that is 35c/kWh or $350/MWh + $10-15 in other variable costs, $500 is a good contribution to overheads
I agree, if you owned it and you got that for it every day for a few hours maybe, but when it sits around for long periods and only runs every now and then, $500 is not what it needs. Very miss priced, not reflecting the actual cost to have such a thing, in your energy mix.
On Thursday there were periods where the diesel generators were receiving more than $4,000 per MWh. The standing capital costs of diesel generators aren’t that high, nor are the maintenance costs. It cost is all about the fuel. It doesn’t take long at $4,000 per MWh to cover the standing costs for the year.
Hello Craig,
I would have thought things should get run to make money?. Are these Genset’s there for simply a backup and to make sure SA has power?.
If that is the case its yet another thing thats very different about how the price in SA is being arrived at.
A lot of this talk feels like re-arranging the deckchairs on the Titanic because there appear to be three principles that constitute something like an Iron Triangle of Power Supply.
1. The grid needs continuous input to meet 100% of the demand.
2. Nights with little or no wind break the continuity of wind and solar input.
3. For the foreseeable future there will be no effective way to store enough power to bridge the gaps.
Conclusion: We need to keep enough conventional power, mostly coal power, to meet the highest levels of demand at dinnertimes in high summer and deep winter, until we have nuclear power on deck.
In the meantime people need to appreciate that wind and solar can DISPLACE coal but they can’t REPLACE it.
The rate of exit from coal is not accelerated by increasing penetration on good wind and solar days, it is limited by the lowest level of output on nights with little or no wind, as a convoy travels at the speed of the slowest vessel, the water penetrates the levee at the lowest point, a chain is only as strong as the weakest link and stock get out of the yard through gaps in the fence even if the rest of the fence is built to the sky.
South Australia has proven you can replace coal with renewables. But getting rid of gas is more problematic.
On the NEM, existing hydro is also available to help meet demand during times of low wind & solar.
You might want to look at AEMO’s ISP to see how we can get nearly all our electricity from renewables. Or you can look at my own simulations here: https://reneweconomy.com.au/a-near-100-per-cent-renewables-grid-is-well-within-reach-and-with-little-storage/
Both Tasmania and South Australia had a week long outage last year and things are only going to get worse as fossil fuel generation is removed from the sytem and replaced with unreliable renewables.
Funny how these outages are becoming more common as we demonise domestic fossil fuel generation and rely on importing fossil fuel power over significant distances from other states.
Why was Market Demand at a 9 year high while Operational Demand only hit a 3 year high?
I think the reason will be those non-scheduled wind farms that now have to bid and be dispatched like their semi-scheduled cousins. That change meant that their output is now counted as contributing to meeting Market Demand; previously their output effectively subtracted from the grid demand met by market-dispatched sources.
In other words there has been a basis change in the last 18 months that increases the measured Market Demand.
On the other hand the Operational Demand and Demand and Non-Scheduled Generation measures have always included demand met by these non-scheduled wind farms.
This does leave (at least) winter 2003 where Liddell will be gone and this new gas-fired dispatchable capacity will not be available. Currently, AEMO’s MTPASA medium-term supply-demand outlook is not highlighting any reliability gap over this period, noting that the ez2view chart below does not explicitly include import capability into New South Wales via the Queensland and Victorian interconnections:
So now let’s take out the 650MW being contributed by the two running Liddell units for next summer and you are in LOR2 territory……and if the contingency then is loss of the interconnector to Qld and things start to get a bit dark. How come there are other, more modern and efficient plants (BW & MP) off in the first week of March? This is starting to look third world stuff. This slow-moving train wreck will not be arrested quickly either. Ironically, the last time NSW was on the cusp (and actually in) this sort of disaster over summer was in the 1970’s when the (then) near new units at Liddell started having generator failures. Market price signals only work when there are some generators left to respond to them!
None of this is terribly surprising. If the asset managers in the stations with announced closing dates are doing their jobs they will have started rationing expenditure on maintenance (both routine and particularly sustaining capital) so that the units are kept safe but will perform a graceful bellyflop on their last day of operation. This means you can expect increasing unavailability and particularly forced outage and derating where you start to increasingly operate on run to failure / breakdown maintenance. If you have no interest in the longevity of your units then this is the cheapest way of doing things. Just guard against catastrophic failures that can kill people and off you go. People outside the industry often don’t understand that a cheap little component can quite easily stop one of these big units. Given that these units have maintenance cycles measured in years and this is the reason you cannot easily change direction once you start going down this slippery slope. Plant engineers have been doing this for decades when they move to retire old power stations – so they know how to do it. What is different this time around is the process is being undertaken across many units simultaneously – many of which would normally have had many decades of operating life but simply they have had their revenue and hence commercial viability destroyed. Inevitably will end in an untidy outcome.
Further to Andy’s comment NSW has nominally 9,500 MW of coal after accounting for the derating of the remaining three Liddell units and yet peak coal output was 7,017 MW. It also has 2 GW of gas which only managed 1.2 GW, not to mention 4.1 GW of hydro which maxed out at 2.1 GW. Combined output from 15.6 GW of dispatchable capacity did not exceed 12.1 GW, and 11.5GW of coal and gas maxed out at 8.3GW so much for the “reliable coal and gas”
You could look at https://electricinsights.co.uk/#/dashboard?&_k=ykdb3l which shows the UK still uses coal most of the time. Technically it still has 4GW connected to the grid but output has not exceeded about 1.6 GW for about 18 months. In 2021 the UK ran for more than 5,000 hours without coal, but even though annual coal share has continued to decline, http://www.mygridgb.co.uk the period with zero coal have actually fallen over the last couple of years, https://www.mygridgb.co.uk/coal-tracker/
It would be helpful if you could provide or point to a simple but clearer explanation of the issues discussed in 13.2 re “hybrid” facilities. For example why would a renewable-battery combination want to be semi-scheduled?
Great charts on the curtailment/constraint side of things. In the USA based on NREL work it seems the DC/AC ratio is known as the Inverter Load Factor (ILF).
With about 35TWh/y of renewables committed including rooftop solar over the next 4 years 19 TWh/y of coal retiring and probably 2-3GW of storage in front of and behind the meter and storage/pumped hydro sucking up negative priced coal and solar power at midday and dumping into the $130/MWh evening market, what role is left for gas. If gas generation is half what it is now in 2026/27 it will be amazing.
In the meantime, heatpumps for hot water and space heating combined with gradually increasing energy efficiency will accelerate, so AEMO’s 2% per year decline in other gas demand will be looked on with wistful fondness by the gas companies in a few years’ time.
It looks like a significant driver of the volatility (in addition to high demands and generation outages in NSW and Qld) were constraints on the New South Wales 132kV network which limited flows from the south on the main 330kV route from southern NSW & Victoria. Check out the N>>NIL_998 and N>NIL_999 system normal constraints for more detail. Hard to say without deeper analysis but could have been a “tail-wagging-dog” or “gateway” effect where limits on a peripheral flow path throttled the adjoining higher capacity path (but I don’t know by how much). If so, another argument for getting the transmission system in better shape to deal with changes in where generation is concentrated.
A lot of focus is rightly placed on high demand situations; my question is whether there is value for “dispatchable demand” in high generation/low demand situations?
I have developed a manufacturing device that converts electricity to plant food; it is profitable to operate even if on only 2 hours a week on average, if the electricity coming in is under $0.02/kWh. It can turn on/off in the 5-10 second time frame.
Is this type of capability of value for a grid operator, or I should just focus on being a consumer?
Hi Dan, as a longtime resident of rural Timor-Leste i am familiar with the unreliability of power, especially the daily blackouts in the rainy season (could be 1 hour, 4 hours or even on occasion 4 days)
For rural communities such as Atauro (or the coffee farming households/ hamlets, villages in the mountains) are you able to suggest any affordable and sustainable solutions?
Hi Peter thanks for reaching out, and it’s great to read about your work at Timor Mountain.
As for your question, I am unaware and unsure of the Timor-Leste government’s plans to improve or augment the transmission/distribution network in the immediate future. So it *may* be the case that consumer-led or community-led solutions are the answer.
In the short-term, lead-acid batteries or other forms of UPS – are ways to increase realibility. We met with one hotel owner who was doing this effectively to have 24/7 power on Atauro.
In the longer-term, perhaps community-based offgrid solar + battery projects are an efficient way to pool resources together to provide reliable and sustainable electrcity supply to individual villages.
Obviously, a lack of capital and a lack of skilled electricians/engineers are major barriers. We have a number of students at UQ who have shown interest in doing further research into this, if you’d like me to keep you updated then please shoot me a message at daniel.lee3@uqconnect.edu.au
c1ue, under the Wholesale Demand Response Mechanism you can be paid to turn off. It’s also possible for smaller consumers to get wholesale-linked pricing through retailers like Amber (so you can turn your plant on and off depending on the current wholesale price) or very large consumers to directly participate in the wholesale market as a dispatchable load.
I think there’s possibly a bigger problem (albeit related to your root cause suspicion).
The question in my mind is what is the impact of rebidding (esp. as the interval decision point/s approach).
It’s an issue that is also seen, with much bigger numbers, in wagering in general but especially totalisators (TABs) as the “jump” approaches there is both an increase in information updates (approx dividends) and number of bets … and rapidly increases the transaction rate (it’s referred to as an avalanche and it doesn’t do it justice especially on big gallop races!).
Efforts to manage “collations” via (sub) aggregation results and new challenges (and ones that the pro-punters hate because they want (must) bet as late as possible with the most information)!
It’s only my opinion (thoughtful/considered but nonetheless still an opinion) that the market (especially the consumer) would be better served if “bids” (rebids) had a total limit or at least a cost to them (escalating as EoI approaches)
Without targetted research it’s difficult to be certain but imho it reasonable to assume that generators (the predominant bidders) are trying to maximise the despatch price. Unrestricted bidding facilitates that!
Kevin,
Thanks – I have looked at this type of setup in Texas.
What I am wondering though, is just how well this will work as curtailment amounts and %time negative/zero electricity prices increase due to duck curve mismatches with solar PV/wind generation?
Large scale industrial users of electricity – their biz models and tech simply don’t work in an intermittent usage setup.
So I guess my question is whether there is value to having dispatchable industrial demand at scale vs. simply turning off plants/generation.
I do know Tesla is trying something along those lines using coordinated Powerwall recharging, but I really doubt just how much dialable demand can be aggregated this way.
There was an outage on a Victorian 220kV transmission line (one of the Thomastown ones) earlier overnight. This implemented a constraint on the Vic-NSW lines. I wonder if that was at play, constraining down Vic-NSW transfer?
Clearly one lesson that should be learnt is that “Summer” doesn’t finish on the 1st of March. It shouldn’t be assumed that scheduled maintenance can start on the 1st of March.
There needs to be greater focus and reporting by AEMO on the performance of interconnectors and major transmission lines. Too often we have seen major supply problems which would have been far less critical, or not even occur had there not been constraints on transmission.
It would be interesting to know what residential demand controls were used in QLD during these periods, such how many and how by how much were residential air conditioners ramped down.
These events show the need for almost all residential consumers to be moved to wholesale spot price linked tariffs, so residential consumers are incentivised to reduce their consumption during these periods of extreme demand.
The problem does not seem to be capacity of the interconnects themselves, after all VN1 has been upgraded and one of the justifications of the Victorian big battery was that it would allow an increase of 150-200MW of Victorian power to NSW by stabilizing the transmission link.
Despite these upgrades imports from Victoria were constrained to zero on the 16th of March and total net imports from Victoria over the last 90 days were about 500 GWh, yet way back in 2010 NSW imported 1,950 GWh in Q4 from Victoria. Further in 2010 NSW prices were only 21% higher than Victoria, now they are almost double, so the cost of transmission can’t be the issue.
How come capacity has been increased 20-30%, the price differential is as high as it has ever been and yet flows are down by two thirds. Methinks either incredibly poor grid management by Transgrid or some seriously dodgy constraint equations in the NEM dispatch engine
‘The joint venture partner in control of one of Queensland’s major power stations has collapsed after a dispute over future funding for the troubled asset. Deloitte turnaround and restructuring partners Grant Sparks and Richard Hughes have been appointed voluntary administrators over a number of IG Energy Group entities, which own half of the troubled Callide C coal-fired power station near Biloela.’
[Editor’s Note – the rest removed so we don’t breach copyright]
‘State-owned energy company CS Energy is considering buying out the remaining stake in the Callide C coal-fired power station after its joint venture partner, [**Genuity**], went into voluntary administration on Friday.
The move is seen as a way to secure the long-term future of the power station which has been beset with problems since a mysterious explosion at its C4 unit caused havoc on Queensland’s power grid in May 2021.’
[Editor’s Note – the rest removed so we don’t breach copyright]
[2nd Editor’s Note – note the Editor’s Note above in the article itself about the corrected name for the JV partner]
Excellent article Dan. Another consideration is what did the developer or owner assume for all these factors when building their business case or arriving at the price they paid for the project. Whatever the effect of each factor, if it turns out operationally to be ‘better’ than assumed, then the project could still be regarded as successful.
As a mere mechanical engineer, a lot of this is barely understood, but my simple question is. Why did we spend $42m upgrading VNI and in the last quarter Victoria exported 350GWh to NSW yet in Q1 2015 we were able to export 1,500 GWh. Back then the price differential was $7/MWh-this year $43/MWh.
This seems to need more than tweaking a couple of constraints in a constraint matrix. It would seem that either the money was spent in entirely the wrong place and Transgrid is being paid some $4m/y for adding costs to consumer bills or that the whole system model needs to be rebuilt
Peter – very reasonable question, no-one who looks at the current situation with VNI thinks it’s an acceptable outcome, but unfortunately it reflects the current market design and getting agreement to change things like transmission access and pricing has proven nearly intractable. If there was a “NEM benevolent dictator” this particular problem could probably be fixed tomorrow – but no-one is in charge of the NEM and the number of stakeholders with wildly different positions on things makes changing the status quo very very slow and painful. Allan.
The difference is mostly due to demand met by “non-scheduled” generators who do not bid their output into the spot market. “Operational” demand includes this component, but “market” demand is exclusively the amount which is supplied by generators through the spot market.
The more telling information is what the demand is in the sunset to two hours afterwards. That is the size of the duck curve which the grid operators will need to ramp up for each day.
Yes, when we have extremely low demand and high renewables (like over the Easter weekend), we’re likely to see little change (fewer renewables constrained off). But that was never the concern. The concern is every shoulder and evening peak since we will have 600-960 MW of bids lost. And it doesn’t have to be 16-March levels of demand for it be translate to higher spot prices.
It should be relatively simple to understand that prices will rise as we saw with Victorian prices after the closure of Hazelwood.
You can’t remove continuous generation capacity and expect a more volatile system to provide the same reliability and thus the same pricing.
It’s only going to get worse as Queensland progresses with it’s coal plant closures.
So Project Energyconnect is adding 800MW of transmission capacity between Buronga and Darlington Point – presumably this will result in decomissioning this section of the X5 line and replacing the X5 constraints with new, looser ones?
A bit of uncertainty is hardly surprising. Will be all about getting the last of the coal yard cleaned up, as many coal bunkers cleaned out as safely possible and maybe even some slops out of the fuel oil tanks used up. Get rid of as much as possible rather than trying to clean it up as part of the decommissioning process.
Good luck for winter guys. Taking another GW of dispatchable generating capacity out of what was already an uncomfortably tight winter last year…..
Why would anybody be “popping champagne corks”?
No doubt the demolition will start the day after shutdown – to ensure there is no way back.
Surely prudence dictates a period of (perhaps 12 months) care-and-maintenance before declaring that the entire grid is stable and maintainable.
That is a rather bland reference to a period where the NEM would have gone black in the absence of remaining coal burners with the wind practically absent from Friday evening through the weekend into today.
At sunrise on Saturday morning the wind was delivering 6% of demand at CP 11%, in the mid afternoon the wind was delivering 4% of demand and RE was 33%, at sunset the wind delivered 3% at CP 8%, Sunday morning it was 5% at CP 10%, Sunday evening 3% at 6%CP, sunrise this morning, 3% at CP 6%.
It all over Red Rover! RE can displace coal, but not replace it. https://www.flickerpower.com/index.php/search/categories/renewables/21-7-intermittent-solar-and-wind-power-can-displace-coal-but-cannot-replace-it
The rate of exit from coal is not accelerated by increasing penetration on good days, it is limited by the lowest level of output on nights with little or no wind, as a convoy travels at the speed of the slowest vessel, the water penetrates the levee at the lowest point, a chain is only as strong as the weakest link and stock get out of the yard through gaps even if the rest of the fence is built to the sky.
As long as periods with effectively zero solar and wind power persist, 100% backup from conventional power will still be required, assuming that we want security of supply. This means that we will have to keep burning coal until nuclear power is on deck.
Four pieces from The Spectator.
A simple demonstration that wind power will not work in Australia due to the need for continuous input to the grid, the interruption of wind power by wind droughts, and the lack of storage at the scale required to bridge the gaps. https://www.spectator.com.au/2023/02/the-iron-triangle-of-energy-realism/
On wind droughts and lack of storage in Australia. When the next coal power station closes, every wind drought will threaten the power supply and prolonged wind droughts will be potentially catastrophic. https://spectator.com.au/2022/07/energy-policy-where-parallel-universes-are-set-to-collide/
The vital importance of “wind literacy.” Wind droughts have been recognized in Australia for a long time but defective advice (neglecting the droughts) resulted in the the worst policy blunder in our history. This is connecting subsidised and mandated intermittent wind and solar power to the grid. https://www.spectator.com.au/2022/11/the-voice-of-energy-realism/
Rafe your comments are valid. Looking at the AEMO dashboard it appeared NSW compensated for the wind drought by using extra hydro on the evening 30/4. But this can’t go on forever given we are at the end of La Nina.
A well designed and operated electricity generation and distribution system will supply customers with a cheap, reliable, and sturdy supply 24/7 365 days a year.
We’ve already had such systems in Australia – large, long life, affordable, coal fuelled base load generators, backed up by quick start, hydro and gas fired generators – more than capable of taking us through the typical peaks of the 24 hour cycle.
Incredibly, we are seeing these “dirty, old, clunkers” being demolished well before life expiry, in favour of thousands of individual small scale “renewable” entities, requiring billions of dollars of taxpayer subsidies, and as the above extensive analysis shows, not quite being up to the vital role they are supposed to fulfil.
It’s worse than that, because the much shorter life of these multitudes of minuscule power generators come with repeated and grossly expensive replacement and no doubt more and more taxpayer subsidies.
When will the blinding reality, that large scale nuclear reactors with long lifetimes – dropped in at locations close to the heavy capacity transmission lines already in existence – are far and away the only plausible way to power our nation into the future.
Smaller SMR units can readily be dropped in at other easy connection points i.e. close to heavy industry, and with existing hydro and gas units used to back-up the system up at times of peak demand.
This article treats our nation’s vital electricity supply system as some sort of grand experiment – let’s keep toying about with renewables, irrespective of the cost – and keep on avoiding the reality that there really is a better way to meet this need.
In addition to the multiple failures of the push to abandon reliable (fossil fuel) generation you pointed out, the pundits claimed renewables would be cheaper. However that claim only existed because the first batch of renewables did not need additional transmission infrastructure.
This reflects on the lack of wind based generation in NE NSW and more particularly QLD. If QLD had the 6GW+ of wind based generation capacity that it should have by now, then there would be far more consistency in wind based generation output across the NEM as the cold fronts sweep across the country.
What batteries have failed? Batteries are more reliable than coal or fossil gas generators.
Once the 100GWh pa MegaPack factory comes online in Shanghai we are going to see a significant increase in the storage deployed in Australia, along with lower pricing.
A major contributing factor was the lack of transmission capacity between Victoria and SA. Transmission capacity was down by at least 600W, which would have had a major impact on the price.
Price is hardly surprising seeing you got into liquids to keep the lights on. look forward to it as reliance on intermittent generation increases. Unless called up well in advance cold thermal machines won’t get there in time and they won’t be there eventually because their business model is broken and they are all bleeding money because of daytime solar. Little business case for GTs without a capacity market. Then their is the small matter of how do you contract for fuel (gas) when you have no idea when and how much you need. Can be fixed with storage (both electricity for short term and gas for long term) but the punters are going to have to pay at the end of the day. But the experts tell us that renewables are cheaper …..really? This is going to end in tears. Looking at the spot market right now, dare I say the missing 1000MW from Liddell might already be showing up in pricing.
My experience in Government and the Industry, across a range of infrastructure sectors, has been (and continues it seems) that politician-pushed programs (rather than technically/economically/soundly-managed programs) so often disappoint.
Great article along with Allan’s and others but it gets to the same point. There are too many constraints. Better modelling might lead to better price forecasts but the actual outcomes for consumers won’t change until some of the constraints are alleviated.
Good thought David. The upside of good models is the ability to prioritise which constraints should be removed and the economic benefit of such a project. Similarly the benefits of DR can be better identified.
Thanks to the author for demonstrating that not only is the grid being structurally weakened by Heywood and subsequent closures and additions, but that this has always been the expected outcome.
This knowledge is essential if we are to have informed discussions about our energy future. Thanks for wording it in non-technical language with minimum jargon.
My take-away: Dunkelflaute is getting progressively worse and our defences against it are progressively weakening.
Lunch time Sunday in Victoria 24 MW wind and at 8.40 am on Monday 2,028 MW of wind ie a swing of 2,000 MW in wind output in less than 24 hours. This is not the way to run an industrialised economy or maintain reasonable living standards. And did anyone notice Battery Of the Nation in Tasmania in AEMO’s database has now been reduced to 1,500 MW which just happens to match Marinus Link (Stage 1, 2a and 3a was removed). With all the Snowy 2.0 cost overruns and delays where is all the long duration storage coming from?
I suggest that this issue was compounded or perhaps even primarily caused by poor investment confidence over the past 10 years. Now with the confidence turned upwards with new targets and incentives, it takes 7+ years to developed PHES projects while it takes 2+ years for solar and wind projects. We did need Snowy II to come on by 2026 but we didn’t need all that capacity at this point. Seems like we will need many more higher capacity battery storage projects between now until 2029?
Dunkelflautes are not getting worse, they just become a serious matter when conventional power capacity runs down to the point where wind power matters, as in Britain and Germany.
Hardly a surprise surely. The real elephant in the room is how (who?) is this going to be paid for. All of the modelling I have seen says you cannot sensibly build enough storage to get through the quite common, seasonal, wind and solar droughts that occur in the NEM (and indeed here in the west also). To attempt to overbuild wind and solar to compensate simply sees these become dispatched generators and hence they no longer have the privilege of free spilling to market. This will make network curtailments look like a picnic. They can’t (won’t) take market risk on this (project becomes unbankable) so who does? Simply their LRMC becomes much higher based on what a proponent reasonably thinks they can get to market. Ditto storage. Unless storage gets some sort of capacity payment (that has to be funded) then their whole business case will assume frequent extreme energy market swings to get an arbitrage. High risk, so who invests? But I thought all this renewable stuff was going to be cheaper? The higher the renewables penetration the less dispatchable generation survives and with the current mindset (that the only answer is more renewables) the overbuild will become worse, capacity factors decline and the system levelised costs go stupid. I suppose it all stops when you get to VOLL. Reminds me of a wise engineering lecturer that once taught me “in real life there is no such thing as the infinite busbar” – we can’t just keep doing this and somehow assume it will all hang together! Trouble is it is up to proponents to decide if they have a commercially viable project and I challenge the rigor going into some of the analysis sitting behind these. Wasted money is wasted money and it doesn’t really matter if it is public or private – it is opportunity lost.
Paul, Thank you for the great waterfall plot.
How far into the future could it be reliably extrapolated?
Kuri Kuri, so 375MW in 12 months time and another 375MW in 18 months, still 2500MW short.
There are a few concerns here about investor risk and some suggestions that the business case for storage is somehow different when it is not.
In the NEM, the business case for generation (including storage) of all types ultimately relies on favourable spot prices. What favourable spot prices look like is different for generators fuelled exogenously (eg coal, gas) than for those fuelled endogenously (ie battery/PHES or similar). To invest, a business case for the former wants spot prices to be high on average if they run all the time (coal) or many short periods of high price spikes for peaking units. This is well understood.
What appears to concern people is that the business case for the latter relies on spot prices to be highly variable – this might seem new (even though PHES has been around for decades?) but the indicator of profitability (future spot prices) is the same. In all cases, generation (and storage) investment business cases rely on forecasts/expectations of future spot prices as they determine which technology is likely to be most profitable.
In relation to the comment above “* note that one thing that might potentially change this to some extent might be when we start to see results for increasing diversity of wind farms across Queensland … however that’s a strong ‘might’, as other analysis we’ve performed (such as Appendix 27 within GenInsights21) suggests that this is unlikely.”
On April Fools day this year there was zero wind in Queensland for 3 hours and on 2nd April zero wind for 5 hours. Queensland wind farms are located from Mt Emerald west of Cairns to Coopers Gap west of Toowoomba. Therefore from the north to the south of Queensland there was no wind. Chemical batteries generally last for 2 to 4 hours and in the situation of no wind across a state could be discharged quicker. Then there is the instance of 204 MW wind output for the 5 states in the NEM which means even diversity of wind farm location across the NEM does not solve problem of not enough generation to meet demand. PS 1st and 2nd April 2023 was a weekend but in the future short and long wind droughts will occur during the working week and that severely impacts our economy, jobs and food security.
Allan great analysis. I believe it’s not going to be good ending.
“Heard about the guy who fell off a skyscraper? On his way down past each floor, he kept saying to reassure himself: So far so good… so far so good… so far so good. How you fall doesn’t matter. It’s how you land!” 🙂
The NSW volatility events on 17 and 20 May and on 16 March were characterised by constrained VNI flows and high (2000 MW) Tumut hydro output. Is the VNI constraint causing the high hydro output or is the high hydro output causing the VNI constraint? Is there a lack of capacity in the Snowy 330 kV network?
Under and over deliveries to target may be manageable with the NEM’s existing 19 GW of large-scale solar and wind. However, imagine the impact when solar and wind farms reach 44 GW in 2030 and 141 GW as per AEMO’s 2022 ISP. Then what if Australia becomes a global green hydrogen superpower as per the Bloomberg/BNEF report yesterday and large scale solar and wind reaches 812 GW by 2050???
Hardly surprising pricing given you are into liquids again. Low sulphur diesel is around $1 per litre (excise free) these days (a lot lower than 12 months ago) so putting this through an old 13,000MJ/MWhr OCGT somewhere gives you about $380 / MWhr just for the fuel. There are some very subtle high level messages from lack of reserve. This is going to end badly people. Liddell is gone, Callide 3 & 4 still off (and how much more could you get south into NSW anyhow?) and it’s not even cold yet. With the announced retirements in coal, sensible asset managers will be rationing expenditure on those as they move to retirement, so it is not unreasonable to expect reliability to decline. Without a capacity market I can’t see people queuing up to build OCGTs (even on liquids) to at least keep the lights on. This is completely analogous to the risk Swiss cheese model. Keep retiring dispatchable capacity and you are removing another layer of diversity and defence. Sooner or later the holes will line up. In the years of high growth here in WA I went through a summer of less than 10% reserve plant margin and it is a very uncomfortable (and expensive) place to be. Clearly, I must be missing something!
While the NEM is a complex machine and it’s very rare that there’s ever a single ‘cause’ for any particular outcome, it could be argued that the systemic destruction of reliable coal fired power plants and removal of a material volume of dispatchable generation is the driver.
Notice how most of the price spikes are in the early evening, the head of the duck curve as it were, when solar (and to an extent wind) is less productive, and people are demanding power for cooking (and heating)
It might have been cold, but hardly peak load stuff. Cold overnight temperatures usually result in high daily energy, not necessarily load. (Dare I say the generators make some money!) What has changed is a not insignificant lump of relatively cheap SRMC generation (Liddell) removed from the market, so it all shuffles up the cost curve. Most energy in the NEM is (in reality) settled by bilateral trading arrangements – particularly so for the vertically integrated gentailers. Retailers in unhedged wholesale positions got scorched last winter you may recall. Now the energy served by Liddell is simply being supplied by more expensive sources on occasions when wind (and solar) are missing or the peak demand goes a bit higher and a higher cost generator bids in and settles the market. Your previous remarks about pricing when liquids appear highlight this. Nothing to see here – your new normal.
The reversal of flow on VNI and the increase of 1464 MW in NSW generation appears to result from the dispatch of ~1400MW of Tumut3 at that time. Upper Limit was already generating 600 MW. AEMO published a document on 3 November 2017 https://www.aemo.com.au/-/media/Files/Electricity/NEM/Security_and_Reliability/Congestion-Information/2017/Interconnector-Capabilities.pdf which notes (Section 2.3) that the nominal capacity of VNI is highly dependent on the Snowy generators’ output. This seems to indicate that the Snowy transmission capacity is not sufficient to simultaneously support high VNI flows and high hydro output.
The concern here is the time of the IRPM ie after the sun has set. So when the chemical batteries are discharged in a few hours and there is low overnight wind then what? The 2 GW Snowy 2.0, 2 GW Borumba Pumped Hydro and 5 GW Pioneer-Burdekin Pumped Hydro will be post 2030. The NEM is desperately short of LONG duration energy storage that can operate for days (or weeks) if need be. The IPCC have said that climate change could cause global stilling in the decades to come. Australia’s east coast is one area that could be impacted by reduced wind speeds. A 10 percent drop in wind speeds could drop wind generation by 30 percent. Will be interesting to see how the proponents of these massive pumped hydros intend to return the water to the top reservoir if the IPCC are right and global stilling does eventuate. The NEM has to be resilient to climate change which requires a mix of generation. Geothermal according to IRENA is 0.5% of global renewables but could it be more. Japan and Korea sit on the Pacific Rim of Fire and Queensland has potential that has so far resulted in failed geothermal projects in western Queensland.
It would be interesting to see on these days of very high curtailment if there was a corresponding restriction on one or more of the transmission links between regions such as QNI, VNI, Heywood or Basslink. It is relatively rare to have both excellent wind and solar conditions across all regions. Today was one of those exceptions.
How can you publish an article on this now when we had record LOW wind generation this week resulting in high prices across the NEM. Especially of note is the non-existent wind in SA, compounded by interconnector constraints, resulting in some of the highest prices we’ve seen in the region in recent months.
This is great news but what is the implication? One thing that sticks out is that total NEM wind gen was well under 1 GW on Monday. Which is a +7.5 GW variance within 5 days. Relatively speaking, that is the equivalent of saying that we lost 2.5 Eraring stations on Monday. What does that mean for system security? There’s all this fanfare for difficult to predict peak wind days, but no mention of times when wind gen is miniscule (which happened five days earlier).
The sub-1 GW wind gen on Monday was at its worst during the evening shoulder period where there is no solar and rapidly rising demand resulting in a measly 15% wind contribution to the NEM. The rest of the supply coming from coal, hydro and gas.
The interesting metric isn’t the absolute maximum of minimum but the % of the nameplate. The min % is far more important than the max. The recent Net Zero Australia plan reckons we need 400-500 GW of wind+solar to power what is just a 25 GW system (+ reserves). That’s a 16 to 1 overbuild ratio.
Generator profits are a combination of quantity and price. Recently truck loads of avocados were found at the local dump as there was no market. People were outraged. However, there were simply too many avocados at that point of time. It would most useful to have the price when wind output was 1 GW compared to when the output was over 8 GW last Friday evening. How many wind farms have a Power Purchase Agreement and how many have a PPA for the expected life of the wind farm?
Don’t forget most of the more recent wind farm builds were motivated by the old RET and the associated penalties for retailers. Business case for the development was driven by this – energy value was almost a bonus. Not normal PPAs as we would know them.
It certainly seems like, under some constraint situations, TUMUT3 is able to ramp up and take advantage of high NSW prices, pushing out cheaper Victorian generation even to the point of causing counter-price flows into Victoria.
The last graph is an eye opener. Presumably due to wind and solar so it’s only going to get worse. How can negative wholesale pricing be transferred to the retail sector?
Good charts thanks as always Dan. I wonder if it’s time to update the threshold for the incidence of spike prices graph from $300. In think this year is being dominated by prices in the $300-$400 range which are no longer reflective of a spike.
One approach to maintain historical consistency could be to use an inflating threshold.
“Giles Parkinson wrote on RenewEconomy about ‘Vales Point coal closure date pushed back four years in latest threat to renewables transition’.”
That is one of the most nonsensical statements I have ever read. He actually used the word, “threat.” Isn’t this the same guy that insists that wind, solar and battery are cheaper than fossil fuel generators? Why is worried then? They’ll just get priced out of the market and the threat of dispatchable 24/7 gen will be vanquished.
Take a high level view and nothing terribly is surprising here. Volatility and slow uptick in average pricing started in 2017 with the retirement of Hazelwood. As wind and solar render the economics of traditional generation unviable and more exit the market both volatility and average price will get worse as holes are plugged with higher SRMC sources (gas and diesel). The “fat” of spare generation is just about gone. Throw in the outage at Callide (though this is tempered by transmission constraints) and the trend is clear. The WA situation is also clear. 200MW of coal was retired prior to the summer of 2022/23 and the coal industry in WA (a non export commodity) lurches towards collapse on shrinking sales volumes, there was more energy produced from gas than coal. (WA already was second only to SA in terms of gas for generation). Indeed for several months it has become almost normal to see diesel burnt on weekday evenings to maintain reserve requirements. WA has a capacity and energy market with regulated domestic tariffs so the price volatility that would otherwise occur is well hidden in the red ink in the Government owned retailer’s books. The actual volume of energy traded in the STEM in WA (the number you have interpreted here as the wholesale price) is proportionally very small.
Inertia is best understood as short term energy storage which is used to absorb/release energy in response to changes in the supply-demand balance. It is not an objective in itself, it is a tool to maintain stability of voltage and frequency.
It should also be remembered that synchronous machines can’t actually deliver much of their stored energy. For example, if the maximum frequency excursion is limited to 0.5 Hz Then a rotating machine can only absorb or supply 1-(49.5/50)^2 of its inertia so a 500 MW steam turbine with an Inertia Constant of 9 or 4,500 MWs of inertia can only convert about 90 MWs of energy. If the frequency change occurs over two seconds it is an average power of 45 MW. A 50 MW battery could supply both the power and the same energy, but if the frequency decline persists, then there is no more inertia available and the governor response firstly has to replace the lost inertia of the steam turbine i.e. return the 90MWs and then start to make up for the shortage of power in the system. The battery is just adding power and energy back to the system from the get-go. Without going into the boring details, for the first few seconds gas turbines are even worse because they have around half the inertia of a steam turbine and engine power drops if speed (frequency) falls so the governor response has to be even more vigorous.
If generation can respond fast enough the system needs very little inertia, in fact inertia can become a hindrance. Many years ago, in machine tool design it was considered necessary that the inertia of the motor should be at least 3-4 times the inertia of the load so to accelerate the load a torque of 5-6 times the nominal load torque had to be applied. Now it has been realized that if the control system can detect errors quickly enough and the power amplifier has a very high rate of response on the power curve, lighter more accurate systems can be built with low inertia motors with as little as 1/10th of the inertia of the load because now total torque is only 1.1 times the load torque so less energy is required to correct errors.
The same is true in reciprocating engine design where engines 50-80 years ago had huge flywheels to stop them over-speeding if load fell or stalling if load increased abruptly, now with high resolution speed and torque sensors, micro second control of fuel injection and low inertia flywheels, diesel gensets, sportscars etc can ramp power up or down almost an order of magnitude faster than their 1950’s predecessors, so better load matching with less inertia
Similarly in a power system if the output of a battery can be ramped up or down by full capacity in 150 ms then it can contribute/soak up far more energy than the inertia of a steam or gas turbine. Some years ago, Eire Grid determined that a 10 MW battery contributed as much to grid stability as a 100 MW gas turbine. Partially curtailing or reducing curtailment of a solar farm can have a similar effect and because wind turbines are asynchronous they can contribute much more of their physical inertia while still maintaining frequency and of course in the case of over-frequency the inverters on wind or solar farms can reduce power safely far faster than thermal generators. Even reciprocating thermal power plants which have relatively low inertia can partially make up for that by having very high power ramp rates as long as the gain on the governor system is set appropriately.
Then there is the load inertia to consider, large fans pumps etc and the inductance of the transmission and distribution system all contribute to system stability. Now many modern motors are connected via VFDs so that decouples their inertia but other systems such as fans or pumps have the added benefit that the load drops as the square or even higher power of frequency, so not only does the system gain the mechanical inertia, but a useful reduction/increase in load.
When all of the above is considered and control schemes are reconfigured appropriately it is likely that in the future, the system frequency will be maintained at or better than current levels with less than 1/10th of the inertia now thought to be necessary
Although your mention of nuclear was not a central part of this article, it still bothers me that its legitimacy is brushed under the carpet and completley ignored. I’m not a big nuclear power advocate but the arguments presented here are weak. The volume of actual nuclear waste products are miniscule compared to the impending toxic solar panel and wind turbine waste entering the global electronic waste scene. Yes, nuclear plants just boil water, but this straightforward design is why they are operational for well over 60 years. You can replace control panels and steam generators with relative ease and keep these stations running for decade after decade. And if you’re concerned with nuclear sources getting into the wrong hands, why don’t you advocate against the use of them in hospitals which are under far less security than a nuclear power plant.
Regarding the minimum demand issues, that really is one of a million issues we’re faced in the future. What is even more pressing is the abolition of all forms of dispatchable generation except for hydro and perhaps batteries if they ever develop their capacity and throughput. Minimum demand days are scary but so are MAXIMUM (scheduled) demand days when we have coinciding wind and solar droughts (days and even weeks in duration).
I stopped reading when I reached the claim “It was good to hear Chris Bowen articulate that nuclear is a distraction …”.
Bowen is a megalomaniac, clearly totally out of his depth and the most incompetent energy minister Australia has ever produced.
We are heading to a grid disaster and Bowen is front and centre of the fiasco.
I am bemused at how near the beginning, nuclear power is dismissed as merely ‘boiling water’, yet by the end the author is calling for “clean steam for electricity to fill the dunkelflaute”. Er, hello?
Even odder is the claim that nuclear power plants have been ’weaponised’ in Ukraine. What sort of weapon has no measurable impact whatsoever? *Fear* of such may be a different matter, but very little of this is well founded. In any case, what on earth has this got to do with Australian scenarios?
SunCable guy, you realise nuclear energy uses steam turbines, right? A central point in the trends this article highlights is minimum demand crashing below zero MW in almost all regions, and the need for flexible generation to work with that. A solution to the problem that twice a day demand will hit crazy lows, but then at two other points in the same 24 hour period it will be very high. Steam turbines don’t like to ramp, and definitely not multiple times a day every day. Simply bringing nuclear into the discussion completely misses the point of the challenges Australia’s grid is facing. Its like saying “delivery trucks find it hard to navigate CBD streets… No worries, aeroplanes can carry heaps of parcels. Let’s use them to make CBD deliveries”.
To quote the article, “How prepared are we for coincide minimum or negative demands that are forecast to occur within 4 or 5 years?” installing generators that require gigawatt level baseloads is pretty much the worst thing you can do to prepare for that.
Australia’s problem is very different to Americas or Europe’s, where they run large interconnected grids. Got too much power? No worries, send it across a border. Not something we can do. The argument for nuclear distracts the point that Australia’s grid issues are not a debate about fuel, we have all the fuels we could imagine. its a debate about generator capability, and how we can balance loads and generation to meet the intra-day demand curve that is getting harder and harder for a typical power system to support. Which is central to a point that Kate has long advocated, that is the need for good system control.
There’s a very good reason Kate is always invited to talk at such events, and that Engineers Australia made her a fellow. There are very few minds as sharp as hers when it comes to the challenges of Australia’s power system.
I invite Mr Farley to review his statements that gas turbines supply less inertia than an equivalent steam turbine, and that mechanical inertia is increased by the use of variable speed drives.
@HowGoodAreRenewables
Plumetting demand periods are EXCACERBATED by your net zero targets. You’re creating a problem by replacing dispatchable generation with renewables (that distort prices due to their LGC-creation) and then blame dispatchables who follow a different set of rules in the NEM. The solution to the problem is to REPLACE wind and solar with dispatchable generation such as nuclear. Don’t try and confuse the issue. Kate blasted nuclear from the introduction as an inferior source of generation in general – way before she discussed the minimum demand problems.
This speech by Kate is just an admission that the system she heavily advocated has serious issues with no solutions. But that doesn’t stop her from pushing those ideas.
Could you ask Kate whether her minimum demand models that go beyond 2050 (a joke to anyone in the investment/commercial industry) is based on Australia constructing 10,000km of transmission lines? How about the problem of realising the monumental task of constructing that much infrastructure? Did Kate assume that all those REZs will magically send their electrons to the load centres and give us those forecast minimum demand days?
I don’t care about how many speeches she makes. She’s one of many people who aren’t involved in the trading/commercial space who make out-of-touch calls with the energy industry.
Keen to understand how LGC-accredited renewable energy reduces demand… i think you’ll find you’re confusing rooftop solar there, which has nothing to do with large scale renewables. LGC accredited generators are suppliers of electricity… literally impossible for them to impact demand. So with that, i again challenge the idea that you can solve a ramp issue with steam turbines. Remove every renewable generator in the NEM… demand is still hitting zero, and you’re still having to implement a 24-hour shutdown procedure on your steam turbine.
Supply is just one side of the story, the other side is changing loads. go back a few decades and all large loads were also spinning machines more or less directly connected to the network. big industrial motors and the like. Now days, they’re more and more connected by VSD’s, and in general loads are far more digitised. That means there is less and less demand-side inertia. AEMO has already started responding to this with changes to its FCAS procurement, so it is very much real. You need smarter, faster reacting generation to meet the demands of evolving loads. How is a steam turbine going to help in AEMO’s new one-second FCAS market? Hot tip… it can’t.
As for the minimum demand forecasts, you can thank AEMO for those. they’re pretty much lifted from the ESOO, as anyone intimately familiar with network planning would have picked up. And hopefully i don’t need to point out that the ESOO is one of the key documents used in future planning for the energy network. Sorry if that’s funny to your investment bro’s, but its reality.
And to say she’s not familiar with trading and commercial.. mate, at least check her linkedin before running your mouth on things like that, and take a moment to understand the core of what the article is about. Regardless of whatever fuel you use, you need network control. Kate’s core points that she often shares with industry is the importance of proper power system control, how Australia’s energy market falls short, and the problems it creates. Forget fuel for a moment, if you don’t have good power system control schemes, you’re not going to have a stable market and you’ll be writing policy that will not has positive outcomes. Many of us in industry will fondly remember being told we need to add synchronous condensers to everything we build… look how that worked out.
So we have one solution that involves paying $11 billion (plus $5b for Snowy2?) and another that costs $6 billion. This is for transmission to supply renewable energy that costs $50 per MWh plus a RET equivalent of $50 MWh. Not discussed is an alternative that would see no transmission costs but augementation of coal based electricity at $50 per MWh
From the VEPC report:
“The Australian Energy Market Operator (AEMO’s) modelling for the VNI-West Project Assessment Conclusion Report (PACR) shows that AEMO expects that for the next four years the wind farms in Western Victoria will waste up to 40% per year of the electricity that they are able to produce because the transmission system is not able to accommodate their production. The picture is no better with solar farms in the Murray River REZ expected to waste up to 40% per year of their production potential over the next eight years.”
Who exactly is to blame for this? A bunch of international renewables developers failed to model power flows properly when making their investment decisions and now the Australian public have to pay for it with the garbage proposals from the VEPC report? The RETs aren’t doing it? Everyone of us is already rewarding them by paying for environmental certs which we all see at the bottom of our electricity invoices.
Where is this magical $50/MWh coal coming from. Finance, Depreciation and overheads on a new brown coal plant will be about $70/MWh O&M $20-25 and coal $12-20
The curtailment situation is complicated. In the past AEMO was not allowed tell new aspirants if anyone else had applied to build plants near a particular substation because it was “Commercial in Confidence”. Also, it was probably felt that a public access system would encourage a gold rush situation where people just “pegged claims” and held genuine developers to ransom. Net result, too much capacity built in the wrong places because the sun and or wind resource looked good
Minor issue but if they don’t deliver power they don’t get LRECs
“We know, because of rigorous research and analysis conducted by AEMO and the CSIRO, the lowest cost replacement for this coal generation is renewable energy from the sun and the wind – backed up by batteries, gas and hydro to smooth the bumps in production.” [AEMO, from the AEMO (initial) Response linked above.]
The roles of AEMO and CSIRO in arriving at a conclusion that “the lowest cost replacement power…” has returned as a very hot issue, with seemingly both expert groups blaming the other for errors or omissions.
I am eagerly watching progress, eg in newspapers and Twitter. No longer is it only models at 30 paces. It is experts at 30 paces. I have no expert opinion to offer here – I am an interested bystander, but my impression is that there is a capital cost hole that must be addressed in the period pre-2030 and the arbitrary exclusion of nuclear power options is enigmatic.
Peter Farley
I wonder how coal plants in China manage since they too would be up for $120/MWh. Fact is that a coal plant that was allowed to contract in the absence of subsidised wind/solar would operate at ~ 90% availability. Fuel $15 (If it was brown coal fuel would cost <$5/MWh); capex $20; O&M $12
Hi Stephen, The SCADA signal should include a quality flag that would indicate the quality to AEMO. We could anticipate AEMO also includes some additional logic, e.g. to check for stuck readings and or readings outside a reasonable range, but I don’t think I’ve seen these documented explicitly.
Using the MaxAvail as a means to avoid negative prices is inappropriate. If your intention is to turn off when prices are too low then simply include your minimum price in your bid. Including a price in your bid that is lower than your real minimum price is misleading and could/should be found to be in breach of the rules.
Hi Greg, I’m not a lawyer so would always defer Participants to seek their own legal advice where appropriate. However, I can offer that the AEMO Operational Forecasting and Dispatch Handbook for Wind and Solar generators indicates in S2.4.4 that the use of MaxAvail is ok, subject to the requirement that bids must not be misleading (NER clause 3.8.22A) and rebids must be consistent with the requirements in NER clause 3.8.22. The other good reference is the AER Semi-Scheduled Generator Compliance Bulletin and Checklist but that hasn’t been updated since 2022, as far as I can tell so can’t yet reflect good-practice use of MaxAvail in this regard, since the change on 7 August 2023.
New coal is a non-starter on the grounds that it’s complete and utter lunacy to contemplate building a new machine to turn large amounts of brown coal into even greater amounts of carbon dioxide.
It’s worth pointing out that there are two quite different “high temperature” risks being discussed here:
1/ Wind speeds may be low at the time of peak demand on hot days (typically 6pm-9pm).
2/ Wind turbine generator capability can degrade at high temperatures (the “mystery DUID” chart in Paul’s post).
The enhanced modelling by AEMO deals with the first type of risk: it has identified that wind speeds and therefore wind generation output might be significantly lower in a few scenarios (“reference weather years”) than was previously incorporated into ESOO forecasts.
It actually appears as though most of the difference relates to modelled wind speeds and output for a single peak demand day in a single reference year – 3 January 2015. See Figures 24 & 25 of the ESOO.
The second type of risk – generator derating at very high temperatures – applies to some degree to almost all forms of generation, and AEMO uses data provided by generator participants in modelling this – see Section 3.2 and Figure 18. Wind has the highest drop in capacity from nameplate levels under “summer peak” temperatures (above 37 degrees). However I don’t think that this derating or AEMO’s approach to modelling it has changed in this ESOO.
It’s also plausible that in the case of wind generation, this type of derating doesn’t matter much for reliability modelling – it’s low wind *speeds* that are the problem. If wind speeds are moderate to high, losing some %age of output due to heat isn’t likely to cause a reliability problem.
It is know that the generation from inverters cause these types of problems. There are IEEE papers on it. Damas et al did a good review on it in 2020. They identified a major source as from HVDC converters as well as series=connected capacitors. One wonders the part the switchyard at Redcliffs plays in this, particularly when SA is exporting.
While I agree with your sentiment that the reactions are over the top, the fact of the matter is that we rarely if ever had generation shortfalls before the ramp up of renewables.
If the energy market is unable to meet the demand from customers, then the energy grid is not meeting its core requirement and we should be reconsidering a new plan to ensure generation supply can meet demand.
Saying it’s okay because we will just shut down large energy users is a bandaid solution. We need chemotherapy to stop the spread of the cancer through the NEM.
Something I found surprising in these tables was the relative effectiveness of short storage in NSW, with 4610 MWh as 2h BESSs said to be equivalent to 4043 MWh as 4h BESS in improving reliability.
This doesn’t make any sense to me. All else being equal, 1 MWh of 2h BESS capacity should be able to do everything that the same storage in a 4h BESS can do and more. Is there something I’m missing that would explain how these numbers can be internally consistent?
Good pickup Tom, I agree that it doesn’t seem logical at all that more MWh of 2 hour storage would be required to match the contribution of 4 hour storage. I wonder if this is some artefact of the way AEMO’s modelling dispatches storage units.
Hi Tom, I noticed the same thing. The low relative effectiveness of the 2hr BESS must be driven by the derating factors applied, as described on p108. 2-hr batteries have the MWh capacity reduced by 25%, 4-hr batteries by 10%. As you notice above, the results above demonstrate that these deratings are much too harsh.
This “game play” of energy battery arbitrage is pure speculation at this point since it is not happening right now. You are only guessing what battery traders may be capable of doing in the future since we are seeing very little of it now in the other NEM regions. The current fleet of operational batteries are still playing the FCAS game even as the revenue pool shrinks further and further with each new battery commissioning. Furthermore, we can safely say at this point that battery “energy contributions” in Queensland are nil.
To say that intra-day arbitrage is easier to manage than very long-duration arbitrage (i.e. up to a year in duration) is not necessarily correct either. A battery operator has a smaller margin of error when trading on an intra-day scale since their arbitrage period is spread across fewer 5-minute dispatch intervals. If a battery trader forecasts low prices in the next dispatch interval and it happens to be much higher, then they have purchased electricity at a higher price and reduced their arbitrage revenue significantly.
On a yearly scale, we actually do know what months give us lower prices and which months give us higher prices. E.g. it is no mystery that summer demand is higher and prices are higher as a result.
The energy (or MWh) of these batteries are tiny which makes arbitrage difficult on the “sell side” and this is compounded by smaller time frames (e.g. intraday arbitrage). I.e. how do you decide when to sell your miniscule volume of energy that you bought earlier in the day? Do you sell it all during the evening shoulder period when prices are sensitive and risk selling when actual prices are too low? Or do you wait until demand hits its absolute peak later on? My point is that the small energy of batteries makes it harder to sell during the forecast high price periods.
I hate to burst your bubble, but energy arbitrage on a short-scale only works if you can predict the market perfectly. And this is obviously not the case. And this leads to a sobering conclusion with batteries. Batteries are only capable of arbitrage over the short-term which has all these major aforementioned problems.
There is a mental confusion in the terminology. There is no need for a one hour battery to discharge over an hour, it can run at a quarter power for four hours or full power for one 5 minute bidding cycle and then ramp down over successive cycles over 6-8 hours if prices fall.
It can sometimes charge at night even buying power at a positive price that is still less than morning peak. It won’t have two full cycles /day but it may have 1.2-1.5.
The beauty of batteries vs pumped hydro is that it is easy to increase the duration as demand for longer duration increases and as the cost of storage falls. Pumped hydro is pretty much stuck once it is built.
If for the foreseeable future we retain the existing gas and hydro fleets and triple solar and wind output, the peak output from new backup/storage will be around 10 GW less whatever demand response is implemented, and for all but a few hundred hours per year wind/solar/hydro/gas will meet demand. So, in the medium term to 2030 or later, we should be focusing on backup power rather than energy. In the current environment it is difficult to see batteries with more than 1.5-2 hours duration being economically viable because on all but a few days per year, the batteries will run at far less than peak power and their output will therefore be spread over the entire evening peak.
The next limitation on storage is excess wind and solar capacity. All reliable grids have significant excess capacity both in terms of peak power but also annual generation. In 2008 the coal and gas capacity on the NEM could have comfortably provided 280 TWh, but in fact provided 195 TWh. If we build enough wind and solar to provide 280 TWh at normal capacity factors then on seven days out of ten there will be excess energy and hydro will handle most of the peaking demand, new storage will be active for only a few hours and will rarely need to run at anything like peak power. We can also see that if you examine the seasonal variation in renewable share on OpenNEM against the running average renewable share, that there is not much variation, typically winter renewable share is about 90% of the preceding summer renewable share. If there is 280/195 i.e. 44% excess capacity then seasonal variations will easily be covered.
Consequently depending on the cost of long term storage it is quite likely that we can easily exceed 83% renewable grid with no more than 10 GW/25-35 GWh of storage
Over 40% capacity factor certainly seems optimistic even for latest-technology onshore wind.
That said, analysis like this based on NEM market data is going to yield an “apples vs pears” comparison for several reasons:
– the existing fleet is up to 20+ years old (with its underlying design parameters even older) so we would obviously not expect that fleet to deliver to new-build levels, not even closely. Today’s electric cars probably perform a lot better than whatever you could buy five or even three years ago. I’m not an engineer but understand that there have been numerous improvements to turbine technology and windfarm modelling / layout methods, as well as in site selection, that should yield better performance under the same wind regime.
– registered fleet capacity always runs ahead of actual achievable maximum output while there are farms being built, commissioned, and extended. To allow for this you need to go through the data and manually back out farms (or adjust maximum capacities) for those still in the commissioning / ramping up stages.
– economic offloading and constraint-driven curtailment will drop fleet capacity factor a bit if the measure is based on actual output; availability is a better numerator if the aim is to compare technical performance before these impacts
Finally, I very much doubt that curtailment or offloading would materially affect minimum coincident output levels across any significant group of farms let alone the whole NEM fleet – minimums will be driven by widespread still conditions in which case curtailment / offloading are almost certainly going to be neglible. On the other hand maximum output levels will almost certainly by affected by these factors.
It’s a valid and worthwhile analysis Paul, and you’re aware of the caveats, but just pointing out for other readers why its results can’t be directly compared with new build capacity factor estimates.
Who cares about monthly wind capacity factors anyway. A monthly figure totally hides one/two/three-day wind lulls where capacity factors are absolutely abysmal (<500 MW total NEM wind gen over evening peaks) and we're faced with actual LOR2 conditions. A purported high capacity factor month can completely hide severe price spike days that took place over that time period due to very low wind gen.
@Allan O'Neil
The bulk of the NEM wind farms are already located in the best wind resource locations in the NEM which is the VIC/SA border. Are the technological breakthroughs in wind the same ones where we just build bigger blades without any conception of their longevity and maintenance costs?
Good question about managing risks of contingency events tripping off SA rooftop PV Paul. There are various system normal constraints which limit flows from Victoria into SA via Heywood under conditions of high PV generation in South Australia for just this reason – to leave headroom for suddenly increased flows if a large block of rooftop PV were to trip because of some disturbance in SA – for example the V::S_NIL_MAXG_1 and _MAXG_2 transient stability constraints. But on a low demand day like today when flows are exports out of SA, these constraints wouldn’t bind as there will be sufficient headroom for imports after a contingency anyway.
Alfred Hill makes a good point about averages, it is the low points that cause the problems.
Also the best sites are taken and if Germany is any indicator the capacity factor falls away over time due to ageing machines, best spots taken and interference by overcrowding.
https://www.theregister.com/2023/09/04/microsoft_australia_outage_incident_report/
The Register has some detail from a Microsoft report, which confirms the power issue occured during the thunderstorms. It notes that only 3 staff were on site to try and get 5 tripped chillers and 2 backup units up and running. Only one restarted successfully.
The loss of cooling created a cascade of problems. “Storage hardware damaged by the data hall temperatures “required extensive troubleshooting” but Microsoft’s diagnostic tools could not find relevant data because the storage servers were down.”
It does raise a question of whether the “public cloud” is a suitable place to house critical infrastructure.
Good illustration of the flexibility of large-scale solar (and no doubt wind) responding – via the price mechanism – to the inflexibility of its small-scale rooftop PV cousin.
It’s worth pointing out that *which* large scale generators reduce output in these events isn’t directed by AEMO for security or similar reasons – which is a common implication of the term “curtailment”.
What’s happening is normal market-based dispatch, driven by the prices at which generators have offered their output. We don’t say that there has been “curtailment” of coal-fired or other conventional generators in these events when they bid their output down to mingen, so I’m not sure why we should label large scale solar or wind bidding and responding to market prices as “curtailment” either.
If AEMO had needed to step in and have distribution companies reduce rooftop PV exports or trip inverters offline for stability reasons, that would certainly constitute “curtailment”.
It will be very interesting to see how this impacts Tasmania’s hydro capacity as I understand their dam levels are already quite low and Marinus Link is not yet in place to provide redundency for Basslink.
Allan, at Squadron we use ‘economic curtailment’ and ‘grid curtailment’ to distinguish between a market response and a binding constraint equation. We VRE are used to semi-dispatch caps ‘curtailing’ our generation and that does not discriminate between a market response or a bidding constraint. We also don’t get to withhold our fuel until prices rise, so I guess it ‘feels’ more like a curtailment.
Just offering some context… Nothing wrong with your comments 😉
One small observation: The battery seems to have been charging.
Now consider recent reports that new apartment buildings must have fast charging facilities installed. Quite apart from the cost of the batteries and their connection, my experience (long ago but not forgotten) in designing fire-resistant buildings is that is a highly technical matter. Combine this with the frequency and severity of large battery fires, it’s time that the community became educated on these matters.
Quite frankly, the idea of large batteries in parked cars and/or chargers in basements of highrise buildings is, in my estimation, somewhat scary.
Can you explain what data you are looking at as the basis for saying ‘The battery seems to have been charging‘.
The AEMO market data I am looking at suggests it was last charging at 12:45 (which is what we’d expect – i.e. when prices are low, in the middle of the day).
Saturday 30th Sept appears some rooftop solar owners in Adelaide were being curtailed- https://forums.whirlpool.net.au/thread/36yz02x6
(not my 6kW 3 phase producing 39kWhrs for the day)
A bright sunny day hitting 30 plus degrees but homes were still too cool for aircon so you could easily see why in some areas.
Hi Genevieve, I have received an update from the team who were looking into this issue. I can confirm that we have conducted a series of tests on behalf of “Smarter Homes” with the primary objective of ensuring the capability to disconnect these inverters in emergency situations, thereby reducing potential risks.
These tests were thoughtfully scheduled during the time of least solar output, which is after 6 PM on Fridays, to minimise any potential solar energy loss.
It is important to note that following the completion of these tests, the system should have automatically reconnected. If this reconnection did not occur, the responsibility for addressing this issue now lies with either the Inverter Manufacturer or the Solar Installer.
Our findings indicate that all other inverters successfully reconnected, except for SolarEdge and Goodwe inverters. This observation suggests that the issue may be attributed to these specific inverter manufacturers. Periodic testing will continue to be a part of our ongoing efforts to ensure the safety and reliability of our network. As such, we strongly recommend initiating contact with these manufacturers to collaboratively resolve the issue.
Hi team, I love all your articles, despite not always understanding 100% of the terminology. I was wondering, is there any tool available, that a casual observer like myself could see live bids into the NEM for wholesale prices? I’d be curious to see for example: “Eraring – Black Coal – 5MW @ $120 / MWh”, and every other generator listed out with the same detail. I’m sure it’s probably a dumb question, and it’s probably nowhere near that simple, but if such a tool does exist, i’d love to use it! I’m eyeing off your “Nemreview” tool’s website, and thought perhaps this is where you get all your fantastic info from.
Thanks in Advance!
I don’t understand why I, as a customer of a PV sales company, needs to deal directly with Goodwe, my inverter’s manufacturer, for guarantee issues and/or the installer, whose contact details I never received.
My commercial arrangement is with the seller.
Until now, I believed that the seller is, under federal consumer law, able to suggest that I communicate with his suppliers and installers, but that required my prior permission.
Re that Whirlpool post ‘smokiedawson’ posted today Sunday 8th Oct at 4:40pm the following-
“Mine went off again today 1030am – 1230pm….so annoying and no notification. At least it reconnected.”
Checking my Goodwe 6kW 3 phase inverter tonight showed exactly the same and it’s the first time I’ve noticed throttling since a new install in December 2022. I still managed to generate 38.3 kWhrs today with 6.64kW nameplate panels and I was heading for a perfect bell curve output at 4.6kW until 10:30 then zero for 2 hours then on again at 5.87kW at 12:30. It’s not the Goodwe inverter problem here as I’m on TOU with those 2 white Dalek communicators on top of the power box
I should add I’m all electric with Origin and 2 TOU meters one for general 3 phase and the other for CL1 for HWS and hence the 2 Dalek communicators presumably.
yes its very frustrating, and I find the lack of transparency and communication not acceptable. Trouble is, I think all the complaining will make no difference at all, as sapn do as they please.
Would be interesting to see what is being offered by coal units for the lower (runback) service. Prices seem to spike around middle of the day when solar is at maximum implying coal units are sitting at minimum load with nowhere to go backwards. Implies to do so would involve stinks like removing a mill from service incurring coasts in expensive auxiliary fuels to do so, or even decommitting a unit that will then be needed in the evening in the stack. Western Australia has been here for a while now. Runback is a bigger issue than spinning reserve.
Limited coal supplies are not a good thing, but probably less of a concern for summer reliability than in winter (cf last year’s winter energy crisis).
Summer demand and price peaks tend to be of much shorter duration, and average demand levels are lower in summer than winter. Provided the power station manages its coal stockpile so that it’s able to operate both units at full capacity over peak demand hours during periods of hot weather and higher demand (which don’t extend over weeks and weeks) – and that coal deliveries don’t fall to very low rates – then I’d be surprised if this leads to reliability isses this summer. But I’ve been wrong before.
This seems to ignore NSWs increased import capacity and declining grid demand. Not only have VNI and QNI been upgraded, but in the past when Liddell was operating, southbound capacity for imports was partially crowded out. With PEC on line and generous generation surpluses in Victoria and SA, NSWs practical import capacity will increase by 1-1.5 GW
Since 2019 grid demand in NSW has fallen by almost 6% and ongoing energy efficiency and behind the meter generation and storage will probably continue the decline. While annual demand is not directly related to the annual system peak, they do follow similar trends.
While AEMO is rightly conservative, it seems to me that this forecast is overly so.
What did Stanwell get fined for the setting for going into Islanding mode that made the recovery from this event much shorter, let alone if the state did indeed get sent black, having it still operable would have been much faster than the old AVON GT at Tarong whining into action or WIVENHOE running out of water due to some failed attempt to start from black again. Still don’t see why CSEnergy continue to get such favorable treatment.
“A decline in intermittency is predicted in regions of Eastern Australia by 2099.”
How can anyone take these types of conclusions seriously? Making forecasts that far into the future with modelling that utilises highly sensitive inputs is suspect.
“Around Townsville” lol you mean almost the entire of NQ, it went all the way out to Julia Creek, Cloncurry, south to Collinsville in the Whitsundays. It is a major city so not all of consumption would have been major industrial users. Either way it was really a non event, had it happened during AM or PM peak hour then possibly it could have been different.
I recall about 20 years ago that the manager of one of the MacGen stations I worked at, bid on FCAS / black start ability seeing it as easy ‘money for jam’. He eventually had it pointed out to him just how costly it would be should the event occur so quickly withdrew that bid.
PEC was purported to provide energy security to SA.In fact on most days it will drive up prices in SA because higher prices in NSW than SA will drive output from SA to NSW.
However on high renewable days in southern NSW, capacity on PEC will be mostly taken up by Riverina solar and Western slopes wind.
With storage being installed in SA and increasing surpluses of cheaper Victorian brown coal it is difficult to see any more than trivial amounts of NSW energy flowing to SA.
You are mixing up energy security with spot prices. PEC will result in greater energy security for SA as it will provide an additional transmission path including inertia. This will further reduce the need for the operation of fossil gas based generators just to provide inertia and fault services.
In addition with PEC in place it will allow more RE to be in the SA grid should there be a failure on the Heywood interconnector such as when the 500kV tower was knocked down in Victoria. This fault significantly increased the total cost of electricity in SA due to reduction in RE in the grid and the increased need for additional grid services to be provided.
The peak wind conditions in SA are largely out of phase with those on the Western slopes, so exports from SA will be good during those periods.
The capacity to handle solar in all markets when solar generation is good is going to be a problem, however when conditions are good in eastern or northern NSW and poor in SA, energy will be able to flow into SA. The same applies for wind.
I agree that in the next few years the scope for cheap energy to come from NSW to SA will be very limited, however as NSW builds out its RE capacity the situation will change.
@Craig Fryer I didn’t realise how simple the solution was! Just build 900km of 330kV lines! Easy and cheap as! I’m sure this dirt-cheap endeavour will be completed by July 2026 and we won’t experience any delays or cost blowouts with our trusty Australian workforce. Furthermore, I’m sure our energy situation will be perfectly fine from now until then.
“The peak wind conditions in SA are largely out of phase with those on the Western slopes.”
Patently false. The same high-pressure systems constantly sweep across SA and into the Western slopes at the same time and give us those nice NEM-wide wind droughts.
“when conditions are good in eastern or northern NSW and poor in SA, energy will be able to flow into SA. The same applies for wind.”
PEC connects to Wagga Wagga which is as far south-west of NSW as it gets. What makes you think generation in northern NSW will easily make its way into SA?
“As NSW build outs its RE capacity”
You mean that genius idea of “over-building?” Which effectively means committing to large-scale infrastructure financing without a return on investment? But maybe it’ll come by once all that transmission is fitted out, right?
The reason why there SA isn’t a net exporter is due to the lack of transmission capacity in NW Victoria going back towards Melbourne. When RE conditions are good in SA and Victoria, energy in NW Victoria flows into SA as some of the RE producers in NW Victoria seem to be able to supply the grid at more negative prices than those in SA. This may be due to older contracts.
The end result is a net reduction of 300MW export capacity from SA to Victoria. This has resulted in considerable reduction in RE exports from SA to Victoria. When transmission capacity out of NW Victoria towards Melbourne is increased, SA will become a net exporter. Potentially EnergyConnect will come online prior to this and may result in SA become a net exporter earlier.
The real limitation here is on northward flows out of NW Vic and SW NSW – as Peter pointed out above. The physical solution for that set of constraints is Humelink.
Might be interesting to look at market prices between SA / VIC / NSW on the same graph. I suspect that once much of the constrained on dispatch in SA ceased courtesy of synch comps etc and the price of gas (including start up cost) into the SA stations simply the RE was sopped up at home and shortfalls met by low SRMC coal out of Vic (or Tas). The market doing what the market does.
As Andy says, while transmission constraints played a part, it is mainly the price difference between Victorian brown coal and SA gas. In 2019/20 when gas was cheap in SA and coal output was limited by outages in Victoria then SA was a net exporter. As soon as Victorian renewable output increased to the point that there was spare coal-fired capacity, then Victoria became a net exporter again.
However over the last 5 weeks SA has become a net exporter while averaging 83% renewables, so it seems that transmission constraints can be managed at least some of the time.
As for energy security SA would have been far better served by a similar investment in renewables/storage.
I think you should be able to adjust for (2) and (3). (E.g. QED reports do). I expect this would be material and give some support to the improving technology claims.
As a less drastic solution than reconfiguring the NEM into a nodal system, how about enacting a higher market price floor for renewable (or inverter-based, or non-slow start…) generators? Snowy Hydro has lodged a rule change request with the AEMC to limit semi-scheduled plant to a floor of $-100: https://www.aemc.gov.au/rule-changes/addressing-access-risks-dispatchable-resources
I’m not sure if that proposal is progressing or what discussions are occurring around it, but at a glance it seems a reasonable approach to limit the impact of out-of-order bidding on interregional transfers.
In a system without any or many slow-start generators it would certainly be preferable to have a significantly higher MFP than -$1,000/MWh. Variations in offer price floors to deal with priority access are also something that ESB / AEMC has been considering – not sure where that has got to, perhaps the Snowy request has got wrapped up in that process. Others have suggested changes to threshholds for interconnectors going into the LHS of constraints. There’s a slightly ad-hoc feel to all of it.
Still waiting for somebody to explain how massively expanding wind and solar capacity, and the transmission to connect it all, is expected to reduce consumer electricity costs.
Allan’s detailed look at settlement residues and constraints is a great explainer of just one largely hidden additional cost imposed on consumers directly from the renewables push, but looking ahead 5yrs it is clear the situation must get far worse.
This is a great article. When I read the nsw energy infrastructure report I gained the impression that this part of the network would be upgraded via:
“South West REZ transfer capacity is delivered through ISP projects. The cost has not been included here as this capacity will be delivered
through Project EnergyConnect, HumeLink and the Wagga Wagga to Dinawan section of VNI West.”
In your opinion Allan, will these projects relieve that constraint? I note via EIS filings that there are a number of large prospective wind projects located East of Balranald that will connect to Energy Connect. Surely there must be a way for that power to get to the load?
Thanks David. As far as projects relieving the constraints, I only considered Energy Connect, as that is actually being built. Humelink and VNI West still seem way too far from breaking ground to think about as providing relief anytime soon. It’s possible EnergyConnect will worsen these specific constraints as it will bring in more flows from west of Wagga and as far as I can see do nothing to improve capability into the rest of NSW.
I personally see Humelink as a project that will be built and more or less on time. Around $600 m of early works capital is already either committed or spent and approved by the AER. In my opinion it’s something that developers and existing stakeholders will find it profitable to take a view on. If you don’t think it will happen on time there are lots of implications. If you do think it will happen (and the target date is 2026) then that’s 2.5 GW of more power that can get to Sydney/Newcastle.
You’re probably closer to it than I am then David. I just hear about lots of landowners and others arguing for it to be built anywhere but on their land or underground or not at all and wonder how it’s possibly going to happen on time, even without the construction delays and cost blowouts that seem to beset just about any major engineering project these days.
At some point in Australia’s energy future, reality and common sense will kick in and everyone of us will at that stage realise we have been saddled with the most glorious of stuff ups in our nation’s history.
That is, we will have destroyed a fully functional, sturdy, secure, reliable, and cheap means of meeting our fundamental need for huge amounts of electricity.
It’s all been predicated on the theory that man made climate change is destroying our Earth and the only way to address this concern is by purchasing and installing horrendous amounts of costly short lifespan “renewables” and batteries. We now have a privatised national electricity grid, in which hundreds of entities all expect to make fat profits via a huge and bewildering accounting system – extensively outlined above – that few people could comprehend.
How can it be that we are destroying healthy coal fired generating plant with great urgency, whilst elsewhere in the world, new long life coal fired plant is being built at a far greater rate than that at which we demolish our own.
A sensible nation (if it truly must reduce its emissions) should have been doing so by building large scale nuclear powered generators at the various sites, where coal fired generation is reaching end-of-life, and operational transmission lines and already exist.
We would be seeking to build small modular reactors at these sites where growing demand for minerals processing and other heavy industry development could be encouraged by cheap electricity prices and the minimised need for new and lengthy transmission lines.
Of course, such a national electricity grid would have been best provided by full government ownership and run by the best qualified commissioners and electrical and mechanical engineers and executives the nation could enlist.
We must presume that such a coterie of altruistic Australian professionals will always be waiting patiently in the wings.
If not, we can only shudder at the potential disaster now looming ever closer.
Interesting comments Rob. I find it bemusing that you seem to have enormous faith in technologies and technical subject matter with which you’re evidently personally comfortable (“healthy (!) coal fired generation”, large scale nuclear, small modular reactors), but prepared to dismiss in such offhand terms other technologies and equally well-grounded areas of scientific and technical expertise (“horrendous amounts of costly short lifespan “renewables” and batteries”; “theory that man made climate change is destroying the earth”; “huge and bewildering accounting system”).
I think Rob’s concerned that the efforts (by state and federal governments) to chase emissions reduction targets using our electricity system is not going to achieve any measurable emissions reduction, but will certainly deindustrialise our economy while it industrialises the countryside.
The only conceivable benefit for doing this at all is if we end up with very low cost electricity and the associated benefits.
I’d very much appreciate some insights into how you think all this is meant to result in low cost electricity for consumers. I think that’s a fair enough question.
Common sense surely dictates that if it actually shuts down in August 2025, the Eraring Power Station should be kept in some form of “care and maintenance” state.
It will be a massive tragedy if the first action following closure was to be the spectacular dropping of its huge chimneys – something that coal’s detractors typically find absolutely delightful.
In the potential event that Energy Minister Bowen’s fanciful meddling backfires on us all, Eraring would become a vital asset whilst new options were needing to be brought forward.
If I may reply to the above response to my earlier contribution:
My reference to “healthy coal fired generation plant” was in reference to coal fired plant (that still has many years of valuable service life), being demolished prematurely. Perhaps a bad choice of words, but we taxpayers are spending billions of dollars on renewables and the essential collateral equipment required to connect them to the grid, while we should be dragging the last bit of useful service life out of our existing plant.
Furthermore, can there be any doubt that horrendous amounts of taxpayer cash are required to facilitate the introduction of intermittent renewable generation into an already functional system and then deal with the reality that this “well grounded area of scientific expertise” comes with short life spans and more horrendous expense to cyclically rebuild it all every 20 years or so.
As for the porridge of cost and profit sharing amongst the hundreds of corporate entities now enmeshed in the grid, I defy anybody to explain how the incredibly complicated financial processes are working for the benefit of all consumers.
Utility solar was and should be the third leg of the stool in the NEM> there is no way that utility solar can be generated and delivered, billed etc for the same cost as rooftop solar.
In effect the battle will be for utility solar + a lot of storage say 60-70% R MW/250% R MWh (R being the rated power) vs wind with 15-20% R/.5-1 R MWh storage for the remaining 50% of the energy after behind the meter generation and storage, hydro and waste to energy take the other 50%
It’s really the joint effect of non-price responsive rooftop PV crashing into the flexibility floor (minimum stable loading) of coal fired generation that’s driving this. With a different synchronous generation mix the issue would be less severe (unless the synchronous stuff was nuclear, in which case it would be worse).
It’s also important to understand that the Vic & SA results also reflect the endemic constraint issues on VNI not just the “wind correlation penalty”.
I found the ISP 2022 storage requirements more reasonable but still only just (my finding it covers up to 1 in 20 year events). Thank you for your analysis – very interesting. I looked at Figure 14 in the Griffith Study and it appears this event is less than the worst “1 in 42 year event” for 2 days of low VRE output. Figure 14 compares % output to average VRE output.
You can also see in this screenshot the VIC-NSW interconnector running counter-price. It’s been doing that all day – I saw over 1GW of counter-price flow earlier – with intervening periods of constraint when the NRM_NSW1_VIC1 constraint set has been invoked due to the accumulation of negative settlement residues.
I am not sure the costs of inadequate transmission from Wagga north-eastward are being accurately captured. Possibly (probably?) the need to activate RERT today would have been avoided if that south-westerly generation was available to the Sydney load centre.
It seems these constraints in southern NSW have reduced flow in both directions in the past. I seem to remember when supplies in Victoria were very tight that there was no flow from NSW. Maybe Transgrid should concentrate on fixing those issues rather than splurging on VNIWest and even perhaps Humelink
I would say that rather than being “disallowed”, what was happening here was simply a case of insufficient transmission capacity between the Snowy region and Sydney to carry all the available southern NSW generation + additional flows from Victoria. Once Snowy bid in its 2400+ MW from Lower and Upper Tumut, Origin 400+ MW at Uranquinty, plus hundreds of MW of solar generating in the area, there was no northward transmission capacity left to carry anything from Victoria. The N>>NIL_4 constraint (on one of the 330kV lines out of Yass) bound solidly, implying that any MW from Vic would have been at the expense of a matching quantity from a southern NSW generator – a zero sum game.
The only fix for this sort of limitation is augmentation of the main Snowy to Sydney flow path.
From the previous article on the issue, it sounds like there is additional transmission capacity that can’t be used due to the risk of a trip on one of the 500kV lines. Should this occur the “excess” energy would flow into the 330kV lines, which wouldn’t end well.
I suspect that the installation of a utility scale battery of sufficient capacity to provide fail over capacity should the 500kV line trip, would unleash additional transmission capacity. The Victorian Big Battery provides a similar service during part of the year for the VNI.
Paul, you continue to provide the most erudite and accurate analysis of all things NEM coupled with in-depth common-sense assessment of the relevant issues. Thank you. Greg
If I recall, this is a side effect of snowy hydro. Some of it is alloVic, some in NSW. Ramping up the hydro in NSW results in some power flow into Vic and vice versa. It might better if the snowy had its own economic region again.
*some is allocated to Vic, some allocated to NSW despite being phyically located on the same transmission network. Comment was truncated by some error.
Similar problems exist with NSW to Victoria flows. During hot weather a few years ago, I was shocked to see ~1000MW flowing into NSW against the price. In this case, part of the problem was Melbourne’s HV network (constrained at South Morang). I suspect we need to add phase shifters or some other means of regulating real power on those 132kV parallel paths, possibly the same on the EnergyConnect loop.
As to comments that these problems are new, they are not. The supply-side issues are from the design of VNI by then state and federal governments. The big change on the demand side is population growth in the urban heat island of Western Sydney.
AEMO’s on the day forecasts for Victorian rooftop PV maximum output have grown from 2,730 MW (forecast early this morning) to a latest forecast of 3,099 MW. Currently running at about 2,900 MW.
I’d say that’s the biggest factor in the falling market demand forecasts you’ve highlighted Paul.
For what its worth, I’m putting my money on the Callide U4 operators making a switching error when doing a test on the battery bank. Things just cascaded from there.
The problem occurs a lot more often than people think. I know of at least two other similar incidents this century. Problem exacerbated when they went from shaft driven to AC driven oil pumps.
Still waiting for somebody to explain how wind and solar reduce retail electricity prices.
I feel obliged to point out that a state’s demand met almost entirely by behind the meter solar sees the transmission grid at very low utilisation. But that grid is not paid for by utilisation, it’s paid for by guaranteed returns on the amount of stuff it owns. So it costs the same whether you use it or not.
And since the grid is growing, AND the grid operators now have responsibility for network stability, the costs can only increase.
It would seem that safe minimum demand was calculated when most generating units were typically 400-700 MW and had response rates of <1% per minute. The loss of one generator or transmission line when only eight were online would be a significant disruption, which would probably trigger further shutdowns, just as the 400 MW loss of Callide C4 led to the loss of 2,300 MW of generation as other units tripped off to protect themselves from instability.
Now however, not only batteries and the newly installed syncons but curtailed wind and solar facilities can respond within milli-seconds. In addition, the tightening of primary frequency control requirements on conventional generators means that gas, coal and hydro plants also respond more quickly to disturbances. Thus, the minimum safe level of generation will gradually decline in Victoria, then NSW and Queensland, just as it has declined from 250MW to 80 MW in SA.
In fact, it is clear that SA can already run on zero gas for hours if not days at a time but the risk of a mistake both in short term load shedding and the long-term political risk to the whole transition are not worth the benefit of saving 80 MW of gas generation.
Similarly, because across Victoria and SA it appears around 600-800 MW of wind and solar was being curtailed and in Tasmania and Victoria 4,300 MW of hydro was only running at less than 500 MW while batteries were actually drawing power, it is quite possible that Victoria/SA/Tasmania could have run at 100% renewables for at least four hours if Victoria's coal plants were flexible enough.
I don’t think the minimum demand issue is to do with loss of a large unit – Victoria is strongly interconnected to three other regions and frequency control wouldn’t be the problem. I’m still inclined to think it’s something a bit more arcane like voltage control, or Victorian AUFLS becoming ineffective because many distribution feeders would be carrying only small net loads with relatively low demand but high rooftop PV generation.
Sorry but this outage schedule looks like nothing more than minimum spend to keep a 1970’s power station struggling along – hardly a significant recapitalization which is what would be needed were the plant to be campaigned in the longer term. A read of the annual report says it all. Red ink in Australia and the second para on page 3 says it all. Major investments will be in “flexible plant” not old colliers. Hardly surprising. The asset managers are doing exactly what they should be doing – spend nothing they don’t need to on an old plant gliding to retirement. Keep it safe and sort of reliable – the trick is to get the balance right between maintenance spend and losses in the wholesale market from unreliability. Expect forced outage rates to go up the closer they get to retirement. And they won’t be the only owners juggling these balls in the air. Best hope they don’t all drop them at the same time.
Thanks for sharing your thoughts, Andy, and apologies for any confusion inadvertently caused.
The above line of thinking followed what’s noted on p3 of the CLP report, specifically:
“EnergyAustralia is bringing forward major outages at Yallourn to 2023 and 2024 to address the ongoing reliability issues while at Mount Piper the main coal supply contract has been renegotiated to include a backup mine. These actions have been taken to restore operational and financial performance.”
… so will watch with interest to see the extent to which operational performance at Yallourn is restored.
Bit confused by the 2023 timelines in this Watt Clarity article.
The Energy Safe Victoria Annual Report is for 2021-22 year and states Cohuna Solar Farm started commercial operations in 2021 and “on 18 October 2021, three minor grass fires were reported at the farm that originated from the solar panel rotational control boxes. These boxes allow the solar panels to track the movement of the sun. We issued a direction to disconnect the tracker boxes as a result of initial investigations carried out after the fire. Further investigation uncovered other issues with the T4 connectors connecting the solar panels and with the T7 connectors connecting the fuse boxes to the direct current harness cables. We issued a second direction to stop the use of T4 connectors. This meant that the farm could not operate.
We worked with manufacturers Enel and Soltec to rectify the issue and ensure the farm was safe to be operated. The T4 direction remained in place until a solution was implemented. As of 30 June 2022,
the direction for the tracker controller boxes was still in place. The supplier has re-designed the product and is in the process of getting it safety tested to show compliance to the relevant standard and
that it is safe to be connected to the supply.”
AEMO databases have Cohuna Solar Farm moving from in-commissioning to in-service in mid-2023.
The Watt Clarity article has 5 outages in its first 6 months of commercial operation including a ESV direction to disconnect equipment on 8 December 2023 following a grass fire in November 2023.
Are the year dates mixed up or was the problem not resolved in the first place? Which means the unresolved problem caused the November 2023 fire which could have had serious implications for surrounding properties.
‘AEMO databases have Cohuna Solar Farm moving from in-commissioning to in-service in mid-2023.’
… but to be clear I have not specifically looked at when commercial operations started.
You say …
‘The Watt Clarity article has 5 outages in its first 6 months of commercial operation’
… but to be clear the chart at the top only looks at calendar 2023 (so starting 14 months after the first fire reported in the ESV 2021-22 Annual Report.
Finally you speak of …
‘the unresolved problem’
… but readers must be clear that I have no knowledge, and am not commenting, about whether the fires in Oct 2021 and Nov 2023 are related or not.
Hi Paul My comments are not a criticism of your excellent article but complimentary of your digging for further information on what is going on with Cohuna Solar Farm and whether the issue is visible to the market and AEMO. My concern is AEMO’s Generation Information Page has Cohuna Solar Farm going into service in mid-2023 which is after the concerns of the ESV reported a year earlier have supposedly been addressed – but have they been addressed? as ESV directed Cohuna SF to cease operations 6 months after the “in service” date for what looks to be the same issue.
According to AEMO’s database on expected closure dates Cohuna SF will close in 2045.
Are the issues experienced by the “single axis tracking” Cohuna SF isolated to Cohuna or are other SAT solar farms experiencing similar issues.
Its unclear as the ESV report states “we also issued a safety alert to warn other solar farms about the importance of maintaining electrical installations and clearing vegetation to minimise the risk of
fire. An owner or operator of a solar farm should have a plan in place that minimises the risk of fire.”
It would be great to know if the issue is design or maintenance related. Design means more solar farms could be at risk of fires.
As an active rural fire fighter who has recently worked alongside Victorian fire fighters in Far North Queensland (yes only a couple months ago FNQ had fires now we’ve had torrential rain and floods), I sincerely hope there is a fire management plan for all solar farms as rural firies in Queensland cannot attend structural fires or fight fires under power lines.
“an elevated risk of insufficient demand to maintain a secure operating state”
Shouldn’t that be
“an elevated risk of over supply to maintain a secure operating state”
Do the curtailed wind and solar farms get paid for production loss?
Hi Peter – I think the AEMO wording is appropriate. Potential oversupply is readily managed – at worst AEMO could direct generation off; on the other hand issues like voltage control or impairment of AUFLS capacity are related to low operational demand, not excess large scale supply.
Wind and solar was not curtailed in most cases – they simply offered their output to the market at prices not low enough to get dispatched. Nor would they want to have been dispatched at -$400/MWh. Generation curtailment (resulting from network constraints) does not get compensated in the NEM in any case.
Jonathan, you arguably should have taken Peter to task for besmirching your excellent article.
Peter: inertia is not kinetic energy. You need to go back to the source, Isaac Newton.
Inertia is resistance to velocity change. A T/G has the same inertia when it is at rest as when it is at 2,970 rpm (49.5 Hz) or 3,000 rpm and the same whether it is generating 50 or 500 MW.
You also appear to be confusing energy and power. In your example, when the frequency started to drop the T/G was delivering 500 MW. At 49.5 Hz, if its AVR took no action, it would be delivering ~490 MW – the drop due to the drop in voltage due to the drop in speed/frequency.
If the drop in frequency occurred in 2 seconds the T/G delivered ~990 MWs of energy to the system (in those 2 seconds) not 90 MWs. How much of this came from a decline in the T/G’s kinetic energy is irrelevant.
Yes, a 50 MW battery only takes a tiny amount of time to start delivering 50 MW, and if we ignore that short delay and any tiny power decline, the battery delivers 100 MWs in the 2 seconds. To deliver the same amount of energy in the time interval you need a 495 MW battery.
But that’s also irrelevant. This is about RoCoF and the inertia of the T/G was in play immediately.
The system is in need of inertia to slow the rate of RoCoF and of power to reverse the decline.
To be harsh, if the decline was caused by the 500 MW tripping a bit more than a 50 MW battery will be needed to halt and reverse the decline.
The battery is good for the secondary response and can give that service from a quiescent state. Whereas the T/G had to be operating and then, if designed to contribute to frequency control, could deliver another 50 MW in a minute simply by AVR and governor action (yes 50 MW of so called “overload”).
Fortunately the boffins in AEMO appear unswayed by your arguments. AEMO’s 2023 system inertia report agrees batteries can contribute to solving the problem of declining inertia, so can syncons particularly if they have flywheels as in SA, and grid forming inverters.
CSIRO has again chosen not to receive expert advice on nuclear costs for their draft GenCost 2023-24 report. Aurecon has again provided expert analysis of all technologies except nuclear. The last time that CSIRO obtained expert advice was from GHD in 2018. For the latest report, CSIRO has attempted to prove by their own analysis that nuclear is too expensive to consider by quoting the cancelled NuScale SMR for Utah Associated Municipal Power Systems (UAMPS) Carbon Free Power project (CFFP) but their analysis is misleading, for example the project was cancelled when the LCOE increased from USD58 to USD89 (at 0.7 exchange rate used by CSIRO = AUD127) but CSIRO quote an LCOE of AUD636. I suggest that CSIRO will need to clarify their analysis for the final report.
“Projected Costs of Generating Electricity 2020”
Nuclear thus remains the dispatchable low-carbon technology with the lowest expected costs in 2025. Only large hydro reservoirs can provide a similar contribution at comparable costs but remain highly dependent on the natural endowments of individual countries. Compared to fossil fuel-based generation, nuclear plants are expected to be more affordable than coal-fired plants. While gas-based combined-cycle gas turbines (CCGTs) are competitive in some regions, their LCOE very much depend on the prices for natural gas and carbon emissions in individual regions. Electricity produced from nuclear long-term operation (LTO) by lifetime extension is highly competitive and remains not only the least cost option for low-carbon generation – when compared to building new power plants – but for all power generation across the board.
The International Energy Agency’s next report, due in 2025, will likely be very sobering reading.
How does the bidding fixed load works? I presume it is required for plant restoration times. will the generator be exposed to any market price risks when bided fixed load?
Stacked under the bids down at -$1,000/MWh in the NEMDE dispatch process, Selva … easiest to think about it as a ‘must run, for technical reasons … I will take any price’ bid.
Commonly used in testing scenarios, and on other occasions like this.
A HPBV is an “optional extra” on a unit intended to permit the unit to survive a major load rejection without tripping the boiler. Basically HP steam is bypassed directly to the unit condenser rather than through the turbine. Not intended for extended use but to manage transients and also assist in start up. Now it is interesting that it is Stanwell 1 – the first unit of what was then Australia’s first heavily automated coal fired station. As such it probably went through lots of start up , shut down and testing cycles during commissioning. Creep is a function of temperature/ pressure and time whereas fatigue can have a few causes. This one is likely to be a result of rapid temperature changes to the pipework around the valve when it opens in anger. Probably has a bit more accumulated use from being unit 1. Probably found an issue when performing metallurgical inspections during their outage (eg a crack or similar). This will now be being managed by making sure the HPBV does not need to operate by (maybe) running at less than full boiler pressure (hence reduced output) and limiting governor action (constant load) until they can get replacement pipework manufactured and get it glued in (no small undertaking). Sounds like good maintenance and engineering at work. Pleasing to see the guys are on top of their plant.
Thanks for helping readers out with some more context, Andy.
Along the same lines, worth reminding readers that Stanwell has TTHL (trip to house load) capability:
1) Which (if my scratchy memory serves me correctly) would be using HP bypass
2) Tested regularly as part of their black start contract, but also used ‘in anger’ on occasions, such as following Callide C4 catastrophe on 25th May 2021 as Allan noted here afterwards.
Dan
Great article very interesting.
Looks like most will exceed their design life but the fall from ~28% to ~20% in CF may accelerate their replacement with larger turbines. Their break even time is longer than I expected and curtailment will probably increase. It would be interesting to see a similar analysis on Solar..
Maybe they were testing whether the inverter is flammable, in which case the lodged rebid reason is entirely appropriate (and the test successful).
But seriously, since when do we demand that rebid reasons be 100% specific about the root cause of the plant issue, especially when the operators need to diagnose, triage, and rebid very quickly? That rebid reason is fine and no one in the market is worse off for its un-specificity.
When VBB and Bouldercome experienced fires, Wattclarity didn’t publish articles admonishing them for not using the word “fire” in their rebid reasons, so why the consternation this time?
Understand the unscheduled 12 MW Windy Hill Wind Farm in Far North Queensland was not selected for your excellent article. However, it could be the trail blazer for what happens when a wind farm is decommissioned as it was commissioned in 2000, three years earlier than the oldest WF in this article.
Also as far as I’m aware there isn’t a wind turbine fire database. Think this should be mandatory as any fire caused by a wind turbine could have have catastrophic impacts, especially if the wind farm is on forested ridge difficult for firefighters to fight. It may also give investors some insight as to what turbines are best suited to the Australian climate.
My submission to the CSIRO pointed out that they should review nuclear technology costs in their entirety (instead of exclusively SMRs, still under development), and that the coal costs (greenfield, AUSC, Japan coal costs) are not representative of a project that would be built in Australia.
If I cared enough to go further, I’d point out that they use the high performing wind and solar as the average, then assume future projects will be both cheaper and better performing (higher capacity factor), while reality shows the opposite (especially with wind) and that over-build must increase curtailment.
Realise your list is curated but would have been interesting to see analysis of the Toora Wind Farm – it predates most of these by at least a year, though is smaller (21MW).
The original Albany Wind Farm (Western Australia) was commissioned in October 2001, after ten years of planning. The wind farm has the capacity to produce 80 per cent of the electricity requirements of Albany. Originally commissioned in 2001 the farm was the largest of its kind in Australia.
courtesy of wikipedia
Pleasing to see commentary and encouragement of understanding of bidding practice, and in particular, uderstanding that plant dispatch directed on the basis of simple variable (e.g. fuel) costs mdels cannot efficently sustain operations. A generator’s minimum bid is the price below which the generator will be losing value. A loss of cash by bidding below fuel cost in one interval may however be anticipated as a risk investment in being ready to generate more profitably in the next or a later interval. Conversely a bid above the minimum runs the risk of missing out a profitable dispatch. To manage these risks, the generator needs to anticipate the system demands and the likely behavior of competing sources.
Good to see commentary and encouragement of understanding of bidding practice. In particular, plant dispatch directed on the basis of simple (e.g. fuel) costs cannot efficently sustain operations. A generator’s minimum bid is the price below which the generator will be losing value. A loss of cash by bidding below fuel cost in one interval may however be anticipated as a risk investment in being ready to generate more profitably in the next or a later interval. Conversely a bid above the minimum runs the risk of missing out a profitable dispatch. To manage these risks, the generator needs to anticipate the system demands and the likely behavior of competing sources.
The weather forecast for the SE QLD doesn’t look that bad or abnormal for this time of year. That still doesn’t mean the QLD grid isn’t going to have troubles. Their lack of utility scale batteries and wind based generation really hurts them at this time of the year.
I’m not sure what question you are trying to answer?
Being ~30MWdc they have multiple PCUs/Inverter stations potentially available, and depending on their reticulation arrangement they can shuffle their collector feeders to continue their 5MW (notifiable exemption status likely) export for hot commissioning on the non-burnt units. And their commissioning test plan would likely be rotating PCUs for that hot commissioning.
I assume you are only trying to ping them for the rebid reason wording, in which case it might still be accurate enough… at least as accurate when FF generators have ‘maintenance’ during low $ periods etc?
Thanks for the details, Sarah … but just for the avoidance of doubt (re ‘only trying to ping them’) it’s not my intention to ping anyone in my articles … just seeking to understand more.
If the NSW is so confident it doesn’t need Eraring then the answer is simple – the NSW Government should bear all RERT costs. Just one RERT event in Queensland on 1 February 2022 which was only 331 MW for a few hours on one afternoon cost Queensland consumers $50.1 million. The SA Black System event which occurred almost at the close of a working day cost SA businesses hundreds of millions of $. How much are NSW considering paying Eraring to stay open?
One little correction Paul – the high price in NSW wasn’t *because* of the forced southward flow on VNI. It was the outcome of a complex constraint situation that was both forcing flows out of Queensland into NSW and setting high prices in those two states, with a further side effect of the same constraints being the flow out of NSW into Victoria.
Meanwhile we have evidence of Energex throttling home energy usage in South East Queensland and loss of power in Moreton Bay Council regions (load shedding?) as solar came off for the afternoon.
So it seems supply was not able to actually keep up with what the consumer was demanding.
Hi V, thanks for highlighting those aspects. The ‘throttling home energy usage’ would have likely been seen by those with home devices (e.g. AC units) participating in demand response programs e.g. ‘Peaksmart’. But I’m aware that there is some consternation from affected renters who don’t own said appliance and didn’t sign up to the program themselves.
There was no load shedding or out-of-market direction to reduce demand at the state-level by the market operator and all market generators appear to have remained online. It is likely the local outages you reference were in the distribution network. Definitely appears the distribution network struggled and in the absence of these outages demand could have been even higher.
Imagine the Queensland demand with an export green hydrogen industry in regional Queensland and the additional demand of desalination in Southeast Queensland. The additional loads won’t have much flexibility to ramp up and down for any extended period of time. I’m assuming green hydrogen will need offtake agreements for project financing.
The Gold Coast Desalination Plant has been operational since November 2023. The December 2023 Seqwater Water Security Status Report states the “Gold Coast Desalination Plant (GCDP) production was maximised during the month of December, in order to support the SEQ Water Grid and slow the decline of SEQ water storages”.
Imagine being dependent on the wind to blow for urban water supplies during a wind drought. The IPCC is forecasting a global still of winds and Australia is a country that the IPCC say will have reduced wind speeds.
The SEQ Water Grid has increased to 72.6% (ie above the 70% drought measure threshold) so assume the Gold Coast Desalination Plant was not operational yesterday and did not add to the record over 11,000 MW peak demand in Queensland.
Always hard to talk about dam building when the media is full of floods and cyclones but unlike solar and wind farms, dams for hydro and water supply take years to plan and build.
We need water to survive and affordable and reliable energy to thrive.
You can have both please check out the proposed 1,000 MW pumped hydro Koombooloomba Energy and Water Project (that can operate in low or no wind and solar) on the Queensland Electricity Users Facebook page.
Update – Its the Queensland Electricity Users NETWORK page for details on the proposed 1,000 MW pumped hydro Koombooloomba Energy and Water Project (that can operate in low or no wind and solar).
The project would secure the water supplies for Cairns for the next 100 years (Cairns needs to build a dam for water supply by 2035).
If the proposed 2,000 MW Borumba Pumped Hydro was reduced to its original 1,000 MW, water could be supplied to the SEQ Water Grid reducing the size of the new SEQ $4 to 8 billion desalination plant that is currently undergoing a detailed business study along with a detailed business case for an expansion of the Gold Coast Desalination Plant.
Energy and water are joined at the hip – water to survive and affordable and reliable energy to thrive.
If I am reading the AEM Dashboard correctly, Queensland gets a lot of power from NSW via the Interconnect?
And that is after NSW buys it from Victoria who seems to be selling to NSW, SA and Tas.
So if NSW have an LOR2 would that mean they are unable to onsell power to Queensland?
Following Alex’ comment, Mick, this looks like a ‘trip to house load’ (i.e. TTHL) test … in order to qualify for SRAS (System Restart Ancillary Services) contracts.
Would potentially be able to see more tomorrow, with ‘next day public’ bids.
So why the massive power disruptions from 22nd to 23rd in SE QLD? No explanation from Energex whose website updates are useless, only telling how many customers are without power and no advise as to how long estimated to fix.
Hi Baz,
It’s clear there were many distribution-level outages, many unplanned and likely driven by the weather and demand circumstances on the day. The supply side – by which I mean electricity generators that supply the electricity market – and the high-voltage transmission network appeared to fare quite well.
Peak dispatchable supply occurred at 6:45 when wind and solar were supplying just 245 MW. 7.9 GW of available coal was supplying 6.7 GW and 3.4 GW of gas, diesel and biomass supplied 2.8 GW and 330 MW of batteries were supplying 125 MW so it appears that the system was well in hand
If this happens next year Qld will have Barron Gorge and probably Kidston 330 MW of hydro, at least 300 MW of batteries and of course Callide C. In total, almost 1.5 GW more dispatchable supply. With some 1.5 GW of wind and more than 1.5 GW of solar coming on line over the next 18~24 months, even if it is only running at 10% still a useful contribution. So assuming Queensland coal manages a bit better than 75% capacity next year there should be no problem
Batteries and pumped hydro are not generators – they store and release, using more electricity to charge/pump than what they discharge. Sadly batteries and pumped hydro seem to be continuously referred to as some type of dispatchable generation.
Absent coal generation and with the uncertain future of gas generation, batteries plus the Kidston, Borumba and massive Pioneer Burdekin Pumped Hydros will be reliant on excess solar and wind to recharge.
Hopefully they can recharge before the next peak demand period.
However, this could be a massive problem in a prolonged wind drought which could go for days or weeks or a heatwave that goes for 2 or 3 days.
Hopefully the sun will always come up but the wind doesn’t always blow…. and batteries and the pumped hydro will be hoping for the overnight wind to recharge.
But in the future Queensland could have new overnight demand in the form of a green hydrogen export industry and desalination plants in Southeast Queensland.
The evidence does not support your concerns. To have a reliable 95%+ renewable grid Queensland needs 3-4 times as much solar and 6-8 times as much wind and about as much storage as Victoria is planning.
Thus between 9 and 4 wind and solar would have been not only supplying all demand but most of the existing “offpeak” demand and still putting 6-8 GW into storage. At6:45 when coal output peaked, wind and solar would still have been supplying 3-4 GW,hydro and storage 6-8 GW. On a few occasions per year, gas might supply 20% of peak demand 10-15% of daily demand and 2-5% of annual demand
Interesting to note that the power outages were claimed to be due to storms but Moreton Bay Regional Council didn’t even experience rain and had a large portion of the power outages.
Also all that solar and wind and it produces nothing when demand is peaking.
Look at the chart in the previous article – PV was generating a significant proportion at the time of peak market demand, and even more at the time of peak underlying demand (which was of course higher, but also earlier and continued for longer).
RERT contract entered into but may not be activated or dispatched.
As stated in your articles this is a long weekend but Queensland is also without thousands of customers due to Cyclone Kirrily that impacted the Townsville region on Thursday night.
The cyclone knocked out 66,000 customers but as of Saturday 9.00 am 20% had been restored with most expected to be restored by Sunday night (great effort by Ergon staff).
However, this means that on a long weekend and thousands without aircon, Queensland did not have sufficient scheduled and semi-scheduled generation to meet demand.
Don’t you think you could have eleborated just a little on what the dew point actually is and why it’s such an important metric in this context? Just c.c-ing someone else’s Twitter post doesn’t really amount to an article.
With all the extra renewables there seems to be no problem. The sooner the coal fired and gas power stations are shut down the better.$1000 MWh would seem to be unsustainable.
As the price has increased over the years there would be some reduction in demand at the margins as some people will economies. That would moderate the increase in demand that we expect from population growth.
Another type of demand management is the closure of firms on the margin of survival in energy-intensive industries. The result is deindustrialization that is happening bigtime in Britain and Germany!
What a mess – and given that the population is growing out of control and there doesn’t appear to be any believable figures for the additional generation (firmed?) necessary for charging more and more EVs, “system blacks” appear unavoidable.
I agree with the conclusions regarding the RRO, but the first section of this article misrepresents history.
Once Northern PS committed to closing, there was only one generator that could have feasibly supplied that large customer load. Realising this, that generator sought to extort the customer as much as possible, demanding a price well above market rates. The customer refused and took the risk of floating on the spot market, and the generator in question punished them for this decision, withdrawing capacity in order to induce volatility. You can argue this was a reckless gamble, but it wasn’t necessarily a bad one – a week of volatility is ultimately easier to stomach than years of overpriced power.
This wasn’t the impetus for the RRO, however. Eight months later, that same generator, still seeking to extort South Australian customers, withdrew part of its capacity for the summer, leading to controlled blackouts in SA. This is what drove the creation of the RRO, the argument being that blackouts could’ve been avoided if customers had given in to the generator’s demands; that AEMO later took the generator to court for misrepresenting their availability seems to have been overlooked.
Many people recognized that this was actually a competition issue – if multiple generators had been competing to win customers, it’s unlikely that a large portion of load would’ve gone uncontracted. The RRO does nothing to resolve that problem and, in fact, makes it worse, as it reduces customer bargaining power. The MLO was stitched on to the RRO in an effort to counteract that effect, though who knows if it’s worked.
As the article points out, none of this has anything to do with actual system reliability, which is a bigger issue than the level of contracting over a short period.
Thanks Teffy, you’ve been part of an experience to which I was unaware.
The period I was referring to was the period immediately prior to the decision to close Northern early and offers by its owners to sell contracts from it. I gather you are referring to the challenging competitive situation after it had committed to close. I imagine that would have been very difficult time indeed for customers.
The general industry narrative, however, that was used to justify the construction of the RRO, referred to decisions of the customer being made prior to the decision to close northern.
In the end, this is all ancient history, based on peoples, subjective recollections of a series of negotiations. my point in bringing it and the neg up is just to show that the RRO was an accident of random history, rather than an outcome of any thoughtful design.
These short minimum don’t worry me as much as the many consecutive days of very low VRE output in Victoria and SA that occurs at least once per “winter” period per year. These events are often also followed or preceded by periods of poor VRE output. These periods are also during the time of year when demand is very high. They are also long enough that most BEVs will have been depleted even when used purely for road use as distinct from V2G.
If V2G becomes a large supply option, then combined with other storage such as residential and utility scale BESS, short periods of VRE won’t be a problem. The technology and economics are generally improving to assist in solving these short term VRE dips.
So according to Teffy- who seems to be mixing up several years – generators stop running for a couple of years to make money. Yeah, righto. Seems like a classic case of a customer expecting the market to stay static as they made a bad call. If you want prices to stay low, lock in a contract. If you don’t the plant that pushes down spot prices won’t run, and prices will rise. It’s not hard Teffy.
The project energy connect RIT-T modelling got much of its market benefits through it causing early closure of SA peaking plant, which lowers total market OPEX. This outcome seems to be confirming that modelling and should be no surprise.
Great, you’re celebrating the fact that the FCAS market will increase when coal generators close down. What about the small matter of the ENERGY that is lost when those coal generators are gone? Funny how little you talk about battery energy. Most battery-pushers, like yourself, always underplay energy and instead love to gloat about nominal capacity levels.
How does your battery future work when Eraring closes next year (which you’re obviously pushing)? Eraring’s daily energy is probably at least 50 GWh all day everyday (taking into account cycling and not running at full output all day). At the moment, the biggest battery in NSW is the Riverina and Darlington Point BESS with a whopping 0.3 GWh of energy! Incredible! It’s safe to assume that whopping 0.3 GWh would be a daily limit since batteries will generally run a single cycle in one day.
I know it’s really exciting that these batteries can enjoy some of that coal FCAS cake but what about the small issue of losing 50 GWh of energy and replacing it with technology that offers 0.6% of its energy.
“How much more comforting it is to know that we depend on millions of interconnected small units rather than a few large units.”
That’s incredible. You’re saying that managing millions of small independent units is way more comforting than several giant units. You should become an engineer! Lets totally ignore the worsening DER problems on the distribution networks such as voltage regulation, phase imbalance and congestion.
There are pictures going around, but unless it is from a reliable source, then we still have to wait. If it is the case then we have had multiple events causing multiple failures with those 500kV towers. In both situations the towers were rated to handled higher wind speeds than those that generally occur around the events. Meaning we would have to either have a fault in the material, design, construction or maintenance of the towers, or we have had two previously extremely rare weather events within a year or so of each other.
I can only recall one incident with collapse of 500 kV towers, namely the collapse of the group of towers near Cressy in Jan 2020 which islanded SA. The 500 kV lines west of Moorabool are on double circuit towers. The 500 kV lines east of Moorabool, which tripped today, are on single circuit towers of completely different design.
There was the collapse of one double circuit 275 kV tower near Tailem Bend in Nov 2022 in a freak storm. A 220 kV tower near Poatina in TAS collapsed due to a landslip in Oct 2022. Back in Sep 2016 a number of towers on different 275 kV lines in different locations in SA’s Mid North collapsed in the storm that led to the System Black.
The recovery to 50.12Hz would have been the initial fault clearing. The roll-over after that likely would be runbacks/control schemes for windfarms. LoyA units may have (mal)operated on ROCOF after the cascade of frequency.
On the mid afternoon of Tue. 13 Feb 2024 I saw both Sydenham T.S. to Moorabool T.S. lines down from several km
away.
The road which is over-crossed by the two lines was blocked by police; they showed me a phone photo of two
collapsed towers; but wouldn’t allow public in the area. The police said six towers were collapsed.
The photo that I saw showed the collapse of the section part way up where the tower bifurcates to make
space for the centre phase.
It appeared that three sequential towers and thus four spans of each 3 ph. lines were not visible
from a distance.
I saw evidence of a microburst in the area with many trees uprooted and others snapped or torn apart.
There were some wet patches of road in the area indicating very recent rain.
The 500 kV lines in that area were built between 22 kV / 12.7 kV distribution lines. There were no aerial crossings.
(There is an underground 22 kV undercrossing outside the MLTS fence in Anakie Rd. some km to the south west from the tower collapse area) I would be very surprised if there were any recent aerial crossings.
The AusNet outage page has been really struggling to load since yesterday afternoon and still this morning. Pretty poor that they aren’t set up to handle spikes in web traffic.
In reference to the frequency deviation, the event lasted ~10 minutes with frequency returning inside the 49.85 prior to 13:20 and then eventually recovering to the 49.9 allowing contingency response to ease. The most interesting part is the swing from nearly under to nearly over frequency back to an under-frequency event.
I think the initial frequency dip was the initial fault and possible reclose, also possible that voltage harmonics on the signal your system was processing made the dip appear larger than it was.
With so much non-synchronous generation on the NEM nowadays, the rate of change of system frequency for events keeps evolving due to ever continuing lack of inertia imo.
My guess is that LYA holds the current Vic Trip to House Load contract with the NEM, and had all four Units selected active at the time of the event (to enhance the probability that the required minimum generators successfully Island to fullfill the contract obligations).
I believe the rapid ROCOF you have demonstrated above probably triggered the detection relays of all four Units, …..and unfortunately all four failed to Island effectively.
Hopefully the battery systems and EDG’s at the plant stood up to the subsequent Station Black.
The escalating loss of load from local distribution outages as the storm front increased in intensity on the outset of LOR3 load shedding would be the main reason for the low quantity of intentional load shed (300MW) and it brevity. Vic’s demand was on track to reach 8,000 MW before thousands lost power.
Opens the door for the trusty ole climate-driven unprecedented storms” gaslight option.
More alarming is the huge volume of reported household phone/internet outages, coupled with loss of 4G internet at the same time, only hours after the event.
This issue alone needs to be looked at, NBN should be forced to comply with minimum battery/EDG supply performance guidelines. This would allow the public to check in on relatives before their land line is lost (like what happened to my 89 year old independent living mother last night.
JDH Consulting are wind loading specialists and have been studying failure of high voltage transmission line towers in Australia, particularly in Victoria for about 25 years. Failures due to small but intense storms such as microbursts are common for a number of reasons, such the wind structure within the storms and the directional orientation of the line. The risk of failure along a line is also dependent upon its length. JDH has been in discussion with AusNet regarding a specific study of line risk for the Victorian network, similar to the extensive studies carried out for ElectraNet in South Australia, following the many failures that occurred there in September 2016.
The article correctly highlights the age of the electrical transmission infrastructure in SE Australia, and the many failures from severe convective winds that are occurring. However, a factor that is often overlooked is the need to design for the particular characteristics of convective winds in comparison with those of larger synoptic scale wind events for which the older systems were designed, often based on methods developed overseas in much different wind climates. In layman’s terms, the convective wind events generate a single large gust that envelopes 3 or more spans, and results in the failure of several towers in a group as we saw on February 13. The result of this is that a convective wind event with a peak gust of say 140 km/h can generate wind loads on a span 50% higher than those from a large-scale synoptic wind event with the same maximum gust. It is essential that design methods for wind loading accounting for these characteristics are adopted for the new transmission lines under construction and planned for the energy transmission.
Another question wotth investigating is whether the wind and solar farms are actually negatively impacting the climate by producing more volatile systems.
What was the response of industrial equipment to the frequency and voltage changes resulting from the event?
How did station auxiliary equipment respond to the disturbance, did this contribute to the trip of LYA?
How did Loy Yang B and basslink ride through the initial disturbance and trip of Loy Yang A (could there be lessons for inverter based and thermal generators)?
Also a partial answer to A2
At 4:27pm local time, AEMO issued notice 114641
RECLASSIFY CONTINGENCY
13/02/2024 03:27:46 PM
Reclassification of non-credible contingency event due to existence of widespread abnormal conditions – VIC region(s).
AEMO ELECTRICITY MARKET NOTICE.
AEMO has identified that a non-credible contingency event impacting the power system is more likely to occur and is considered reasonably possible because of the existence of widespread abnormal conditions namely severe weather in the VIC region.
45 minutes later, there was a trip of all four 500/220 kV transformers at the Hazelwood Terminal Station.
Market Notice 114643 “Significant power system event
– Victoria Region – Hazelwood transformers”.
A side note, during the LOR3 period I was unable to access my DNSP (Ausnet, offline altogether) or the AEMO (Gateway timeout) websites. The latter did delay my response as an Amber customer. Other parts of the state suffered large scale communications outages. I question whether our telecommunications and IT infrastructure is suitable for market participation or even advisory of outages during severe weather events. I also wonder about the effect of communications outages on virtual power plants or demand response.
Safety factors incorporated at the time these existing transmission towers were being designed were doubtless higher than those applied today. Nothing wrong with that since today’s material quality standards are superior and support lower safety factors. If heavier cables have been strung, that would test any ageing components.
The Sydney Harbour Bridge, West Gate Bridge, etc, should probably be reassessed against today’s operating circumstances with heavier trucks and more of them etc.
Who is auditing the condition of this ageing infrastructure?
– Automatic changeover for Y protection inoperative, configured such that both X and Y redundant protection are powered from the same DC supply.
– Switching sequence has risk of interrupting DC power to plant, removing redundancy from critical systems, and all protection, was carried out while plant is running.
– No recovery/rollback plan or procedure for loss of AC power to, or failure of, the battery charger while DC system is operating without a battery is mentioned. Restoring DC supply from one of the batteries would have improved the situation.
The accompanying video also mentioned that operators did not ask Calvale to isolate the generator as they feared a steam valve leaking by. However, reverse power would suggest the grid was not providing a mechanical load on the turbogenerator, so isolating it isn’t going to make the situation worse.
AEMO did invoke a constraint to procure 450MW of mainland raise reg – possibly to cover overspeed cutout at windfarms? I could see VIC bringing in a protected event category similar to SA – not sure how they would operate system more conservatively eg SA limiting flow on their one AC interconnector
The tree’s in background look fine begs the question the quality and the integrity of the steel and design of the tower’s I Wonder where they where made and designed??.
With regard to the 2.3GW generation loss of LYA, given the 1GW of lost load downstream of the transmission failure, we should regard the low frequency event as a loss of 1.3GW.
Re Basslink/LYB/(and YPS) riding through the voltage dip and subsequent low frequency, I feel it is likely that LYA tripped due to the operation of its equipment that facilitates its Trip to House Load market ancillary service contract with AEMO?
I believe there is an electrical/magnetic component involved, as current is passed through the cables it essentially turns it into a “motor” which the geomagnetic storm twists the towers, generating a great amount of torque on the frames.
It reads similarly to how the University of Queensland accounted for the performance of their St Lucia BESS, where a significant source of revenue was the cap contracts they were able to forego as a result of being able to manage their own risk with the BESS.
It shows our energy system is becoming more and more fragile as we remove reliable coal fired power stations and replace them with distributed renewables that require significantly lengthy transmission networks and interconnectors with other regions that significantly expands the area of risk.
Network charges on our Electricity bills have been rising ever higher ,year on year to such an extent experts and many in the know have termed all the money being charged as ” Gold Plating the Network” suggesting …consumers are and have been paying too much money in their bills to justify the cost of keeping the grid and networks operating optimally.
NOW…this article suggests that major network infrastructure…Electricity Distribution Towers…at least in Victoria is averaging an age of 50 years+ and as such in some areas is reaching it’s expected life span!
SO… Why have consumers been paying higher and higher prices for Network Charges that are supposed to be needed to keep the Network/Grid in peak shape…and yet have major aspects of that grid / network reaching their life span ? Where has and is all that money for gold plating gone ?? And where and what has it been spent on ?
You CAN’T charge Network / Distribution costs that are supposed to keep things in peak condition , up to date and modernised …justifying the high charges being paid …YET here is an example of Network Infrastructure that HAS NOT been maintained and /or replaced since or before the Introduction of our Decimal Currency in 1966!!!
Brett, the investment has largely been in new lines and upgrading existing lines to connect renewables to the NEM.
So we have had billions spent and the network is now less reliable than it was before because we have much more distributed power generation that relies on flimsy line infrastructure.
It doesn’t show that the system is becoming more and more fragile at all. What bit of six of ten coal generators going offline don’t you understand.
Three years ago, a failure at one 400 MW coal generator in Queensland caused 2,400 MW of other generation to go offline and 400,000 customers to be disconnected.
In this case Victoria had sufficient reserves so that even though 2,960 MW of its coal generators were offline, within two hours of the incident all load shedding had been reversed and it was already exporting power.
The average transmission distance is falling as rooftop solar has near zero reliance on transmission and all of Northern and Western Victoria and even western Melbourne is closer to generation than it has ever been. More interconnectors allow diverse transmission paths so in fact reduce risk, particularly as the average load on interconnectors is declining
SA and Victoria both had massive blackouts across the state because they don’t have sufficient reliable generation and strong winds brought down the distribution network. Victoria was extremely close to having a system black event and things becoming significantly worse than they were. The risk of this happening again in SA, Vic and Tasmania is much higher as we rely on thousands of kilometres long transmission systems to get power from distributed renewables and don’t have sufficient synchronous generation located close to load centres.
On the other hand QLD had a catastrophic failure at an actual generation plant which of course affected the other units at the plant taking them offline. Meanwhile the majority of the state was unaffected. The Callide failure was not caused by a weather event and the risk of it happening again is extremely low.
Capacity revenue maybe from
“
The latest tender with Victoria and South Australia will target 600 MW of dispatchable capacity with 4-hour equivalent duration across the two states, in line with the Australian Energy Market Operator’s 2022 Integrated System Plan.
Bringing this new generation online will help cover the loss of capacity from the scheduled closure of South Australia’s Torrens Island B power station in 2026 and the 2028 retirement of Victoria’s Yallourn Power Station.”
That is an insanely large amount of money to produce a system that is less reliable than the fossil fuel generation system we had in place up to 2010.
And does it also take into account the shorter lives of these renewables which means we would need to be spending a significant portion of that capital outlay again every 10-15 years, in addition to having to transport and dispose of the wind and solar components?
It isn’t a particularly large amount of money, replacing the coal we had in 2010 almost all of which will have to be replaced by 2035 would cost about $100-$120bn in today’s money not including new mines or wate supplies or coal rail tracks. Combined with the operating costs at today’s coal prices would mean that the plants have a breakeven of $120-180/MWh. Firmed wind and solar can comfortably make money at $100 without suffering 2GW trips.
Peter Farley, you’ve done it again!
“Firmed wind and solar can comfortably make money at $100 without suffering 2GW trips.”
That’s right, once we have 10,000 km of new transmission lines reaching the REZs, we can enjoy $100 energy! I’m sure that transmission lines will be both dirt cheap and timely. That maths definitely adds up.
And I’m sure that this high quality transmission line will not increase our chances for transmission line failures the next time there is a freak storm.
I also like the way you implicitly blame Loy Yang for what happened on the 13th 😉
Hi John, Great article.
I note elsewhere on WattClarity they mention: ” Sydenham (SYTS) No. 1 and 2 500 kilovolt (kV) lines tripped following failure of six 500 kV towers (three on each of the two 500 kV circuits).”
I would be interested to know how far apart these groups of towers were, and whether the collapse occurred within a few minutes of each other.
Peak winds in these microscale events are usually very small, temporally and spatially.
I would have thought two lines should be enough redundancy. Is it necessary to increase the separation between lines?
All solutions sound expensive!!
With regard to your highlight 2) of Dundonnell under Event2, the fist paragraph of Section 1 Overview of the report contains the sentence “Initial review indicates Dundonnell WF tripped as designed due to operation of the South West 500 kV special control scheme.” By my reading the reference to the special control scheme does not occur elsewhere in the report.
Relating to the loss of Stockland Hill, there appear to have been fire service notices issued about a fire in that vicinity.
There are five big variables which are difficult to quantify which will mean that the transition is different and almost certainly quicker and cheaper than the above model suggests.
1.How much of the current “offpeak” demand which keeps coal plants going at night will be shifted to daytime “solar soaker” rates
2. What is the optimum expenditure on existing hydro eg the upgrade of Clover could be just the start of hydro upgrades.
3. Will the perception of grid unreliability drive overinvestment in behind the meter generation and storage, which will supply 20%+ of total demand and clip peak demand by 2-5 GW
4. Will detailed economic analysis drive developers toward different configurations of wind and solar farms optimized for output in suboptimal conditions rather than just adding storage to conventional designs?
5. Will the TSO’s suddenly discover dynamic rating, reconductoring and storage as transmission to speed up transmission capacity increases
My understanding is that Stockyard Hill WF was asked to turn off to enable the fighting of a large grassfire in the area which was closer to the turbines itself.
Thanks for the comment. I don’t know the details of last week’s failures at Anakie and they have not been made public as far as I am aware. However, in previous events, several adjacent towers have failed as a group.
On your second point, the storms are not that small. A ‘footprint’ of peak winds in a ‘typical’ convective downdraft has rough dimensions of 10km along-wind x 3km crosswind. Those dimensions would be large enough to cover two parallel lines, unless they are separated by several kilometres.
Furthermore there may be several storm ‘cells’ – e.g. in SA on 28/9/16, there were at least three separate cells resulting in 22 towers failing or with severe damage.
The AEMO report suggests load shedding in Melbourne was to manage system constraints, mentioning thermal overload on a 220kV line from Geelong to Moorabool terminal station taking power from generation in Western Victoria to loads Melbourne. This situation complicates management of dispatch.
APD is possibly the participant referred to in section 7 of the preliminary report
“To manage power system security for credible contingency events, AEMO issued a direction to a participant to maintain power system security.”
since “a credible contingency event – the loss of the MLTS A1 500/220 kV transformer – would have resulted in the synchronous separation of South Australia and the APD smelter from the NEM”
@Peter Farley
1) None. With all the GIGAWATTS of capacity already in the NEM that your lobby promotes, you’re not making a dent in off-peak demand. Battery ENERGY is almost negligible when compared to coal plants.
2) Optimum expenditure? How about we use Snowy 2.0 as a proxy and go with $10 billion and a 10 year lead time.
3) Perception of grid unreliability? I guess what happened in 2022 was just a fantasy of mine. Clip peak demand by 2-5GW? You’re literally just making up numbers now.
4) LOL!!! All the best wind and solar regions are already taken. There’s nothing you can do about it, pal. That’s why your lobby wants the heavy 10,000km transmission line project subsidised.
5) “Suddenly discover”? That sounds like very sound reasoning to scrap all coal. Hopefully we’ll suddenly increase our transmission!
AEMO directed Mortlake to run from later on Tuesday afternoon. This would have been to manage the risk to Portland Smelter in the event of synchronous separation of the south west 500kV network from the rest of Victoria. As we saw in February 2020, Portland can’t be securely supplied from South Australia in those circumstances.
I was interested to read Paul Simshauser’s research about dynamic transmission constraints. The main thing I took from this, and I probably took the wrong thing, was that if the wind is blowing then more transmission on a line is possible because by definition the line will be getting wind cooling.
I guess the transmission constraints through the NEM don’t generally reflect dynamic max load as opposed to hot day static max load?
It’s interesting to me that because wind and solar receive an AGC set point developed from a wind / solar forecasting model, that any variation in weather conditions (sudden strong winds, calm spells, clouds) cause forecasting errors, which require FCAS to correct, which increases cost and risk.
I will certainly be interested in seeing how this changes in future iterations with transmission requirements modelled – as you mention, the results will likely be very different.
I would note that if you are designing a system like the one your model has chosen, where there are subregions that are substantially underserved with local generation and require near-permanent access to generation that is physically remote, that leads to critical transmission infrastructure that has to be substantially overbuilt, to allow for some transmission paths to be out of service for scheduled maintenance while still remaining resilient to events that trip some of the remaining paths.
To incorporate transmission I think the offshore region(s) will need a definite location as well.
It would also be interesting to start the VRE portfolio with existing/committed VRE projects as well rather than completely ‘from scratch’. This may not change much, of course.
I thought this was interesting on the LOR3 condition (from section 5 of the preliminary report):
There was enough generation in the Victoria region to meet demand, however, due to constraints on the network,
this generation was not able to supply the load. At 1420 hrs on 13 February 2024, an actual LOR3 condition was
declared and AEMO issued a direction under section 116 of the National Electricity Law (NEL) (deemed a clause
4.8.9 instruction under NER 4.8.9(a1)(2)) to load shed 300 MW of load in the KTS/TTS/ROTS metropolitan area to
maintain the system in a secure operating state.
There was some information on this in section 2.3 of AEMO’s preliminary report into Feb 13:
Prior to the event, AEMO was managing a sustained low frequency event including the loss of approximately
500 MW of generation from Stockyard Hill WF that occurred at approximately 1135 hrs and had increased the
minimum frequency control ancillary service (FCAS) regulation raise requirement. Constraint set
F-MAIN_RREG_0450 was invoked between 1200 hrs and 1355 hrs on 13 February 2024 to ensure mainland
regulation raise requirement was greater than 450 MW.
One thing to keep in mind for bids in the second half of the period at MPC is that from 1435, AEMO applied a manual override to set the price at MPC. That has to affect bidding behaviour.
Tristan,
You discussed curtailment but did not take into account the period of Dunkelflaute.
Batteries will not provide power for periods of weeks or months. It requires pumped hydro. Even Snowy 2.0 will only provide 2GW and then only for a week.
Boston 2022* paper showed that during Winter 2010 there would have been significant blackouts, even with 10xSnowy2.0 for storage.
It also showed massive curtailment.
Other analyses I have seen like this start in July 2010 or 2011, but this paper shows peak plants (or massive batteries) could be on standby for 10-15 years.
Both Queensland and NSW look like they’re heading towards very high levels of operational demand this evening, with levels at 3pm around where they were at 5pm yesterday (NEM times), and yesterday was a fairly high demand day.
The reasons for this are very clear in the market data – the output drops were Moorabool WF being constrained off by transmission limitations around Moorabool / Geelong, which preceded the 500 kV line trips (and worsened thereafter). The availability of the plant remained high until about 3pm, but in any case it was mostly constrained off from 12:40 NEM time onwards
From the market data, Berrybank 2’s ups and downs are clearly constraint driven (like Moorabool WF), although it was persistently below target output (by up to 10-15%) between 11:00 and 13:00.
On the other hand Berrybank 1’s availability dropped to low levels then to zero in the half hour following the main event and stayed there for all but a couple of DIs until about 20:00 when it recovered.
Interesting article — I’m a long-retired EE who worked in and around generating plants for 33y. Five of those years were in fossil and nuclear plants with 11 units. It never occurred to me to ask what the time interval was between the unit trip and the generator trip, nor do I recall ever seeing any records that would provide me with that information. Thanks for answering the question I never thought to ask. (USA)
Perhaps the Moorabool Wind Farm (and the Berrybank WFs) tripped on excessive wind speed at its location. This wind storm subsequently hit the 500kV towers causing their demise.
This will be interesting when subsidies for rooftop solar are removed the amount of systems will reduce over time as they start to fail creating more demand for the house hold that didn’t consume and for not being able to supply the network what it needs
Using AEMO’s own Operational Demand and Rooftop PV tables gets a value of 15,088 MW for the half-hour ending 15:00, close enough to your 15,100 MW Paul. David Osmond’s figure of 15,400 MW from OpenNEM seems very high, perhaps it’s based on 5 minute data and a different PV source (eg APVI).
I made some fairly detailed comments on a LinkedIn post by Geoff Eldridge of Global Power Energy that went into the basis of AEMO’s Underlying Demand figures and how they might differ from others’.
And that perhaps only the worst performing / highest cost to maintain unit at Eraring will switch off ‘some time’ after 18 Aug 2025, rather than all four being switched off on 18 Aug 2025, i.e. Eraring will stage the closure of it’s four units from August 2025, over perhaps even years.
Is there a tool where capacity factors for individual years or overall for all wind farms and individual wind farms that could produce a lust of the selected criteria?
Dan – Great graph.
So we need to double solar/wind in Apr and May, BUT the network curtailment will be an issue all months AND the economic curtailment will more than double in September to December. Or it’s not that simple?
Considering “operational demand” and “underlying demand, what is the data at “Annual generation capacity and peak demand – NEM” showing us? 2024 peak is looking low in this data which agrees with the analysis presented here. From that data, the peak demand was in 2010-2011, it is difficultto tell by looking so i downloaded the csv data to Excel. 2019-20 came close but fell short of 2010-11 by a considerable amount. Depending on what the overall peak is for 2024, I would say there is a downward trend in this data, flat at best.
There are other charts too. For example, “Annual electricity consumption – NEM” which paints a similar picture, and this one shows a significant downward trend from 2010.
These are all nice and pretty graphs. The real question should be “What is happening with the NEM to cause prices to rise at close to 100% over a period of flat or falling demand?”
But a major factor in the long term increase in NEM prices (from the 1999-2012 period) is the increasing linkage and exposure of generation costs – specifically fossil fuels – to international energy prices. Another factor has been the retirement of ageing coal-fired capacity. Replacing old, fully depreciated capacity with new generation of any form is generally going to require higher prices to justify the new investment.
Assuming we get annual cycling equivalent to two full depth cycles, 120 half depth cycles and 200 at quarter depth mostly in the mornings, that means that simplisticly twenty-two GWh of storage would eliminate the problem.
Now let’s say 10 GW/18 GWh of batteries were available as well as 4.5 GW/400 GWh of new pumped hydro, we should be able to eliminate at least 75% of the curtailment. That would incidentally increase renewable share from 38.7% to 40.7% plus whatever rooftop solar curtailment is avoided.
Very interesting to see how the calcs would fall out if solar capacity was tripled and wind doubled
Did SA batteries respond to the sudden 348 MW ramp over 4 dispatch intervals? Because if there was a “possible” LOR3 event later that afternoon then a battery with a Capacity Investment Scheme Agreement would in the future have to have some of their capacity reserved for that possible LOR3 event.
Dan,
Thanks – another thought-provoking article. It was a difference world 7 years ago, pretty much halfway through it’s design life and 100MWh for $56-90m.
How much have the cost of grid scale batteries dropped since then?
Interesting to compare these to the Kurri Kurri peaking station. It will have taken 30 months IF it is completed by the new scheduled date, December this year. It will have cost $1billion but will produce the same energy capacity as all current 20 batteries in 3.5 hours. (2.4GWh)!!
Project construction time is highly dependent on project size, which should explain some of the spread in the FID-to-COD timeline between projects. Also, a lot of the projects with FID in 2021-22 were probably affected by supply chain issues which definitely led to significant delays. It would be helpful to include the project size in the table above
Hi Craig, for the article I used my own definitions for each of the phases which I stated in the article.
The fire at VIC Big Battery was in Sep 2021 (which I believe happened during testing). In any case, by my definition it was fully operational in Nov 2021.
Thanks for the discussion Ben particulaly around the assessment process. Although a bond of a “few hundred thousand dollars” is not to be sneered at its tiny relative to capital cost of a project and, in my view, nowhere to provide confidence that projects winning contracts will actually be built. Building confidence that projects actually will be built is in a way the entire point of the scheme.
I also wonder whether greater storage depth really is more valuable. Would you rather have 1000 MW of 1 hour storage or 1 MW of 1000 hours storage?
Thanks David. I thought the same re bond. Indeed just the cost of preparing a bid would run to seven figures.
I think the modelling approach proposed sounds quite good for valuing the depth of storage to meet the needs of the NEM, certainly much much better than just dividing by 4 hours in the RCM, which I have heard (but struggle to believe) is also what is used in the much bigger capacity markets of the US.
Your concluding question may be even more rhetorical than you intended, consider this: Battery A can do both roles as the system requires it, but not vice versa.
Excellent insight from WattClarity as always.
One note is that the Vic-SA tender did not apply the four-hour minimum. Whilst present in the consultation draft, it was amended to 2 hours as the tender opened.
Would battery discharge by VBB be able to provide the increased import limits on VNI under SIPS when the Moorabool – Sydenham 500 kV lines were lost?
You refer to the V>>NIL_MLGT_MLGT_R2 and V>>NIL_MLGT_MLGT_R5 constraint equations. The names of the constraints suggest that they relate to the Moorabool – Geelong 220 kV circuits, which could overload during the 500 kV outage.
Further reading for the technically inclined. This is an open access paper.
Title: Modeling geomagnetic induced currents in Australian power networks
Journal: Space Weather Journal
Issue: 10.1002/2017SW001613
Publisher: AGU Publications
Great summary Ben. It seems the Federal Government were left with little choice but “to do something” to drive to 82% by 2030. Given this reality, the CIS is the best of the options on the table and as you say at least preserves much of the wholesale market functionality.
No-one actually knows what the safe level of inertia is, but in an inverter based system with distributed hierarchical control, the safe level may well be 1/10th or less of what traditional systems considered necessary. Inertia is actually an instantaneous but very small energy store. Inertia does nothing unless there is a frequency change. If there is a frequency change of say 0.3Hz, the same as occurred when LYA tripped, the energy available to stabilise the NSW grid is (1-(49.7/50^2)x I = 1.1% of the inertia.
When inertia is 20,000 MW.s, that is 239 MWs available. If the frequency change happened over 3 seconds the available power is about 80 MW. i.e. a 100 MW battery could provide similar fast frequency services as 20,000 MWs of physical inertia.
A key question is how to ensure all batteries don’t inject 80% of their available power at the same time otherwise there would be serious overshoot and potentially cascading instability. This is not a risk with physical inertia, but on the other hand batteries can arrest a far larger frequency excursion than physical inertia can.
In addition there is synthetic or virtual inertia on wind and solar farms which is also potentially significantly larger and more controllable than synchronous inertialarger
Great article! I see batteries are not providing intertia before an event however the FFR would provide a level of effective inertia. The question is how AEMO plans to consider this.
Wouldn’t like to unnecessarily speculate but it is quite usual for pairs of adjacent units to share a certain amount of “common” plant and /or interconnections intended to provide reliability and the ability to maintain both units in service whilst minor maintenance activities are undertaken. Given it was simultaneous trip (and the unload seems to be vertical so either an electrical protection operation or turbine trip rather than combustion where the unloading rate would be slower) it looks like it is a secondary system of some sort. Recovery was good. Whenever these Toshiba units- or indeed most large units – trip from high load (normal metal temperatures) they need to cool for a bit before steam can be re-admitted (so called mismatch temperature). So it appears there was nothing really to “fix”. Just sort it out and get the units back on. It does happen, fortunately very rarely, but is the joy of working at the pointy end of a production operation. Something most commentators will never experience let alone understand.
It may be coincidence, but the increase in energy arbitrage revenue seems to roughly correspond to the introduction of 5 minute settlement in October 2021.
Hi, I’m looking for information regarding the turbine availability (no. of turbine available for power production) provided by the Wind farm owners to AWEFS. As far as I have searched this information is not publicly available for research purposes. Am I right? If so, are there any proxy for the same?
Thank You.
2/ The part owner of C4 has a court order for a new enquiry
3/ The Australian Energy Regulator is taking legal action
4/ A industry standard Castell interlock System had not been fitted to the unit breaker to stop the isolation of the 220 VDC control supply when the C4 unit was in service.
5/ The original forensic enquiry by a structure engineer is on going
Uhh? Does anyone have a frequency plot for Tassie for event 1? A near 50% system load rejection is pretty huge. I assume the load was shaken off due to a severe voltage depression from switching onto a fault? Generation did not run back by 500MW so assume Basslink imports stopped. Dare I say a zone two distance protection operation from the opposite end so a slow clearance? What happened at 1.39pm? Generation went down so did this imply Basslink started to import again? Would be interesting to pull this one apart a bit more if someone has the operational data.
“so in the hypothetical future grid that’s heavy on wind and storage, such as we analysed in Appendix 27 within GenInsights21, you’d at least have had some forewarning to stock away what energy reserves you had available to be drawn down through this period”
Sadly the amount of storage you would need to cover even a whole day of low wind is astronomical in cost, and if you can’t recharge it because of low wind then each successive day it gets harder and harder to keep the lights on.
Great article Paul:
Based on modelling (Boston 2022* – Figure 1) the wind drought of June 2017 was small compared to 2006,2008 and particularly 2010.
There is now good hindcast ERA5 wind data for 40 to 50 years, so updating this paper would make sense. ERA6 will be available next year.
Climate change may also decrease wind speed so the future requirements could be tested against the various RCP models.
Nothing to see here I suggest. Well described by David Leitch “Seasonality and deep storage in a highly decarbonized NEM” published on your web site Sept 10 2020, and seasonally highly predictable. Have a look at some of the modelling work done by Geoff Bongers et al from Gamma Energy on the NEM. Basically says the same thing. You cannot sensibly afford enough storage to get through these doldrum periods. Perhaps you could get Messer’s Bowen, Albanese and Bandt to start reading some of this material. They might learn something!
April to June is a common time for slow moving highs to sit over SE Aus bringing very stable (and rather pleasant) weather. And it is astonishing how widespread such conditions are: from the Eyre Peninsula to north NSW and down to Tas. Also affects offshore wind around the 4 SE states. Geographic diversity of assets within this huge area doesn’t help.
You will however find solar was somewhat increased during this period compared to the week before, which is to be expected in this weather pattern. A silver lining to the non existant cloud, at least in daylight hours.
Fantastic article, Ben. Could I please query one detail though. If the interim unserved energy reliability standard is 0.006% and the old was 0.002%, then the interim would be less conservative than the old. But, I think it’s simply a typo, and the interim standard is 0.0006%.
Frustrating to hear that a Rule change can’t repeal this because it’s in the NEL. Hopefully some traction can be gained with customer representatives.
> In the wake of South Australia’s devastating 2016 grid outage, authorities moved to boost grid resilience in this way, building big batteries, more renewables and new interconnectors, while Australia’s energy market operator AEMO changed how it dealt with windfarms if grid issues occur.
None of these proposed solutions would have helped what took place that day. This is just another article pushing renewables at any cost.
> Big batteries: the maximum dispatch of big batteries in Victoria during the load-shedding period was 20 MW! The two hours of $16.6k spot prices had an average of 5 MW battery dispatch! That is quite literally nothing.
> More renewables: there are already plenty of renewables in Victoria. Stockyard (500 MW) tripped due to an unrelated issue a couple hours before the transmission towers fell. 1 GW of other wind generators tripped due to the transmission towers falling. How much more wind over-builds will be satisfactory for you?
> New interconnectors: an extremely expensive proposal with very long project lead times.
I also love the sophistication of your climate analysis. “We’ll have warmer air in the future which means we’ll have more moisture which’ll probably mean we’ll have more intense thunderstorms.” Top notch analysis!
Now a tower lattice truss could well be designed to handle extreme wind all over it but maybe not a wind driven object seriously impacting a single member. Buckle that link and you introduce a chain reaction among load members bringing down that tower and similarly more interconnected ones with the resultant shock loading.
Some interesting analysis.
But worth emphasising that our existing hydro assets can dramatically reduce seasonal storage requirements.
A mostly wind & solar grid will have large deficits in winter (& large surpluses for Sept-Mar).
A lot of this winter deficit can be filled by existing hydro.
An average, NEM hydro discharges 1,500 GWh more in winter than summer. That immediately reduces seasonal storage requirements by that amount. But inflows are even more biased to winter, 6 TWh vs 2 TWh. So by simply restoring hydro discharge to be more seasonally similar to natural inflows, then we can reduce the winter deficit by 4 TWh.
During the years of the carbon tax, hydro discharged 6.5TWh more than inflows to take advantage of the temporary high prices, making it clear that our hydro assets are able to substantially shift generation from month to month & year to year to maximise profits.
Also worth mentioning that the Blowering & Hume reserviours at or near the bottom of the Snowy scheme largely decouple irrigation water requirements from Snowy hydro generation.
This is not a new phenomenon. There is a 220kV single circuit line in WA that has been screwed up and thrown on the ground by thunderstorms at least 3 times in 35 years that I know of. And this is the point. Microbursts are not dissimilar to tornados involving high velocity swirling winds. Towers are designed for dominant lateral and axial loads. Tortional loads from swirling winds tend to screw them off at any “skinny” point in the structure – a swirl does exactly this pushing wind loads on the conductors one way on one side of a tower and the other way on the other side. It would simply be aesthetically and economically prohibitive to attempt to design for these extremes. What it does challenge however is the validity of considering dual circuit towers as genuinely N-1 designs and indeed in this most recent Victorian case, putting single circuit lines immediately adjacent to each other in the one easement. Microbursts are usually very localised events touching down and following a bit of a path and then disappearing. Fully acknowledging the difficulty of getting line easements these days, but a few km of separation for critical network interconnections is a very good idea if you want to chase reliability.
Hey Paul!
We see the future contracts are trending up significantly after the report came out, it’s quite contradictory that the report is indicating lower system price, but the market is reacting in another way. Any thoughts?
Hi Paul/team
Interested in your BoA preso, is this available to view (clicked on it and just the header slide opened)?
Fully understand if just for clients.
Many thanks,
Graeme
There are many challenges ahead. Have any of these predictions taken into account the steady rise in EV car charging during the night when the system will only be able to use gas for the expected “firming ” periods?
Fascinating article, thanks. As a retired engineer, now an outsider to the industry, I very much appreciate this free (to me) trusted communication.
I note the increasing scale of BTM energy, combined with talk of future industrial BTM generation for larger users. Does this tend to move supply from the green zone to “off the picture”? Maybe plot the estimated BTM generation as energy on the first chart, though that is already cluttered.
In the précis above, it is interesting to note the Abstract referring to “low-cost intermittent wind and solar resources”.
This “low-cost” claim is highly disputable. For instance, at the outset, the cost of a renewables based system must include the cost of all the required firming or back-up needed to make such a system work. In effect this amounts to building two independent generating systems – each capable of maintaining the system when the other isn’t.
Furthermore the renewables option must include the full cost of building and maintaining a vast web of new transmission lines needed for the express purpose of collecting the outputs from a huge number of generating units.
Note that these new lines collect electricity rather than distribute it, and they also inefficiently lay idle for much of the time.
Then there is the reality that short life renewables require perpetual rebuilding at unknown expense. Even the replacement of thousands upon thousands of worn out turbine blades presents as a huge expense item.
The gas based back-up generation system will produce a high level of emissions, perhaps far greater than anticipated.
However, the worst aspect of this discussion is the absence of any reference to the emissions free nuclear option.
A large, inflexible coal fleet in the presence of rapidly rising intermittent renewables can be
expected to confront a rising number of negative price events (Nelson et al., 2022)… prices fall to negative levels with solar plant capable of producing to the negative of ‘renewable certificate prices’, and rooftop solar exports being indifferent to wholesale clearing prices.”
The model seems to include an assumption that REC’s will be with us for ever, along with increasing negative price intervals and cost pressure on retail tariffs. Since VRE is often touted as the cheapest generation, why not do away with REC’s immediately, if not sooner? That would reduce market volatility and energy cost to the often poorest members of society customers.
Just to be clear, your definitions of network curtailment factor and economic curtailment (offloading) factor use maximum capacity in the denominator, in simila fashion to capacity factor?
That’s absolutely fine, but every other presentation of these curtailment or offloading metrics that I’ve seen (eg AEMO QED and ISP, Energy Co REZ stuff, my own posts) uses available energy as the denominator – because this basis translates directly to the percentage reduction in potential output experienced. Numerically, using available energy in the denominator yields curtailment / offloading factors which are three to four times larger than the values in the Global-Roam results based on maximum capacity.
It would be worth highlighting this difference for readers, since most will be used to seeing the higher values derived as percentages of available energy.
Thanks Allan. Yes correct, this was done so that I could display the data as a 100% stacked area graph along with capacity factor (a metric that a wider audience would be familiar with), but I completely agree with your point and don’t want to confuse readers who might have thought that those numbers are artificially low – so have added in a note to make this clearer.
This is what happened at Callide C4. They tried to re-invent the wheel! It was ignorance. Just remember ignorance is just the lack of knowledge. Large 220 VDC Batteries are not designed to be connected together under loaded conditions because of the large fault currents produced when used for large critical loads in a power station. We are lucky it was only a coal fired power station
Indeed, the modification of the Callide C4 design, which included the removal of the Castell key interlocks, unfortunately led to a catastrophic failure. The attempt to improve the system’s redundancy and reliability inadvertently resulted in the complete destruction of the steam turbine, incurring significant costs.
The incident underscores the importance of thorough risk assessment and testing when implementing design changes, especially in critical infrastructure like power plants. It’s a stark reminder that while innovation and improvement are necessary, they must be approached with caution to ensure safety and reliability are not compromised.
The lessons learned from the Callide C4 incident will undoubtedly contribute to safer and more reliable design practices in the future. It’s a costly but valuable lesson on the potential risks of modifying established systems.
The Castell interlocking ensured that that the critical circuit breaker which needed 220 VDC to operate was locked open before doing any switching on the critical 220 VDC supplied from the unit 220 V battery, the unit should have been out of service. C3 must have been out of service when it’s 220 V battery was replaced
I don’t think we can assume that the shrinking quantity of bids in the 0-$300/MWh range necessarily increases reliability risks in the NEM. It would do so under the condition that “everything else remains equal” but this isn’t the case here.
There is a reducing quantity of generation priced in the market according to the variable cost of fuel in the price range 0-$300/MWh and this has occurred because we have added VRE priced at zero (or lower) and storage priced according to opportunity cost. These new resources tend to hold a good share of their bid quantities outside the normal price range and yet prices still tend to land within it. If anything, prices outside this normal range are much more likely to be below than above it.
More research is needed to understand bidding dynamics but outcomes suggest bid quantities are being shifted between bid prices above $300/MWh and below zero to achieve this outcome. If rebidding is occurring frequently (every five min for some resources) then even sudden and unexpected shocks might be corrected rapidly (ie in the next 5-min interval).
Apologies if what I wrote was not clear, but I was not focusing (just) on ‘reliability risk’ when writing this article.
The much broader pattern of prices in the NEM these days (which was happening anyway, but was also bumped along by 5MS) leads into different types of risk – including commercial risks for a range of different participant.
You mention storage in particular, but that’s only a small percentage of the volume of energy offered in bidding. There are also other reasons why bid prices have changed for a variety of different assets.
It seems that the answer would be some storage, perhaps 15-20% of windfarm peak capacity for two to three hours on the upstream side of the transmission constraint and a similar amount of storage near the loads. The load end storage limits peak demand on the transmission and the generator end limits both economic and transmission based curtailment
If I participate in the FCAS service, but happen to encounter a situation where I cannot provide the quantity as bid, will there be a fine? Is there any documentation on the penalty regulations? Thank you.
If I participate in the FCAS service, but happen to encounter a situation where I cannot provide the quantity as bid, will there be a fine? Is there any documentation on the penalty regulations? Thank you.
This statement is made in the concluding remarks section: “Detailed analysis of demand side reductions and the impact of electrification was not considered in this study.” Given the potential scale of household batteries and other demand side resources, I’d be keen to understand their impact on the modelling outcomes.
I believe the context here is gas network demand-side reductions (hence the mention of electrification, which will reduce domgas demand). It’s clear from this that heavy reliance on gas peakers is going to require either an increase in the storage capacity of the gas network, gas demand-side reductions, or more probably both in some measure.
To my mind, this should add some urgency to the gas-to-electricity household transition.
Wind was lower than often the case in the morning, but not the afternoon – when the longest peak prices were sustained. Only VP was substantially below. Eraring was not substantially below, it was half mast when 2/3rds mast on average.
Eraring might “normally” be at “2/3rds mast on average” Bruce but this was very far indeed from being an average situation. Prices were in five digit territory and in those circumstances Eraring would almost certainly have been running near full tilt if it had all four units available – but it had only two (again, not a “normal” state of affairs).
well I am still of the view that if we got on with it, it would be less of the problem. What ever the new supply source is I should have thought there was little doubt than unless demand is falling you can’t close old supply before you build new supply. The fact is that NSW has been ridiculously slow at getting new supply built given the age of the coal generation, the propensity of the coal supply to be throttled by flooding or competitive export parity pricing and the every increasing forced outage rate that requires $100s of m of maintenance capex every year.
It remains the case that getting new transmission built and more projects through the planning process is the only realistic and achievable answer. And it is realistic and achievable.
Hi Bruce and Allan,
I think it’s worth taking note of the amount of transmission line outages at the time of the price spikes. Appears there were 3 x 330kV lines out. Wollar West – Wellington, Armidale – Tamworth and Collector – Yass. These outages were causing constraints on generation and interconnector flow too. I think they were all planned outages but some work may have taken longer to complete, hence the overlap.
Good points from both David and Trevor (and thanks Rob); those transmission constraints were definitely a very significant factor, and of course sufficient new supply – both energy and capacity (ie fully dispatchable MW) – should be built before existing stuff gets shut down. It’s also worth noting that there was plenty of undispatched coal-fired and gas-fired capacity available at the times in question, it was simply offered at very high prices.
I think Bruce’s main point (and David’s) about the growing gap between the rate at which the ISP assumes new stuff can get built versus the build rate we are actually seeing is the critical thing here.
Not quite sure I understand the logic here. The so-called network curtailment is almost irrelevant. It was never 100% of the wind farm capacity. You can only curtail something if it was capable of generating anyhow and none of the “Southern Tablelands were producing anything. If there was any wind they would look like (say) Bango Wind Farm #2. Touching a transient stability limit probably just implies the transmission network was getting full with Thermal and Hydro MW to keep the lights on and would no longer be N-1 fault tolerant for significantly more coming from this region.
Hi Andrew, the second chart is meant to show that the wind conditions for the wind farms in the Southern half of the state were still generally better than for those in the North-East – but NEMDE was constraining those wind farms down (to 0MW except for Bango Wind Farm 2) largely because of the N::N_CTYS_2 constraint equation.
As mentioned in the article, Green + Red = Availability and Grey = Unavailability.
ID fans are downstream and maintain a small negative pressure in the boiler, precipitators/baghouses to keep things inside. FD fans provide the bulk of the mass of hot (ish) combustion air to the furnace to ensure complete combustion occurs. Generally, controls to around 3% residual oxygen in flue gas stream to prevent CO formation. PA fans provide hot air to the mills to dry pulverized coal and transport the milled coal from the mills to the burners. No PA fans = no fuel.
Perhaps the more correct headline for the SMH and the Age might have been “Origin Energy throws NSW a Lifeline” ? So aside from all of the renewables running late – assuming 2 years is enough to get them back on track, who is building the 1500MW of OCGTs to keep the lights on during the cloudy, windless weeks in the future?
I would be curious to know how the aggregate wind yield compares to the aggregate energy consumed in restarting the turbines during these periods and what the ‘net’ might be (if I understand the operation of these machines correctly)?
This particularly long period(s) of low wind output is significantly increasing the wholesale prices in Victoria, SA and Tasmania. It has been compounded by lower hydro output in Tasmania due to below average rainfall over the hydro catchments for around six months. A further compounding factor is the lack of solar in Victoria, both utility and behind the meter. While relative solar output at this time of the year is lower, Victoria has disproportionally lower utility and rooftop solar compared to SA, NSW and QLD. Additional solar output during these low wind events would significantly lower prices during at least some of the days.
assessing the wisdom of recent agreement re NSW and Origin requires an analysis of what the renewable energy contributions actually are, and how much projected input is likely to be realised, what other measures are available to the Gov to secure renewable energy ( including regulation and legislatory measures regarding investment, taxation etc) and how the grid itself can manage energy collection and distribution. I am keen to to see all that put together.
Given the prices prevailing I can’t see too much economic curtailment happening. Good coin to be made. Bottom line is the wind simply isn’t there. So how about someone doing a quick integration under the curve to work out how much wind overbuild, solar and storage (GWhr + MW)) is needed to keep the lights on for the last few weeks once all the nasty black and brown curves aren’t there….and then do a quick calculation of the costs. The number will be eyewatering. Then perhaps a visit to Canberra.
NSW has gone out to tender for 1 GW of LONG duration energy storage – in NSW the term long duration can be as short as 8 hours. What energy source will charge the 1 GW of long duration storage when wind droughts go for days or weeks and coincide with low solar output?? Its not just a case of the price to charge, its whether there is energy to charge. Sadly, the general public think batteries and pumped hydro are a like for like replacement for coal and gas generation. A position the general media is only too happy to promote.
It wasn’t just low on the Friday. It was low for the whole week, in fact, the lowest it has been in five and a half years.
I have now kept detailed daily and weekly records for wind generation for the last 295 weeks, more than five and a half years, and this most recent week ending yesterday, 26May was the lowest power generation across a whole week that there has ever been. Because that Nameplate increases regularly, then it stands to reason that delivered energy (in GWH) to the AEMO grid increases. So, the only comparison we have over time (taking that Nameplate increase into account, now an increase of 6108MW from my start point of 5301MW back in October 2018) is that all important indicator of Capacity Factor, (CF) where you actually can compare week to week.
That CF for the last week was just 10.33%, the lowest that figure has ever been for a full seven day week. The long term average CF for all wind generation here in Australia is just 30%, so when it is this low, then questions need to be asked. Only twice in those last 295 weeks has the weekly CF been below 20%, and the lowest prior to this week was 17%, and that was four years ago now.
Capacity Factor has been falling consistently every week now for more than 30 weeks, a loss of half a percentage point, and that’s pretty big considering that the full maths calculations are across 295 weeks. 30 weeks back, it was 30.41%, and now it is 29.90%, and the last time it was lower than 30% was also that four years ago. Perhaps the more worrying trend in all of this is the most recent 52 week CF percentage. As more efficient and larger turbines come on line, then you might think that they would tend to raise the CF figure. However, the exact opposite is happening, with that recent CF percentage for the last year now coming in at 28.7%, a full 1.2% LOWER than the overall long term percentage.
And hey, has nobody put together the fact that wind is low when those huge High Pressure weather systems hover over that area where two thirds of all Australian Nameplate for wind power is located, in SE South Australia and Central Western Victoria. Have a look at power generation for wind this last week, and then correlate it with the Synoptic weather maps. that looks a little like an ‘Uh Oh’ moment to me. Why place so much wind in that one area, and then watch as the bottom falls out of power generation with every ‘High’ that comes across.
Anton Lang.
I expect this will be short lived as a blocking high is forecast to sit over Australia from Thursday night so Friday and saturday should see the wind die down again.
Economiser leaks are lots of fun! By the time they show up as wet ash in the ash hoppers the high pressure feedwater has usually cut adjacent (finned) tubes. If you don’t spot it early and shut down then the cut tubes fail and the leak gets really big, really quickly seen as a big feed flow / steam flow split. Hopefully it is an accessible spot (economisers are usually jam packed, multi layer locations) or it will be cut your way in and weld your way out. Somewhere there will be a boiler Engineer about to have a very busy weekend sorting this out.
There is of course also some “transmission” component to the economic curtailment – specifically the limited interconnector capacity between regions. Eg if exports from SA to VIC are at the dynamic maximum capacity of the VIC-SA interconnector then SA will be economically islanded, resulting in apparent “economic curtailment” of generation that could other be dispatched to meet some out-of-region demand.
General scale predictions of wind are fine as can be seen from my comment on your article from Wednesday 29 May now being realised. Just this morning we have already seen 3GW of wind capacity reduction between 8am and 11am.
The reality is that wind and solar are unreliable. You can’t have a system where you just randomly start producing 3 Hazelwood equivalents of power and then 2 days later it all just dissapears.
We are on the precipice of collapse of the NEM because of how volatile the energy generation mix is so everyone is producing to ensure the lights stay on, which drives up wholesale costs which flow o to retail costs.
Halleluiah! We are now starting to get on the same page. The bottom line is you simply cannot expect to get base load energy generation out of inherently intermittent, essentially unpredictable, sources of energy without simply gross overbuilds of capacity and storage that will have to be paid for by someone. I suggest people go and read carefully the work done by Gamma Energy Technology that displays the results as exponential cost of decarbonisation for the NEM. (the reason any of this is being undertaken, remember people!). This is why simply comparing LCOE between technologies is simply silly. You need to look at the total system cost of energy supply to get the correct (least cost) mix of primary energy sources. Near zero carbon emissions simply becomes a constraint on that list of suitable technologies. Wind and solar certainly have their place and contribution, but the blind ideological belief that we can achieve (let alone afford) a 100% renewable power system is going to take us to a dark place.
Renewables are still cheaper when factoring in storage. The CSIRO report made this clear. In a 100% RE NEM, wind and solar are not the only energy sources. You are ignoring the not insignificant hydro component. There will be additional energy sources such as off shore wind (higher capacity factor than onshore), solar thermal (built in storage), possibly wave, geothermal and more.
Australia is not alone in experiencing lower wind output despite higher installed wind capacity.
From Reuters article 15 Feb 2024:
“Power generated by Texas wind farms dropped by 22% in January 2024 from the same month in 2023 as low wind speeds continue to stifle output across the main power system in Texas, the largest power market in the United States……
As wind power is the second largest source of electricity behind natural gas in Texas, the drop in wind output so far this year has forced utilities to sharply increase generation from fossil fuels to balance system needs.
Combined output from natural gas and coal was close to 50% greater in January 2024 than in January 2023, underscoring the enduring importance of fossil fuels within the ERCOT system despite the ongoing build-out of renewable generation capacity.”
Data from the Electric Reliability Council of Texas (ERCOT)
What comes first:
1./ The removal of the legislated ban on a nuclear industry, or
2./ A Royal Commission where all the financial aspects of nuclear fuelled electricity generation can be put and thoroughly tested against those of a renewables based system.
Australia is becoming bogged down in a highly politicised faux discussion wherein truth is the first casualty while elsewhere in the world,evidence that supports the nuclear option is becoming undeniable.
Each passing day sees huge torrents of taxpayer dollars flowing fourth.
Is it really flowing in the best direction?
In the 18:30 interval wind generation in SA was 13 MW (out of a max capacity of 2,732 MW)
So a capacity factor of just 0.4%.
This is why LCOE should include interconnectors as well because all the batteries in the NEM wouldn’t have been enough to keep the lights on in SA if they didn’t have fossil fuel generation.
Actually it’s more an illustration of why LCOE analysis of any kind is a complete waste of time if you’re trying to understand whole-of-system economics. Semi-arbitrarily adding various system-level costs – real or imaginary – into technology-specific cost measures is a fool’s errand.
CSIRO’s GenCost reports have done the industry a disservice by pretending that this approach is meaningful or even feasible. Don’t take this comment to mean that CSIRO’s data on underlying cost parameters is necessarily wrong, just that they have gone down an unhelpful path in then trying to shoehorn other costs into LCOE calculations, in response to the clamour from people who don’t understand how to do a system-level analysis.
Conversion of open cycle gas plants to combined cycle should be part of the Capacity Mechanism as it enables 50% more electrical energy from the same amount of gas. As batteries are displace open cycle gas peaking plants for the morning and evening peaks, the new role for gas will be for high efficiency gas plants operating continously for several days at a time during wind droughts. Their lower marginal cost will enable them to cost-effectively charge peaking batteries at night and during solar hours. As an example, Mortlake Power Station was designed for upgrading to combined cycle. It is one of Australia’s newest gas power stations, and has its own gas pipeline to the Iona storage facility, so is exempt from the pipeline constraints described in the paper. Conversion of more gas generators to combined cycle could well overcome nearly all the gas capacity contraints outlined in the paper.
There was an unusual drop in Pelican Point output from 490 MW at 1600 to 81 MW at 1620 before resuming full output around 1800. Bolivar, Snapper Pt and Hallett all responded immediately to the change.
Yep, PP clearly had a very significant wobble (technical term) as its output was running up towards full capacity and that together with the 50 MW import constraint across Heywood, which is here every day for the next week and bit, drove the $9.5k/MWh spike.
In your 15 March 2024 article “Need for speed: How long has each battery project in the NEM taken to deliver?” you recorded Tailem Bend Battery as being partial operational from 27 October 2023. To date I have not seen any activity in the AEMO data for the battery.
Hi Alex, I checked again and from what I can see in the AEMO data there was non-zero SCADA output for the generation-side DUID for that battery at 13:25 and again 16:30 on the 27th October 2023. This could have been very early testing or a SCADA anomaly as it doesn’t look like there was any other activity until last Tuesday the 3rd of June 2024 – where it looks like it output 42MW at around 18:00
CS ENERGY CALLIDE C4 https://www.csenergy.com.au/ArticleDocuments/251/CS-OHS-53%20-%20High%20Voltage%20Electrical%20Isolation%20and%20Access%2008%2023.pdf.aspx
4.1.1.4 is the way the CS Energy has ensured that they will not switch off the 220vdc again with unit in service
What has CS Energy changed with their 220v batteries and what happened to the castell interlocking key systems that ensure that the UCB is open and the unit is out of service. Where are the critical HV isolation instruction sheets. Who turned off the 220vdc? The C4 unit operator involved would have PSD poor guy. What has CS Energy done to him? He was left in the dark! CS-OHS-53 is available on Google. Snowy hydro HVSI is a good information source
Thanks for covering the Dunkelflaute Paul.
It is now common to see a loss of 6GW of generating capacity across the NEM in the space of 24 hours due to blocking high pressure systems.
This is going to become more and more of an issue as fossil fuel generation is wound down and can’t be relied upon to meet demand, especially in times of drought and/or night.
The rains are late and storages are closing in on the ‘prudent storage level’. Storage levels are the lowest at this time of year since at least 2018. Reference to the Tasmanian Economic Regulator’s website – link hopefully included. [https://www.economicregulator.tas.gov.au/about-us/energy-security-monitor-and-assessor/tasmanian-energy-security-monthly-dashboard]
Something very interesting technically going on here. Sounds like one are of Tassie is electrically very weak – not many synchronous machines running and a line fault (lots of reactive needed to clear it – Basslink won’t provide that) and it all got the wobbles. MW swings you don’t want. Sounds like some power system stabilisers not doing what they are supposed to (or not enough of them in service).
Hydrogen plant in SA ready by the end of next year! LOL…Seriously, AEMO? A new type of power plant is going to be up and running from scratch in 18 mths? has it even reached FID yet? Even the SA government’s own website says “early 2026” which is I suspect still very optimistic.
So let me get this right. We have wind at about 10% capacity factor for 12 hours, no solar (its night) and the nasty coal and gas have probably delivered 250GWhr overnight. So, let’s just assume we ended the 18th with full batteries – each with about 250MW of inverters and 2500MWhr of storage and we have 100 of them – the lights might still be on when the sun comes up…..and then we need to not only need to supply the higher daytime load but also have enough spare to recharge the batteries to do it all again. And this is before we try to “electrify everything” and as a minimum shift an increasing proportion of liquid transport fuel energy into electric energy stored in EV’s ready for the daily commute. Even if the 127,000MW of wind proposed (shown in the AEMO forecasts) gets built and it behaves like this we have spent a truckload of money and we still have a problem. I must be missing something again.
You’re right Paul that 30202 MW is a record for NEM-wide AM demand, at least in the last 15 years. But not far above the previous record from last year of 29996 MW on 21/6/2023 at 7:45. Andrew makes an interesting point in his comment about “how will this be done in future?” My view is that the “electrification of everything” will be part of the solution rather than part of the problem. If I’ve got an EV in the driveway with 50kWh in the tank, wouldn’t I be ready to sell some of that for of $0.70/kWh (=this morning’s spot price of $700/MWh)? Governments should be legislating a new FiT to motivate people to plug in their cars like we all plugged in our rooftop PV. They should not be spending a truckload of money to duplicate infrastructure that people will end up buying anyway (at least, a version with four wheels attached). Forget 100 grid-scale batteries, and imagine a couple of million cars.
We were extremely lucky that this occurred during daylight hours when solar production was able to offset the complete lack of wind in the NEM.
Having said that, these blocking highs do tend to occur on a somewhat regular basis (another one forecast for Saturday/Sunday) which is why the claim of just building more and dispersed renewable capacity is not a reasonable answer.
Related is that gas deliverability from Iona storage is itself dependent on storage volumes once the threshold limit is reached. Deliverability doesn’t simply stop when it’s empty but begins to decline well beforehand.
Plus the ability to refill storage is in all cases much slower than the ability to empty them. So a single day of low gas demand and storage filling cannot offset a day of high demand and storage withdrawal, the refilling process is considerably slower.
They’re seasonal storages, they’re not capable of refilling rapidly so once depleted it’s a slow process to refill.
You ARE on the Right side of Oz National energy grid history – please keep it up for the benefit of the technical understanding Oz citizens though now a little past it!
The coverup continues, the contact American technical writer is being extra diligent and taking all necessary precautions!
“The safety of our people and plant is our priority,” Mr Busine said. (The rebuild accident rate is unknown)?
Why is the forensic engineers report secret?
The critical 220 VDC Battery was disconnect during a unauthorised switching by persons unknown apart from a flash cartoon video
The C4 unit operator was disconnected from his controls, and left in the dark.
Unit C4 has been impacted by the availability of overseas based specialists needed to perform critical work, is more contact technical writer coverup.
The Contract Technical Writer gained most of his experience working for Boeing so he is use to big wreckage.
The cost of this debacle is top secret.
Of course what further complicates the use of historic wind data is climate change. Meaning that the probability of low wind periods could be changing over time. Of course 30 plus years of data is still going to be much more meaningful than two or three years.
A further complicating factor is that as more wind farms are installed, more wind energy is removed from the system meaning an overall reduction in generating capacity.
The law of conservation of energy makes it pretty clear we can’t just use the wind to spin blades and not have any negative impact on wind patterns.
The 66 MW Barron Gorge Hydro Power station is just coming back on line since Cyclone Jasper in December 2023 BUT due to a very long and good Wet Season in the highest most reliable rainfall region in Queensland (if not Australia), the 88 MW run-of-river Kareeya Hydro Station in Far North Queensland has operated almost 24/7 since Cyclone Jasper ie its basically operated as a baseload rather than peakload power station for over 6 months and continuing to do so as the Koombooloomba Dam that supplies Kareeya Power Station is at 90% capacity. Its over 60 years old so its printing clean $ for its owner – the Queensland Government.
Hi folks,
I’ve got a much more basic question – how are interconnector import and export limits calculated by AEMO? How much relates to interconnector physical limits / design and how much relates to NEM demand / supply?
And why are the import and export limits often different to each other? For example, on the VIC-NSW interconnector right now, the export limit (VIC–>NSW) is 183 MW and the import limit (NSW–>VIC) is 18 MW.
Going out in time we can also look forward to the combination of low wind and old wind farms that need. 1. replacing and or 2. lots of maintaining coupled to a financial depression. But I am sure Hydro will not be in a once in a 10,000 year drought at the same time.
The C4 220VDC was isolated with the unit in service, a written high voltage isolation was not used, the insurance has not made a payment to CSEnergy,
The RTS insurance will be self insurance by Qld Gov, their is a problem with staff turnover, there is a problem with operator training, maybe the CS Energy mess should be given to Snowy Hydro as surely they would not turn off the 220VDC. They provided a picture of the C4 220V battery, is that small battery able to supply all the critical services for the C4 unit for one hour with battery charger not available. Does the employment of a extra 40 Engineers help, remember the C4 unit operator did not know that that the 220VDC had been isolated
Most years in the non-summer months, we typically have two periods of low wind output combined with low solar output which last around 5 days. However this year has been different as it has been low wind output combined with low rainfall on top of a drought in Tasmania, meaning much less hydro output from there. Conditions from a year like this are one of the most challenging to solve in a future 100% RE grid as the typical projected storage and up scaled RE won’t solve it.
There are several factors that will assist, but resolve the problem. Victoria needs much more utility scale and rooftop solar in the mix of RE generation. It will require a significant over build of solar due to the poor performance during the winter months, but this appears to be one of the cheapest options.
A doubling of capacity of the QNI flows into NSW is potentially a good solution, but not cheap and complex on the NSW side of the connection. The addition of BESS at multiple points would be required to increase the carrying capacity of the existing links in addition to upgrades of substations. This of course assume QLD continues to build out VRE sufficient to have excess.
Additional wind and solar in Northern NSW is another option, but of course this is where there is great resistance to this, particularly wind.
None of that fixes the problem ofhe wind not blowing across the NEM at pwriods when solar isn’t producing.
The fact that we have over 11GW of wind energy capacity and it was producing just 88MW shows oversupply of capacity doesn’t fix the fundamental flaw of renewables meaning you can’t guarantee output.
I would suggest your sketched trendline overstates the case, since it is assuming the 5-day period of very low wind extends out to 9 days. A more reasonable comparison to see “what would have been” if the large burst of wind generation at the end of month hadn’t occurred would be to extend it at the rate of the overall May average?
Thanks for sharing this report. it outlines a common scenario in many industries where risk competency relates mainly to personal risk ( PPE and site conditions ) and not to the hidden risk of unmanaged process risk. In my many years of visiting many mine sites/generator sites / and general industrial areas, I have always been conscious of both risks. However, it is not the case that these two areas are taken seriously enough. Hopefully, this report will be taken up with all industries waking up to the combined scenarios stated.
We are seeing more anomalies like this due to the weakening magnetic field, shift of the magnetic poles and being near the peak of the 11 year solar cycle.
Multiple market notices for unplanned outages in Northern/Central NSW yesterday:
– MN117313 at 05:41: unplanned outage of Lismore – Coffs 330 kV at 0511, influencing constraints on Directlink
And then after load shedding (from 18:17):
– MN117323 at 19:41: unplanned outage of all 3 cables of Directlink at 18:45 (- but it looks like flow was restored by the next DI)
– MN117325 at 20:01: unplanned outage of Eraring – Newcastle 330 kV at 19:40
Speculation: some correlated transmission failures around Lismore, with load shedding in that sort of area and limited impact on overall NSW prices or supply/demand balance.
Macarthur Wind Farm has produced nothing since 6 June, which at 420 MW is one of our largest wind farms. Surrounding wind farms have produced well when the wind has blown, and other wind farms connected to the same 500 kV power line haven’t had the same issues.
I’m curious how the logistics of firing up for the first time in 5 years works. Do they have to start by hiring in engineering and operational staff on temporary seasonal contracts?
It is interesting that the wind drought is caused by a succession of high pressure zones which in turn minimised cloud cover, so while wind output was down across the NEM by 10% probably 13-14% adjusting for capacity changes, solar output was up 20% so combined wind and solar output was actually up 5%. The big problem is that hydro output was down 22%, 30% in Tasmania and 40% in NSW so total renewable generation was down 0.1% for the quarter
So CS Energy’s process safety management systems were not where they needed to be and the critical 220 VDC supply to unit C4 was turned off by maintenance staff without the operators that where in charge of the power plant operation knowing. So the insurance payout is lost and the cost is many $100M! It’s time that management was transferred to Snowy Hydro as they are the only public authority with the ability of managing power plant without wrecking it by ignorance(ignorance is only the lack of knowledge).
The event this morning comes a day after planning approval was granted to the Limestone Coast Energy Park battery. Will having that storage online allow the 50MW constraint on the Heywood interconnecter to be relaxed?
Hi Tristan, Thanks for the question. It does indeed seem like it could support the grid in that way, being quite large – 500 MW / 1500 MWh. Perhaps a SIPS (or similar) contract might be entered into. We’ll have to keep an eye out. Noting the anticipated timeline for operation is in 2026.
Do you have any view on how the marginal benefits of energy connect compare with where we would be had the original $95M been spent on Riverlink in 1997?
Thanks Mark. Many would argue that the situation would have been easier, cheaper, quicker, etc. if things were started two decades earlier.. I don’t disagree, but hindsight is 20/20.
A few what-ifs and other questions worth thinking about:
– How things would have panned out had Riverlink been in place before the renewable development supercycle (and the system black event)?
– Would it have actually cost $95m?
– Were the expected benefits the same back then, as they are now? (and how easy was it to predict these when the original Riverlink didn’t go ahead)
– Are all the benefits measurable/model-able?
No, or at least I think it is still too early to call it. The trailing 30 day levels are still below that of last year. The other factor in this wind drought weather pattern, a lack of rain particularly over Tasmania, is still resulting in Tas Hydro using fossil gas for generation and importing significantly from Victoria. Even though the key catchments in Tasmania seem to have good rain in the last couple of weeks, it would appear the catchments are still a bit dry and the dam levels low for Tas Hydro to want to return to normal operations. While this situation continues, it will result in higher prices in the southern regions of the NEM.
Here’s hoping we continue to see good wind levels and rainfall over southern Australia.
A major difference in projects cost blow out is whether its govt or private funded. Your statement of solar and wind farms not running over budget is because the estimate is real and there’s little room to keep throwing money in to fund the project. Compare that to govt funded projects where the real price is rarely disclosed, as politicians don’t want to admit the real cost as it will be hard to justify and won’t win votes. Hence cost blow out from original budgets.
Interestingly Snowy timeframe is near a decade, $12Bn in costs + $6Bn in federal buyout of state ownership + Humelink sums to approx $20Bn. This equates to ~$10Bn/GW…. kind of makes the argument against Nuclear null and void.
I’m certain that the masses do not understand the magnitude of the energy storage problem. Pipe dreams like Snowy Two will continue to crop up until the people understand the magnitude of the problem and grasp that even when complete this project will not provide a sensible backup to the wind and solar power grid.
As a simple example. A fuel tank in a typical car may hold 60 litres of petrol. Petrol will have an energy density in the range of 9.3 kWHr/litre. So that tank will hold 558 kWHr of energy.
Lifting one litre of water 680m will provide approximately 6800 Joules of potential energy, which is 1.9 WHr.
So to get the same energy storage as per the petrol in your tank, (ignoring all losses), would require lifting nearly 294 thousand litres of water up the 680 metres of Snowy Two.
Petrol really is an amazing dense fuel. If you want to replace petrol with electric propulsion, how many cars will be allowed to recharge each week on Snowy Two? And remember, you have to RECHARGE the Snowy Two scheme between discharge events.
I’m certain that the politicians who are pushing this scheme either won’t know the above or won’t listen to you if you tried to teach them. If the truth is known, the bubbles bursts.
It doesn’t seem unreasonable at all that 60 litres of petrol can raise 294 tonnes of mass 680m when used efficiently – after all, you don’t expect that driving your 2 tonne car up from Penrith to Hazelbrook will use much of your tank, even when it’s burning that fuel very inefficiently in an internal combustion engine (where the bulk of the energy stored in the fuel is going out the tailpipe and radiator as heat).
If you’re talking about replacing petrol with electric propulsion, you need to take that efficiency difference into account. The petrol vehicle is going to get about 700km out of that tank, so about 800Wh/km, whereas similar sized electric vehicles use about 160Wh/km. So drive the EV as far as the petrol vehicle goes on one tank requires less – only about the energy stored in 58,000 L of raised through the 680m head.
Also 294,000 L sounds like a large amount, but it’s almost exactly what the existing Tumut 3 station can pump uphill *each second* (the downhill generating flow is considerably higher). The Tantangara top reservoir in the Snowy 2.0 scheme holds 800,000 times that quantity when full.
More batteries will help a little, but those 4 to 5 day low wind output events are going to be the largest challenge for SE NEM to transition to 100% RE.
The quick answer about highest 7-day average is yes, by a long way. Midnight to midnight, 19-25 July averaged over 5,950 MW, 600-odd MW up on the previous high in July 2023.
Someone raised a point about minimums. Those have increased as well – the Q2 2017 calm period saw a minimum 7-day average of 420 MW. This Q2 the minimum 7-day average was 1,320 MW.
Can someone outline how semi-sched and batteries are treated in this calculation please? For very little variation in Market Demand, the IRPM seems to be almost precisely out of phase with maximum generation. Not unsurprising, but if this IRPM is calculated assuming charged batteries and predicted wind then once you get down around 10% you really are only a hiccup away from a problem – particularly in a region unable to be supported via a transmission constrained market. Gets even more precarious if some of your assumed generation is coal that has been decommitted for a few days and is now cold.
I did miss poor Peter Hannam’s mention of emissions in the body of his piece about power price changes, so have updated my LinkedIn post accordingly, but I reckon my broader point still stands – it is super weird that such a significant change of direction – one that seems partly systemic beyond week-level fluctuations in wind output – isn’t getting more coverage!
“Large-scale government-initiated projects unlock many smaller privately-funded projects. In the case of transmission, this can be in terms of actual project approvals – but in a more holistic sense, many of these mega-projects eventually play a role in enhancing investor appetite and ensuring long-term confidence in reliability.” Whoa, Dan. And your evidence is ….?
Is the battery in SA flat yet? It’s supply proportion of the local grid has fallen from 6% to 1.5%. Is it being kept partially charged to allow for a black start?
Due to the extension leads from Vic being out or seriously limited, SA is operating like an island. And if you’re exposed to the going rates. OUCH.
Glad Bowen has plans for the rest of us that don’t look like an extension of SA’s current woes. /s
So CS Energy’s process safety management systems were not where they needed to be and the critical 220 VDC supply to unit C4 was turned off by maintenance staff without the operators that where in charge of the power plant operation knowing. So the insurance payout is lost and the cost is many $100M! It’s time that management was transferred to Snowy Hydro as they are the only public authority with the ability of managing power plant without wrecking it by ignorance(ignorance is only the lack of knowledge).
So CS Energy’s process safety management systems were not where they needed to be and the critical 220 VDC supply to unit C4 was turned off by maintenance staff without the operators that where in charge of the power plant operation knowing. So the insurance payout is lost and the cost is many $100M! It’s time that management was transferred to Snowy Hydro as they are the only public authority with the ability of managing power plant without wrecking it by ignorance(ignorance is only the lack of knowledge).
CS Energy spent $38m of taxpayers’ money hiring lawyers over the past two years, including to fight the release of the landmark report into the company’s woes. The investigative report into the 2021 Callide C explosion compiled by forensic engineer Dr Sean Brady, cost taxpayers a further $10.4m.
Richard Baum.
CS Energy spent $38m of taxpayers’ money hiring lawyers over the past two years, including to fight the release of the landmark report into the company’s woes. The investigative report into the 2021 Callide C explosion compiled by forensic engineer Dr Sean Brady, cost taxpayers a further $10.4m.
Regarding the black line marked “total volume produced” on the “Trended Aggregate Volume…” graph, is that the total energy that was actually delivered after bidding?
Giles is a hack. He always blames fossil fuel generation for every outage or price spike despite them actually being caused by the lack of renewable energy generation.
Glad to see you shining a light on his lies but sadly his true believers will likely never see your rebuttal.
It’s fascinating how people will see what they want to see in any given piece of data or analysis.
Take the dispatch interval highlighted in Paul’s bid chart, ending 18:00 on 30 July. Putting aside how much supply capacity was renewable or not, there were 37,281 MW offered to the market and only 31,144 MW needed to be dispatched. However the proximate reason that the spot price went over $15,000/MWh in every region was that of those 37,281 MW offered, 6,273 MW (17%) were priced above $10,000/MWh.
Maybe the technical and competitive dynamics leading to this bidding behaviour would be more worthwhile subjects for commentary than whether Giles Parkinson is a hack or the level of renewable generation at the time?
It would seem the lack of competition from Victoria (or NSW via EnergyConnect) contributed to the high pricing. Of course more battery operators combined with additional storage capacity would have assisted as well.
Not sure if I’m missing something here but in the CS Energy response to Brady’s report it appeared that the key interlocks have now been removed to allow 2 batteries to be paralleled on the unit 4 dc switchboard. Given this was done quickly it presumably did not include upgrading/replacing the unit and station dc switchboards to allow double the fault current. Does this mean these dc systems are now in a much less safe state than their original design? It would seem to me that changing over from unit to station dc supplies with the unit online would never have been allowed for in the original design. This is unit protection after all. Just wait for the next outage. Did the dc electrical design documentation allow for dc switching with unit online? If not was a dc trip while replacing battery chargers with unit in service in 2021 an entirely foreseeable risk? And Brady states that the dc system relies on the charger, with the battery only providing a backup to the charger. I’m having trouble with this when dc oil pumps must always be available in an emergency. If a unit trip removes the ac then the dc oil pumps are the only thing preventing a catastrophe. It’s a long time since Callide C was designed, but maybe someone can reassure me that CS Energy’s “additional redundancy in the battery charger system since Unit C4” hasn’t exposed it’s people to far higher risk?
I thought it was the downramp of solar @ 5~6PM in South Australia that initiates the price volatility. Maybe I have to hate fossil fuels more in order to properly understand.
Fantastic article as always Allan.
I’m curious about how you estimated the state of charge of batteries that don’t report them.
I’ve heard that the 5-minute Initial_MW or other 5m data for batteries doesn’t accurately represent the average charge/discharge amount over the 5 minute period.
You mentioned calibration against the Torrens Island battery. Was that to determine a relationship between the 5-minute Initial_MW data & the actual 5 minute average, or something else? If so, and if you’re happy to share, I’m curious what you found out.
Thanks David. The charge estimation is based on:
– taking the mean of SCADA (InitialMW) values at start and end of each 5 minute interval as average interval dispatch, to yield energy consumed or sent out in each interval (yes the reality is more complex if Regulation FCAS is being provided by a battery but this ‘trapezoidal’ approximation is not too bad)
– applying a conversion loss rate to both charge into and discharge from storage
– assuming a constant rate of ‘parasitic’ losses (auxiliaries etc) from stored energy
Playing around with the last two rates yielded a calibration against the published TIB storage levels which then seemed to yield reasonable results for the other batteries after picking a start-of-day charge level for each that left them with minimal energy in storage at the time their availability fell to zero in the evening.
It’s only a rough estimation that wouldn’t necessarily work over an extended period but I’m happy it’s consistent with the behaviour of the batteries over the day in question.
In Victoria Newport Power Station contributed nothing through the morning and evening price spikes, while Mortlake was well below full output. Are gas pressures low through inadequate withdrawl capacity from the Iona storage?
The state and federal governments have been forced to re-enter the business of electricity generation, storage and transmission and they are making a mess of it. Look what happened when CS Energy’s Callide C4 steam turbine was destroyed by ignorance (they turned off the critical 220vdc supply) and problems will continue until training,examination and advancement of critical personnel is controlled by a central authority. God help us if nuclear power stations are required.
Critical personnel at CS Energy lacked experience and knowledge of process safety.
Callide C4 management allowed unauthorised untrained maintenance staff to carry out the critical switching that destroyed the C4 turbine.
Their claim for an insurance payout is lost.
What is the total cost including lost income, damage, legal , enquiry and repairs?
The coverup continues, what is the history of the critical 220VDC batteries, what happened to the critical key interlocks that have been used for over 100 years to avoid what happened at Callide C4.
220V batteries should not be paralleled because of possible large fault currents, switching has be done offload, that is why key interlocks use to be used and only authorised trained operators should carry out these critical switching proceedures with the unit out of service.
This makes me wonder if the gas generators aren’t doing a similar rough-and-ready estimate of their battery competition’s State-Of-Charge level. It seems like the optimal strategy for them might be to provoke the batteries into emptying before the market loosens, allowing them more pricing power with less competition remaining with available capacity.
You’d hope the battery bidding would assign a certain amount of variance to future market outcomes, not just take the pre-dispatch forecast as the 100% expected price (ie you only have to assign a 4% probability for a future $15,000 price to have a better expected value than a $500 price available now).
You can just imagine where this would go had Eraring closed as scheduled – its energy overwhelmingly would have moved to OCGTs resulting in higher capacity factors. People need to understand that gas is not the same as coal – coal stockpiles are easy – managing instantaneous demand flows in a gas system is difficult and will cost money. Pipelines, compression and localised storage. LNG storage starts to look sensible. You need a very big battery to store the energy equivalent of a 50,000 tonne tank of LNG. Otherwise you end up in liquids. And the logistics and costs get eyewatering.
We still appear to be in conditions with blocking high pressure systems that are resulting in lower than average wind conditions that are typical for this time of year combined with low rain fall in Tasmania. At least we are starting to see some short periods of very good wind conditions, but regions like SA lack the storage capacity to fully take advantage of these short periods. Lack of transmission capacity is further compounding this issue.
Hi there, sorry if this is a dumb question but why do gas storages on AEMO’s GBB dashboard continue to empty when there is a net surplus of supply over demand? For example, right now (Tue 8 August 14:35) there is supply of 5,721 TJ and demand of 5,580 TJ giving a net supply of 132 TJ. Yet Iona, Moomba, Roma and Silver Springs storages are all emptying with an aggregate rate of -76 TJ. Shouldn’t these 2 numbers be the same unless gas is being lost somewhere in the system?
Hi Andrew, I probably can’t give you a firm answer but, initially, two reasons come to mind. Yes, there is a thing they call UAFG – unaccounted-for gas – that represents fugitive losses along the network ~3 – 5% ish. Secondly, The AEMO note on Demand in the GBB (fur the current day) suggests that some calculations are performed for VIC and some demand for gas powered generation data might be excluded due to the methodology.
Thanks for your thoughts Matt. In general some of those factors could impact regional reserve levels and prices of course. But in these specific cases of multi-region spikes:
– there weren’t binding constraints on the relevant interconnectors (which is why prices spiked jointly).
– there were no instances of actual LOR1 let alone LOR2 declarations.
– if you’re contending that being “close … to LOR1 warnings” would be sufficient explanation for prices near the MPC then I strongly disagree.
– hydro plants had enough water to offer the availability that they bid into the market. Are you seriously suggesting that the opportunity cost of undispatched water was a 5 digit offer price?
– for gas, see water.
Of course you’re welcome to run and present your own analysis of the individual region cases including interconnectors, I’d be genuinely interested in your results and conclusions.
Would be interesting to unpackage these events to see who / what was the marginal cost generator. This is the problem with energy only markets. Batteries have no hope of paying their way without a very large spread on their energy arbitrage. Gas turbines have to cover their start up costs (loss of life of very expensive hot gas path components) over a relatively short generating window the duration of which the generator needs to predict before committing to run. This will only get worse as coal plants are pushed out of the market. From their min gen position, they can ramp to max relatively cheaply. Once they have to be decommitted it takes a long period of predicted generating time to amortise a restart. This is where the gentailers have inherent advantage.
Thanks for your comment Andrew. I’m not sure that the identity of the marginal generator is all that important here – there were signficant volumes offered at 4 and 5 digit prices from all fuel types other than solar and wind, and by more than one participant. I think that’s the key observation; which specific generators happened to be marginal in the high price intervals doesn’t really add any information, since there were clearly a number of participants offering at very similar prices. Allan
I don’t understand why we are getting this continual rhetoric of coal plants being “inflexible”
Why do coal plants even need to be flexible when they have contract volume secured for months/years in advance.
A lot of Renewable advocates really get upset by the advantages low-cost scheduled generators have.
Maybe they should strive/mandated to be just that, low-cost scheduled generators instead of this semi-scheduled nonsense of doing half the job.
There have been a few papers published by Griffith University regarding Australian energy, most interestingly, P. Simshauser, 09/2021, “Rooftop solar PV and the peak load problem in the NEM’s Queensland region”. It is definitely worth a read in my opinion.
Mother Nature trying to average out the wind conditions by reminding us of the power of the wind.
Interestingly wind generation in Victoria in Q3 has continued to perform below average for this time of the year, while SA is close to average. Tasmanian dams and catchments need more rain, but I’m sure they could get by without floods.
It is time for a Royal Commission into what happened!
Was the critical 220VDC supply modified from original and who planned the process that allowed it to be turned off, with the C4 unit in service.
The C4 unit control room operator did not know that the 220VDC was switched off.
Why the need for a coverup, the fancy SC employed by CS Energy, commissioned the so called enquiry to avoid public disclosure of what really happened.
CS Energy should be given to Snowy Hydro
What is the total cost including increase cost of supply to the grid of CS Energy failure?
For those not following too closely, note that it’s a different unit than Callide C4 (the one that blew up)!
Callide C3 is the same as C4 its 220VDC switching happened with the unit out of service.
C4 was wrecked by ignorance (which is a lack of knowledge)
Casstell Key interlocking has been used for over 100 years to avoid what happened to ensure that the critical 220VDC battery was always available to stop the turbine motoring.
Beleaguered state-owned CS Energy spent $38m of taxpayers’ money hiring lawyers over the past two years, including to fight the release of the landmark report into the company’s woes. The investigative report into the 2021 Callide C explosion compiled by forensic engineer Dr Sean Brady, cost taxpayers a further $10.4m.
I think we are unlikely to see many, if any, new utility scale solar farms announced that don’t have a matching battery of roughly 4 hours at the rated output of the solar system. Even with all the storage announced there is still going to be an excess of solar production on the weekends at this time of the year. What is more important is to look at the levels of curtailment on week days and how that compares to the capacity of the existing and announced storage.
So what does that do to LCOE (seeing that is the metric everyone loves to use) given it is the same amount of energy produced now servicing a lot more capital? From an owner’s perspective it is probably cheaper to simply forgo the revenue and work on a slightly lower capacity factor when making the FID. If the marvelous capacity investment scheme is underwriting things…..who cares? Certainly not the investor…..
An acid test of the CIS will be whether it underwrites any utility PV or focuses on wind/batteries/PHES for now. If the former, we’ll know it is flawed. There’s no need for more utility PV until a few more coal plants have exited and even then it’s likely rooftop PV will have already filled the gap since it’s competing against retail prices not wholesale.
In the recently released GenInsights Quarterly Update for 2024 Q2 we pose a question about whether the Capacity Investment Scheme should only support VRE additions if they are fully integrated hybrid units.
Unfortunately, that’s not my understanding of the way that it’s structured at present…
A most interesting generating frenzy. Paul as Tasmania’s wedge tailed Eagles are an endangered species how about a regular story on their deaths at our glorious windfarms.
You should check on the 350,000 MWh storage of Snowy2 on page 47 of ESOO 2024. It seems the whole water resource used to turn this storage into 2,200 MW over 160 hrs is not fully recyclable and Tumut3 will be impacted
Firming capacity is classified by AEMO as shallow (less than 4 hours energy storage), medium (4 to 12 hours) and deep (over 12 hours). The Capacity Investment Scheme is primarily enabling shallow energy storage.
The government owned deep storage projects of Snowy 2.0 Pumped Hydro (2,200 MW) and Borumba Pumped Hydro (2,000MW) are unlikely to make their indicated full operational dates of December 2028 and September 2031. A Final Investment Decision is yet to be made on Borumba PH yet AEMO has classified it as an anticipated project for the ISP.
Its not just IF we have enough firming capacity its also IF there is sufficient medium and deep storage and what IF medium and deep storage is not operational by 2030 when the Capacity Investment Scheme has enabled another 23,000 MW of Variable Renewable Energy.
A question:
Last week the Victorian Government published its Cheaper, Cleaner, Renewable: Our Plan for Victoria’s Electricity Future which included “nation-leading storage targets of at least 2.6 gigawatts (GW) by 2030 and 6.3GW by 2035. Curious to know what are the medium and deep energy storage projects that will be operational in Victoria by 2028 ie by the time the 1,450 MW Yallourn power station closes?
Why does the ESOO state that “These systems [rooftop solar] do not tend to contribute to lowering the scale of peak demands during extreme hot conditions in the summer..”?
Geoff Eldridge wrote on here in February about the new NEM peak of 38.6GW on 22/02/2024, of which ~8GW was coming from rooftop solar. Finding this fact hard to square with the ESOO statement.
I think you’ll find that both are talking about different measures of demand.
1) Geoff is talking about ‘Underlying Demand’ or ‘Native Demand’ (not quite the same as each other, but similar)
2) The AEMO most cares about ‘Grid Demand’. Because rooftop PV has already shifted the time of peak Grid Demand to evening periods after sunset (in summer and winter), adding more rooftop PV has no additional benefit of reducing peak Grid Demand further.
“The Callide Unit C4 incident was the result of the simultaneous failure of key electrical equipment and system back-ups in a complex series of events that could not have been anticipated, with some of the contributing factors being traced back to the original design of the power station,” Mr Busine said.
What is the total cost? $50M on coverup, maybe $500M on rebuild because the Insurance has refused a payment all because CS Energy allowed maintenance staff to switch off the critical 220V battery with the C4 Turbine in service.
The C4 operator in the central control room did not know that the 220V battery had been isolated.
It has been an incredibly rough five days in Tas. Never seen or experienced anything like it. The greater Launceston / central north has been decimated. Saturday night saw at least one transformer explode in Launceston. Sunday night has been the worst by a long shot. We live beside a main feeder in the country which managed to ride through many dips in voltage as the Hadspen substation was managing other feeders from the distribution connection point that were going to ground. In the end, despite an apparent re-close attempt, the main feeder that passes us finally failed. Just to finish things off, 45 mins later, lightning struck the dead feeder just near us, generating a current rush to ground via the surge protection on the transformer causing the HV fuses to drop. Current then found its way into our house via the in ground phone line (yes, land lines are still a thing!), then back to ground via our network and local earth stake! We are the sole customer on this trannie and don’t expect power to be restored for a week. As of 0230 this morning, TasNetworks call taker said they now have 42 000 customers powerless. Would be interesting to see how much our frequency fluctuated down here given all the chaos that has ensued tonight.
While these are one off events, the trend is definitely there. There are unlikely to be anymore large utility scale solar farms that don’t have a very large battery with them. This of course is due to the large amounts of “excess” solar in the grid, the lower cost of batteries and the change in regulations that now allows hybrid systems.
The smart money is, or should be, building MEM-attached [battery] storage that store power when prices are negative, and kick in during 4 to 9pm when we all pay the time-of-day premium tariff.
I wonder if anyone is doing this?
The time to build solar farms was back in the 2010s or earlier.
According to Prof. Simshauser’s modelling, rooftop PV is basically forcing the shutdown of coal plants, otherwise the NEM will become unstable, (or evening tariffs are going to get very, very high to compensate!)
I really think there needs to be a weighted standardised measure of temperatures that better reflect their impact on electricity demand. I know all too often I look back at the average temperatures for a month and find they don’t really reflect how hot or cold some of the periods were during the month.
For example a single hot day doesn’t place create as much demand for cooling, compared to five days in a row of the same or even slightly lower temperature.
Thanks Linton for this article. Importantly, the reduction in gate closure has allowed improved forecasts to be used by NEMDE. AEMO’s own figures quote an 11-14% improvement in MAE to their ASEFS/AWEFS forecasts and I know a significant part of this improvement is due to being able to use later site measurements as inputs to the forecast.
At ETSI we see an even greater improvement when moving from a 60-second to 10-second gate closure. The result of these accuracy improvements is downward pressure on regulation FCAS utilisation, evidenced by AEMO not having increased regulation procurement volumes despite the large increase in VRE capacity (admittedly PFR has helped frequency control around the PFR Control Band).
What happened on AEMO’s page under Renewable penetration for Sept 5th Thursday ? Somehow black and brown coal wasn’t recorded at all and so the % was over 90 % which is false.
“The Callide Unit C4 incident was the result of the simultaneous failure of key electrical equipment and system back-ups in a complex series of events that could not have been anticipated, with some of the contributing factors being traced back to the original design of the power station,” Mr Busine said.
What is the total cost $50M on coverup, maybe $500M on rebuild because the Insurance has refused a payment all because CS Energy allowed maintenance staff to switch off the critical 220V battery with the C4 Turbine in service.
The C4 operator in the central control room did not know that the 220 battery had been isolated.
Readers need to remember that ‘Installed Capacity’ can be a significant overstatement of aggregate windfarm capacity physically able to generate, because for each new windfarm connected, its nameplate Maximum Capacity arrives in AEMO’s systems in one hit whereas actual physical capacity can take up to 12 months to fully commission.
So ‘Installed Capacity’ can be well ahead of actual capacity on the ground, especially when larger new windfarms are being connected and commissioned.
As a consequence, Global-Roam’s calculated fleet “capacity factor” can be materially understated because it is comparing apples and pears – generated output from actually commissioned capacity versus ‘Installed Capacity’ not all of which is actually commissioned and able to generate.
The impact on fleet capacity factor, compared with using an apples and apples approach (eg by excluding newer partially-commissioned farms from the calculation) can be of the order of 2-3 percentage points – eg 30% vs 33%, which is fairly material.
So three years to procure a cooling tower rebuild, new; turbine, alternator, transformer, switchgear and associated electrical auxiliaries.
So I would call this 30% of a new unit, but like only 5~15% timeline of the full brownfield power station extension eg. the original building of C3C4, with the whole development approvals, environmental approvals etc…
So how does this factor in the “little bit optimistic” view that building a mature technology nuclear PS on a existing Brownfield site as “eh, I say we can do in a short 10 yrs eh??”.
Can we have a fact based deep dive into the topic of timeline of major infrastructure planning in the NEM please?
“The Callide Unit C4 incident was the result of the simultaneous failure of key electrical equipment and system back-ups in a complex series of events that could not have been anticipated, with some of the contributing factors being traced back to the original design of the power station,” Mr Busine said.
This is classic technical writer cover-up “gobbledygook”!
$50M is the cost of the legal advice and inquiry which was meant to be secret by claiming legal privilege after maintenance staff isolated the critical 220V battery supply with the C4 Turbine in service so the insurance was voided!
The C4 unit operator “who most likely now has PDSD” in the principal control room did not know that the 220V battery had been isolated, so the C4 Alternator motored to complete destruction!
If you take a look at the participant UIGF, it would appear that they are using their actual output rather than a forecast of their actual capability given the current/forecast irradiance and cloud cover. More importantly, they appear to be shading this/biasing it lower (possibly only around gate closure?), presumably to optimise their causer-pays factor – critically at the end of an interval where they are subject to the 85MW limit/have a target for that level and a semi-dispatch cap flag, their self-forecast will be somewhat below 85MW, the following interval they will receive a target less than the constraint limit, they are no longer subject to a semi-dispatch cap, and with the flexible nature of targets for semi-scheduled plant they can operate up to their fuel availability… This particular situation highlights that perhaps a more robust approach is required to impose a maximum output on semi-scheduled plant. Arguably biasing one’s self-forecast on the low side to enable provision of “raise” PFR (or just raise in general) is not necessarily bad for the system, and it wasn’t done with the intention of gaming the constraint – but I look pretty dimly on a self-*forecast* of *unconstrained* fuel availability that appears to be based largely on actual output and not actual fuel supply ie irradiance and cloud cover in this case…
Great update. My observation is that the green band reduction must be constrained by the number of baseload units available. As more coal units are retired, less volume can be bid $0-300.
This exposes the truth in the idea that ‘cheap’ renewables displace the most expensive generators in the bid stack ie merit order effect.
What’s actually happening is the mid-merit baseload (coal) is being forced out.
The evolution of this trend means the end of baseload generation and much higher system costs.
Clearly more government intervention is required – either to force more floor price generation into the market, or force more baseload.
CS Energy Ltd’s annual report 2024 is not available today Monday 23rd September 2024, it might have a total cost of the Callide C4 outage caused by the maintenance staff’s isolation of the critical 220VDC supply with the C4 unit in service?
What is the total cost of wrecking the Callide C4 steam turbine in 2021, cause? maintenance staff isolated the critical 220VDC supply. maybe the annual report 2024 for CS Energy Ltd.will supply some information.
What a thought-provoking article! 🌟 It really highlights the struggles many of us face with “smart meters” and how misleading the data can be. It’s frustrating to see our energy-saving efforts not reflected in our bills! 😩 I love how you pointed out the importance of having accurate data, especially for those of us with rooftop solar panels. 🌞 It’s essential for making informed decisions and maximizing our sustainability efforts. Let’s hope energy companies take note and start improving their systems! ⚡️ Keep spreading awareness! 🙌
2nd Oct 2024. What is the total vast cost of switching off the critical 220VDC battery with unit 4C in service? The C4 Operators in control of unit 4C had not been informed!
This is worth a closer look at the rate of roll off of both load and frequency and then a look at satellite imagery for cloud cover. One that people didn’t see coming I suspect. Has already happened a couple of times her in WA where unforecast cloud banks rolled in from the ocean and cruelled rooftop solar production. Chewed up all system spinning reserve and then some.
Looks like there might be a minimum load condition at about 13:30 AEDT today Wednesday 9 October? Prices on the AEMO NEM dashboard are showing as being ‘under revision’.
CS Energy’s Ltd Callide C4 Management allowed maintenance staff to isolate the critical 220V DC Battery supply with the C4 unit in Service without informing the unit C4 operator who was in control. He was left with his control panel switched off.
Process Safety proceedure had been removed by CS Energy’s management’s lack of knowledge or understanding
A critical insurance payment has been lost as it is impossible to insure against a lack of knowledge or understanding and behaviour that shows a lack of good sense of judgement.
Technical Writer gobbledygook has been used to try and hide what happened
CS Energy’s Ltd Callide C4 Management allowed maintenance staff to isolate the critical 220V DC Battery supply with the C4 unit in Service without informing the unit C4 operator who was in control. He was left with his control panel switched off.
Process Safety proceedure had been removed by CS Energy’s management’s lack of knowledge or understanding
A critical insurance payment has been lost as it is impossible to insure against a lack of knowledge or understanding and behaviour that shows a lack of good sense of judgement.
Technical Writer gobbledygook has been used to try and hide what happened
CS Energy’s Ltd Callide C4 Management allowed maintenance staff to isolate the critical 220V DC Battery supply with the C4 unit in Service without informing the unit C4 operator who was in control. He was left with his control panel switched off.
Process Safety proceedure had been removed by CS Energy’s management’s lack of knowledge or understanding
A critical insurance payment has been lost as it is impossible to insure against a lack of knowledge or understanding and behaviour that shows a lack of good sense of judgement.
Technical Writer gobbledygook has been used to try and hide what happened
Its not just the stock of towers. Next year is the midway point of the Solar Cycle which is traditionally the most active ie potential for transformer damage. This would appear to be an issue for Tassie and maybe Victoria. As an energy consumer advocate I’d be interested to know the status of transformer stocks and if engineers have plans to increase stocks when mainland states become dependent on Marinus Link/Battery of the Nation to keep the lights on. Maybe the supply chain lessons of COVID have already been forgotten.
Yes, it seems unfortunate that they can’t at least use the windfarm and BESS alongside the diesel generator, which would seem helpful in the evening peak.
* Feb 2024, 120 km/h winds near Anakie, VIC, 6 towers, 530,000 homes, thunderstorm
* October 2024, Menindee, south of Broken Hill, to Tibooburra near the Queensland border, 7 towers, extensive blackouts
With respect to comment about Prominent Hill – both Prominent Hill and Carrapateena Mines are supplied from a separate 275 kV transmission line ex Davenport, and not the same 275 kV line that normally supplies the load to Olympic Dam (also ex Davenport) – ie there are 2 275 kV lines / 2 completely separate systems, so I’d assume both Carrapateena and Prominent Hill mines are still connected and operating as normal.
A good summary of battery opportunities in the NEM but I’m curious as to why you think there is an immature contract market. OTC trades have existed since before the NEM started in 1998 and the ASX started trading futures in 2001. I can appreciate that getting contracts of sufficient tenor is a challenge as most customers don’t contract past 3 years whilst project owners seek contracts as long as 20 years. Surely, this is a mismatch of the price and volume risk that a customer faces as compared to that of the project owner more than any fundamental problem with contract markets per se?
Might be helpful if people read the “Lessons Learned” document prior to pounding keyboards. Appears the project was all about studying the stabilising potential of a grid forming BESS in a fringe of grid radial system with relatively high renewable energy (inverter connected) inputs – hence ARENAs involvement as a research project. It was not about keeping the lights on. This is entirely consistent with what would otherwise be a somewhat odd combination of a 50MW inverter and 50MWhr battery (ie one hour of storage at rating). Indeed, I would be surprised if without major re-engineering of protection systems in the town it would be possible to run on the grid forming battery and renewable energy sources because of a very low faut level. Once a synchronous generator (or generators) were running it is possible provided the dynamic response of the power electronics does not end up pushing the generator off. Hence why things are being taken very gently. Sounds like sensible and prudent engineering practice to me. Not a great situation for anybody, but it appears the battery was never designed to keep the lights on, despite what the media people might have written. The best it might do is accept the risk of fault level issues whilst the GTs ran up and synchronised following a line trip – say 10 minutes.
The major transmission lines *never* have the towers replaced throughout the their life because it’s so hard to get the lines out. The job to replace towers and restoring lines overnight (as required by AEMO) has to be done with temp towers. It is an enormous undertaking.
The second part of this statement is a carbon-fibre, kevlar reinforced bow to draw. If only it were that simple.
“This equipment will ……….. allow renewable energy to be transferred between states which will help provide cost of living relief by delivering access to cheaper clean energy.”
Dont forget the initial event can often create harmonics on the system in the region that the frequency analyser is sampling at. Can create initial false high/low values.
Thanks for this Warwick, I wonder what happens to to liquidity in the futures/swaps market as coal exits, surely it is difficult for renewable energy owners to contract with retailers because they in most cases already have PPAs, and besides cannot be sure enough of volume and price realised to take that side of the trade.
The by-now-familiar constraints on transmission from southern NSW towards the load centres north seem to be playing their part in this volatility as well.
Terrific explanation of the events, and a nice outline the linear program behind NEMDE by zooming in on “simple” constraints. Perhaps not the points you intend to make, but it does serve as a reminder for
1) how it is becoming more common (or is already the baseline case) for it to be inaccurate to analyse electricity markets with whole-of-market bid-stack / merit-order framework. Assuming away binding constraints is assuming away reality.
2) that the system-wide impact and consequently economic value of injections at different locations (and by different technologies) at the same time within a NEM region can significantly differ. (I’ll spare everyone a rehash of the outcomes and incentives that follow from zonal market designs — it was probably all best understood and explained 20+ years ago! https://scholar.harvard.edu/whogan/files/nezn0227.pdf )
Hi Paul, I am a retired AEMO System Operations Control Room Operator. The statement that it was a “Small Frequency Wobble” is grossly inaccurate. The system Demand increases by approximately 2400-2600MW in the 30mins to 1400hrs. The Frequency in the previous 30mins to 1330hrs was struggling to get to 50hz. I have not looked at what Turbo-Generator plant was dispatched but it seems to be insufficient for the rate of change in demand. In other words the system does not have enough inertia. If the largest generator was to trip when the frequency was sc low, the remaining plant with active governors probably would not be able to pick up the loss of generation because it could not even cope with the increase in demand. I suggest that the system was “Insecure.”
I think you and Ian are both correct. Extended cloud band playing havock with the system load, and a lack of system generation inertia. And for my 2 bobs, the almost constant mid daily state to state tie line constraints to near 0 MW transfers cant be helping either.
So are you saying that if the constraint equation wasn’t in place and Victoria which at the time had negative prices and was only running its coal plants at 35% capacity was exporting to NSW instead of Tasmania, prices in NSW and Queensland could have fallen 80-90%
Re the openining quotation: It may have been Pierre de Fermat on the wonderful proof of his “last theorem” that he’d found but didn’t have space to write in the margin of his notes – ultimately the proof took Andrew Wiles hundreds of pages of extraordinarly abstract concepts that very few humans are capable of understanding. Or perhaps Eugene Galois who scribbled out everything he’d discovered about group theory the night before he died in a duel.
Phase shifting transformers, allowing a ‘dial a power transfer’ control (assuming Heywood and VNI are functioning). I guess daytime negative prices and solar curtailment in SA are about to be far less common.
How quickly did the load return? Pretty normal for big hits on the transmission system (voltage depression) to “shake off” load (things like voltage held contactors etc releasing and require a restart). These days can also shake off generation in the form of small rooftop PV inverters that fall over on low voltage – then restart after a few minutes. This will get worse as system strength (particularly fault level) declines.
They won’t be doing anything they don’t have to in the run down to decommissioning. These outages will be driven by statutory inspection stuff for pressure vessels and piping including testing of protection systems and the like. Can probably get away with an extension of time for the last unit if nothing untoward found at the last inspection if the thing is being decommissioned. Probably planning it nicely to decommission plant all due for maintenance as any good asset manager will do. Just don’t change your mind and ask them to run longer at the last minute (even if the lights are going to go out)!
Those network constraints also limit the contribution of some NSW generation as well – Uranquinty gas generation, wind in the south-west, Snowy scheme hydro can all be affected depending on how the constraints bind.
Yep, all trying to squeeze into a smaller sized pipe to Sydney. Hence the comment at the bottom of the article about potential further reductions on already reduced VRE
Yep and with more power trying to flow around the Bannaby – Mt Piper – Wollar – Bayswater 500kV route to get into Sydney from the north it’s possible the Bayswater-Liddell bottleneck could limit contributions from an even wider set of NSW generators. I don’t see how AEMO would let the Avon – Macarthur line 17 outage proceed if it was having any material effect on reliability.
Between the relevant STPASA runs a number of NSW transmission outages affecting lines in both the north and south of the state were removed or deferred from AEMO’s schedule. Looks like that played a significant part in reducing the amount of projected load shedding.
I’ve had this tagged to read for a couple of weeks and have finally got around to it – very sensible commentary from Origin/Frank (as we’ve come to expect). As Origin have commercial benefit from both traditional and future generation tech they can speak truth to power without it seeming that they’re just talking their book.
A question on your Section B – did anything ever turn up in the media outside AFR and the Aus? How is this call to arms or wake up call not being picked up by a broader network? He’s the CEO of an ASX20 company speaking at a premier non-political think-tank event and still seems like it’s hard to get cut-through, doesn’t do much for one’s confidence that we can help the discourse mature!
Hi Dan – Thursday is looking really tight, with a ~16h LOR2. Curiously, about 5h of that LOR2 is sitting at exactly 0 MW reserve under the latest STPASA. Do you or the WC team have an explanation for what’s providing that balancing – it looks almost to be a market loadshedding mechanism that acts before forced loadshedding?
G’day Tom (and Peter who has also commented below). We’ve fielded the same question from someone else – and a member of our team also observed that behaviour.
The STPASA process description says: “The linear programming formulation attempts to optimise reserve sharing between regions along with other constraints which are described later in this document to maximise the amount of reserve that is available to the system as a whole.”
We are not 100% sure, but another member of our team suggests that it could be that the model judges that there is benefit to leaving one region fairly ‘thin’ on reserves if the excess can be of benefit to a neighbouring region.
And now followed by a notice 121083 to the Waratah BESS (which is still commissioning): 96 MWh to be held in reserve. I was wondering what we’d see today from the commissioning BESS, and it seems that Waratah at least will be in action (to be joined by the Capital BESS?).
My wife works for Transport for NSW. Office staff have been told to adjust aircon temps up to 24-26deg and are not allowed any plugged in electronic devices at their desks etc.
It’s known for us as the “Government Energy Action Response” and in their words “is designed to manage and mitigate energy demand during critical periods”. I do know the idea was borrowed off the NSW government.
Participating agencies sign up and are notified by DCCEEW when to activate the plan.
For us, it involves raising the air con temp, turning off unused office equipment and all the external facing blinds are closed
`The best response would be by the NSW Planning Department to stop approving apartment and office towers and by the Immigration Department to reduce the number of immigrants who fill these towers. 2 towers with 750 dwelling units generate a peak load of 5.2 MW
It would be interesting to see an overlay
of proposed you transmission developments to see what extent these pinch boys in the network would have changed the outcome. What role did wind help yesterday
Hi Dan AEMO has stated it is going to look deeper into the change of use in gas in the future for the 2026 plan. The UQ Gas and Energy transition research centre is doing some great work on this matter. I attended their annual presentations on research work yesterday. I am confident the change can be managed however it comes with a cost for gas storage and low utilisation of gas production. There is option feasibility and optimisation studies required. I think this area is the remaining missing issue. Through in biomethane/biofuels.
Thanks Ian – I was also at that event last Thursday afternoon so sorry I missed you.
Agree with your thoughts. If we go down that path, I think there is still some work to do to communicate to the wider public that the cameo/insurance role of gas is worth having and paying for.
What do the 3 [Zone:…] terms represent in the Predispatch-RHS? Is a zone an aggregated value of generators in a region or is a zone expanded to a series of individual generator/load terms? All the generators/loads named in the Dispatch-RHS are in the locality of Yass/Canberra/Tumut or in SW NSW.
Yes you may very well be on to something there! There are a number of other constraints with similar constructions. Although, in this case, the magnitudes still seem a bit different.
The difference between dispatch and predispatch here was only about 200MW so in retrospect I think there’s more in play here than just the one constraint. Negative residue management playing a part too.
Thanks Ben! And yes, gas fired gen was also curtailed. We see it more often for the solar and wind because there is generally more transmission network between them and the load center, so more flow pathways to be managed with network constraints. But all gens are connected to the network at some point so there’s potential for certain network conditions or settings to impact all types of gens, at times.
Nice work Paul. It’s also notable that despite the work done to provide guidance on how participants should classify outages in MTPASA, some of them routinely seem to ignore that guidance.
Without naming individual generators, there are cases we’ve seen of various units extending planned outages (sometimes for significant numbers of days and/or multiple times) but continue to label the additional outage period as a ‘Basic Planned Outage’, when the correct classification would obviously be ‘Extended Planned Outage’.
Other NEM participants and spectators could perhaps infer these generators are:
– unaware of AEMO’s guidance
– aware but unable to interpret or implement it,
– aware but choosing to ignore it, or
– trying to avoid scrutiny of maintenance slippages.
Thank you for drawing attention to this issue. I think it is a matter of great concern, not just now, but into the future. There are many very large VRE projects under construction and announced for Southern NSW, plus EnergyConnect plus a likely VNI-West. While HumeLink has been announced, the capacity of it just doesn’t seem any thing like sufficient to meet the needs of the future, let alone what additional announcements of VRE are likely for Southern NSW.
I am surprised that we have not had any announcements for a tender for one or more batteries to provide virtual transmission capacity on the existing transmission and plans for virtual transmission for HumeLink.
This might be improved by including EnergyConnect and observing if NSW would have been helped or hindered by SA demand / generation profile at that time with Eraring gone?
EnergyConnect won’t assist with the transmission bottlenecks between southern NSW (Wagga / Yass / Canberra) and greater Sydney. Humelink will ease those limitations but construction of that project hasn’t commenced yet.
It looks like there might be some issues with VNI flows since the Initial MWs were following exactly the same as Target MWs from 13:35 to 14:40.
Perhaps, AEMO has lost visibility of actual VNI flow reading, and NEMDE only could calculate combined demand of NSW and VIC but not individually.
With Initial being the same as Target, NEMDE saw/used the flow into VIC increasing = NSW demand decreasing and VIC demand increasing. The correct Initial value seemed to have come back at 14:45 with the NSW and VIC demand corrected.
I suspect batteries might prove the key to keeping the lights on in 2027. About 3GW/10GWh committed or anticipated for NSW, primarly 500MW/2GWh Liddell, 850GW/1.7GWh Waratah, 415MW/1.7GWh Orana & 275MW/2.2GWh Richmond Valley.
The class action covers a period before Callide C4 was wrecked by CS Energy Ltd “management failure” that turned off the critical 220V battery with the unit in service and the alternator motored to complete destruction.
The damage insurance of the alternator was voided by ignorance and the RTS insurance is self insurance.
Total cost maybe $500,000,000 The coverup continues!
All the above assumes an engaged and sophisticated energy consumer. This is unrealistic given the generally poor levels of energy literacy of most residential and small business consumers. The consultation overview states it is seeking insights from “electricity consumers (industrial, large and small-scale)” – not sure what their definition of “small-scale” is….. if they truly wanted insights from the customer classes that consume the bulk of grid supplied electricity why didn’t they just write “residential and small business” electricity consumers? Just another consultation to add to the myriad that electricity consumers and their advocates are already expected to participate in for free. No wonder Australia’s electricity system has failed electricity consumers…. because it has poor representation from electricity consumers at both a state and national level. So tired of the worn out phrase of we put consumers first. Nothing could be further from the truth.
Just on constraints in that forecast, there is a second relevant constraint set due to a current outage on Liddell-Tomago line 82 (set N-LDTM_82). This isn’t due to finish until 16:00 on Monday so overlaps with the Avon-Macarthur line and Kemps Creek transformer outage set (N-X_AVMA_KCTX) highlighted in the article. The joint effect of these outages would be significant limitations on flows into Sydney from both the north (QNI) and south (NSW southern generators and Victorian imports). There is no way known that AEMO will permit the Avon-Macarthur / Kemps Creek outages to begin on Monday. So the forecast currently looks worse than it will really end up being (unless there are generator failures between now and Monday).
Looking at that chart, there’s no denying that the peaks and general grid utilisation these days is slightly lower, or possibly the same, as when compared to the grid back in its heyday, <=2010.
And that's with an extra 4 million people in Australia since 2010, presumably all using electricity.
When I see terms such as “extreme weather” and “catastrophic fire danger” used to describe a day which perhaps MAY be the hottest in just 5 years, the only thing that gets hotter than the projected 45 degrees, is my collar out of a sense of frustration over the increasing trend to invoke such hyperbole when discussing weather.
a/ large scale wind and solar being below forecast doesn’t directly affect market demand – they are part of the fleet that supplies that market demand (except for non-scheduled plant which is a small fraction of wind/solar).
b/ but large scale solar availability being below forecast could well correlate with small scale rooftop PV also being below forecast – this does directly affect market demand, as explained in Paul’s post.
As an (irrelevant to the topic) side bar, the pronounced saw tooth profile of the AEMO small scale solar is interesting. Any idea what gives rise to this? Given these are <100KW and the sun rises continuously, one might expect a smooth profile.
It’s just an artefact. The solar data is a half-hourly estimate. When superimposed on generation data with 5 minute granularity, the sawtooth pattern results.
One more key takeaway Paul – if we add rooftop PV estimates to operational demand to get underlying demand (ie a measure of total power consumption from all sources), then yesterday’s half hourly values peaked at about 10,600 MW, higher than market demand in Jan 2009 when there was no rooftop PV to speak of, and almost certainly higher than peak underlying demand in Jan 2014 when there might have been a hundred-odd MW of rooftop PV output, if that, at the time of peak market demand.
Since yesterday was not the culmination of a multi-day heatwave here in Victoria, this does point to a material rise in underlying peak consumption – not surprising given population growth etc.
Can someone point out which column in ‘Figure 4.9’ represents ‘Liquid Fuel’ as appears in the AEMO dashboard for South Aust? FWIW, another name for ‘Liquid Fuel’ would be?
The methods and assumptions document published alongside the report (section 10) notes that “share of fuel type analysis included only coal, gas, battery storage, hydro, wind and solar. Other fuel types such as liquid and biomass were excluded.”
MTGELWF is connected to the grid at a sub-transmission level, on the Powercor distribution network.
It has a permanent 100MW export limit, imposed by Powercor as a condition on the connection agreement. This was lifted from 66MW sometime in 2023.
Two additional important points on this:
– the constraint is invoked because of a pre-existing planned outage of Marulan-Yass 330kV line (Transgrid Line 5), scheduled to run over 11-19 Jan. This is the same outage that led to some very high prices in NSW and Qld in early November 2024.
– when the NRM constraint clamps NSW -> Vic flows to zero, it’s true that this forces down generation in southern NSW. However this doesn’t reduce net supply to meet demand in NSW, because any higher generation in southern NSW would simply flow south into Victoria – there is a hard restriction on flows from southern NSW towards Sydney because of the Line 5 outage.
A very informative article Oliver,
I really enjoyed reading all three.
No doubt I’ll read again.
However, I have one query
“Players will bid in their contracted level of output at SRMC.”
I can’t see any evidence of that in MMS Bid-data.
CFPSs tend to bid around $-900/MWh up to 80% of available capacity. which is $950 BELOW their SRMC.
My understanding is they do this to secure volume to dispatch to fill their (mostly base load and peak) contract swap positions.
During the day when the spot price is below their SRMC, they have a bit of fun where they shift volume into higher price bands to intentionally run “short” to boost day-time revenue.
(I also imagine they realise some benefit of consuming LRCs as well)
Then during peak/high wholesale prices they outbid the other thermal plants and run “long” with their 20% or so marginal capacity to boost peak revenue.
Certainly seems to be a big incentive for them to invest in ramping capacbilities.
It’s Win-Win happy days for them.
My view is that Contract positions along with Gentailer control of the few powerful scheduled assets seems to be the biggest driver of bidding behaviour.
The underlying cause looks like a sudden drop in northern NSW large solar output, particularly at New England Solar Farm – a “system strength” constraint out of nowhere dropped its output target from ~370 MW to zero at 1400. The N::N_MNYS_2 constraint also played a role because the only NSW generator that could ramp quickly enough to cover the drop – Tumut 3 – sits on the wrong side of that constraint. I half suspect a rogue input to the system strength constraint could be the real issue.
Some further digging shows the system strength constraint came on because of a forced outage / trip of a transformer at Muswellbrook in northern NSW that took some QNI-related lines out of service – that’s the fundamental villain of the piece.
Hi Allan & Paul, thanks for the commentary. I’ve seen this happen before when TUMUT3 ramping upwards causes the NSW interconnector to flip 180 degrees (once was a couple days ago, once on 21/11/24). As touched on by you Allan, I think because TUMUT3 as a large positive term on the LHS of N::N_MNYS_2, N::N_2_5_2 (among others), when it ramps up, NEMDE seems to choose to flip the interconnector (force it’s target flow negative), as opposed to constraining other generators on the LHS? It seems unusual to me but intuitively, given the congestion around Marulan-Yass, energy from TUMUT3 has nowehere else to go but southwards. To make things even more confusing, the price setter for 14:05 shows TUMUT3 price setting with their -$1000 bid… but for an increase in -139MW. What is the price setter file implying here? that to meet a marginal increase in demand, the marginal cost arises from the cost of TUMUT3 turning off? Thanks guys
Update, Jan 2025: AEMO has decided to increment the maximum requirement volumes for the VF Raise (R1) service from 425 MW to 500 MW from 1:00pm (AEST) Monday 20 January 2025.
Unless you want to vacuum up LRCs to get your Green Stamp, I really cant see any value in Renewable PPAs
which for buyers have anti-hedging characteristics.
Looking forward to your analysis on the benefits Gentailers get, two-shifting their scheduled assets and the effect on the value of independent retailer contract positions.
Batteries anyone?
QLD was too slow to install batteries, even just to provide grid stability services, let alone bulk storage to help with evening peaks. Today QLD really needs more batteries than they have or have been announced.
It will also be interesting to see how much the lack of transmission through Wagga Wagga makes this situation worse.
Actually, at the time of peak demand*, rooftop PV contributed nearly 2,800 MW of output Paul!
* That’s underlying demand of course, which hit something like 12,335 MW. Due to the high penetration of rooftop PV it’s inevitable that it won’t contribute much at the time of peak market demand – because market demand will only peak when rooftop PV is nearly done for the day.
Hi Paul, in reporting the peak loading, you need to factor in the administrative load curtailment in effect (which has an associated economic loss) in order to obtain fair comparison of events. Over-loading the grid is one thing, having to shut down the economy is quite another cost!
I must be out of touch as I have never seen QLD demand (10,897MW) exceed NSW demand (~8,175MW) by 2,722MW.
It would be interestung to see a monthly peak demand comparison between these 2 NEM regions over the past few years.
Thanks for this. Many had been wondering if PEC1 had actually started operation, rather than just doing some testing. It would appear that PEC1 combined with the transmission restrictions around Wagga Wagga is going to make it very difficult, if not impossible, for SA to increase their percentage of RE in their grid, until PEC2 comes online. Even with PEC2, progress in SA’s percentage RE may not occur until transmission problems around Wagga Wagga are resolved.
So the TL;DR is that imports to SA will be higher at least sometimes when more expensive and SA large solar (marginal producers) will be curtailed more? A followup on this when there is more data will be interesting.
I thought PEC included phase-shifting transformers to allow the energy transferred on the link to be controlled? Is that only going to be actively used once PEC2 is commissioned?
I sort of agree but it’s not a data error. A 0.12 hz frequency swing on a 26GW system is a bit under 900 MW based on 4% droop and a bit of vagaries in hydraulic governor systems.
It is a data error – go plot the frequency vs demand. Over-frequency sitting at edge of PFR band from 16:35 when NSW demand starts going crazy, breaks through PFR band around 16:55… and then starts dropping back to normal over course of 17:15 -> 17:20. Ie AEMO had a data/forecasting error such that they were over-procuring generation earlier in the afternoon, and the -1000mw demand drop was taking it back in line with actual demand – over that dispatch interval the surplus generation situation reverts to normal, with frequency dropping back to ~nominal.
Pretty much spot on HR10 – between DI 15:55 and DI 17:15 something screwed up AEMO’s calculation of NSW actual demand from metered generator outputs and interconnector flow levels. You can verify this from the relevant MMS tables. Between 15:55 and 16:30 the calculation was too low, then up to 17:15 it was too high. The problem got fixed for the 17:20 DI which explains the sudden “demand drop” for that interval. These errors correlate reasonably well with frequency being on the low side of the PFR band between 16:05 and 16:30, then on the high side until 17:20.
That’s the highest Victorian 5-minute TOTALDEMAND recorded for any non-working day – more than 300 MW above the previous high on 28 Jan 2018. So it’s consistent with the underlying picture of demand growth revealed by the Q4 record Victorian demand we saw on 16 Dec 2024.
A roughly 4pm there was a facebook comment on meuqld facebook page saying that they had come to an agreement. I suspect up until the agreement they were running the units at min generation in goodwill?
I think we wait for the investigation. Vestas quote a withstand of around 180km/h for this model of turbine. Either there was a specific issue with this tower (the failure seems to be at the top lift joint of the tower) or it will be a very localised wind event beyond its design capability – not unusual in thunderstorms. The photos clearly show the proximity to other turbines that survived – though perhaps not without damage. We have seen the enormous power of swirling winds on transmission towers in recent history and this is likely no different.
Thanks Roger … you must have better eyesight than me (and or saw better quality photos) to see a flange. What I saw was just what’s above in the article.
The top approximately 10 metres or do of the supporting tower appears in the ABC’s photos. It seems, as commented above, that the bolted joint failed. This structural failure makes me wonder whether it could be a common fault across a specific design, or perhaps a construction fault due to a single contractor’s work.
Presumably we will soon know the results of the investigation.
Unless lightning struck a blade whilst rotating & caused a massive imbalance causing a failure of either 1) the boss that the nacelle bolts to OR 2) a tower segment weld. Many questions – was it feathered & locked (braked) at the time or still operating ? The pic of the top of the tower whilst ‘clean’, still not detailed enough to see of 1 or 2 above. Nothing in the wreckage to indicate either (on the photos I’ve seen) – what looks like a nacelle bolt pattern I think is the root end of a blade. But, if the winds were only at the speeds mentioned, unless a blade did indeed have an utterly catastrophic failure, that’s a worry that the tower has dropped the nacelle. That 1 & 2 are back running seems to point to it not being a fleet tower issue and maybe a blade has shattered and smacked the tower. The root cause analysis will be a good read.
The coverup continues, CS Energy wrecked Callide C4 turbine by switching off the critical 220vdc supply battery with the unit in service, the unit operator was left in the dark and the turbine was destroyed. What happened to the interlocks that have used for over 50 years with critical 220vdc batteries
AER should be paying the $500,000,000. lost because they allowed this to happen by not having procedures in place to control such process procedures.
Guys, I realise you do not cover Western Australia to any extent however your assumption re rooftop PV are spot on. I suggest you start using the BOM satellite cloud maps and the answer will become obvious. We see enormous rates of change in operational demand on hot cloudless days as the sun sets (gets worse as we move past the summer equinox – remember we do not have daylight saving in WA) and the evening peak arrives. If you get a hot overcast (hence humid) day in Perth then the operational demand comes up in the middle of the day and keeps growing. These become big energy days for conventional generation and as we back out of coal, we end up in more gas than can be delivered so hence into liquids. Batteries are there but if there was no spare solar / wind around to charge them it simply implies more fossil primary energy. Train wreck arriving soon!
Portland smelter has two potlines so when this happens you’ll see roughly half then no load (like you see in the charts). You can change power usage a little but there’s a bit of a window to operate in to be make aluminium and not just heat. The Dec 16 outage looks pretty long for an aluminium smelter, probably close to the max they would allow. Based on the load dropped then returned it was one potline for a bit, then the next pot line for a bit with some overlap – so roughly 1 hour per pot line.
Alcoa Portland Aluminium does do RERT – not sure if AEMO activated it then though?
Thanks Michaela, that looks right. Seems one of the potlines draws about 300 MW and the other a shade under 200 MW. The load profile on the 16th fits with the smaller one being tripped for one hour, then the second larger one going off roughly half an hour later, also for a one hour period. As you point out, they probably wouldn’t want a line off supply for more than an hour as they get more temperamental to restart the cooler they get. Allan
PS RERT definitely wasn’t called on any of these dates.
The first point to consider is that making comparisons with 2008 is fraught with difficulties. For starters, this article overlooks the massive reduction in Victoria’s energy intensive manufacturing base. Brand names like Holeproof, Dunlop, Hard Yakka, Fletcher Jones were all Victorian manufacturers in 2008. Ditto intensive energy users like Ford, Holden, and Toyota have subsequently packed up shop. Another big load that has disappeared is the Port Henry smelter at 360 MW. A additional chunky +40MW load is the Altona oil refinery that has closed since 2008. Further, there are all of the upstream suppliers, component makers, and contractors that have subsequently disappeared. The list of loss of manufacturing business goes on and on. Green activists look at the reduction in energy usage and jump with joy. I look at the reduction in energy usage caused by loss of manufacturing and see an economic calamity unfolding in slow motion.
The second point to consider is how the network is having to do double duty boomerang parenting for solar PV owners when a bit of cloud comes across on a hot summer’s day. Customers are quite happy to install a solar PV system, hold up the middle finger to networks by completely undermining revenue when the sun is shining. But then, like wayward children, expect to be able to return to the network in force on a hot summer’s afternoon, when clouds come across or the sun goes down, and demand there is network capacity spare for them. It is the equivalent of an _infrequent_ customer expecting Qantas keeps seat 9C free on every 737 in their fleet for them, on the off chance that they want to book a flight sometime in the next few months. In the real business world, this practice just doesn’t happen. Yet politicians and regulators are encouraging this behaviour. Having idle network for just a few hours every few years is inefficient CAPEX in the extreme. Perhaps it is time for a new rule that specifies a domestic battery of a certain size is installed with every solar PV system.
Making comparisons with 2008, 2023 or any other year since NEM start isn’t “fraught” at all. It’s just data. If you want to get into reasons for changes, let alone your own personal views on their merits, that’s your prerogative but it wasn’t the point of the post.
But on your last point, why wouldn’t we also make a rule to require purchasers of (say) air conditioners or electric vehicles buy a generator of a certain size?
Then I have missed the point. There was a “remarkable” conclusion made that population had grown by 30% but peak demand hadn’t. Yet, the analysis overlooked “the why”. Surely a big part of the “why” is the large base load and large industrial consumers that had closed in Victoria.
By contrast, the population in Queensland has also grown by about 30% over the same period. Yet its system peak demand has grown by about 4GW. Obviously, a big factor is the increase in base load due to LNG and coal mining, the upstream suppliers in machinery and manufacturing, and the population growth who assume comforts like AC are a mandatory in the home.
As for the point about air conditioning and electric vehicles, on households without solar PV systems, they are generally paying their fair share. Particularly AC units that have long daily run times. (AC that only come on for a few hours per year are a burden.)
From a grid perspective, EV on TOU or load control are one of the best customers have around as they suck up electrons in the middle of the night to improve network utilisation when it would be otherwise idle.
While the wholesale electricity market is an important part, 45% of the electricity bill comes from TUOS and DUOS expenses. Declining network utilisation (which is a function of short and spikey peak demand) drives additional network CAPEX that everyone must pay for.
If the frequency is wandering it is clear indication that there is not enough governors active on the system. For turbogenerators to have effective governors the machine must be above minimum load and below maximum load. From my experience as a AEMO Load Dispatcher (retired), AEMO needa to apply some constraints to get better regulation of frequency.
Governors respond in less than 6 sec and adjust steam flow to compensate for change in steam pressure (rise or fall) to keep turbine speed at 50Hz. Fuel flow is determined by Steam Pressure and Flow rate and is slower than the Governor response and should not change until the governor demands more average steam flow which takes more than 60 sec. So the problem is that System frequency is on a slow trend up/down which is causing the fuel flow to oscillate in response to the slow change in frequency. This will be found to be caused by largish slow load changes which were not traditional in load caused by consumer demand (diversity) but changes in large solar and wind generation caused by cloud patterns and wind variability. As solar and wind farms in common localities get larger the problem will get worse. Have you ever watched the output of your rooftop solar panel for a few minutes? You will notice the power level change by up to 30% for very little change in solar brightness. So imagine a 1000MW solar farm with some wispy high cloud passing by, the change in output can be around 300MW. This kind of change system wide on a diversified basis will become more noticeable as large solar farms in a common locality become more prevalent (eg. on the Darling Downs and New England areas)
Absolutely, Ian McG.
Trends to more variable (wandering) grid performance is a serious matter.
Meanwhile, records are being set and AEMO’s messages are about the past; mothing done to reduce “wandering” to historical levels.
This will end in tears.
This is an excellent article. Consumers used to ‘simple’ energy tariffs are sheltered from the reality of a highly volatile underlying energy market. This can only be done by retailer hedging or self-hedging (production). When the methods of self-hedging (coal fired power plants) disappear, then some other source or ‘supply’ for the financial market will need to step in OR retailers will need to pass on this increased risk (cost) to consumers.
Is this an opportunity for long duration storage (and a revenue stream for Snowy)??
Adding my appreciation to Aaron’s above – thanks Allan, this is a great piece. I genuinely had no idea that the shadow price associated with constraint violations could flow through to dispatch prices in this way.
One question: the parallel flow paths of V-SA (over Heywood and PEC1) seem unusually likely to drive this sort of situation where a constraint might have a lower-bound greater than its upper-bound, and would necessarily then be violated. It seems reasonable that this could happen even under relatively normal system conditions and cause violation-set price spikes out of nowhere.
Does this actually happen anywhere on the NEM? Or is AEMO good enough at their constraint formulation process to make these sorts of infeasible constraint combinations very unlikely?
Great info. I’d love to see an update on these charts with time of day for peak demand included.
Knowing what time of day the peak demand has occurred, from before rooftop solar was measurable to now, would give us a good idea of the real impact of rooftop solar on peak demand.
The problem actually started about 1pm. Note the slope on the operational demand curve starts to get very steep. A quick look at the BOM satellite images shows a nice bank of cloud pushing up the gulf then sweeping over Adelaide. The joys of uncontrolled, weather dependant generation.
It would certainly be interesting to perform another study on this since we now know the impact of distributed PV means yesrly grid demand increases are closer to 0%.
It is also interesting that the 2015 study mentions that 20 to 50% of revenue is earned in the top 20 days of the year, and now we know most of that is in the 5 hours between 4pm and 9pm.
Where does the number finally selected come from, do you know? Is there a financial model somewhere that tries to predict how often we’ll bat the MPC and how much revenue will acrrue during those outlier events and this is meant to be assumed that generators can bank on that equivalent to a capacity payment?
When I write that, it seems convoluted…but I’m not sure I’m wrong (definitely not sure I’m right either).
Is past wind performance a reliable indicator of future performance given the IPCC has said Australia will face slowing winds in the decades to come and the type of wind turbine chosen to be installed today is based on past wind characteristics?
I suspect that forecasting wind and solar separately is not the right approach. For example although wind output in May June 2024 was down 24%, all renewables were only down 13%.
Wind stilling is a genuine issue, if it is evenly lower speeds it is not a big problem, taller towers and larger rotors solve the problem. If it is extended periods of overcast twith low wind, more like a Northern European Dunkelflautte then significantly more generation and storage is required.
Up, up and away … Market Price Cap to increase by $2,800/MWh (or 16%)
Hi Paul, so what does that mean for consumers?
Also, big picture, what is your opinion as to whether the price of electricity, paid by the consumer, will be more or less under a 100% renewables scenario?
Cheers
Great article. One minor point:
The frequency diagram seems to have ‘low’ and ‘high’ back to front assuming the fulcrum and the horizontal bar are rigidly coupled together.
Supply up or demand down = frequency rise (not fall)
Supply down or demand up = frequency down (not rise)
Hi Robert, thanks for reading and commenting! I also took that view, initially. The reason I say initially is that upon reflection (in the course of writing the article) I realised the diagram represents the balance as see-saw. Adding supply (supply up), makes that side heavier on the see-saw, and that tips the frequency pointer to the high side.
“On one hand, proposals like the Federal government directly intervening to build more state-owned generation (including nuclear power stations) does not bode well for a more competitive market. ”
Really ? It certainly would. Its a new competitor capable of filling and offering hedging commodities and contracts.
Having a very low SRMC there is no real benefit in nuke plants two shifting to chase extreme pricing to fatten contract positions and rape the market.
Warren/Allan this is in conflict to many of your articles about hedging and contract positions.
With nukes I see a less volatile market, forcing cooperation between intermittent sources to produce competitive hedging services.
A less volatile wholesale price. Ah yes, Flow Power, with a Retail arm like Amber that profits from volatility by offering retail tariffs based on a function of the wholesale price, under the guise of their consumers producing a less volatile load profile than the spot market, and profiting from the difference.
Aha even more let’s check the references, there they all are. The Super Power Institute and Climate 200 crew with Simon HAC/Malcolm Turnball.
The Solar farms are probably being limited (down) due to Spinning Reserve requirements, and/or Fault Level requirements, that the system needs to maintain reliability, for such a case as expected reduced solar output due to weather conditions.
Sorry guys, can’t get my head around the graphs. The small solar is obviously some sort of derived number from perceived total load and generation and imports. Bass link is not clearly shown here. The abrupt drop in small solar (to zero) corresponds to your fall in “demand”. Faults (big ones) will shake off small PV due to voltage depression but not all of it. Will also shake off load (voltage held contactors releasing and also inverters on VSDs stopping but the load reduction does not seem to correspond to the timing of these reported faults. Wind and hydro had been moving around prior. What was Bass link doing? ie are these market pricing driven load changes or real time responses to system load changes caused by disturbances? I note there did not seem to be a lot of Hydro on at the time so I’d be interested to see what the system fault level was at the time. Big influence on clearance time to faults and voltage nadir during a fault (and hence load shake off). Sorry, not enough info here to pull this apart.
Hi Andy, thanks for the thoughts. Not enough time for more detail but worth noting that the NEMwatch snapshot was a real time view and, as such:
1) Shows the most recent estimates for rooftop PV … which, by virtue of the slower cadence of the data, is always behind the 5-minute updates for ‘InitialMW’ (SCADA snapshots) of the bigger generators.
2) A trend for Basslink Target Flows and limits is accessible in NEMwatch, just was not shown on that image.
When time permits in future, might post an ‘after the fact’ analysis.
Do you have to have an Australian energy retail license to participate in Very Fast FCAS (FFR)? I have invested in a community battery being built in Goulburn and I am wandering if, as shareholders, can participate in FFR with only a small and/or large BESS.
Hi Emilio, you have to be a registered ‘participant’, and there are different categories of participant to facilitate different involvement in the markets (including the VF FCAS markets). More information can be found on the market operator’s (AEMO’s) website https://www.aemo.com.au/learn/market-participants/electricity-market-participants
Thanks to the tireless work of Energex we had power restored two days earlier than first predicted. We are so grateful for there massive effort after cyclone Alfred.
One would imagine that these rules will quickly incentivise wind to add or at least contract with storage or even syncons to smooth their output. It may even make sense to add capacitor storage as 10-20% of expected output for 5-10 minutes would be enough to virtually eliminate risk over the next bidding period
You can really see the limitations on fault tracking caused by the lack of smart meters in QLD. The lack of data from smart meters means that Energex is only making guesses on where many of the outages are, and the related impact. Simplistically this can be seen by the many overlapping areas, but also by the fault zones not following street patterns. I looked up where I used to live to look at the outage impact and it was very clear to me that this was the case.
On your comment about volatility vs spreads, I’ve also noticed a lot of confusion on the topic – everything from questions about ‘typical daily volatility’ to revenue trajectory assumptions extrapolated from maybe 1 or 2 data points (e.g. the whole-of-station LYA trip).
I’m not sure if this is standard terminology, but I’ve started using ‘variability’ and ‘volatility’ as consistent terminology to separate periodic/forecastable outcomes (e.g. daily or seasonal cyclic trends) from unpredictable scarcity events. This has helped a bit in explaining some of these concepts to clients, such as why we might expect the incoming wave of shallow BESS to tamp down variability and short-duration volatility events but maybe do less to offset longer-duration volatility.
VRE tech needs to start to deliver profiles with hedging value for retailers
or consumers will continue to provide with higher charges.
We’re already seeing CFPS falling “short” of their contracts with failures,
chasing high profits two-shifting their contract positions.
When are VRE/batteries going to start offering caps or even a new evening peak style hedge commodity?
They are still not prepared to take the risk.
oh well. I guess my retailer will be paying inflated prices for base and caps and I’ll be paying.
Kinda makes state run baseload nuke plants with a very low SRMC, who’d would not be interested in two-shifting their hedge position,
quite appealing.
The LGC concept is fundamentally flawed once renewable energy sources become a large enough % of the market for them to end up being curtailed. If a business enters into an agreement with a counterparty building a wind farm and said windfarm ends up being curtailed for whatever reason (for example if lots of wind farms happen to have lots of wind available on a given day and all end up being partially curtailed) then the total number of LGCs produced over the year may well fall short of what was contracted. Now you can argue the generator should have to procure, but this will be a commercial nightmare. Renewable energy proponents only want risk free (take or pay) market position with an offtaker to keep their bankers happy. The only guys able to do this are the big energy retailers who have balance sheet capacity to underwrite this sort of risk. The cliff in 2022 was the end of the RET bubble – stuff signed up to before COB 2020 all came on line and that was that. This is the problem with intermittent generation – if it is not there then industry is in a hole looking for energy and when there is a glut it may well get curtailed – so industry doesn’t get the LGCs it wanted at the end of the year. Smart thing might be to just get rid of semi scheduled generators and make it such that the generators do whatever they want or need to do to become scheduled generators. (Effectively this is already happening with so called solar hybrid projects.) They then contract accordingly. Trouble is I suspect the resultant LRMC numbers will be ugly so the projects still won’t work.
It is going to be a very complex market to model in terms of estimating how much energy a particular battery might have left in storage, and how much capacity there is in the transmission network connecting these BESS.
I don’t believe there will be that much more competition in the evening peak market by 2027 and certainly not 2025. Fossil gas and hydro are still going to set the price during much of the evening peak, even while fossil gas use drops significantly. A major problem is the transmission networks, not just between states/regions, but within states/regions. We will increasingly see cases where there is capacity and energy available, but the transmission network is operating below its rated capacity, so these batteries can only have limited impact.
SA hasn’t seen as much benefit as should have from the additional BESS in the last 12 months due to a lack of competition.
Have they considered a possibility of being able to feed into the grid from under-utitilised home battery systems?
Say I install a 20kW home battery while my kids are still living at home and staying up to 1am playing their PCs and running aircon, etc, but after they’re moved out, all that battery capacity would simply not be needed. Is there any way to tie that back into grid export?
Including Callide C3, that appears to show that the entire coal fleet of Australian generators had an equivalent forced outage rate, EFOR, of ~5% for the period in question.
On this occasion, when the clinker detached, it resulted in a pressure spike within the furnace. Our initial assessment is that repairs are required to cladding and lagging on the upper section of the boiler.
Pressures spike?
Is this technical writer gibberish for tube leak?
Was it a furnace loss of ignition?
It would appear that the trippimg of the 500kV line caused the trippimg/off-loading of an aluminium potline smelter at Portland which in turn caused a drop in system demand and thus a rise in frequency. When did the smelter return to service?
Hi Robert, correct – looks like it was off-loading rather than a trip at the smelter. From this morning’s data it looks like the smelter began its return to normal loading just after 08:45. Noting that time is before the line was officially returned to service (as per the market notice, 09:48).
The link to the report appears to be broken. I came here after watching a recent Youtube video by Zoe Hilton then going on a search for the report she refers to. No luck so far.
On this occasion, when the clinker detached, it resulted in the release of steam from the ash conveyor water system. The released steam extinguished the flames of all four fuel pulverisers (mills), prompting an automatic unit trip, in accordance with the unit’s safety design.
Is this more technical writer ignorance, and another CS Energy coverup?
Was it caused by oil torch ignition by the unit operator after the loss of the PF mills?
Good article – although I question some of the conclusions drawn around the role gas CAN play or rather the assumptions used to draw these conclusions. What would be interesting to know to qualify some of the conclusions drawn in this article are the following:
– How much additional gas could be brought into the east coast gas system by increasing gas exploration? Given Victoria’s moratorium on gas supply – one would assume a decent amount without considering the additional gas production potential of QLD and NSW that could be acheived from further exploration. How far would this additional gas, combined with a gas reservation policy (which in my view is desperately needed and stupidly never implemented) go to meeting the gap?
– Taking into account the above, how much additional renewable energy would be needed to then fill the gap over the next 15-20 years until nuclear is available? The statement above that the Coalistion’s plan assumes that renewable energy industry will collapse in “a few years” seems overdone – what is this based on? If it’s simple the frontier economics report, well we know that is just a model and is probably wrong just like the ISP is probably wrong.
– My gut feel here, it is critical to continue to accelerate renewable energy development, gas development and start nuclear development to meet our future energy needs – the politicisation of energy is preventing thisand I think its risky to draw a conclusion that gas supply won’t be enough and nuclear will never be built. This is self fulfilling rhetoric that doesn’t focus on solutions to the problem, and adds to the polarisation of the debate. This article, without saying it seems to imply that the only solution is full scale renewable energy deployment – which while I agree this is part of the solution, discounting other potential parts of the solution (a.k.a gas and nuclear) is risky rhetoric, particularly given the massive uncertainties we face over the next 50 years in not just energy markets but international markets and politics.
But as the research also says, “operators often make strategic decisions to scale back asset management and capital investment in ageing coal plants, opting against major upgrades due to their limited remaining operational life and the shortened window to capture a payback on upgrade investments”.
If the planned closure is pushed back, won’t that likewise delay the onset of this ramp-down in maintenance, thus deterioration will be similarly delayed?
Great to see some knowledgeable analysis of the event, its so difficult to decipher the media reports! Please keep us updated if more information becomes available. I’m fascinated to also know how their blackstart process was supposed to respond and why its taken so long to restore power.
There is so much going on in that video. I had to watch it multiple times.
It really shows there are far more regular constraints than I was aware of. I wasn’t aware of problems in NE Tasmania.
It also shows how large and wide spread the transmission problems are in NSW. It will be interesting to see if some of these new batteries assist with this issue. Given I haven’t heard of much in the way of grid forming inverter batteries for NSW, I suspect the impact of the batteries will be some what limited when it comes to transmission problems.
It has been reported that they had 5 blackstart units at hydros across the country, but three of these were on outages or breakdowns. If true, that implies their contingency grid management wasn’t the best.
This is just another example where electricity markets ignore financial opportunities for players to make bids out of what policy makers imagine to be a physical forecast. It isn’t entirely unusual to find rules that effectively bet market participants will ignore good financial decisions and where, when they don’t, things go awry rather than self-correct. This isn’t a good design strategy. A well designed market is one where players operating in their self-interest coincidentally operate to maximise social benefit (thank you, Adam Smith).
Another example where this ‘mistake’ occurs is the day ahead market in the US. One of the primary selling points of a day ahead market is that it forces market participants to submit bids that best approximate their output and demand in each trading interval of the following day. This provides an incentive for participants to spend the goldilocks amount of money on forecasting because they have to pay or be paid for quantity differences in real-time at real-time prices.
However, it turns out that wind farm owners are better off deliberately bidding below their mean output forecast day ahead because of the financial effect of the curve in the supply stack (they lose less money if they overbid than if they underbid). Their actions cause the day ahead schedule to represent an optimal financial position and to diverge from a physical schedule, which upsets the market/system operator and policy makers no end who were hoping to kill two birds with one stone (bid).
Great article, thanks Dan. My feeling is that the balance is a little bit off here. I think the MPC should be lower, but the APC should be higher.
Main reason is that I would like to see policies encourage more consumers to participate in the spot market. I view the high MPC as a deterrent for consumers exposing themselves to the wholesale price.
There have been several instances over the past few years where the MPC has been hit due to local transmission constraints rather than genuine supply scarcity. Whereas the triggering of the APC would be more linked to genuine supply scarcity or energy resource issues.
Hi Dan – I think you’ve covered a lot of ground here and that is to be commended. However, doing so can tend to trigger some conclusions and convictions in people that are shaky when you consider them in more detail.
Here is just one example. We need to understand how economists differ on the matter of market power to better understand the problem. US economists tend to use SRMC and variable cost interchangeably and any differences are considered an abuse of market power (they tend to ignore the presence and operation of hydro generation for some reason). Efficient pricing should account for opportunity cost but this is often misunderstood.
Market power in the US is considered such a scourge that even relatively recent attempts to allow generators to offer according to opportunity cost of fuel in ISO-NE have failed. Similar attempts to allow small generators to vary from variable (fuel) cost (“small fish swim free”) in Ercot also failed, which is why they adopted an Operating Reserve Demand Curve to ramp up prices when generation becomes scarce. This takes places despite the fact that consumers and society is better off if scarce fuel stores are priced according to opportunity cost.
Also, the incidence of very high spot prices (ie above $10K/MWh) needs to be considered in the context of the event that caused scarcity to occur. They are often due to sudden and unexpected events and this is where rebidding rules and trading behaviour needs to be well understood. I’ve published articles on Watt Clarity about this issue and which rules need to be reviewed.
Thanks Greg — very useful points, especially the context around how different markets handle market power. Appreciate you sharing the examples; it’s given me a bit more to think about for future analysis.
CS Energy was refused a payout of Insurance in May 2021 because of ignorance when the critical 220V C3 battery was isolated without the unit operator being informed! How will EMG help?
Maybe a Royal Commision would!
“We’ve not delved further into why intervention pricing was so high in the 2017-2019 years in this article, and whether it relates to the directions for system strength gaps in SA from late 2017 onwards”
If the NEM at the time was about 21GW then yes, the bump is about right for the loss of 300MW odd. Interesting the relatively slow recovery. Must be about the deadband setting. I thought the magic batteries were bidding in for small signal raise????
Perhaps check out “veracity” vs “voracity” although perhaps the winds were “voracious”.
Good to see this result, and a safe prediction is that the continuing commissioning of a number of large wind farms means that new peaks are likely over the next few months.
Wouldn’t it be simpler to require semi scheduled generators to become scheduled generators by having (say) 30 minutes of storage behind their network connection so they can actually comply with their offers to the market? I say 30 minutes because this is probably enough time for a gas turbine or hydro machine (the most likely things needed for a big block of additional energy) to be dispatched to fill the hole when they bid down (or out). Surely just putting the risks and costs of non-performance where they belong. If they want to contract with a third party to provide the service to them great – but no double dipping.
It has dropped back down to around minimum load with prices dropping below $0/MWh!!
Old coal fired units need to operate as constant designed base load units,to run them up and down will wreck them as they approach the end of designed life.
The coverup continues?
Now it’s 50% Sev.en Global Investments and 50% Queensland Government.
Who is paying for the repairs to Callide C power station as there has been no insurance payout for the loss of Callide C4 turbine when the critical 220 DC supply was isolated with turbine in service and the turbine motored to destruction?
VP6 was clearly taken off “hot” ready for restart. If you bring the unit loads down really low before desynchronising the throttling losses chill the incoming steam to the turbine which is hard on the machine in terms of stress management on thick walled components. Better off tripping from a higher load as this will not cause through wall temperature differentials that Toshiba machines don’t like but still sufficiently cool to be a sensible target temperature for restart.
MP1 clearly tripped over its shoelaces whilst unloading. MW came off pretty quickly (but not dead vertical) so probably some form of protective trip.
I have successfully installed a 2kVa ‘run of the river’ micro hydro system for an individual off grid (at that time) resident downstream of the Tinaroo dam in NQ. Unfortunately the 1980’s flood washed everything (including the house) away. The principle is very sound and could be repeated. Where can I find an organisation to discuss this idea.
John Pietila
0421 240 241
Brisbane
I have just been exposed to WattClarity and I can see I now have a new best friend in the Australian Energy Space – great articles, awesome raw and analysed data and a treasure trove of information. Thank you.
Certainly expensive in Victoria, after a cool cloudless day, those with home batteries and wholesale pricing must be making a mint.
In terms of generation, hydro is at 5.4GW batteries are at 1.3GW, wind is just above 900MW. However brown coal appears to be lacking. Unit 3 at YWPS is out of service and Unit 4 shut down mid afternoon. Unit 1 at Loy Yang A is also out of service, while unit 4 is operating below normal capacity since mid afternoon. Both units at Loy Yang B have also dropped in output coincident with the high prices, which they also did yesterday. Cannot think of a physical reason to be down by 30% in mild conditions, perhaps there is a financial or FCAS reason?
The AEMO has a report on this incident, root cause was one phase of a circuit breaker failing to open while taking a generator out of service. This tripped YPS No.1 220kV busbar and as a side effect, YWPS unit 2 lost auxiliary power supply. Some screws holding the trip coil in position were found to be loose on the phase of the circuit breaker which failed to trip. The faulty circuit breaker was an Ausnet asset.
Victoria as of writing is sitting at the market price cap, the other states are not far behind. It appears the shortfall in capacity is in Victoria and New South Wales, with 1100 and 550MW respectively of demand supplied from other states.
Thank you for writing this thought-provoking article. However, I’m struggling to reconcile the idea that BESS traders and their auto-bidders are simply making poor decisions and consistently losing money. The paper you linked mentions that the provision of contingency FCAS costs $10/MW/hr, which raises a couple of questions for me regarding their methodology:
a) Could the BESS be holding other underlying financial contracts? For instance, is it possible they are strategically withholding capacity as part of a broader hedge against wholesale price movements? Do we have any visibility into the terms or pricing of such contracts?
b) How is BESS Capex impacted by charging and discharging patterns? Is this accounted for in their analysis?
It would be helpful to really understand the bidding behaviour of BESS so DR participants can identify and exploit rare opportunities when they arise.
Liam. Loy Yang likely operating fully hedged or “short” with a low-cost contractor(s) filling their residual contract margin.
Contracts are complex and obfuscate apparent operations. Wholesale market price often gives very little indication of direct commercial outcome.
The Spot price is more representative of bidding outcomes based on underlying contracts and contractual obligations.
Thanks for your comment. I agree, it’s unlikely to be a case of poor decision-making. My article was intended to draw attention to the consequences of whatever is driving this behaviour.
a) Protecting a 3–4-hour hedge seems like a plausible explanation. Unfortunately, the NEM is not ‘perfectly competitive’ in the economic sense, as details of such arrangement are not, to my knowledge, made available to all market participants. I’m also not clear on how modern tolling arrangements might result in this behaviour.
b) Battery replacement costs are certainly a likely factor. That said, with battery prices continuing to fall, most analysts would argue that it is best to maximise revenue in the short term rather than conserve capacity.
Greater transparency around out-of-market contracts would help all participants understand how these deals influence market outcomes.
Kind regards,
Paul M
I agree with the conclusion that the story was largely a beat-up. Your chart shows that average daily Vic gas generation so far this year is below last year.
As to whether the peak would have been higher with more gas~fired capacity? Even at the extreme prices of last week there was some capacity undispatched, because it was only offered at even higher prices.
You are right.
It was another beatup. There was a severe wind drought for three days and poor solar as well but gas peaked at about 75% capacity and averaged 26% for the week. On most days where gas generation is required, 2 GW/5 GWh of storage would displace it entirely. Further as wind and solar capacity increase coal and hydro utilisation will decline during normal operating conditions, leaving even more capacity to fill the gaps in wind and solar
I can see why there is an argument for geographic diversity of supply in provision of these services (assuming this better ensures power system security than geographic concentration).
But I struggle with the concept that ‘technology diversity’ is required. These contingency services, particularly very fast frequency response, are very inexpensively provided by BESS (in pretty much the same way that provision of inertia used to be effectively costless in a fully synchronous system) – I don’t see any good argument at all for forcing inclusion of higher cost FFR technologies into the supply stack.
The article seems to be arguing that we won’t have enough of these services at some future date if economic competition drives DR providers out of the FFR markets, but given the level of investment in BESS, that seems a very long bow to draw.
Hi Allan, thanks for the comment.
Just to clarify, the article isn’t arguing that we’re heading for a shortage of FFR supply. The current low prices reflect an abundance of BESS-based supply, and projections from Modo Energy suggest this supply will continue to grow rapidly in the NEM.
The case for technology diversity is more about risk management. Each technology brings its own physical response characteristics and commercial behaviours. With the rise of algorithmic bidding and increasingly strategic dispatch from BESS, those commercial behaviours may become a system risk in their own right.
My argument is simply that it would be prudent to maintain some technology diversity within the FFR, and broader FCAS, stack. That diversity could be at risk if prices continue to clear at very low levels, potentially leading to over-reliance on a single technology class with tightly correlated incentives and response profiles.
It is interesting that overvoltage was suspected as the cause.
What are your views on the fact that there is increased solar weather activity recently (for the last few months) that could have induced voltage in power lines?
A similar event occurred in August 2003 in Northern USA and Canada, and this was also during a period of increased solar weather activity, although, according to Google’s AI, it was ruled out as being the cause.
In the higher latitudes, solar flares can alter the earth’s magnetic field which in turn can induce DC currents in transmission lines. These DC currents can cause a transmission line to trip out of service due to such currents appearing to the protection systems as earth fault currents. This has happened in Canada on very rare occassions. In Spain no transmission lines tripped until after the power system was in a system crash spiral.
In Spain the lack of power sources with voltage control incorporating stabilising circuits meant that steady state voltage oscillations
were undamped and grew to such an extent that over-voltage casued the tripping of 2 large solar PV farms that then led to the collapse.
The large amount of solar PV stations with Grid Following inverter systems that do not control system volatge but only control reactive power output according to a fixed power factor was clearly a problem.
Lots a batteries with Grid Forming Converters and voltage stabilising circuits should fix the problem.
David’s model is a copper plate model. As such, it could prove what is impossible; but not what is possible.
For example, if a battery capacity of X GWh in a copper plate model was insufficient to get through a particular low wind night, then it would be impossible for any real world placement of X GWh of batteries to service that night. But the fact that X GWh IS sufficient in a copper plate model to get through some night tells you close to nothing about the possibility of finding a suitable configuration to get through that night. David’s not an idiot, he knows this, but he still repeatedly misrepresents his model as being relevant to real world feasibility … it isn’t.
Used Co-pilot to translate and summarise in 1 page – could ask for a full translation if you want one:
Here is a **1-page summary** of the **non-confidential report from the Committee of Analysis on the April 28th Electricity Crisis** (“Informe no confidencial Comité de análisis 28A”):
—
### **Summary: April 28, 2025 Iberian Electricity Crisis Report**
On April 28, 2025, at 12:33:30, the Iberian Peninsula experienced a **complete voltage collapse**, disconnecting Spain and Portugal from the European electricity grid. This unprecedented event triggered a formal electricity crisis declaration under EU Regulation 2019/941. The Spanish government established a multidisciplinary **Committee (28-A)** to investigate the causes, impacts, and propose corrective measures.
#### **Key Findings**
– **Multifactorial Origin**: The blackout was not due to a single failure but a **cascade of events** involving voltage instability, system oscillations, and inadequate reactive power control.
– **Five Phases Identified**:
1. **Pre-event voltage instability** (9:00–12:00)
2. **System oscillations** (12:00–12:30)
3. **Generation losses due to overvoltage** (12:32–12:33:18)
4. **System collapse to zero voltage** (12:33:18–12:33:30)
5. **Supply restoration** (12:33:30 on April 28 to 14:36 on April 29)
– **Root Causes**:
– **Insufficient dynamic voltage control** due to limited synchronous generation and inadequate reactive power absorption.
– **Unusual oscillations** (notably at 0.6 Hz) possibly triggered by specific photovoltaic installations.
– **Overvoltage-induced disconnections** of renewable plants, leading to a self-reinforcing collapse.
– **Delayed or inadequate response** from some generators and infrastructure operators.
– **Cybersecurity**: No evidence of cyberattacks was found. However, **vulnerabilities and poor configurations** were identified in several digital systems.
#### **Restoration Efforts**
– Power was restored to **99.95% of demand by 7:00 AM on April 29**, using a strategy of **islanded grid segments** reconnected progressively.
– Support from **France and Morocco** was critical in re-energizing the grid.
#### **Recommendations**
– **Technical**: Accelerate deployment of voltage control technologies (e.g., synchronous compensators, FACTS), enforce compliance with voltage regulation standards, and improve grid observability.
– **Regulatory**: Update operational procedures, strengthen oversight of shared evacuation infrastructures, and enhance coordination with European partners.
– **Cybersecurity**: Implement NIS2 and CER directives, improve authentication, logging, and network segmentation, and expand coverage to smaller operators.
– **Market Design**: Reassess the impact of quarter-hourly market intervals and interzonal gate closure times on system stability.
—
Would you like a translated version of this summary in Spanish or a more detailed breakdown of any specific section?
Great article Linton. Appears that the count of forced outages may have stayed at the same level over the last 5 years but the EFOR % number has more than doubled over the past 5 to 10 years. Thats suggests that frequency of the events is steady while the durations of the outages are much longer.
Hi Ian, yes your thinking aligns with a feature Paul identified in 2023 when looking a the trend of unavailability. The trend was increasing and the recent extended forced outages were contributing to that.
Does the Δ300 metric risk excluding the ten Queensland 350MW Hitachi units if they’re operating at partial load? Gladstone runs 280 MW units, so is entirely excluded from this analysis.
Does a Δ200 metric risk including planned shuts or things like clinker deloads?
Hi Alex, you’re right. The measure would miss unit trips when not running higher than 3ooMW. And smaller unit, yes, wouldn’t make the cut. A Δ200 metric could be some thing we explore in the future!
Further to Linton’s reply Alex, we chose the Δ300 metric in order to make the trips large enough to feature more clearly in what Linton’s going to analyse in a follow-on article … frequency changes.
What I found interesting was all the asynchronous plant was supplying the grid at fixed power factor. They were using the synchronous plant on to do the dynamic VARs control but the grid wasn’t dispatching them to do it. – Their static equipment had all already been taken out in an unsuccessful attempt to lower the voltage. The report authors don’t seem to understand capability curves of generators, but that may be an auto-translate issue of word choice.
From my thinking, for consideration on this topic of frequency and voltage…
– Traditional turbines have more trouble turning when there is more load on them, making frequency drop.
– Inverters, theoretically, can continue to pump out 50Hz even when past max load, the voltage just cannot keep up.
– Flywheels that are not generating will follow frequency regardless of voltage, including traditional turbines at minimum (steam) pressure.
So in theory, couldn’t a grid that is mostly/completely inverters and flywheels have a grid collapse while the frequency is still within bounds?
And the graphs above showing that the frequency drops less on big generation loss seems to back up this theory.
It’s really quite amazing how widespread Aus weather conditions are-high or low wind speeds seem to occur simultaneously across half a continent; some of these farms are over 2000km from each other!
I never realised this until the advent of wind generation. Presumably applies to the nearby coastal waters also.
Various lack of reserve forecasts issued and cancelled for the first few days of next week. Hard to keep track of the situation from the market notices alone. I assume associated with forecast uncertainty provided by the east coast low.
High voltage and cascade failure, reads like too much reactive power. It will be intetesting to find out why this occured, as it depends on other components (Statcom, SVC, reactors, capacitor banks, HVDC converters …) as well as the generation sources. Transmission and distribution lines themselves can act as reactive sources, it may be there is insuffienct capacity to sink reactive power under the configuration in use that day.
The physical power system was under no major stress: there were no LOR conditions and reserve margins were around 20%. There was over 3,000 MW of available but undispatched (ie spare) gas and hydro capacity through the high priced periods in the NEM excluding Queensland (where prices were not so elevated).
So we didn’t get the extreme prices due to the power system running short of capacity, “wind collapsing” or for other hyperbolic reasons. They arose from a more involved and opaque set of factors involving the ownership of peaking generation in the NEM, the market contracting strategies of a wide range of participants, which physical capacity happened to be offline (and yes low wind is part of this), and how this all then played out through bidding behaviour.
This article exposes the vulnerability of the NEM to load shedding (or just as damaging for consumers high wholesale prices) when coal is retired and we have insufficient gas powered generation and medium and long duration energy storage. The Capacity Investment Scheme’s continued support of mainly 2 to 4 hour batteries needs a radical rethink before its too late. Wind droughts are real and could go for weeks. The wind drought in this article lasted for a day. Even with the NEM’s current coal and gas generation capacity, what would happen if there was a 5 day wind drought combined with low solar and moderate demand in 2 or 3 states of the NEM?
Jennifer, on what basis are you saying there was insufficient gas capacity? There was more than enough bid available, its owners just wanted more than $17,400/MWh to turn it on. Same for a big slab of hydro.
Hi Allan My comment states “exposes the vulnerability of the NEM to load shedding (or just as damaging for consumers high wholesale prices) WHEN COAL IS RETIRED and we have insufficient gas powered generation and medium and long duration energy storage”. My comment implies we got through the event last week because the NEM still has coal generation. But the Capacity Investment Scheme is incentivising 23 GW of more variable renewable energy by 2030 backed up by 9 GW of “dispatchable” clean energy. The article shows how 2 to 4 hour batteries can’t plug the shortfalls “as storage capacity became exhausted, contributing to price escalation”. The CIS needs to radically rethink its definition of dispatchable as coal is retiring and new gas generation is not included in the CIS.
Fair enough Jennifer – of course if we don’t have enough dispatchable capacity in future then reliability will suffer and prices will be high. And of course a range of dispatchable capacity types, not just short duration storage, will be required. No-one who understands power systems would disagree.
I just don’t see the connection to an event where prices went high for reasons other than insufficient dispatchable capacity – of which we had plenty on the day in question.
“Congestion, connection delays and security constraints. Because revenue is divided by output, declines in revenue due to such curtailments is worn by the investor. Again, this is necessary to encourage the investor to locate and operate in a way that minimises these risks.”
The documentation specifically builds ‘curtailment & congestion’ into the revenue calculation ie the project is allowed to claim MWh lost to curtailment.
It’s included in the calculation of the contract MWh –
Nominal MWh = sent out MWh + curtailed MWh, if the curtailment is from AEMO.
That says to me these CIS projects are protected against curtailment from anything to do with the network, which includes curtailment to protect the grid against minimum demand. This will naturally be the biggest problem for solar projects.
This fits well with the premise of the CIS – to massively overbuild wind and solar.
The fact that the wholesale price in Victoria was significantly higher in 2025 than 2024 means there is a management problem. 2024 was one of the worst years for wind conditions on record, so to have higher prices is really saying something went wrong.
“Several planned transmission outages”
I think it is time that the exact timing of these outages needs to be made more flexible. The cost of delaying these works by one or two days until wind conditions were better, might have been small compared to the additional electricity costs to consumers. Some one needs to run the numbers.
Is the limitation with VNI, or with transmission around Wagga Wagga? I was under the impression that it was the later. It is simply the reporting system that makes it appear as though it is VNI.
It seems all renewables were operating in constant power factor mode (general practice as per FERC and ENTSO-E in US and Europe is to control to unity power factor according one AI site) They did not participate in dynamic voltage control mode and hence did not do much to help with the overvoltages. The full non confidential Spanish gov report recommends that market rules be changed to allow renewables to be remunerated for taking part in dynamic voltage control. One wonders if the blackout would have been avoided if the renewables had been in this mode.
Don’t we want to know the frequency of events per numbers of units? If the number of units declines so should the number of events. Removing the closed units won’t account for that underlying trend.
Hi Steve,
There’s a multitude of stats that we could calculate. In the end, we aimed to understand whether or not we are seeing fewer ‘events’, compared to the past. Some drivers of fewer events are closures (but we eliminated that in the second chart), potentially longer durations spent unavailable, and potentially more time spent operating at low loadings (below 300MW).
In the next article (that Paul referred to) [ https://wattclarity.com.au//articles/2025/06/frequency-change-associated-with-coal-generation-events/ ] the aim was to see how these events impact system frequency.
Is work on the Tungkillo-Tailem Bend line the reason for a wierd AEMO dispatch interval? At 5:25pm EST the AEMO dashboard had Vic price $193.82, SA price -$19.97, NSW price $138.48. Both the SA and NSW inter-connectors had strong counter-price energy flows. From Vic to SA it was 620MW on the AC inter-connectors, which was at its limit, 15MW Vic to SA on the DC connector (at its limit), and from Vic to NSW it was 1088MW, again at its limit. Being close to 5pm SA time, work on the transmission line may have been finished for the week and the line re-energised, but as this capacity was made available to NEMDE it couldn’t find a stable solution.
I’m definitely going to have a look through the report, but i have noticed the problem nobody seems to address so far is the diminishing operational demand as distributed PV keeps increasing, and now the government is pushing everyone into getting solar batteries. I’m certainly not complaining about that, but surely that is going to affect the grid electricity price and ROI of any new large-scale grid-attached generation projects?
Kate Summers speaks a technical language suitable only for power engineers. However she speaks in such a wonderful and passionate fashion that even amateurs like myself cannot help but be entertained by. It’s no disrespect to her fellow panellists to note that her star shone more brightly.
Also of great interest to me was the session on REZs which was of course the session on transmission.
Suppose at Eraring we have an accident similar to what happened at Callide C4 in 2021. Do we really have the renewables and storage to replace this coal plant at full capacity? What would happen during a dark, cold winter week as just experienced end of July?
My website is on oil. But I got interested in power supplies because everyone assumes petrol and diesel cars can be easily replaced by EVs. (The real problem are actually our diesel trucks!)
“This distributed PV output offset underlying demand growth across the daytime hours and led to an
overall reduction in operational demand (Figure 6). NEM-wide operational6 demand averaged 21,760 MW, a
reduction of 151 MW (-0.7%) from Q2 2024.”
Looks like the trend continues.
It begs the question why would major grid generation projects of any type, green or otherwise, be attractive for investors when the opportunity for sales keeps reducing?
Perhaps it is simply the replacement of the fossil fired plant creating this need?
Perhaps the government also backstops dispatch targets?
Just on Waratah’s maximum discharge to date Paul, where you mention that it doesn’t include any Raise FCAS enablement volume – the battery isn’t yet registered to deliver FCAS, I guess that will come later in its commissioning programme.
Isn’t the question not whether the NEM can operate on 82% renewables by 2030 but whether ALL consumers, particularly the largest consumer group SMEs (now called Business Mass Market by AEMO), can afford their power bills from a NEM with 82% renewables. After all, if energy consumers can’t afford their power bills then the plan has failed the National Electricity Objective. SMEs don’t get government bailouts like large energy users (eg steelworks and smelters) and can’t even access hardship programs. Over 5 million Aussies depend on a small business (under 20 employees) to put food on the table and a roof over their heads. This is not just an engineering problem.
Re. KKGT – there appears scant info on their gas supply – contracted &/or spot, TJ/d?
Nothing on Snowy website, etc – they talk up the lowering of costs…?
It shouldn’t be too difficult because the amount of grid electricity generated and sold keeps falling every year, so the 82% target becomes relatively easier to achieve as time progresses.
But don’t ask how the rising cost of grid maintenance gets paid…
So it looks like the Basslink trip is the cause of the demand drop. Basslink was carrying ~330 MW southward at the time; there’s a Frequency Control Special Protection Scheme (FCSPS) set up in Tas to manage the risk of Basslink tripping. In this event the FCSPS immediately dumps an amount of Tasmanian load sufficient to offset the loss of imports.
The 292 MW scheduled demand drop is about the size of lost Basslink imports, indicating that it’s likely the result of operation of the FCSPS after Basslink tripped from 330 MW.
The annual report suggests that the STATE Government paid lOrigin between approximately. $230 Million and $3OO million which are availability payments and consistent with reports on State Government financial support.
Thank you (both) for publicizing this event, which is potentially (pun intended !) Impactful on Tasmanian community activities.
Dead silence emanating from public media and Government.
Strengthens the case for a new cable link to mainland Australia?
I am still waiting for a public explanation (by a responsible authority) explaining intemperate cable overload protection adjustments, possibly in response to desire for increased revenue, directly related to previous BassLink cable failure.
Thank you, Dan, for sharing such a comprehensive and insightful perspective on the unique energy challenges facing Timor-Leste. The contrast between reliance on fossil fuels and the urgent need for reliable, sustainable energy access really highlights the complex balancing act many emerging economies face during the transition. Stories like these remind us how critical tailored solutions are—both for improving quality of life and supporting long-term economic resilience.
This situation will continue while the political imperative of (uncontrolled) renewables at any cost exists. I think that every grid scale solar (and possibly wind) installation should be required to have a battery system capable of producing and absorbing power to limit the rate of change of power generation so that other generators have a chance to react.
As Australia’s push for net zero is looking increasingly unlikely (and too expensive), the prospect of extended use of coal should go way beyond trying to 2shift large coal fired units that were never designed for this purpose – progress to much smaller capacity units that can be more suited to allow dominance of solar power when available. These should not be high tech supercritical units, but basic design – with the cheaper fuel costs, super high efficiency should not be an issue (it never was in the coal fired units on which we still rely). What politicians will have the courage to go down this track – not the current mob! Best regards
Its interesting to note that Victoria’s aggregate electricity demand decreased from 52,976 GWh/year in 2002 to 46,885 GWh/year in 2019 in spite of a population growth from 4.8 to 6.5 million.In that period solar/wind generation increased from zero to 17% of total generation.At the same time average annual wholesale electricity prices increased from $35.44/MwH to $124.88/MwH which appears consistent with the increased volatility as shown in the graph above.
Would be very interesting to have a look at a longer period – before and after. There appears to be a frequency oscillation here with an approximate 15 minute return period prior to the one at 11.45am taking off. The exceedance occurred at 11:55 but it all started 10 minutes before that.
I feel sorry for all those who have to monitor frequency excursions and evaluate Unit response to prove compliance nowadays. Just about a full time job!
I am interested to know how much of a negative it is for Australia that we are working to such a failing plan. Not necessarily that high levels of wind and solar are unworkable, but that we cannot implement at anything like the touted rate. We don’t understand when coal plants can close, we don’t understand when we will need more gas fired peaker plants etc etc. A plan that is plausible would surely keep costs down.
I’m Ok with accepting predictions that Eraring’s closure is likely to be delayed. I’d prefer if they came with some detail of how much NSW is falling short of replacing it, particularly when the comment is ” … or perhaps in 2029 “. And what is the potential uptime and reliability of plant four years of more more after the originally planned closure? I think there is an emerging “real world” gap presupposing that it can indefinitely continue to ramp up and down and be relied upon.
The Elephant in the room is: Does the system have sufficient Inertia for reliable operation? If the large heavy generators are shutdown and replaced by Solar, Wind and Batteries which contribute little or no inertia because the inverters cannot remain working during severe frequency excursions, then the inertia required to contain frequency decline will not be there. Low inertia results in steeper frequency decline during loss of generation because there is not enough governors on the system to restore frequency or even to control the frequency at 50hz. The result, as we are seeing even at this early stage, is a wandering frequency and an inability to identify what equipment actually caused the frequency excursion.
“..4 different analysts quoted in voicing their concerns about the unrealistic notion that Eraring can close in 2027 … or perhaps even in 2029”
Bit of an exaggeration that 4 analysts are voicing concerns specifically about Eraring, given only Tom Allen is directly quoted in the article on this. From the article:
– Jarden are quoted as doing modelling of Eraring staying open beyond 2027 with no normative statements that this is what should happen.
– Neither Matt Rennie or Greg Elkins mention Eraring at all in the their quotes, just the 2030 and 2035 targets.
Is it possible to have an article that summarises minimum synchronous generator requirements across each of the NEM regions as they are current set, and how they are planned to reduce – assuming that is actually the case. Feel free to remind me I’m attempting to simplify a complex set of rules
Thanks for the idea Paul, it’s certainly something we want to cover so I’ve added this to my to-do list — though I can’t say exactly when I’ll be able to get to it.
Great work Paul and Linton in cleverly using the available public data to unpack the likely causes of these events. Very interesting and I think important towards understanding frequency resilience. Publishing this sort of investigation may plant the seed towards institutional reporting which can access more data.
A quick look back at combined availability levels for the affected wind generators shows that this constraint could bind about 10% of the time. The energy potentially curtailed might be something like 2.5% of available output. Not insignificant.
In practice the figures will be something lower because of other constraints and price-based offloading.
Great article. Brings some balance to the discussion. Too many are focused on the output of “aging and unreliable” coal generation and ignoring the limitations and transparency around VRE output… which is absolutely necessary when navigating the impact of extended wind droughts.
Nice work. I like to think it should be possible for market 4s data to feed and automatic viewer for frequency events to filter the larger causers and performers. A significant factor not being considered though in the frequency observation space is whether there actually is any controller steering frequency back to 50Hz because the mainland is actually at 49.980 or 50.020 Hz six times more often than the midpoint of 50Hz so whilst less >or< NOFB events are occurring the consolidation is starting to resemble continual slow variations between the slightly wider range than the AEMO PFRR PFCB but is actually rarely settling suggesting alsupply and demand are never well balanced. In my recollection of days pre NEM NSW frequency was much steadier.
The 4-second causer pays data for 11th September does not appear to have yet been uploaded (as at 15th Sep) to the AEMO webpage. Is there an issue that you know of that could be causing this delay?
Great remarks Paul, and really supporting the view that 2033-35 is where the rubber really hits the road with coal retirements. Not only is that where capacity is forecast to fall off a cliff, but those retirements are also technical retirements as much (more so?) as economic retirements: Bayswater, Vales Point, Gladstone (if still active?), Callide B (if the extension sticks), and Tarong will all be climbing the far end of the bathtub curve around that time and there’s very little public information to be had on what the options look like for delaying that.
I think it’s underappreciated how different the technical EOL aspect is from the discretionary, policy-driven pre-EOL economic retirement discussion that has captured attention, and I also wonder how well QLD and NSW are doing at recognising that a retirement wave hitting both states near-simultaneously may limit the ability to rely on imports to solve local capacity problems.
I hope you’re not verballing Oliver with that “90% of the coal fleet disappearing” comment Paul 😉
That’s a strawman argument. Anyone who understands the very different availability characteristics of wind and thermal generation would never expect the output profiles of the two to be comparable.
With respect to the differences between coal and wind, I’m sure Oliver understands the difference:
1) His presentation on the second day was focused on explaining the size and depth of storage required to meet firming requirements (with a focus on the QLD region)
2) So (my sense is that) Oliver’s comment was a way to help the audience understand the size of what will be required in a VRE + Firming world without coal.
Thanks for sharing.
The pages dedicated to clarifications and fragmented state of the ‘GFM’ environment goes to show how important it is that International Standards be set up as soon as practical. It will be a hot mess in years to come dealing with the variety of vintages, manufacturers and capabilities.
From my reading of the report, they say the models say they should work well – excluding “niggles” like close-in fault current behaviour. As those in the power industry know, there is a massive difference between models and actual proven near 100% reliable performance under myriad fault behaviours – many which weren’t covered in contingency planning. The non-credible events – with Callide being a good example.
Just how much importance is to be placed on preventing blackouts? Would you stake lives on the risk that they won’t fail to do their job? I wouldn’t.
This table about yesterday’s, today’s and tomorrow’s problems is a very good planning tool. “out of scope” is routinely used to shovel away the real problem
This is just the usual posturing we are seeing from industry claiming they will shut down and then having the Government throw money at them to stay open.
Very blatant that they choose to make this announcement with the new QLD Energy Plan to be released in less than a fortnight.
To which extent were recent NSW market interventions caused by calculations as described in:
26/09/2025
“A minimum level of synchronous generation must be in service to ensure the stable operation and protection of the New South Wales transmission network. Transgrid and AEMO have jointly agreed that this requirement is met when the equivalent of six large synchronous generators are in service. Under these conditions, the network is considered to be in a satisfactory state. To be considered secure, the network must still have the equivalent of six large synchronous units in service after any credible contingency— typically the loss of the generator with the highest contribution” https://www.aemo.com.au/-/media/files/electricity/nem/security_and_reliability/congestion-information/2024/transfer-limit-advice—system-strength-nsw.pd
Suppose Eraring were to shut down, would that number be (12-4)=4?
What other criteria are there for the minimum of rotational, synchronous inertia in the grid? How about imports?
To which extent were recent NSW market interventions caused by calculations as described in:
26/09/2025
“A minimum level of synchronous generation must be in service to ensure the stable operation and protection of the New South Wales transmission network. Transgrid and AEMO have jointly agreed that this requirement is met when the equivalent of six large synchronous generators are in service. Under these conditions, the network is considered to be in a satisfactory state. To be considered secure, the network must still have the equivalent of six large synchronous units in service after any credible contingency— typically the loss of the generator with the highest contribution” https://www.aemo.com.au/-/media/files/electricity/nem/security_and_reliability/congestion-information/2024/transfer-limit-advice—system-strength-nsw.pdf
Suppose Eraring were to shut down, would that number be (12-4)/2=4?
What other criteria are there for the minimum of rotational, synchronous inertia in the grid? How about imports?
Good thoughts, but it may be even knottier. Rio needs Gladstone PTI committed to close GPS while Powerlink need a GPS retirement date before 2030 to justify Gladstone PTI. So in a way, GPS is forced to announce retirement before that retirement is feasible if they want any chance of the retirement becoming feasible later.
I think the early closure of Gladstone has been speculated for many years. The author doesn’t even mention the subsidy the station receives via CS Energy. Nor do I personally think system strength was the major reason for the Eraring extension. And although the firming for Boyne Island is not yet known I suspect QLD plans to build a new gas plant and that coal from other plants will provide the firming while it’s being built. That’s not a solution I favour but what do I know.
Its getting hard to ignore the rollout of batteries even for the most conservative of “experts”, but what is clear is we are still in the high growth phase so forecasting where we will end up is little more than a guess, and then you have to reestimate what power and energy looks like in such a system.
The inbuilt conflict between government’s ultimate responsibility to maintain power supplies and the commercial decisions of companies and investors on the timing of closing aging power plants was, in the case of NSW, created by the very government having privatized what was and should have remained public infrastructure in the 1st place. With this privatization, the State government wanted to rid itself of the difficult transition away from coal. This opportunistic approach has backfired.
Great article. The Author could include some other points or solutions. Most coal power plants in Queensland are Government owned so I don’t think this will be a systemic/repeat issue. Other solutions could include: (1) other Queensland coal power plants can have their life extended, or alternatively (2) another coal plant could close instead to improve the economics/operation of Gladstone.
It appears that all projects have been reported on the basis if their peak energy sent out (nameplate) capacity, not reduced by capacity factors. This leads to a loads served estimate that is wildly overstated – something in the order of 3 times overstated. The number of hypothetical houses served is thus overstated by the same factors.
Surely, after more than a decade of optimism, reality should be applied to all Ministerial statements and other public guidance papers.
IMHO, the promised 6.6 GW of new power gen will actually deliver only 2 or 2.5 GW, with lower reliability than the Australian Eastern States Grid provides currently.
I note sync condensers in the mix. They obviously have some doubts about grid forming inverters being able to completely meet grid strength needs. I wonder what their black start plan is?
I wonder if there is a way of calculating how efficient the fleet was at capturing the high prices with the benefit of hindsight? Then you could track this statistic over time to see if there is a trend.
Looks like someone is either unfamiliar with the dead time of the wind systems, or they’re using a square wave to join the dots rather than the triangular wave used in the charts above.
Frequency plot consistent with the above – near vertical drop implying an electrical fault / trip. Would be interesting to understand the amount of load and solar shaken off with the voltage depression.
A terrific article. Thanks. The science/maths of “emergent” effects is little understood and could lead to counterintuitive behaviours. But very, very interesting.
Interesting. Thanks. Isn’t about time we stopped approving large solar farms unless they have at least 30min battery backup to hold up output until scheduled generation can ramp up?
I was at the All Energy conference on the first day last week and there was no discussion about All Energy at all it was just about renewables and batteries.
They need to include all forms of energy in their discussions.
The number of batteries on show were enormous.
When are all these battery farms getting recharged? Have we got this much solar power abundance. Why not shut the coal and gas fired power stations down now and save us from future hailstorms and cat.5 cyclones,rising sea levels and 50 degree temperatures and bushfires from hell
If the frequency signal was measuring fine, the error could have been in the AGC delivered set point telling generators to increase output, or in the controller’s understanding of actual supply.
Another error source may have been the automated wind and solar forecasting system? Could that have predicted less wind and solar about to occur?
If the frequency measurement was ok, could the error have gotten into the demand forecasting which in control systems is a feed-forward input to the controller?
Good series of articles Dan, all seem to be pointing in the direction of why local control of key electrical variables is necessary and showing the weaknesses of centralized control. Controlling a global system means if something goes wrong – it goes wrong everywhere.
If the idea is successful and 15 GWh/day is moved from evening-night demand to soak up curtailed wind and solar even if it pushes up the current midday price by $30/MWh to replace coal and gas at $130/MWh that is a saving of $1.5m/day in energy costs for the retailers.
Further for customers it will make switching from gas and immersion hot water heating to heatpumps more attractive, which will reduce household energy costs even more. One heatpump hot water service can replace almost two gas heaters and two immersion heaters with no net increase in electricity demand.
The alternative is to invest about $3bn in batteries to store the curtailed wind and solar and add that to the RAB
The reality is most people’s demand profile is morning breakfast and then afternoon/evening for dinner, dishwashing and showers.
Ultimately most people can’t shift those demands so what will happen is people will use their airconditioners in the middle of the day and maybe do a load of washing but then get slammed with higher pricing for the morning and evening when the majority of their demand is.
Quick suggested correction:
“that’s an additional 8.5 TWh/year additional demand”
This is out by a factor of 10, I’m assuming you forgot to multiply the annual consumption by the efficiency loss of batteries (around 10%?). It should be closer to 0.85 TWh/year of additional demand (still a lot, but a little less concerning!)
Not often a Wilson transformer fails – especially not 2 out of 3. I wonder if they were wrongly specified (eg can’t handle the harmonics), or damaged during testing? Be interesting to get more “inside knowledge”.
Oh this is an interesting failure mode. It highlights there’s a risk when a large amount of data entry is done by a single provider a single update/patch/issue could heavily impact a lot of units. Might require some thought around ensuring diversity.
Hey Dan, nice paper
Did you look separately at daily max demand / demand at time of minimum reserve? And bucket by max demand?
Anecdotally AEMO was consistently under-forecasting QLD evening peak demand through Q1 2024 – you might hope there was a human in the loop to nudge the forecast in the right direction, it took their model several weeks at least. Constantly had a forecast lower than prior day’s outturn, despite similar or hotter weather…
It’s much harder to unpick now given the impact of rooftop solar, but you did used to see demands outperforming AEMO PD for some time before the forecast changed to something more realistic (see Allan’s forecast progression chart https://wattclarity.com.au//articles/2017/02/initial-analysis-sa-load-shedding-wed-8-feb-2017/ )
I have made an attempt to create the visualisation you’ve described and it confirms this thinking. Will post as an article in the coming days/weeks (when I find time!)
Walgrove setting prices despite plenty of other available generators and very little volatility in the FCAS markets. Keen to get our detailed WattClarity treatment on this blip! Interestingly the same $14k got triggered again on the 21st morning.
Very interesting dynamics on the constraints.
N>>NIL_85_86_S definitely hit max at 12:10, but I think N::N_CNLT_2 is an interesting feature.
Based on dispatch constraints…
N::N_CNLT_2 was at capacity the entire time, but around 12:15 it suddenly had the RHS drop by 300MW. https://nemview.hgn.id.au/graph/dispatch_constraint/N::N_CNLT_2
I would guess this had a ripple effect combined with…
I also notice the NSW Wind was mostly unconstrained up to 12:15, and constrained down about 60% for the rest of the day.
Interestingly, since the start of 5MS there have been relatively fewer occasions where a region has both hit the MPC and fallen deep into negative territory (well beyond “negative LGC” levels) in the same day. With the MPC making a step change up from July 1st — a 16% increase for FY25/26 — the gap between the market floor and cap has now widened beyond CPI too.
Hi, I noticed that previously both the output and input DUIDs had a MAXAVAIL value. This makes sense because the charging capacity might be offered differently to the discharging capacity.
But the bi-directional DUIDs only have a single MAXAVAIL.
Is the MAXAVAIL considered symmetrical n for the bi-directional DUIDs ie same MW for charge and discharge?
Hi Ben, MAXAVAIL is a parameter submitted in the bid and bidirectional units submit a 2-sided bid for energy. The DIRECTION field in the bid distinguishes the ‘side’. It will be gen for production or load for consumption thereby allowing for a different maxavail on each side.
It looks to me like the lack of battery capacity close to Sydney contributed to the delay in pushing the price back down. There was only a 250MW increase in battery output at the time of the price spike on Wednesday, which was insufficient to cover the increase in demand in the Sydney region. Of course the lack of transmission capacity within NSW was setting this situation to play out, then dramatically compounded it when the demand spike occurred. Clearly relying on only one large BESS close to Sydney is not looking like an optimal solution. Even at full capacity the Waratah Super Battery would not have fully handled this sudden change in supply.
I think the implementation of the Negative Residue Management (NRM) constraints needs to be reviewed. The current process is producing unnecessary supply/demand shocks on the grid, which often occur at times when the grid is under strain. It would make far more sense to stage the introduction the NRM constraints over several intervals to avoid the shocks to the grid.
On NRM constraints, 1) they do cut in progressively, IIRC they don’t try to cut back counterprice flow by more than 100 MW per dispatch interval; 2) they don’t reduce supply available to the high price region, since they target reduced exports out of that region.
You write that “The rapid movement of clouds can cut rooftop PV output with little notice — and because this generation is largely invisible to the market operator, it adds another layer of uncertainty to their forecasting models.” but a reasonable proxy for that data is available via the large and growing network of home weather stations that measure and report insolation.
For example have a look at the Ecowitt map at https://www.ecowitt.net/home/map drill down to my weather station at https://www.ecowitt.net/home/index?id=168614 and you can clearly see the history and realtime data for insolation. Aggregate this up by postcode with a multiplier for estimated rooftop panel capacity and you are in business!
The web site will even provide the data in tabular format and you can export it. To automate this would be a very small project that could yield a relatively reliable forecast model.
I reckon you’ll find that the sudden curtailment of wind and solar shown in your charts will coincide with Snowy rebidding a lot of its Tumut generation capacity down to MFP in order to dispatch against the high spot price. By 11:15 those generators were running at over 1,600 MW, having been at 0 MW until 10:45. They sit lower in the firing line of the 16_8_39 constraint than some of the southern NSW wind and solar generators, which would have been first cabs off the curtailment rank.
Thanks for the article. It was interesting to read the following statement, “Gladstone has already been operating at minimum stable load (Pmin) almost daily for the past year. In effect, the plant is already mothballed.”
There is nothing unusual about a unit at GPS running at “Pmin”. Historically this may have been after evening peak demand. It also was dependent on which units on line were contractually supplying Boyne Smelters. The other units would at, or be close to, Pmin. The change is that this Pmin period has shifted and extended each day.
Regarding reliability, it should be noted that reliability of a power station depends on the resources available, including reasonable continuous funding for maintenance, asset refurbishment/replacement and associated periodic shutdowns. Since privatisation, GPSGOS (Gladstone Operating Services) has had a funding model based on reliability, not MWe (export). This funding was fixed and eroded dependent on having the contractually agreed number of units on line. This can be a challenging environment when operating a coal fired power station that has been operating since the 70’s. Particularly when you don’t have control over the quality of coal you need to burn.
Notwithstanding the solar forecast issues, isn’t the immediate problem that primary frequency response and regulation FCAS are not providing sufficient response to avoid these frequency sags, whatever their cause?
It seems arguable that AEMO ought to be enabling a lot more regulation volume, at least during solar ramp periods. This would have the physical benefit of keeping frequency closer to target – which after all is the key objective here – but also a market benefit in that it would raise the cost of regulation FCAS being incurred by the under-performers via frequency performance payments, so might prompt them to forecast better (or at least more conservatively)?
It could take a very long time to get any changes to the semi-scheduled generation category and rules. As far as I know AEMO could vary the volumes of FCAS enabled with very little lead time provided there is a reasonable justification for it. Which there certainly seems to be here.
My understanding is that the AEMO is exploring this as a possible first step … because (as you say) it would be a relatively simple/quick approach.
I’m wondering if doing that would be enough, as the Semi-Scheduled capacity continues to scale, and the mix of Scheduled capacity changes (perhaps with aggregate MW installed declining from current levels). Various people have looped me into different conversations where various options are being considered – at some point might have time to write about some of these.
The cost of coal is irrelevant. Burning it increases the cost to our future. Enter that of our children and grandchildren are such a large amount that will dwarf any current costs.
It would help the average non electricity espért to understand this article if you spelled out what the many acronyms stand for and what they mean , in layman’s terms please ?
What the data shows, so far, is circa 10% reduction in peak output compared to the average for the previous week. Exactly in line with the fire involving a fraction over ten percent of the area covered by panels.
Monitoring for a longer period will give us a better picture but suggesting that there is no long term damage seems premature at least
Elav may not be such a good measure of availability of solar plant. ELAV reflects the number of inverters in operation – which may, or may not, reflect the maximum output available if damage to an installation affects strings only.
A week has gone by and peak output is still sitting 10% below its recent maximum. I can find no update from Lightsource BP
Regarding multiple generator contingencies, how do you suppose this will be implemented? Would a second layer of runway pricing be implemented from within the tripping scheme? My brain hurts trying to think about what a fair scenario would be.
These coal stations are getting old with nothing to replace them.
Based on NSW SCADA tonight (so may not be trustworthy), I think there could be 3 of 9 operational coal turbines in derating after 7pm tonight.
A NSW LOR2 of only 80MW remaining tomorrow came an hour later.
Worth noting against this post for future reference (though don’t take as official values)…
NSW market demand maxed around 16:00 AEST at 13,000 of the NEM wide 30,000.
NSW estimated underlying demand maxed around 14:00 AEST at 16,500 of the NEM wide 39,000.
A quick look on the Zoom Earth cloud cover shows increasing cloud cover over Woolloongong, Sydney and Newcastle from 11:40am on into the afternoon. Could that be the cause? I would say a good chance.
Thank you Paul for keeping us all updated over the holiday period, especially since Christmas Day was the big MSL day.
However, the focus is now on Victoria and SA for 1st January. Waiting to see what battery in Victoria receives directions if AEMO deem it necessary for MSL management.
“Whilst this notice speaks to ‘forecast MSL1’, the lowest level of System Load forecast here is below MSL2 level, as well”. Paul think your pickup here may not be the first time an MSL2 has been classified as a MSL1.
I believe the run to late 2022 was related to speculation the Albanese/Bowen team would use LGC demand mandates as tool to deliver their decarbonisation ambitions. Once the expanded CIS was announced in Dec 2022 that story began to fall apart, and the introduction of the REGO scheme has made it apparent the government has little desire for a meaningful certificate price in future.
A classic example of one or more participants being able to exercise pricing power through the event. Have a look at which participant/s had online capacity at batteries or synchronised gas units – ie not fast-start GTs etc – offered at MPC!
Large Solar = includes any unit for which AEMO publishes SCADA readings on a 5-minute basis (e.g. InitialMW). So typically Semi-Scheduled and a few larger (Registered but) Non-Scheduled units
Small Solar = AEMO estimate of rooftop PV, which uses a sample-and-extrapolation method, with the extrapolation being governed by aggregate MW capacity of units registered in the SRES part of the MRET … so units no larger than 100kW.
So Medium Solar = everything in between … which, readers will understand, is a growing scale for which there is no visibility.
While not a direct explanation for batteries on the 10th, I note there were minimum load concerns from 11am (NEM Time). Demand was not as low as forecast, possibly due to shading of solar in Melbourne by bushfire smoke.
“132659
MINIMUM SYSTEM LOAD
10/01/2026 04:17:29 AM
Forecast Minimum System Load (MSL1) condition in the VIC region on 10/01/2026
AEMO ELECTRICITY MARKET NOTICE
AEMO has detected that there is an elevated risk of insufficient demand to maintain a secure operating state in the VIC region on 10/01/2026.
Minimum System Load (MSL) conditions are forecast when demand is less than the relevant MSL threshold. The advisory MSL thresholds are:
– MSL1: 2116 MW
– MSL2: 1616 MW
– MSL3: 735 MW
The demand is forecast to be below the MSL1 threshold for the following period:
From 1100 hrs to 1430 hrs 10/01/2026. Minimum demand is forecast to be 1783 MW at 1230 hrs.”
A little background on the VBB System Integrity Protection Scheme (SIPS):
The scheme reserves 250 MW of the VBB’s 300 MW capacity between ?November & ?March each year to provide AEMO the ability to run transmission between Snowy/NSW and Melbourne (the Murray-Dederang and Dederang-South Morang 330kV lines) at higher capacity levels. These are parallel circuit pairs and normally have to be run below a level that would not rapidly overload the remaining circuit if one line in a pair trips. The SIPS scheme allows AEMO to run the lines to higher ratings by triggering the VBB to rapidly discharge at 250 MW if there is a line trip: this extra supply near Melbourne reduces flow on the remaining line to a level that allows sufficient time for market redispatch if the tripped line remains out of service.
The VBB discharge events discussed in Paul’s article show the scheme operating when different lines on the Murray-Dederang-South Morang corridor tripped, presumably due to bushfire smoke or lightning.
Note that the market bids are “housekeeping” to reflect what the battery has physically done in response to a line trip: the SIPS scheme operates outside the market and doesn’t rely on the 5 minute bidding and dispatch cycle.
There was:
132687 RESERVE NOTICE 10/01/2026 06:39:59 PM
Actual LOR1
11 out of 12 coal units were operating. Despite plenty of solar power, they started to ramp up already in the early afternoon to reach around 7 GW for the whole evening
Of interest in this period, I think I saw ‘games being played’.
Based on the past days, NSW large solar drops to practically zero at NEW 19:15.
But I watched the P5 forecast through the hour up to this time, and there was bidding of way higher than that to the end of the P5 hour. Like bidding that 500MW will be produced at 19:45. Obviously, the very last bid of an interval was much closer to real output to avoid penalty, which means after 19:15 was zero.
But how can a forecast be trustworthy when there is such large amounts of false data in it?
I wonder what took up the slack between the forecast and final?
So youre speculating that the transformer will come online then?
Because that’s not testing, testing would have already been done offline. Soaking for a second time and then taking load is proving.
Nice one Michaela! Captured the two steps well, and more clearly showing the first unit impacted frequency just on or moments before 16:55. Suggests, also, some latency in the timestamps of the MW readings.
The obvious question here is whether it was merely coincidental that immediately after the Callide C trips, aggregate off-target level for semi-scheduled solar leapt to nearly double the level it had hit in any previous interval on the day?
Or did the frequency disruption due to Callide somehow trigger some solar farms into underperforming over the next 5 minutes? – I can’t quickly think of a mechanism, but it does seem strange.
Given the evacuation of staff, the ‘control system fault’ may have been caused by a problem with DC systems and CS were concerned about a repeat of the events of May 2021 when C4 turbine disintegrated.
I noticed that in the https://nemoutages.com.au/mtpasa, QBESS will be out of action until 2029/01/13 which is a fairly decent time away without market participation given that its only been fully operating at full capacity around Jan-Feb 2023.
To clarify my earlier comment: I’m basing this observation on publicly available MTPASA data at the time of writing. As far as the eye can see with the data available to the public, QBESS shows as unavailable until 2029/01/13. I acknowledge there may be additional information not reflected in public datasets.
RE : wind farms dont like the heat… Not sure that is explained correctly.
Are you sure it’s not the air density that is lower during hot conditions that leads to reduced output from the wind turbines – I’d be surprised if their components care about being hot…
As a aviation pilot , hot weather doesnt bother the engine but the lower air density can double our runway requirements….
Air density
Fan forced cooling in nacelle for gearbox oil & electrical components (or feathering of blades to reduce output)
Reduced electrical efficiency at high temps
Air density
Fan forced cooling in nacelle for gearbox oil & electrical components (or feathering of blades to reduce output)
Reduced electrical efficiency at high temps
Would really like to see a similar modelling for the hunter transmission project and northern inflows and wether this will eleviate flows into sydney. Looking at closures of coal in this region epecially with recent news of extention of closure dates.
Accumulated aggregate state of charge is a good metric.
I expect that when EnergyConnect is fully commissioned the SA volatility will flow into all the other states. Will be interesting to try and track that.
Widespread acknowledgement that increased renewable electricity generation is driving down wholesale power costs. No mention of politicians agreeing with this. Hopefully the current quarter will continue to reflect the increased storage capacity, solar and wind generation. And if so, what effect will continuing wholesale power costs at lower levels have on retail power costs. Just noting that jumps in retail power costs due to removal of state and federal subsidies have contributed to recent increases in cost of living.
I was surprised by this, but as a fairly new battery owner and Amber customer benefited. Looking at Open Electricity trying to work out what might caused it, significant exports from NSW to VIC seemed to occur around this time while VIC started curtailing Wind and Solar. (I know there are a lot of other factors I can’t see and am not aware of that may have contributed.)
For those in control of such a large response, it is in your interest to go to an effort to voluntarily become part of the scheduling process as, by directly influencing price, you will moderate these spikes rather than having to react after they are published, or, when restoring, you can avoid provoking a new spike.
This “analysis” is embarrassing to see on Watt Clarity. You guys are better than this.
The article is fine until the author seemingly decided it was too difficult to work out the impact on coal generators and instead decided to assume that every coal generator in NSW would run at ~50% load when prices are at their peak. If the headline wasn’t based on this final assumption, it wouldn’t be such an issue.
I’d love to see an analysis diving into the dynamics of greater battery capacity on the grid. Including higher daytime pricing, reduced emission intensity (less gas for peaking) and coal retirement projections. It seems that batteries won’t have a huge impact on coal (compared with their impact on other generators). There won’t be enough of them to justify bidding lower than coal to stop coal from matching demand (minus wind) during the night.
Date: February 2026
From: A Retired Power Plant Operator (20+ Years Experience)
Subject: Formal Record of Technical Negligence and the Path to Professional Safety
1. EXECUTIVE SUMMARY
As a former Marine Engineer and 20-year power plant operator, I am providing this record to ensure the 2021 Callide C4 explosion is never repeated. The incident was not a random technical fluke; it was a consequence of bypassing engineered safety barriers for production goals. To restore safety and insurability to the Queensland energy grid, we must move away from “in-house” company training to a Statutory National Standard for plant operators.
2. THE TECHNICAL “SMOKING GUN”
The 2021 disaster resulted from two specific, documented deviations from core engineering safety principles:
The Interlock Bypass: Maintenance protocols bypassed physical Castell Key Interlocks. These “trapped key” systems are designed to make it physically impossible to isolate the 220V DC protection supply while the turbine is online. By bypassing these, the turbine’s “emergency brain” was killed while it was generating power.
The Lack of HV Isolation: Management opted for “live” maintenance on critical protection circuits to avoid a planned outage. Under standard engineering practice, High Voltage (HV) Isolation is the only safe method for this work. Without it, the unit was left with zero redundancy, leading directly to its kinetic disintegration.
3. THE ECONOMIC COST OF NEGLIGENCE
The “white-wash” narrative was dismantled by the following financial and legal facts:
Total Financial Impact: Taxpayers face a bill exceeding $1 Billion for repairs, rebuilds, and grid stabilisation.
Insurance Crisis: Following the failure, CS Energy became virtually uninsurable; premiums surged by $40 Million annually.
Legal Penalty: In February 2025, the Federal Court issued a record $9 Million fine against the operator for breaching National Electricity Rules.
Thanks Bill. I’ll go back to Reneweconomy. By the way other than the podcast I have no professional relationship there. I contribute to put something back after a fairly successful career in in investment banking research. I have a huge amount of time and respect for Wattclarity and the quality of its work and I’d hate for it to be diluted with my “reneweconomy” tier. Very happy to have my points debated, never happy but will always accept where I am wrong and actually over a very long period of time have mostly kept towards the front of the analysis curve. If you want to be early you will be wrong. Perhaps batteries wont have any impact on evening coal volumes any more than solar had an effect on daytime ones. I would be open to the point that charging the batteries has an impact on midday volumes but equally there will likely always be plenty of low cost solar. Still I abandon the watt-clarity field. It’s not that I mind criticism Ive had a fair bit of it in my career and from very very senior people, but I really don’t want to upset the quality diet that Wattclarity has. regards
Seriously guys – are you suggesting that all electricity consumers should engage in this process in order to get their electricity bills under control.?
Why not just go off-grid and be done with it? (And do it at one’s own expense?)
And why not reflect on the incredibly generous taxpayer cash splash that drives this behaviour.
Then there are all those consumers whose residential circumstance preclude using panels and batteries but are being obliged to help fund them anyway.
And if this circumstance is not fair and just as it stands, pray tell what taxpayer subsidies will flow when all existing batteries and panels (and wind turbines) reach end-of-life status?
Let’s be honest – most people just want reliable and cheap electricity whenever they need it!
And an electricity grid based on nuclear power is the only honest way to deliver that!
A really insightful analysis.
A few things which would have been great to see in addition:
1. Energy inflow breakdowns to battery, load and grid export on Sankey diagram.
2. 12% energy lost from battery?! Much higher than I would have thought (and what is generally touted by industry). Is this the efficiency reality, or are there other factors contributing to “battery losses”?
3. Hot water being the largest energy consumer is surprising too. Resistive system? Not much distance travelled in EV? AC used infrequently or otherwise efficient building envelope (or moderate climate)? Some discussion on this would be illuminating.
4. Any import earnings (i.e. getting paid to consume)? Or are the negative wholesale price instances needed to overcome the various variable network/market charges just not there yet?
5. In Fig 9, another useful comparison would be to add the upfront capital expenditure to the solar + battery curve in a new data series and adjust all $ numbers for inflation. Could include another series with un-subsidised capital cost as well (if known). These additional series would show the overall payback period (eventually) and should include the costs of the EV & charger (if counting the fuel savings in the BAU data), the 2nd battery, and any other infrastructure, efficiency or fuel switching changes that have been made. It feels like the true payback period might be a while off yet, but that’s not accounting for any non-monetary value gained from the whole exercise.
Can we see any trend in battery charging costs either increasing or decreasing?
I ask because higher charging costs will translate into higher selling price and associated higher wholesale prices eg will batteries eliminate negative price periods?
Hi Ben — I’m not sure I agree that it’s valid to think of discharge prices as ‘charging cost + a margin’ in a cost-plus sense; both charge and discharge outcomes are mainly set by system conditions and volatility, not by what the battery paid earlier.
In the animation in the article, we can see that charging prices sit in a much tighter band than discharging prices, which strongly suggests that the relationship between the two isn’t linear and the spread is often dominated by what happens on the discharge side.
My view is that AEMO was entirely correct to provide pure capacity factors. What is actually achieved is of course different. In ITKs LRMC modelling we adjust for capture ratios as we call them which allow for both economic and physical curtailment.
And although we are talking wind in solar it’s the same thing only more so. And we have what used to be called an inverter load factor but now I think ACDC ratio is the more common term.
Within reason the underlying pure wind capacity factors will remain constant but economic and physical curtailment and availability may well change.
Wind traces are generally produced using one of the major climate reanalysis suits (ERA5, MERRA2, BARRA-2). I’m not sure which one AEMO uses but many studies use ERA5, which significantly overstates annual capacity factors. The Australian BOM’s BARRA-2 is much better. We published an analysis comparing actual wind farm output versus modelled output using the reanalyses above and found that all the reanalyses can produce good results at times, but the difference between modelled and actual is highly variable. Modellers should be aware of the gaps between actual and modelled.
Validation of BARRA2 and comparison with MERRA-2 and ERA5 using historical wind power generation https://doi.org/10.1071/ES24028
The failure at Callide C4, which resulted in a $500 million loss, was attributed to a combination of technical failures and systemic organizational shortcomings, with the primary responsibility resting on CS Energy, the plant operator. Forensic engineer Dr. Sean Brady’s report, released in February 2025, concluded that CS Energy failed to implement effective process safety practices, including the absence of formal risk assessments for critical switching procedures, which led to the catastrophic failure of the Unit C4 generator on 25 May 2021. The report specifically highlighted that the decision to carry out a switching sequence with the unit online—despite known faults in the 220VDC switchboards and the absence of redundancy—was a direct result of inadequate management oversight and a culture that did not prioritize process safety over cost savings.
The failure was triggered by a cascade of events: a voltage collapse in the new battery charger, the loss of the AC supply due to a false arc flap protection trip, and the failure of the automatic changeover switch, which had been disabled since January 2021 due to a low-consequence classification.
CS Energy’s processes were found to be deficient, with no formal requirement for risk assessment of operational changes, and a lack of a dedicated process safety team from April 2019 to July 2020.
The company’s shareholder mandate, which emphasized cost reduction and debt management over safety, created competing pressures that undermined safety protocols.
The insurance company refused to pay the $500 million claim, citing the failure to follow proper procedures, which further underscores the organizational negligence.
As a result, Callide Power Trading, a subsidiary of CS Energy, was fined $9 million in February 2025 for breaching the National Electricity Rules, with the court emphasizing the need for a deterrent penalty to prevent future failures.
The responsibility for the ignorance that led to the failure lies with CS Energy’s leadership and governance structure, which failed to enforce safety protocols, despite repeated warnings and known risks.
Could the problem be that the power electronics or cabling is undersized with insufficient heat removal capacity and the thermal overload alarms come up? The plants may have also been built down to a price point and the cooling was downsized as it was regarded as just a nice to have.
There are two effects. Air density at 40C is 10% less than at standard testing temperature of 15C so energy in the airstream is correspondingly 10% lower and all machines,cooling sytems and electronics are less efficient, particularly cooling systems as the cooling effect of a radiator is dependent on the Delta T and the mass flow which are both lower.
Ummmm – can I suggest you find some representative sites to say what the wind is doing rather than the temperature. Dare I suggest the temperatures became very high when the wind dropped and you will note the error declined rapidly with temperature decreased….dare I say when a breeze came in and the windies actually did something. Simply highlights the vagaries of trying to forecast the output of these things. We see this characteristic all the time over here in the west in summer – on hot days the easterly winds die and nothing happens until the sea breeze (south westerly) arrives. On really hot ones there is no sea breeze so we get nothing until the easterlies start up again in the late evening / night.
Particularly striking (and I think underdiscussed?) is that transport fuel security & 1-in-N-year electricity system resource adequacy look to have the same least-worst solution: giant tanks storing a few dozen PJ of diesel to hedge domestic transport needs against foreign conflict *and simultaneously* hedge dual-fuel GTs against gas network constraints through long periods of high residual demand. That’s two rare and uncorrelated events, so there’s no reason you can’t use the same assets to do both simultaneously while deploying renewables and EVs as fast as possible to reduce fuel demand wherever possible.
“…the coal trains are (I believe) all electrified…”
All coal trains in the Hunter Valley run on diesel, whether to the Port of Newcastle, or to the various coal fired power stations (Vales Point, Eraring and Bayswater).
the big thing im wondering about is what impact battery will have on FIT’s because as you noted the wholesale price of energy is down to $0 in the middle of the day
even a blind man can see this as easy arbitrage but my question is how much battery can be installed and slurping up that cheap $0 excess solar before the cheap excess solar runs out?
if you have say 10gw of batteries slurping up cheap daytime solar but only 5gw solar being output are we going to see daytime prices smooth out to nighttime prices?
will FIT’s start to go back up again and a restart of the solar/wind boom where batteries want more cheap renewables installed so they can keep the easy arbitrage going?
Less about “what happened at Macarthur Wind Farm” than “what happened in the operations area(s) meant to be overseeing its market dispatch?”. Its bids appear to be made on a set-and-forget basis, and no-one noticed that AEMO’s AWEFS availability forecasts somehow got stuck at ~80 MW for nearly 18 hours, well above what it could actually produce.
The duration of a battery is just its inverter size. A 2hr battery can operate as an 8hr by simply discharging at 0.25 of its max power, but it can ramp up to cash in on price spikes. For example, on 5/6 April the NSW price was ~$75 all night, so Eraring discharged at half power from 5pm to midnight, so it was effectively a 7hr battery. I don’t think we’ll get many 8hr batteries because their ROI will be slower. Having a smaller inverter is cheaper, but it limits your ability to cash in during price spikes.
In a cooperative system it would make sense to have 8-16hr batteries. But we have a competitive system where they need to take advantage of negative prices to charge (typically 4hrs per day) and peak prices to discharge (typically 1-2hrs per day).
Price volatility will be with us for some time as whenever batteries get on top of things, that will trigger another coal plant to retire, which will set off another round of volatility.
“Pumped hydro highlights the issue. It provides long duration storage, synchronous system services and asset lives measured in decades, but much of its value is not captured in project revenues.”
Hmmm. I wonder what other low-emission generation technologies might be favoured by the called-for “market design changes that deliver long dated revenue support aligned with asset lives”.
Seems to me to be insufficient governor action. Add up all the non-conformance in solar and wind generation and compare to steam and hydro generation units off Target. I reckon the system is too light on inertia. We are headed for a serious system event (reference Spanish system collapse)
550+MW of over performance? Call it what you like, but I call it Nonconformance. 550 is a huge matter, particularly if all the governor controlled units are near minimum load. This is serious territory that cannot be spoken about lightly. I hope the Under-frequency load shedding system will save some sections of the system on the way down.
Not quite. The answer to your observation was in the 10th April write up. It talks to a 17 day combined outage. Not surprising – maybe a bit of scope creep since April. Lots of common plant in these sorts of stations and it talks to switchboard replacements etc which is never quick or easy. Also talks to a control system replacement so common plant also has to be done.
Although PEC-1 operation has seen clear increases in the maximum export and import limits for V-SA, a similar chart showing average limits applying in dispatch would show reductions in both directions, particularly on imports: flows from South Australia into Victoria.
This counter-intuitive outcome results from PEC-1 inclusion causing V-SA to appear in a number of regularly binding constraints that it wasn’t previously subject to – for example V^^V_NIL_KGTS. The reasons for this are outlined in detail in the article you linked above.
Quite frequently these limits have forced counterprice flows out of Victoria into SA when the “natural” flow direction would have been opposite, one reason why average flows have trended more strongly into South Australia since the advent of PEC-1.
I would be curious to understand if there are any requirements from Solar OEMS to provide AEMO access to there API for data on behind the meter solar and Grid data, the amount of curtailment occuring ect. understand and modelling this is imporant and would make the Soalr inverter OEMs more accountable for clear and concise data. If its not it should be in the pipeline or being written into the standards otherwise made mandatory to gain CEC approval and rebates. Making the chinese OEMS accounatbe and provide clear metrics, when normally that stuff is hidden.
Let’s not worry about the price, suggest worry about where the energy is going to come from on days like this when all of the nasty black stuff isn’t there anymore. It won’t happen every day but these events are far more common at this time of year than the green lobby would like you to believe – and the wind drought goes right across the bottom of the continent. And the quantity of energy needed to plug the hole is far more than Snowy#2 and a few batteries will provide. Who is building the GT’s and the gas (liquid??) infrastructure needed to fuel them? Business model will be challenging.
@Andy
Snowy 2 will have the energy for days like this, just not the power (batteries by 2030 vice versa). IMO the NEM needs a capacity system for gas so it gets built even if it barely gets used, and I’m not sure if the ESEM will perform this function. I don’t think gas fuel will be in short supply at this time of year (Iona currently at 102%) given batteries are taking away its intraday shaping role but we’ll see what happens..
Using my experience with Domestic Battery traded on the Electricity Market through my supplier, the trading software is not informed on the severity of the lack of solar for a General Rain Event. My 13kW solar panel could only manage around 1 kW in the middle of the day, so the solar could not satisfy my domestic usage, let alone recover the battery sales from the previous evening. I was forced to intervene and charge the battery in the “Off Peak (1100-1600hrs) period to have secure supply during the next 2 nights and ensure the battery did not reach a discharge level that could damage the battery.
It seems to me that the major battery operators should take notice of the forecast size of rain events coming across the country and not be so aggressive in trading capacity the night before. That way they can remain competitive for unforeseen System Events and trading the next evening if the rain event is prolonged for a couple of days.
My experience is weather forecasts are useless beyond 4 days ahead. Can you do an analysis that looks at RMS error in estimate vs days ahead. I suspect you’ll see 3 days ahead is where it gets fairly reliable.
For a grid without coal, 48hrs accuracy is really all you need. Running 2GW of gas for 48hrs is a lot better than 10GW for 10hrs.
Good points Mark. We also need to keep in mind that weather-driven uncertainty in demand likely exhibits relatively rapid growth beyond the same 3-4 day horizon too. With an energy-constrained supply side it’s going to be those first 3-4 days in the projection that are most important for reliability over the succeeding days anyway.
What will be interesting is whether the current spot market mechanism will be enough to bring the gas turbines online early. The batteries should foresee the VRE shortage and withold capacity in hopes of cashing in on the price spike. That should increase the spot price, but will it bring gas turbines in early enough? 48hrs @ $400/MWh is cheaper than 1hr @ $23,200/MWh
Is your NMSE normalised to nameplate rating? I.e. a 100MW unit with 0.2 NMSE has an average 20MW error in its prediction. If that’s so, then anything over 0.3 is bordering unusable for estimating shortfalls. If you’re estimating 50MW, but on average it’s either 80MW or 20MW, then it’s not a very useful forecast. CLRKCWF1 being at 0.4 after just 2 days is a real problem.
I would have liked to have seen the 0-2 day numbers, even though it would be crowded.
Most of them “stay in their lane”. MEWF1 is an outlier as it starts in the upper band of poor predictions, but after 6 days is in the lower band of better predictions. Nearby KABANWF1 does the same thing. Both are low on your first “figure of thirds”, so they’re wind farms that spend 60% of their time below 40% capacity. If they’re below 40% most of the time, then predicting the output within 33% after 6 days is easy. WAMBOWF1 also fits this, but MCINTYR1 does not, even though it’s also a “consistently low” unit.
Excellent point about overestimates being more problematic than underestimates. An underestimate results in a little gas be used when it wasn’t needed. An overestimate could result in load shedding. Maybe it’s better to plot both average underestimate and average overestimate.
Could you pick 3 wind farms with high, low, and typical predictability. Make a plot with y-axis showing error in estimate from -60% to 60% nameplate, and x-axis showing days ahead forecast. Average overestimate (estimate – actual > 0) will of course curve upwards (because we’re only averaging positive numbers) and average underestimate (estimate – actual < 0) curves downwards in the negative axis. Plotting every 6-day forecast would get too garbled, so ideally a colour density plot of how often the error in the estimate is at that point. More red area below the x-azis in the density plot would indicate systematic underestimate.
Hi Mark – Thanks fore the comments. The nmse is the mse normalised with the mean square of the actuals. Would love to dig into this further, along the lines you suggest… when time permits!
It would be great if you could do an article looking at the SA yields over the past 2 weeks. There were a few points where total output was below 10MW, this would be a record low in the context of the total installed capacity.
Should all home batteries be called “price responsive”? Most are essentially off-grid with grid backup and don’t respond to price signals at all. They’re effectively removed from the grid, except for some solar exports in the afternoon and some imports during cloudy weather. Only VPP/Amber and other peak-bonus plans (Globird Zero Hero) are price responsive.
Hi Mark, the two halves of the slide are different datasets:
The left side (400,000+ batteries, 11.2 GWh) is the total fleet installed under the Cheaper Home Batteries Program since July 1st 2025 — all of it, regardless of how it’s operated. Unfortunately, DCCEEW do not regularly report these numbers on a power capacity basis (i.e. GW).
The right side is AEMO’s separate “Unscheduled Price-Responsive Resources” (UPRR) dataset, filtered for behind-the-meter batteries, since January 1st 2023. This is a defined category that captures resources observed to be responding to wholesale price signals (Amber, etc.). AEMO have only recently started publishing these numbers in order to track how these active resources affect its demand forecasts.
So no one was calling all home batteries price-responsive — the 2.8 GW is just the price-responsive subset of behind-the-meter batteries. Apologies if that wasn’t clear.
Ah ok. The numbers vaguely lined up (420k installs × 5kW export limit = 2.1GW, plus a bit extra for 3-phase systems). What’s the stat for total home batteries? Most recently reported numbers seem to be just CHBP only.
It seems like AEMO must be factoring the whole inverter size into that 2.8GW number. I’d use export limit rather than inverter size. Seems wrong for AEMO to claim there’s 2.8GW on tap during a high price event when most of that will be blocked by export limits. Obviously it gets complicated with dynamic limits. Also they will likely be self-consuming some amount, but that’s pretty much a given that they will always self-consume regardless of price.
Hi Mark, good catch — off the top of my head I don’t have a total home battery number beyond what DCCEEW are reporting, and I’m not sure if AEMO publishes a UPRR battery count (vs just capacity).
Better data about behind-the-meter storage is needed, in my opinion.
To clarify, the ISP doesn’t produce generation mixes based on modelling outputs. The ISP starts with demand and generation mixes decided up front – all inputs.
It was a “complex chain of factors” and “unforeseen circumstances” that could not have anticipated that the 220 VDC was isolated!
They had removed or never installed the mechcanical interlocks that had been used for many years to stop just that from happening in most power stations!
Simple, but super insightful chart, Paul, thanks so much. It emphasises how challenging it is to optimise batteries through periodic dunkelflaute. So easy to get wrong, yet I suspect will set our future critical reliability limits.
I wonder if 2-3 day-out market forecasting systems are as good as they could be? They are obviously crucial to the process.
Well thought out, and indicating risk and outcome considerations often left out of corporate and government analysis. And absolutely not sensibly considered very often in political dialogue bereft of long term considerations e.g. claims like “net zero by 2050 is a proven failure” being made in 2025. In the context of this article I specifically worry about “…Gas-powered generation is the reserve capacity the system leans on in precisely the conditions that produce unserved energy, yet gas supply and transport limits are too often left at the edge of the model. ” If I look one and then two decades out, I think there is much less gas being used in homes, small businesses and probably a number of large users. Removal of uneconomic retail gas networks is already occurring at a small scale while states are banning new connections and/or requiring higher up-front payments. Some are also subsidising the move off retail gas with a general intention of ending it. How far do these trends go before the economic rationale of maintaining long distance pipelines for gas becomes undermined noting the growing age of much of this infrastructure. Where is the new gas burning backstop generation going to be built and what supporting sources of gas will exist by then. Anywhere south of the Qld border seems doubtful. Has anyone really modelled the possible futures of gas transmission and distribution over the next 10 to 40 years? If we only need gas for backstop generation ie. rarely but with high volume for days, is it at viable cost even with large subsidy for the pipelines and storage as well as the generators?
I wonder what would happen if Fed & State policy conflicted, how would that get resolved in terms of AEMO modelling the ODP ‘while meeting government policy’? Eg. Fed Net Zero 2050, Qld extends life of Kogan’s Creek to 2052.
Would be good for AEMO to produce a counterfactual of ODP with only national carbon targets but no other gov policy (eg. No offshore wind in Vic, no state targets).
Should be no real surprise. The WEM has seen about 3 weeks of low renewable output- a bit longer than usual for this time of the year but increasingly the norm. the couple of great days were cloud cover and no wind. Every scrap of remaining coal ran base load to make the energy stored in the batteries. Just as well the carbon filters weren’t on. For SA however it does beg the question what is going to keep the lights on when TIPS B and Osborne aren’t there. Where is the market signal to put the capital up for OCGTs to deal with these sorts of events? Whilst your topic here was which was worse, 2019 or 2026 this is still less than a 1 in 10 event and happens pretty much annually to some extent. But you won’t make money out of a power station that runs only a few days a year without a capacity market.
Hi Paul,
To assess the recent wind drought, I selected a capacity-weighted mean location across current South Australian wind farms and extracted ECMWF IFS 100m wind speeds at that point. After applying a 4-day running mean, the recent period came in at just under 3 m/s — and looking back through the 9 km resolution model record, you’d need to go all the way to 2017 to find anything comparable.
The ERA5 reanalysis, while coarser at 31 km resolution, extends the record back much further. There’s a 5-day lag in availability, but it should be valuable for putting this event in historical context — how rare is a wind drought like this, and how often has South Australia seen something similar?
*Using the Capacity-weighted mean location is a huge simplification, but it’s quick and practical.
I will provide the similar ERA5 plot when the recent event values are available.
This is really interesting. The three BESSs are using grid forming technology with a contract with Transgrid to provide system strength in that region. And they were directed to disconnect due to system strength.
Whilst BESS growth does impact the intraday price trough, by far the larger driver is that LGCs (which VRE offer their energy at roughly the negative of) have dropped from somewhere around $50 to well below $10
Absolutely a factor in raising prices, LGCs falling from ~$50 to under $10 does make VRE bid less aggressively negative.
But that gets you shallower negative prices, not positive ones, crossing zero needs an actual buyer bidding positive, which is what the charge-bid data shows batteries doing in QLD and VIC.
Both matter, but I’d push back on “by far” for the positive crossing specifically.
For QLD vs VIC, I’d argue latitude is the biggest factor. QLD seasonal demand tracks quite closely with solar output – highest demand in summer, low demand in winter. Lowest is late May and early Sep, so not quite a perfect match to seasonal solar output, but certainly an enviable position to be in. To a large degree, rooftop solar is meeting all their demand growth and seasonal variations. VIC is opposite with highest demand in winter (and growing due to degasification [is that a word?]) and low-ish demand in summer (barring the occasional heatwave). Summer and winter demand are about equal in SA and NSW, and that is where fuel mix and interconnector dynamics are the main differences.
Agree that latitude is one of the factors. A big part of it is also that VIC and SA have highly correlated wind patterns (and a relatively high amount of wind capacity). VNI limitations too obviously, amount of ‘baseload’ coal in the fuel mix, etc. are other reasons why we commonly see a north vs south divide.
Very interesting thank you,
what is your opinion of the eventual flattening of the 6pm afternoon peak, due to BTM battery systems, and what effect that will have on retail prices?
Thanks Freddy. The impact of behind-the-meter storage is hard to quantify but the reported numbers coming out of the Cheaper Home Batteries scheme point to the fact that it will/is going to have an impact. Some of the AEMO numbers suggest installations of behind-the-meter storage aren’t uniform across regions (i.e. QLD installing at a much faster rate). I covered this a little bit during my presentation to the EUAA back in May: https://wattclarity.com.au/articles/2026/06/reflections-on-supply-and-demand-across-two-very-different-industry-conferences-last-week/
Re: retail prices — some impact, but obviously retail prices are made up of a lot more than just the wholesale component.
Re the comment about developing a large number of CCGT plant, I believe many of the existing open cycle plant will be quickly converted to CCGT once the full details of the ETS are know. Plant such as Braemar in Qld, Uranquinty in NSW were constructed with this in mind.
That’s a likely scenario.
Even some of the smaller and/or older plants have been designed with the potential for conversion in mind (e.g. Quarantine), though the efficiency of some, post-conversion, might be too low to warrant the investment (compared to the option of newer, bigger, higher-efficiency CCGT).
Your pictures above are very colourful and good for people who know nothing, but those of us in the industry know your analysis is extremely simplified and shallow (eg a)’with holding capacity’ is worng, as their available capacity remained the same – should say output instead of capacity, b) no mention of constrained off output from Tumut due to high import flows into Victoria constraining 65/66 lines etc).
I commend you on taking electricity to the people because it will only become more ‘in their bface’ once CPRS commences. Keep it up, but to maintain industry credibility, please make sure it is correct and not sensationalised (which I know gets the media grabs).
1) You are correct in that we do try to simplify what’s a very complex industry, in order to make it more broadly understandable. At times this does mean that we can omit certain details – whilst we try to ensure this is not the case (especially if the details are core to what we are discussing), we do apologise for cases (such as this) when it does occur.
2) I believe you are referring to what we wrote specifically about what happened on 17th March 2008 (from p19 of this report). We tried to confirm this on email with you, but did not receive a response. In this instance, we recognise that it is possible that the output of the Snowy capacity – moreso Tumut, on the north side of the 65/66 lines could have been restricted as a result of constraints on the lines between these stations – however this was not checked, as we were focusing moreso on SA at the time. (For anyone else interested in learning more about this issue, there’s a mountain of reports and submissions on the issue on the AEMC’s Website, in the past couple of years, leading to their decision to abolish the Snowy region and allocate Tumut into NSW and Murray into Victoria for reasons related to Apol’s point – i.e. that constraints within the Snowy region could impact so dramatically on market outcomes and yet not be very understandable, except to those intimately aware of their detail).
3) We also appreciate your comment about the term, "withholding of capacity".We appreciate that that term is incorrect (because, as you say, the capacity is still offered to the market, just at a higher price). It is for this reason that:
(a) we tried to make this clear in the report (such as on p11 of this report where we state that "about 70% of their generation appears to have been bid above $9900")
(b) we are very careful to use the different term "economic withholding of capacity" to clarify. Based on our experience, this term is one of a few (others include "generator strategic bidding" and "generator market gaming") that have become widely used in the NEM, and more broadly internationally, over the past several years. Though, we infer from your comment, the term is perhaps not as broadly used as we had first thought.
Once again, we do appreciate your comments, as they help us to understand how others view the explanations and comments we provide. We look forward to continuing to enhance the quality of the service we provide, into the future.
Regards
Paul
PS – it’s been pointed out to me that NEMMCO were about 7 months quicker than us, publishing this press release about the fact that a new record NEM-Wide demand was set on Monday 28th July 2008.
Also, we’ve been told that:
(a) There is a discussion on nem-wide demands in section 3.3.6 of the 2008 SOO.
(b) It provides similar results to those quoted in our article and notes a significant difference in the level of diversity between summer and winter.
(c) More infomation on diversity is provided in chapter 13 of the SOO.
Hence, we probably need to be running a peak demand picking competition in winter 2009 as well. More on that later…
PPS – in Winter 2009 we did launch a “Best Demand Forecaster” competition in winter for the first time!
You can also check out the range of entries we received this time, as what we asked for was slightly different than during summer.
From the Herald Sun website:
“National Electricity Market Management Company spokesman Paul Bird could not say exactly how many homes were affected.
“We instructed the transmission companies in Victoria to reduce the load because of that unusual event,” he said.
Energy and Resources Minister Peter Batchelor said electricity supply was “very tight” but there was “just sufficient power” to get by.
“Subject to no major plant failures and other catastrophic interruptions, we think we’ll get through today, the tightest of all days,” Mr Batchelor said.”
This is a conflict of information with what you have reported – does this mean that the media is covering up the fact that load was shed before basslink tripped due to the lack of reserve?”
Hi Kieran,
Thanks for your excellent question.
We’re still pondering this one, to try to work it out. What we can see from the NEM-Watch snapshot included above is that:
1) The Market Notice about Basslink trip was published at 14:14 (i.e. first) and the LOR load shedding in SA at 14:26 – both of these times are NEM times, we believe.
2) The Market Notice itself refers to load shedding starting at 13:50 (i.e. half hour beforehand). We thought that this may have been a local time in SA, but that does not make sense, either, as SA is ahead of NEM time by 30 minutes in summer.
3) We can also see (snapshot not included) that the load started dropping in SA at 14:00 and that this continued to 15:00 (NEM time).
4) In total there was a reduction of 200MW or so – so it is possible that some of this was not the INVOLUNTARY curtailment, and might have just been a coincident outage in ETSA’s distribution system, which was under strain all day.
Hence we’re not sure what the situation was …
Paul
Thanks for your excellent analysis of NEM spot market behaviour. It would be even more worthwhile if you could complement this with information and discussion about related derivative trading.
Hugh
Hugh,
Of course you are aware that the derivatives tend to trade on a quarterly basis for forward exposures, not in real-time along with spot market events.
d-cyphaTrade provide a comprehensive weekly market wrap of futures activity on their website (http://d-cyphatrade.com.au/market_wrap).
Joe
Hugh,
Apologies for the slow response.
We realise that the wholesale market is only 1/2 of the puzzle, and are working on ways in which we can shed more light on the operation of financial market trading (both OTC and ASX), but are not at a position where we can say much more yet.
Cheers
Paul
The two photos in this article were both taken on the same camera and accurately reflect the differences in switched on lighting in Brisbane at the time.
The reason that they use different exposure times is that the fifteen second exposure shots I took during earth hour were over-saturated by the reflected light from the cloud cover.
The only alteration performed on the photos after they were taken was some slight angle correction due to my camera not being perfectly level when the photos were taken.
I was personally quite surprised at the level of lighting reduction during earth hour. Even after earth hour had finished there were less lights on in the city than there were the night before.
Disclaimer: I am personally skeptical of the benefits of earth hour and did not participate in the event, except to take these photos.
Guys, this is a fascinating analysis. Keep up the good work.
Thanks Brenna,
Unfortunately only get time to do this kind of thing in between our “”real work”” of software development. Hence, posts will continue to tend to be sporadic!
Cheers
Paul
PS – a couple of our friends at NEMMCO have pointed out a couple of things following from this article:
1) Technically, this is the 4th time the CPT has been reached (as both SA and VIC reached the CPT as a result of the heatwave this summer – as we noted here)
2) NEMMCO has provided this briefing paper on the Operation of Administered Pricing in the NEM – amongst other things, it clarifies how the pricing in other regions can be affected, due to the operation of the interconnectors. For instance, in the case above, prices in VIC could be affected because of flows over Basslink, meaning VIC also experiences a defacto cap, in some scenarios.
3) NEMMCO has also noted (with reference to this Power System Operating Procedure) that they remove the cap if the following logic holds:
If NEMMCO has applied the APC and at the end of that business day, for Trading Interval 0400 hrs, the Cumulative Price is;
< CPT, then NEMMCO will not continue the application of the APC
> CPT, then NEMMCO will continue with the application of the APC
If at the Trading Interval 0400 hrs the Cumulative Price is less than the Cumulative Price Threshold, it is of NEMMCO’s opinion that determination of the price exceeding the Cumulative Price Threshold during the next trading day cannot be accurately predicted due to market volatility and as such will not be utilising NER 3.14.2 (c) (3). However, NEMMCO will exercise this option under the appropriate circumstances.
Hope this helps!
Paul,
Thanks for the feedback, but I can’t let you get away without at least one comment. You mentioned reviewing the accuracy of the load forecasts. I suggest you have a look at Appendix C.
Regards, Anon.
In relation to the chart shown above with respect to AGL’s Torrens Island plant, this report in the SMH will be of interest.
In his article, Brian Robins makes reference to this investigation at the AER.
Please note our view of a few principles of how an energy-only market should operate.
As a PS to my post above, Andrea has requested we make a donation, instead, to her favourite charity, the Permaculture Research Institute, which we were very pleased to do.
Andrea explains:
In that we’re firmly focused on making the world a better place, in our own small way (by making the electricity market more understandable – hence facilitating greater efficiencies, higher long-term sustainability and lower prices) we’re more than happy to help Andrea in this way.
I think Joe or anyone else who makes NEM price comparisons when spruiking benefits or costs for consumers is very brave to do so. The only reliable data source on retail electricity prices for all States in Australia is the CPI survey.
The electricity price indices included therein:
1. are based on incomplete coverage (being capital city only); and
2. do not by definition allow for price level comparisons between different states.
With respect to Joe Tripodi’s contribution at the conference, I would like to add that I was particularly pleased to hear his focus being on ensuring that (through the structure of the sale process so designed) the amount of competition in the market would not be diminished, and may be strengthened.
This was noted on two points in particular:
1) The need to ensure at least one new entrant in the market, and reserving the right to IPO Integral-Eraring if this is not achieved in trade sale.
2) Splitting Delta and MacGen trader rights into 4 parcels. As Joe noted, the co-insurance arrangements he spoke about would be one key to making this plausible (though there were no details provided of the specifics).
Philosophically, as a company we believe that the competitive market is (on balance) a better chance of delivering sustainable benefits to energy users – hence we see this as a positive step.
Might not be an “ideal” way to extract the NSW Government from it’s current conflicts of interest, but stands a good chance of being workable, in principle (though depends on the substance of where responsibilities lie under the gentrader model).
Will be very curious to see:
1) How many foreign bidders actually emerge after the global road-show earlier in the year.
2) Whether any large energy users team together to buy out one of the gentrader rights on issue – as I speculated a couple of years ago now.
I was speaking to a client earlier today about this event. They indicated that VOLL was hit by a combination of very high temperatures in Sydney combined with the trip of a unit in NSW (I do not remember which one, though from the chart in NEM-Review I have to assume it was one of the Eraring units).
I also got the impression that as this is still September, there is a number of units that are still out for maintainance that would have helped with the load had this occurred in December or January.
And by September I of course mean November (actually I meant to type “Spring” but “September” came out instead).
I would just like to clarify one of the points I made in this article.
Why does interconnection increase economic efficiency?
In general terms, the more interconnected a network and the larger a market, the more efficient it becomes. Connecting Mount Isa to the rest of the NEM will allow Mount Isa to purchase cheaper power during off peak periods, and will provide strong price signals for demand management during peak times. From the point of view of the generator at Mount Isa, exposure to the NEM makes them able to operate more efficiently as well, by allowing them to take advantage of high price periods.
A quick PS – I missed this Market Notice that AEMO issued this morning, which explains the tightness of supply/demand.
1) Taking one transmission line out of service reduced the transfer capacity into SA by half, hence exacerbating the shortfall.
2) Note the alert level was actually LOR2 (i.e. more significant).
See this notice in the list shown in NEM-Watch above (I just did not pick it up earlier!)
——————————————————————-
MARKET NOTICE
——————————————————————-
From : AEMO
To : NEMITWEB1
Creation Date : 12/11/2009 07:03:49
——————————————————————-
Notice ID : 28695
Notice Type ID : RESERVE NOTICE
Notice Type Description : LRC/LOR1/LOR2/LOR3
Issue Date : 12/11/2009
External Reference : Actual LOR2 South Australia Region – Thursday 12 November 2009
——————————————————————-
Reason :
AEMO ELECTRICITY MARKET NOTICE
Actual LOR2 South Australia Region – Thursday 12 November 2009
Refer Market Notice ID 28684
Due to the planned outage of the Heywood Moorabool No 1 500kV line in the Victoria Region, a credible contingent event may result in a separation of the South Australia region from the remainder of the NEM.
Load may be shed automatically as a result of low frequency in the South Australia region.
There are sufficient capacity reserves in the South Australia region but the technical characteristics of the capacity reserve generators may delay response to the contingency.
LOR2 declared from 0655 hrs until further notice.
Alan Honecker
Power System Operations Manager
——————————————————————-
END OF REPORT
——————————————————————-
In your description of the 10 year look ahead (ESOO) you have
“Because this document is published only every 10 years, these forecasts are provided in the form..”
You have missed a bit – it is published every year, looking ahead for 10 years … – or something similar.
Might be worth changing to avoid confusion.
Cheers
Tim
P.S Really enjoy your updates.
Thanks for picking that up, Tim!
Silly mistake – have corrected now.
Cheers
Paul
Paul
There are several problems with load forecasting and predispatch for the 1 in 10 year event or even more extreme event.
a) AEMO do not appear to adjust the predispatch forecasts for humidity and wind speed. In contrast the Victorian gas industry does. (So AEMO has access to the skills set from VENCorp to make the approariate adjustment to the process.).
Over the years the Reliability Panel in their Annual Market Performance Review present information relating to variations in predispatch forecast and the actual dispatch in Figures 5 to 10 in their draft Report 2009. They appear to have convieniently exclude errors associated with peak loads. Several of the earlier AP Performance reports indicated significant predispatch errors. Each year they claim to be reducing the number and size of errors.
b) There is limited information relating to size and type of airconditioning for less than 160 MWh consumers across the NEM. My local Harvey Norman sold $ 180,000 worth of air-conditioners on the 6th of Feb 2009, the next summer they will add to peak demand along with the new consumer added to the system. Energy Australia is one of the few distribution businesses who penalise consumers for moving from one to three phase.
One can also interpret the data provided by ETSA from their Demand Mangement trial that 75% of the air-conditioners in SA need to be replaced so that they can effectly load control the business!
Thanks for the comments, John,
The analysis above (and what I have added today for the other 4 regions, and linked above as well) just relates to the MT PASA and ST PASA forecasts.
As I noted in my “part2” post, there are a number of things about these forecasts that I don’t fully understand.
The issues you raise with respect to pre-dispatch are additional to these (and relate to an even shorter time-horizon forecast), and I have not spent any time thinking through these.
Certainly I agree with you that one of the largest uncertainties facing everyone in the NEM is the fact that there is no central reference that indicates how much new demand capacity (particularly AC loads) has been added in each distribution substation area from one year to the next. As far as I am aware, no-one really knows the full picture!
Cheers
Paul
Finding ‘categoric’ data that quantifies Earth Hour results is difficult, and the variables on any given weekend often mask events as narrow as this. However, taking a look at the last 6 consecutive Saturday demands in each NEM sector does give me a little more clarity as to ‘apparent’ effects.
I wont post the graphs and spreadsheet here, but suffice to say that NSW overall appeared to have a subtle 5%-10% nibble, while Victoria and SA shared a larger, longer 10%-15% bite.
Compared to similar views of previous events it seems that Earth Hour is losing its Mojo in Sydney and waning elsewhere in Oz.
Maybe we’ve already achieved so much lighting efficiency by changing over to CFL’s that we barely register a change from turning off lights anymore.
Enough of my highly scientific analysis – over to the experts.
With Australia’s expanding population it is very hard to imagine that this reduction in emissions would turn into a trend. For this to happen the inevitable increase in demand for electricity would need to be more than offset by improvement in efficiency (unlikely) or by improvement in emission intensity. To my knowledge we haven’t closed any coal plants so presumable the drop in emissions has come from reduced capacity factors. As demand rises capacity factors for coal plants will increase again. New gas plants will reduce the emission intensity further but I would bet that emissions will return above 2008 levels before long.
Hi Kim,
You’ve given us the numbers but do you know the story behind the numbers? It was a mild day with little indication that such spikes would eventuate so I would be really interested in the factors contributing to this seemingly inexplicable outcome.
Cheers
Swee Fong
well done Paul – an excellent synopsis. You have obviously been paying attention at Roman’s conferences (which I think are very good also). I look forward to reading the next few chapters
cheers Jim
Thanks Jim!
Paul,
A very good start.
One of the big problems with electricity transmission and distribution regulation is the Propose-Respond model that the AER has deveoped and had enshrined in the NEL (I think) whereby the DB’s can put forward whatever they want by way of Capex and Opex and it is up to the regulator to try and determine whether it is valid or not.
Unlike non-monopoly businesses these businesses do not have to demonstrate and capital restraint.
What they seem to do is underspend on capex for the first few years of a 5 year regulatory period, then overspend in the latter years.
They can keep the underspend as an “efficiency gain” and then they get to reset their asset base and recover the over-spend as a P 0 adjustment when the new regulatory period commences.
I believe this is called regulatory gaming.
There has to be a better way.
If users challenge an AER decision then we have to foot a huge legal bill whereas if the DBs do the same they have nothing to lose as any legal costs just get funded by an increase in their Opex which we pay for anyway and on balance they more often than not get a better outcome as the Appeals tribunal has no idea about reality it would seem.
The system is well and truly skewed in favour of the DBs.
Thanks Peter,
I am still trying to get my head around some of the aspects of the concerns you have expressed (plus some others) about the revenue reset process for DNSPs and TNSPs.
It seems to me that, compared to 5 years ago (when it seemed like it was not much of an issue at all) it is certainly top of mind at present.
Paul
PS – just noted that the new AEMO website has this additional information here about the price spike on the 29th June.
PS – today the AEMO has published this communication to explain yesterday’s interesting events:
I will look forward to seeing the final report – and (if time permits) will post it here as well.
In case you were curious, these notes are normally copyright & confidential AEMO, so I ran this past Brian and he said it was ok.
The price hike in 2007 was due to two powerstations reducing output due to a government restriction on taking cooling water. I wonder if privately owned powerstations would have been so quick to cut production in lieue of finding alterate cooling methods? Consequently, it seems that the government owned power stations were the big winners during the drought. Perhaps while the Queensland government retains business interests in the power industry, they should recuse themselves from legislating where such legislation could be seen as favouring their business interests.
Great stuff Paul –
Some very interesting graphs.
Thanks Derek,
Would be interested to learn more about your particular areas of interest, so will contact you offline to understand this.
Cheers
Paul
It’s like you know what I’m thinking Paul?
Sitting at my desk trawling through 100+ page reports at Melbourne Business School on the final stage of my MBA researching wind and embedded generation and up pops your graphs on the changing generation mix! A quick cut & paste, a bit of text and referencing (of course) and I’ve only got 39,500 words to go. Sweet. Thanks mate. Cheers Marc
As always, Marc, glad to be of service.
Cheers!
Paul
I’m speculating whether a privately owned power station would not have found alternate methods for cooling, if it meant that in doing so, they would not have to shut down due to drought inflicted water restrictions.
So whats the cause of the current volatility? Unbelievable.
Hey Paul
Great work on posting this article so promptly!!!
I’m studying power engineering at university, and I regularly visit your site. I really enjoy reading your reports on events within the NEM. Shame it wasn’t so hot down here in Vic, otherwise I also think that the NEM-Wide demand would have been clearly broken. Do you think that with most schools up and running now that the first term has started, the demand is likely to be higher due to extra HVAC load?
Also, do you know if the maximum supply available in S.A is limited by generation / transfer of power into S.A, or limits within the S.A transmission/Dist. system?
Furthermore, I would like to know whether you do a student edition of your NEM-WATCH software, for personal use.
Thanks
– Kieran
PS – this morning the Age reports that it is at least 170,000 people without power
PS – also Powerlink’s tweets clarify that the 275kV network is ok, but the 132kV network has some issues.
A lot of residential load was disconnected very early on Wednesday night (5:30pm where I live) when there wasn’t much wind yet. This would seem to suggest that the Ergon system really is quite fragile. I will need follow this up with Ergon once the dust has settled. With the rediculous hikes in prices they have been charging in recent years, I expect a corresponding improvement in service.
In section 2 you state that the difference in peakiness is primarily a result of weather patterns in the regions. I’d add (as you state in the final bit on Tasmania) it is also quite strongly influenced by the proportion of load being consumed by our major industrials. And maybe rather than Tassie having cooler summers we can say ‘less extreme weather changes’ 🙂 Hey, I’m looking for a job for Tourism Tasmania! And being a bit pedantic – it is not the cooler weather but the lack of air conditioners that impacts on consumption – the cooler weather makes it less important to buy a/c in Tassie but the lower average income in Tasmania also impacts on the number of people with a/c. Just some patriotic thoughts!
Tim
Hi Tim,
Thanks for the pointers.
With respect to electricity, Tasmania is a fascinating place. Four big loads account for about 60% of the average demand shown in the table above, all with some flexibility to provide some form of DSR to varying degrees. Predominantly (zero SRMC) hydro with relatively large storage plus a single interconnector that is very large in relation to the demand in the region.
In terms of a tourism ambassador, sign me up as well. Lost count of how many times I have been there, and have enjoyed every minute.
Cheers
Paul
Paul
Note that the NEM demand you use (TotalDemand in MMS) is actually a 5-minute forecast rather than an actual, and excludes demand served by significant non-scheduled wind farms in SA. If you include those, and use the half-hourly average demand then you will find that the NEM-wide peak summer demand was 35,568 MW which occurred for HH ended 13:30 on 1 Feb 2011.
Just a thought
Thanks Ross,
The demand data we show in NEM-Watch (and hence use in the competition) is the dispatch target, as you note, for a couple of reasons:
1) The time stamp of the demand target aligns with the time stamp of dispatch pricing and the decision was made back when we first built NEM-Watch (in 2000) to ensure these were aligned.
2) In these early days, there were fewer other measures of demand available, as well.
The other difference you note is between 5-minute data and 30-minute data.
I know that the different measures of demand can confuse NEM observers, so I’ve tried to explain this a bit here.
We’ve made a note to look into what we can do to also include the metered numbers, as well, in some way.
Thanks, once again
Paul
Have you included non-scheduled generation, a classification which grew substantially during those years when the energy growth appears to have slowed?
Great catch, Ben
You are correct in that the numbers shown include only Scheduled and Semi-Scheduled generation, at this stage.
When I get a moment, I’ll update the post with another chart including the Non-Scheduled generation.
Cheers
Paul
PS Ben – had a chance to have another look and so put these additional thoughts together here.
The more I look, the more fascinating it is.
There are a number of reasons quoted for a flattening in demand growth – which has had the most impact, and will it continue into the future?
Two comments/questions on these data:
1. Do you keep track of degree days for winter – say with 20C base temp perhaps? Or at least average ambient (not daily min/max as most commonly reported out of BOM) Degree days give a bit of insight into seasonal variation in heating load (for winter)
2. Is it just possible that the pink batts rebate is also having an impact at the margin? There was a lot of insulation put in over a short period before this program sank. I’d guess the impact would be to mitigate both aggregate heating load, and possibly peak heating load, if only a little.
Hi Jim,
Thanks for the questions:
1) Yes, we do have a record of weather data (including temperatures) – however we don’t currently have the time to do anything with them (such as comparing winters to try to understand why demand is declining). It seemed as if it was roughly a “normal” winter – though we have not analysed the data to back up that sense.
2) I’m sure that the insulation would be having some effect at reducing winter demand (and also summer) – it’s one of many contributing factors we have identified and will post about when we have time. The more challenging question is how much effect they have had – or even how we could determine this!
Paul
Hi Paul,
Another interesting article.
Just a couple of ideas you may want to add to your list (if you dont have them). Some are a bit left field but you may find something interesting in the data. The correlation between the demand data and:
– ASX Market Cap
– The AUD
– Severe weather events
– Variances to weather forecasts
– Demand Response products available
– Demand based retail contracts signed
– Co- Tri-Generation systems installed
– Number of Business Consumers
– Inverse Correlation to median business electricity prices.
Thanks Kenji
A few things for me to think about (yes, some a bit left-field!)
Cheers
Paul
Good morning
Additional (albeit minor) reasons reducing electricity demand (apologies for the Queensland bias):
– Increased Demand Side Management e.g. Energex’s Cool Change Trial – A 13 per cent reduction in peak energy demand was achieved on the hottest day of the 2009/2010 summer (www.energex.com.au/__data/assets/pdf_file/0015/32352/Cool-Change-Cool-Change-Newsletter-November-2010-with-form-number.pdf)
– Household types and composition. Increasing densification of capital city populations is changing the historic relationship between population growth and electricity demand growth. IPART’s survey of residential energy demand illustrates the difference between house and unit electricity use by number of residents, as well as the impact of adults (>15 years old) versus children.
– Industrial strucure and growth. Low growth in energy intensive industries (e.g. manufacturing) and high growth in service industries (typically low energy intensity) has reduced business electricity demand growth.
– Resdiential high-draw devices. There has been sizeable growth in high-draw devices (primarily air conditioners) over the past 15 years, but the market is now close to saturation point in some categories (what is stopping the 20% of high-income households that don’t have an air conditioner from purchasing one?), which coupled with increased efficiency of high-draw devices, would be expected to reduce growth in electricity demand from these devices (closer to the growth in new dwellings).
Evan
Hi Evan
Thanks for the thoughts – and apologies for the delay in responding.
I am curious to delve further into the detail, in order to understand the significance of these factors, and others people have noted. When time permits, I’ll be doing this with the database we’ve accumulated of sub-regional demand histories across the NEM – which should allow the benefit of (to some extent at least) looking at the effect in different sectors of the economy, and with respect to different weather patterns, etc….
Regards
Paul
Good day,
If we are referring to forecast maximum seasonal demands, it could an overly optimistic forward projection influence mainly by outlandish economic forecast, the need for capx investment into aging network, influx of migrants, the weather, etc, etc.
Notice that post GFC period, the economy did not shrink but grow at a slower pace, which means that the behaviour on energy utilisation has changed. This is inline with what Paul has mentioned on energy efficiency measures and retail price of electricity. However not forgetting that in recently 1-3 years, the weather is cooler not any warmer (especially in Queensland for where i am from).
Thanks Ken
Appreciate your points.
The numbers I have looked at, to date, have been NEM-wide. To some extent this means that regional patterns (e.g. cooler summers in recent years in QLD) have their effects masked.
Regards
Paul
There is another rather prosiac reason for a weak energy demand in 2010/11: the natural disasters over the summer, most particularly the Brisbane River flood which supressed Qld demand for over a week. It would be possible to estimate this effect.
On “embedded generation” a few points. Of course, what NEM commentators colloquially describe as “demand” is a misnomer, it is actually a measure of generation, and then only a subset of that-that which is easily measurable. And the “demand” excludes some generators, but includes losses in generating, transmitting and distributing electricity, which all change in turn with the pattern of dispatch!
The easily measured generation is all “scheduled” and “semi-scheduled” generation. This is most generators with connection point capacity>30MW and some larger windfarms built prior to 2010. AEMO publishes the large non-scheduled wind generation separately, so that “demand” can be recovered, but as you correctly point out there are still many other non-scheduled generators for which AEMO has no data, such as PV’s and co-generators below 30MW.
On the terminology, the rules define “embedded generation” as any distribution connected generation (which includes the 400MW Tallawarra!)-so I think what you mean is “non-scheduled”. We have lots of embedded scheduled generation, such as Tallawarra.
On attempting to estimate the trend of demand for the future, the AEMO CEO presented on this matter at the ESOO launch: http://www.aemo.com.au/corporate/0007-0003.pptx
Thanks, Ben (once again) for contributing
The ESI is full of commodities that are difficult to measure in actuality!
As noted to Evan (above) I am keen to delve more into the history of zonal data we have in order to at least be able to identify the extent to which the trend observed on a NEM-wide basis (with the approximation of real demand) is similar across the 26 zones for which we have data NEM-wide.
With so many data sets to analyse, it should be able to provide some insight into the extent to which various factors are at work – e.g. residential -vs- industrial -vs- commercial, north -vs- south, etc…
In terms of estimating demand for the future, I would also like to have time to complete this type of retrospective on forecasts. Perhaps AEMO might include this in the data it releases in 2012?
Regards
Paul
Hi Paul,
i don’t have numbers for you but have a book called “Brown Power – A jubilee history of the State Electricity Commission of Victoria” It is a self published work from 1969 and gets back into the halcyon days of brown coal expansion that lead to Loy Yang etc. There are many interesting aspects to this, not least the intent and associated policy and practice of promoting load growth. Linking this with the nature of large thermal plant there was a natural desire and deliberate process of growing what we currently call ‘base load’. Of course you know what baseload is, but it is an artifact of both technology and socio-politcal intent. Baseload is an outcome of historical priorities and triumph of one story (coal) over all competing others. Hence the title of the book.
Long story short, I’m contemplating revisiting this book over the next few weeks. If so, what would your review questions be in light of your current historical investigations?
Jim Castles
Hi Jim
Thanks for the question.
Unfortunately I can’t really comment in detail, not having read the original edition to understand its context.
Obviously there has been no new brown coal development since the VicPool market started – though my understanding is that utilisation rates increased, and have remained high. I’m not sure how this fits into the context of the older book.
Regards
Paul
Hi Paul,
thanks for providing a bit of fun and entertainment over the Christmas break, and you’re right: I really do need that barbeque!
But you are wrong in saying that I only submitted 9 guesses. In accordance with your Terms and Competitions, I provided a guess for every single day until 2 January 2012. But your analysis only includes guesses until 31 Dec 2011.
You wouldn’t want to deny us of those valuable last two guesses on 1 and 2 January now, would you?
Hahaha, gotta love people who work in the energy compliance area and read the fine print…
Cheers
Reinhard
Hi Reinhard
Thanks for picking this one up (and please accept my apologies for the goof).
I have updated the listing, and the analysis on this new posting made today.
Like many, I will be very curious to see where peak demand lands this summer (and I don’t even stand to gain a BBQ!)
Regards
Paul
An excellent chance to test the theory of http://en.wikipedia.org/wiki/The_Wisdom_of_Crowds
Maybe we should take the average of those with multiple entries and then the average of each individual’s pick and see if the “crowd” is indeed wise?
Tim
Hi Tim
Thanks for the comment.
It is an intriguing book – it’s one of the ones in our office library that has been read and reviewed by others in the company (and I appreciated the Wikipedia reference).
In updating the results table to correct my omissions, I also included another chart at the bottom designed to draw out, as much as possible, the Wisdom of the Crowds in this instance.
Interested in your thoughts?
Paul
Tim,
I would certainly be curious to see the result of such an analysis. If you find the time to do so, let us know the outcomes.
Cheers
Reinhard
Gee, who would have thought that we’d have such low demand this summer!?! I can see my bbq ambitions going down the drain. But maybe we’ll still get a bit of a summer for the remainder of Feb, and temperatures will hopefully go up a bit…
Some wind energy advocates have argued that it is wind that is driving down spot prices – they like to refer to the “merit order” effect. Paul would you like to comment on this?
Hi Martin,
Thanks for the comment.
When I have more time, I might post a bit more about this.
In the meantime, the simple answer is that everything that’s having an effect of either:
(a) reducing demand, or
(b) netting off any generation that bids above $0/MWh (i.e. thermal generation and other)
would be having some effect on lowering prices across the NEM.
My suspicion (without having looked at the numbers in detail) is that the volume of generation from wind is still too small to be having the most significant impact in this respect.
I suspect that any effect of wind might still be overshadowed by such greater impacts as:
1) Lower demand, which has been offset by a range of factors, including those linked to in the article above; and
2) The availability, for a couple years more, of cheap “ramp up gas” from coal seam gas fields in QLD, providing a cost incentive for open-cycle plant to run more heavily loaded than might “normally” be the case.
3) It’s possible (though I would need to check the data) that the higher production volumes from hydro, noted in this article, would be having more impact (because, through significant storage, hydro output can be targeted at times when prices might otherwise be high).
It’s my understanding that part of the enduring challenge with wind is that there is still a degree to which diversity patterns are not so great as to overcome the situation where localised wind patterns are independent of peak demand times, or peak price times. Interestingly, there’s some charts that surprised me for the UK here, showing less diversity than I had envisaged in that part of the world.
That said – through the remainder of the decade, as the increased MRET target continues to deliver more wind farms, the effect of wind on reducing peak prices will increase (all other things being equal).
However, if this was the sole objective of wind farm development, there’d be plenty of people who’d point out that there would be much cheaper ways of achieving the same goal – it would seem to represent a Pyrrhic victory, if done for just that reason.
Does this help?
Regards
Paul
Hi paul
we are studying the impact of the renewable wind and solar on WECC Western energy coordinating council and PJM e.g. the whole USA and N america. We see a big increas in CC generation and cycling to have 20-50% penetration of wind in the USA/ N america.
stay in touch
Thanks Steve
Your comment is not clear – are you saying that you’ve seen an increase in cycling of thermal plant attributable to increased penetration of wind & solar?
Paul
Thanks Paul for your response. It would be good if someone with your electricity market knowledge could write something on the effects of wind on electricity prices in Australia. A number of renewable energy advocates are pushing the line that wind is driving down wholesale prices. I would like to see an analysis from a less biased source.
Thanks Martin,
I have inferred from your comment that you do see us as a source of objective analysis and commentary. This is certainly what we strive to be.
I’ll add your suggestion to our list of “things to do”.
Regards
Paul
The comparison table illustrating the changes does not include the NT. As a resident of the NT I would be interested in seeing the before & after affect of the CT
Thanks & Regards
Pat O’Connell
Hi Pat
The NT is not part of the NEM – so we can’t help you there, in terms of what’s happening up that way. Perhaps the esaa might be able to help?
Paul
Pat, Paul,
NT is a fully regulated system so Power and Water Corporation would have to apply to the NT utilities commission for an increase to retailer prices based on the increases to their cost due to the carbon price. Their website is http://www.utilicom.nt.gov.au/ where you may find some more info.
Cheers,
Ariel
Was the new MPC effective from 1st July? I couldn’t find any info from AEMO’s website. Any link or news about the change? Thanks.
Hi Ray
Yes it was effective from 1st July. We were surprised to see it, too – there was probably an update from the AEMO or the Reliability Council, but we missed it…
Paul
We most of the SA wind farms not producing because of wind ….. Or because they were off line? My hair was ruffled by a breeze in Adelaide yesterday!
Have not had time to look at this one, Mike, but will try to remember to do so, next time I open up NEM-Review.
Regards
Paul
Paul
despite your accurate comment at the beginning “it’s really too early to make any significant observations”, it is still interesting to watch what happens. But, as you say, too early yet to read too much in to it. Will watch though, and hope others do too.
Agreed, Rupert
Just because it’s early days does not make it uninteresting. Indeed, perhaps the opposite is true, which is fine – so long as we don’t run ahead of reality with conclusions.
Paul
Hi Paul,
As always, very informative insights. Will be looking forward to a follow up review after a full month of trading.
Thanks Kenji
Will look forward to posting more, as time permits.
Paul
Nice work Paul … Interesting as the wind increases it’s voilitility will drive more cycling and higher costs and more emissions. Remember wind is not free. We calculate wind integration costs in terms of extra cycling costs. This can be done for the carbon tax effect too.
Steve Llefton. Steve.lefton@intertekmcom.
Thanks Steve
Have not had time to look, yet, at which of the thermal plant are cycling more because of the intermittency of wind.
Had not thought of the increased emissions (from the cycling thermal plant) that would result from running at sub-optimal heat rate, but it’s not likely to be of the same order as emissions saved from wind production (as it’s not the brown coal plants that are cycling).
Concerns about cost, though, have been raised by others – but more with respect to capex for other build required to provide capacity when the wind is still (whereas you’re talking about increased O&M because of more creep and fatigue degradation etc, I think?).
Paul
Paul
Thanks again for another insight into the changes. With respect to the downturn in generation, we Engineers often overlook the economic perspective of cost elasticity of demand. The first time I was reminded of this was back in the 1980’s when NSW introduced the first inverted tariff. Fifty or so years of increasing public electricity use (thanks to price decreases) suddenly stopped!
That could clearly be part of what is happening now and, if cost (Carbon plus regulated costs) is a larger component than we expect, the downturn could accelerate. Not a win for renewables necessarily, but a win for energy efficiency and conservation.
Thanks Peter,
As you imply, reductions in demand due to higher costs could be significant – and could well be more significant than the effect of embedded generation of any type.
Will look into this further, at some stage.
Paul
It’s not difficult to see why wind is only given a capacity credit of under 10% – often much lower.
Thanks Martin
There’s some interesting charts of the European (UK and Spanish) context of the correlation of wind production across multiple sites in this report here:
http://escornwall.com.au/index.php/2011/05/jonathan-dennis-final-report/
Paul
Yes please take the time! This looks interesting. Are the peaks actually maximum NEM wind capacity?
It’d be good to also understand it on a state by state level to perhaps see how dependent it is on regional weather patterns.
Thanks Adam
I will put this on the “things to do” list and return to it when I can.
I suspect that reducing to individual states (or other areas) will highlight a greater individual variation of output from various wind farms.
Paul
Wattclarity – exactly as the name suggests
Would also suggests that rising power costs are a combination of infrastructure building, peaking demand leading to higher cost generation options etc etc. Carbon Tax may come into play with respect to cost increases but the data will reveal that in time.
Thanks John
It is nice to know when people appreciate how any of our products do help to “make the energy market more understandable”.
In relation to energy costs, you might be interested in a series of posts I started 2 years ago about the concerns of energy users, which included some indication of the drivers of higher costs that were apparent even at that time.
You’re not alone in pointing a finger at the rising cost of TUOS and DUOS charges, resulting from increased Capex allowance in the most recent AER regulatory findings. There are a number of reasons for this, some of which have been discussed by others elsewhere.
In addition, other factors are also at play, including the following:
1) The costs imposed this quarter by the introduction of the carbon tax (which we’ve started to look at recently); and
2) The rising cost of the various renewable schemes and subsidies (both federally in terms of the LREC and SRES schemes, and state-based in terms of attractive feed-in tarrifs).
3) Possibly there is still a hang-over from the period in 2007 when the drought drove up spot prices and retailers locked in hedges at these inflated prices which are still washing through into the retail space. We’ve not had time to investigate this.
As we’re striving to maintain our
Hope this helps.
Paul
Just a question or two:
is there anything akin to shorting in this market?
if so, who is taking a short position and are any particular generators being targeted?
are there any accurate/reliable figures showing the total energy generated from all sources, as opposed to this figure for demand (in the NEM context) that you have used here?
Good work and thanks, Steve
Hi Steve
Thanks for the curly one – I’m not aware of this happening in the NEM, and can’t immediately think how it could happen.
Maybe one of our other readers can shed some more light on this?
Paul
Hi Steve,
Being a cash settled futures market, electricity traders sell futures just as easily or as often as they buy futures. There’s no physical supply obligation attached to an Australian electricity futures contract, so it’s not like buyers or sellers of electricity futures can “corner physical supply” via the futures market. There’s no limit to the size that the cash settled derivatives (futures and OTC) can grow to as a multiple of the electricity consumed in the physical pool market. Ultimately, any futures positions held open as at contract expiry (i.e. at the end of the relevant quarter) are cash settled against the AEMO pool price average across the underlying quarter. It’s still the marginal generator(s) that set the pool price every half hour during that period, regardless of how many futures contracts have traded or at what price those futures contracts traded. Hope that helps.
Dean.
Thanks Paul for the great summary.
If anyone would like to do further research on the Australian Electricity Futures & Options Market we freely publish a weekly Market Wrap (here: https://d-cyphatrade.com.au/market_wrap) and a bi-annual Energy Focus newsletter (FY12 can be downloaded here: https://d-cyphatrade.com.au/newsroom/energy_focus/energy-focus—-financial-yea).
Additionally, the online historical Data Centre includes all trades since Sept 2002 is available via an annual subscription here: https://d-cyphatrade.com.au/products/data_centre.
If anyone has any queries about the d-cyphaTrade ASX Electricity Futures & Options market please feel free to contact us on toll free 1800 330 101 or email info@d-cypha.com.au.
SSL Error
You attempted to reach d-cyphatrade.com.au, but instead you actually reached a server identifying itself as asxenergy.com.au. This may be caused by a misconfiguration on the server or by something more serious.
An attacker on your network could be trying to get you to visit a fake (and potentially harmful) version of d-cyphatrade.com.au.
You should not proceed, especially if you have never seen this warning before for this site.
Hi David
Thanks for pointing this out. The ASX bought out d-cyphatrade from the previous owners earlier in 2013 and renamed it ASX Energy.
Paul
You say:
“With reference to these daily production figures we see that average daily output as high as 1540MW has been seen from wind farms in 2011, which would have represented a daily capacity factor of the order of 70% – clearly a windy day.”
It would provide balance, and actually more important to point out, that there have been times when all wind farms in the NEM have produced no power. In fact, they have drawn more power than the generate.
In May 2010, there was near zero generation over a period of 4 days, under 10% capacity factor for 10 days, and 65 5-minute periods where the NEM wind farms were drawing more power than they generated. It is the minimum that is critical output that is critical, especially at times of peak demand.
Thanks Peter
The intermittent nature of wind generation means issues for the NEM at both ends of aggregate output:
1) As you note, some times the wind barely blows (even across multiple stations and regions) which means that, from a capacity point of view, only a small percentage of this capacity can be treated as firm at peak demand times.
2) At the other end of the scale, it’s often windy at night which has caused negative price events (mainly in SA) which has imposed costs on all generators – even the uncontracted wind farms.
These reinforce that wind power is primarily an energy-play, and not a capacity-play.
Paul
There has been heated debate since the report in the “Weekend Australian” (6/9) that wind energy is avoiding very little CO2 emissions and doing so at a cost of $1,484/t CO2 avoided. I would like to know how we, in Australia, can say how much emissions reductions can be attributed to wind generation?
Australia has no actual measurements of CO2 emissions from its power stations. We have crude estimates of emissions based on long term average fuel consumption. AEMO quotes the same emissions intensity for all units in the one power station no matter at what state of efficiency the unit is operating at the time and no matter whether the plant is ramping up or down. So clearly, the emissions intensity figures are very crude estimates and do not allow for changing intensity as the output and efficiency of the units changes.
We don’t even measure the fuel used in most power stations, let alone at the frequency needed to estimate the emissions avoided by wind generation.
We need at least the emissions data quality and frequency that Ireland has http://www.eirgrid.com/operations/systemperformancedata/co2intensity/ or, better still, we’d need what the US EPA requires for emissions measurement and reporting. But we’d need the information at preferably 5-minute intervals, or at worst 15-minute intervals, for every unit to allow us to determine how much CO2 emissions are avoided by wind generation.
Here is a recent analysis of the emissions avoided by wind generation in Ireland: http://joewheatley.net/emissions-savings-from-wind-power/
CO2 savings per MWh of wind generation (relative to savings at zero wind penetration) are:
Wind Penetration CO2 Savings
0% 100%
10% 90%
20% 77%
30% 62%
40% 47%
At 50% wind proportion of total generation, 1 MWh of wind would be avoiding about 32% of the emissions from 1 MWh of fossil fuel generation, all else equal.
In fact it’s probably worse than that. I understand the Irish emissions intensity calculations do not include the extra emissions from fossil fuel units, for example: starting up and shutting down; in spinning reserve mode (i.e. not generating), powering up and powering down to back up for the changing wind power, etc.
Ireland’s figures apparently do not include all the extra emissions attributable to backing up for wind. But Australia has nowhere near the quality of emissions intensity data that Ireland has. So how can we make any reliable statement about how much emissions are being avoided by wind generation? Can we state what is the cost per tonne CO2 avoided by wind generation?
Hi Peter,
Thanks for your detailed question.
I did not read the article you mention, but understand that there are a number of views being promoted on aspects of:
1) Whether the emissions reduction from wind is large, or not; and
2) Whether the same emissions reduction is expensive, or not.
In the analysis we publish (and tools we provide) we strive to remain technology agnostic – and just to present facts, such that people can make up their own minds in terms of judgements such as large/small and expensive/inexpensive.
In the 5+ month period analysed above, for instance, Metered Wind Production in SA totalled 1,591GWh, whilst total SA demand over the same period totalled 5,795GWh.
This means that (based on these numbers) wind supplied 27.5% of the total demand over the period – and, as the chart above shows, oscillated to a significant degree over that period between cyclic high and low points. Both factors reinforce the fact that wind power is an energy-play moreso than a capacity-play. Both are important in the context of the broader energy supply industry.
Determining what emissions reduction resulted from this production is not a precise science as one needs to hypothesise about “what might have otherwise been”. At an extreme we might look at the emissions factor of Alinta’s Northern Power Station (whose output has been curtailed due to a combination of factors, including declining demand and increased wind production) and calculate an upper bound of emissions reduction as follows:
0.95 tonnes/MWh sent-out x 1,591GWh = approx 1.5Mt abated.
Is that not much, or a lot? There will be a range of views on this.
In terms of the cost of this abatement, this is analysis that we’ll have to leave for another time.
Does this help?
Paul
Hi Paul,
Thank you for your quick response and for your upper bound calculation of the emissions avoided by wind generation in the SA grid.
You say:
However, you’ve estimated an upper bound but not a lower bound or best estimate. I suggest it is really important to present this information objectively and with proper balance. The costs and benefits of wind energy is an important issue. Australia is committing to and spending billions of dollars on wind energy, apparently without objective analysis of the costs and benefits. So I’d urge that the estimates for CO2 emissions avoided and the CO2 avoidance cost be calculated and presented objectively.
Your upper bound estimate assumes 100% of the wind power displaces coal power. Many studies show this is not the case. The paper I linked to in my previous comment is one example. There are many others from Europe and North America that reach similar conclusions. Australia does not have any emissions measurements, so we are guessing. We don’t even have accurate emissions estimates, let alone at the 5 to 15 minute time spacing that is needed to do the equivalent of the study of the emissions avoided by wind generation in the EirGrid.
I’d urge you to balance your comment by also presenting the lower bound and the best estimate.
I would suggest the emissions avoided should be compared against three baselines:
1. the average emissions intensity of the SA grid, excluding wind power
2. the average emissions intensity of the fossil fuel generation that would have been the case now if the government had not mandated and subsidised wind power – if government policy had not mandated and subsidised wind power, the SA grid would have more CCGT and less OCGT, and the OCGT would be running more efficiently (lower emissions intensity) than they are, so the average emissions intensity of the fossil fuel generators in the SA grid would be lower than it is now.
3. Baseline #1 with the emissions embodied in the wind farms included.
Below I show an example of how I suggest the calculation could be done for the first case above. However, please note that numbers used in my example below are fictitious. I’d urge the analysis be redone using the best available numbers.
Assumptions and calculations:
CO2 emissions avoided:
SA grid average emissions, excluding wind = 0.7 t/MWh
Wind power displaces 65% of emissions at 27.5% wind penetration (“Emissions savings from wind power”: http://joewheatley.net/emissions-savings-from-wind-power/ )
Emissions avoided by wind generation = 65% x 0.7 t/MWh = 0.46 t/MWh
CO2 Avoidance cost:
Additional cost of electricity from wind generation (this cost should include all the hidden subsidies and hidden cost transfers to the fossil fuel generators) = $60/MWh
CO2 abatement cost with wind generation in SA = $60/MWh / 0.46 t/MWh = $132/t.
I’d urge that the best available numbers be used to calculate a lower bound and best estimate.
Hi Peter
Thanks for your second comment.
I appreciate what you’re seeking – however this is not something that WattClarity will provide, for three main reasons.
1) First and foremost, an answer to a question “what would emissions have otherwise been?” is, in truth, unknowable –
2) Of course it would be possible, through rigorous analysis, to progress down the asymptote to deliver an answer more refined than my crude upper bound estimate (above):
3) More fundamentally, however, this is not an area of debate WattClarity will engage in.
There may be other places you can find the assistance you’re looking for.
Regards
Paul
PS you asked about time-based emissions data, which is available from the AEMO for the Australian NEM – though I suspect it has been generated with the use of the same gross simplifying assumptions that are (to my knowledge) also the case for time-varying unit efficiency calculations right across the world.
Paul
Thanks for publishing this interesting analysis. Quick question – what definition / source of system demand are you using for plots 1 & 2 above? With nearly 400 MW of non-scheduled wind capacity in SA, obviously wind production and scheduled demand are going to show an exaggerated negative correlation.
Good pick-up, thanks Allan
One of our friends at AEMO contacted me offline with the same question and I have been investigating further since that time.
It appears that the “Metered Demand” figure we update in NEM-Review falls under the “Scheduled Demand” classification, described further here.
http://www.aemo.com.au/en/Electricity/Market-and-Power-Systems/Dispatch/Regional-Demand-Definition
As we find the time, we’ll add in the updated figures in additional charts above.
Regards
Paul
What would be a great addition to the graph would be a line showing the operational capacity (the theoretical maximum that the bars are falling short of). At the moment the temptation is to guess it as the envelope of the peaks; but that ain’t necessarily so.
Also plotting the generation data as bars makes it hard to see how much low-generation there is. Could the same data be plotted as just points?
Thanks Jon
Will make a note to do both of the above, when we have a chance to return to this analysis.
Regards
Paul
What about the impact of vertical integration of the majors? Could this be contributing due to the natural internal hedges they have by owning generation assets or rights to generated power due to privatisation?
Hi Chris,
Yes, that could be a factor.
Uncertainty about the medium-term landscape that will apply for carbon pricing might also be an influencing factor – along with the 3 other factors mentioned above.
At the end of the day, it’s a market – hence we won’t know for sure the motivations of each and every participant who trades (or doesn’t, as the case may be) all of the time. All we can do is think through possible contributory factors.
Regards
Paul
Paul,
Interesting times indeed. There were years not so long ago where a little price excursion such as $217/MWh would not have merited a post here.
Interestingly, this price would not even be able to trigger the $300 strike price of the ASX d-cyphatrade futures cap. Perhaps we will see a revival of $100 strike caps as a standard hedge instrument now that this is looking like a more sedate norm for the market.
On the other hand, the underlying structure of the market has not changed and all that is required is a few more plant retirements to bring the supply-demand balance back to something that can again generate price excursions in the $1000’s on a regular basis. I know we had one in VIC last week so they do still happen.
Thanks Ariel
Interestingly, it was only 5-10 minutes after posting this that the price did spike (about $1600/MWh in the 16:35 and 16:40 dispatch intervals).
It’s overly simplistic, but one might say that “all it took” was for the demand to rise a little bit more (to 8490MW – so about 45MW higher than it had been prior to that, much smaller than the plant retirement you proposed).
Regards
Paul
Overall there is still plenty of capacity in the NEM, and while there has been a step change after 1 Jul 2012, spot prices are still historically low. I accept high price events are function of a properly working spot market however I wonder whether there were greater forces at play here.
Hi Mohsin
Thanks for your comment.
Yes, there were a number of other factors at play (the 855-871 constraint being one of them). We have other tools that help our clients to understand more about these factors, and the interrelationships between them.
Occasionally we find the time to post some more detailed commentary (such as we did on 2nd July) about these more detailed factors.
Regards
Paul
Thanks Paul. In essence these things come down to generators taking advantage of their market power in favourable circumstances, which I would also do if I was in their shoes!
Hi Mohsin
It might surprise some people – but, on at least some occasions, generators don’t need to “take advantage of their market power in favourable circumstances” for the price to spike.
Paul
Paul
Spot on with the ENSO (el nina etc) summers. Great analysis and very interesting!
Interestingly 2010/11 was a very strong La Niña but a year later it was relatively weak but seemed to have similar effects – just that one peak demand period in NSW. Also noteworthy that ENSO has a much less of an effect on VIC and SA than NSW and QLD which bucks the trend again in 2011/12
Cheers
Thanks, Martin
I note hot weather forecast for VIC later next week, but PASA forecasts are relatively modest (because of the holidays).
We’ll be very curious to see what happens when everyone’s back to work, and schools are in, after Australia Day.
Paul
Paul,
Finally, it’s good to see someone post a sensible description of the issues in play in the observation of lower spot prices.
Further clarification about demand is now available in the AEMO resource here:
http://www.aemo.com.au/en/Electricity/Market-and-Power-Systems/Dispatch/Regional-Demand-Definition
It appears that the “Metered Demand” figure we update in NEM-Review falls under the “Scheduled Demand” classification in the AEMO document.
Everyone still on holidays here in SA…
Except for those few who were very keen to win the BBQ, it seems!
Paul, thank you for the analysis. Are you able to talk about the impact of the 855-871 constraint? I’d love to watch you identify it as a culprit in the stubbornly high prices.
Paul
Thanks for the analysis. You note that the current prices are not unprecedented and were seen when the drought was on in 2007 and generators withdrew capacity because they had limited cooling water. Why would prices reflect drought conditions when there is no drought?
I saw with interest the Courier Mail report http://www.couriermail.com.au/news/queensland/energy-firms-playing-games/story-e6freoof-1226553917675 which makes more sense to me and reinforces my view that generators have market power when there is congestion in the transmission system and withdraw capacity to increase prices.
This is why we introduced a rule change to stop this happening
Regards
David
Hi David,
The courier mail article was a very one sided view to stir up public opinion. The QLD generators have been losing money for most of the past 16 years – that “lost money” ends up subsidising the private retailers with generation capacity rather than the QLD government. So all Queenslanders are worse off than the states who have sold their assets. The oversupply of generation has been suppressing prices along with a carbon price that makes some generating units uneconomical. The idea of the carbon price was to make it expensive for coal generation so that they would eventually shut down and be replaced with generation with less carbon – ie gas. You would think with all the profits from the Carbon Tax that most of it would be put to creating cleaner energy funded by the federal government – like a national Solar or Thermal program.
Hi David
Likewise – thanks for the comment, which is really three in one:
1) You ask ”Why would prices reflect drought conditions when there is no drought?”.
2) You note that Paul’s subsequent article ”makes more sense to me and reinforces my view that generators have market power when there is congestion in the transmission system and withdraw capacity to increase prices.”.
3) You also reference the MEU’s Rule Change proposal, which follows from your perspective above.
Does this help?
Will post more analysis when I have a bit more time…
Paul
Some sensible arguments Paul. You might like to highlight that it was not just the drought affecting AGL hydro, Snowy hydro and QLD generators but also bidding behaviour in Q2 2007. Interestingly, NSW which was less affected by the drought than other regions usually had the highest spot price in most dispatch periods. The ETEF (Electricity Tariff Equalisation Fund) was still in place and the surplus was large (incentivising generators to increase spot prices). Also, the price sat around $90/MWh for most of that quarter until the last week, when it was continuously around $200-$300/MWh. Noticeably, on July 1, prices returned to historic normal levels. Given trading bonuses are based upon financial years and there wasn’t a drought breaking storm on July 1 maybe there were other factors at play…
Paul
Thanks for the further clarifications. Addressin your and other comenst in order
1. I accept that generators are having a hard time and I personally beleive that the energy only market has a flaw and that a capacity market would be more equitable and stable. I also agree that if there is a reduction in demand, the logical solution is that some generators will shut down, reducing available generation and that this should increase prices offered by generators.
What I don’t accept is that some generators can use their market power (a la AGL TIPS in SA during 2008 2009 and 2010). It is inequitable that some generators can have this market power but others don’t. Using network constraints to drive up power prices is not how a competitive market should operate as it is inefficient and creates significant other problems. Basically generators shpould bid their underlying costs becasue this results in the most efficient dispatch
2. I guess the difference between the iron ore example and electricity is that if i don;t like the price from RTA I can buy it from someone else and I know the price before I buy. In the case of electricity, I have used the electricity before I know the price and I don’t have the option of another source. I accept I can pass the risk to a retailer but the fundamantal issue still underlies the cost to my retailer and I pay it to accept that risk.
You comment that we have a market so we should accept market outcomes. I accept that but what I don’t accept is that some generators use market power to reduce the competitive pressures that ensures an efficient outcome. When bananas were high priced, many consumers didn’t buy bananas. When electricity is high priced I have no option but to pay the price. In today’s environment electricity is an essential commodity – bananas and iron ore are not – without elecricity the whole country stops.
To use market power to drive an inefficient price outcome is not what a market is about. Eminent economists Newbery, Tirole, Joskow all agree that using makret power is not efficient and have advocated various ways of preventing it
3. Whilst the MEU response to the draft decision not to make the rule change is quite definite, the response from the AER who have objected to the use of market power in the electricity markets is equally damning of the AEMC position.
So the debate about generator market power rages
Regards
David
Hit 45.8 hottest since 1939 with many other NSW records also falling. But heat across NSW was a bit below par with northern parts cooler. So the state’s overall temp fell well short of record levels!
Correction: meant to say hotter than 1939 when it hit 45.3!
Thanks for the added context, Martin!
Cheers
Paul
This is a great tip particularly to those fresh to the blogosphere.
Simple but very precise info… Many thanks for sharing this one.
A must read article!
Here is my blog post – diets that work
It is interesting that after the DSR in SA the load again peaked with no further pricing response.
Has DSR changed generator bidding behaviour??
I’ll be happy to answer this one for you, one-on-one, Peter!
Will call you in the next day or two – or call me 07 3368 4064 if a particular time suits.
Regards
Paul
SPAusnet also declared Monday 18 February 3pm to 7pm as a critical peak demand day, so this could have been a contributing factor.
Hi Mohsin
Thanks for this note, as well – yes, it would probably have had some effect.
SP Ausnet notes here that:
1) they expect the reduction in peak demand to be approximately 3-4% of their (distribution area) maximum demand – hence perhaps as much as 200MW.
2) Given that this DSR would have needed to be in place between 2pm and 6pm NEM time, it’s one plausible explanation for the dip in demand seen occurring around 2pm on the chart above (i.e. before the price spike).
However, we can also see that the demand climbed rapidly back up to the same level, preceding the spike. It seems likely that this was non-related demand (e.g. distributed air-conditioning load) and that the 400MW of DSR seen immediately after this price spike was more directly in response to price exposure (either directly at the energy user, or indirectly at a retailer – calling optional DSR).
Regards
Paul
Paul,
Thanks for the analysis. Be aware that SPAusNet’s Distribution area is only one part of Victoria, probably about a fifth of Vic’s distribution connected industrial load. Their submission to the Power of Choice Review claims their CPP scheme achieves 70-80MW load reduction.
Thanks for the added points, Ben
At a more holistic level, I think we’re seeing the results of how Demand Side Response (in a variety of forms) is proving more widely accepted as a means of managing risk (physical and financial) than was the case a decade ago, or so.
Paul
Re the IRPM I followed the link to the Nem Watch glossary. Unfortunately it no longer provides a formally definition of the IRPM nor a formula. Would you let me know where I can find one?
Regards
Hi Romek
Thanks for pointing this out – this information provides an explanation of the family of bar charts shown:
http://v8.nem-watch.info/help/islands/graphs.asp
IRPM translates the measure of RPM (traditionally used in a longer-term planning context) to real time:
IRPM = (Total Available Generation – Total Demand)/(Total Demand)
Within an Economic Island we substitute:
IRPM ei = (Total Available Generation – Net Demand)/(Net Demand)
where Net Demand = Demand in regions within the Economic Island + Net Exports, which are supplied by the local generators.
Does this help?
Paul
Shouldn’t the drop in brown coal production be attributed to the carbon price? Yallourn was back in action not that long after the carbon price came into play. Sure, the initial drop was due to Yallourn but it’s been 9 months already. If the drop in brown coal was only due to Yallourn then brown coal should have bounced back by now.
Hi Tom
Thanks for the question – and apologies for this belated response.
I’ve noted a number of people stating “the NEM is a complex place” the past couple weeks – the challenge in your question is that it’s not just one factor that is delivering change (and varying degrees of change dependent on participant, technology, commercial and other considerations).
Regards
Paul
‘Official’ is such a meaningless word, Macarthur has been dispatched for energy since late last year. Congratulations to the boys and girls at AGL, Vestas and Leighton on their hard work!
Thanks Peter,
With reference to NEM-Review it looks like it’s already produced about 371GWh over the period – so already contributing to reducing emissions.
Cheers
Paul
>”Already contributing to reducing emissions”
How much emissions has it abated? How much will it abate per year? How do you know?
Will it be more effective at abating emissions than wind power in Ireland (EirGrid)*? How do you know?
What will be the abatement cost ($/t CO2)?
* For those who don’t know, wind power in Ireland is about 53% effective at reducing emissions (17% wind energy proportion of total energy).
Joe Wheatley (2012) (in press) “Quantifying CO2 savings from wind power: Ireland” http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf
Hi Paul. I really appreciate your input into the electricity supply debate.
I have been reading the gushing reports about Macarthur wind farm and was wondering if the wind might come to the party. Probably a bit sad that it didn’t show up for the paperartzi.
A debate I am having on various blogs is the issue of dispatchability as the renewable advocates claim that baseload is a myth (read Mark Diesendorf) and even so, wind power can supply baseload (contradiction accepted).
So your comment about wind farms being a type of generation built for energy production, not capacity supply is what I’m trying to explain to these rather over zealous advocates who don’t seem to understand the importance of capacity if we are going to have a reliable network. It continues to be a tough sell for those that don’t want to really understand how electricity networks actual work.
Thanks Martin
The industry was already a complex beast pre-market, then the market came along and the complexity (and data load) increased. Add in a social drive to increase renewables and lower emissions and the complexity has ratcheted further.
Part of the noise in the debate reminds me of the story of blind people gathered around an elephant trying to describe what it is by the different components they each feel.
Paul
Hi Martin,
I agree that capacity reserves are vital to a reliable network, however peak capacity is currently not a barrier to increased renewable installation. It will be a barrier in several decades perhaps.
Currently the main barrier is cost of RE vs. incumbent fossil fuels. As the price drops more and more RE will be installed at least to the point where oversupply will become the next hurdle. This is where the price will be unprofitable for significant periods of time for coincident RE generators (e.g. with 20 GW of PV noon will be the cheapest part of the day).
RE will not be able to mount this hurdle without energy storage or more flexible consumers. If these requirements are met then back-up capacity becomes an issue. It’s pretty easy to solve too – build OCGTs.
May I remind you of the current situation in SA? 25+% Wind power and no new peaking capacity added. Maybe when they get to 35% and the remaining coal plants shut down there will be a need to build some more peakers.
On your point about baseload; any technology can supply baseload, it’s just that no one cares to. If all the baseloaders were gone we could still run a peaker at night.
Clear, easy-to-understand (considering the complexity of the subject!), insightful.
I’m interested in gaining more information about the report.
Thank you,
Dana
Thanks Dana – will provide you some more information when it’s released.
Paul
Paul,
With regards to your last points on Northern: We all get that wind is intermittent, and if average wholesale prices have fallen and the consumer is receiving that cost reduction then any market volatility is purely a problem for the retailers. Maybe instead of focusing on Northern which is mostly hobbled by the carbon tax you could be asking why several important peakers and intermediate plant weren’t made available for generation over the recent weeks? Summer was pretty weak and no-one had to push their assets hard. I count over 1 GW of unavailable gas plant. The entire point of some of these assets is to provide capacity yet they did not deliver. Maybe there is something in holding on to that gas…
You also brought up externalities for wind (of which neither example exists); why not talk about the health costs of coal generation? 😉
Cheers,
Tom
Thanks for these questions, Tom
This is the second lot from you I have not yet responded to. When I do find some time to look more thoroughly, I’ll be sure to post more on WattClarity.
Regards
Paul
thumbs up
Thanks Tom
It’s a pity that the offpeak load isn’t staggered. I guess ETSA have no incentive to do so.
It only takes 4 hours to charge a hot water system from cold, and 2 from warm, so there’s plenty of play in the timing.
Possibilities like these intrigue us, Tom – and we’ll look further into them when we have the time. At present we don’t know what the barriers might be to prevent this happening.
Regards
Paul
Note re Morwell Power Station – they confirmed last year that they would no longer operate to generate electricity into the grid – they are only operating one turbine (enough to supply on-site demands from briquette plant). The understanding was this is/was due to higher costs associated with carbon tax, but also combined with poor efficiencies, lower demands etc.
They received Govt funding to keep briquette plant operating for 2 years – industry noise is that they will close June 30 next year (2014) when funding runs out.
Thanks for the added note of detail, Jon.
I could vaguely recall something like that but could not remember the details.
Cheers
Paul
1) re Energy Brix power station: In June 2012 Martin Ferguson gave HRL – the ower of the Energy Brix plant – $50 million for a “a two-year restructuring package to allow the Latrobe Valley’s Energy Brix Australia Corporation to maintain briquette production while regional businesses that rely on briquettes transition to a cleaner fuel source”.
See http://minister.ret.gov.au/MediaCentre/MediaReleases/Pages/BriquetteRestructuringPackage.aspx
2) It is worth mentioning that a central reason why the carbon price has had little effect on the dirtiest coal fired power stations is that their owners have already been paid $1 billion in cash payments in 2011-12 from the $5.5 billion ‘Energy Security Fund’ and more subsidies to keep them going are on offer.
See http://www.climatechange.gov.au/en/government/initiatives/energy-security-fund-cash-payments/eligible.aspx for a list of the company by company payments.
Thanks for the comment, Bob.
Cheers
Paul
Amidst the fuss that switch to gas is an essential transition, have always been doubtful about this on suspicion that power stations extract brown coal for free, no $$$ per tonne royalty, resource rent depletion tax etc etc.
This being so, why would they switch, unless taxpayers make gas available for free? Any chance you can confirm or deny this suspicion?
“It’s a pity that the off-peak load isn’t staggered.” – –
I agree – how much discount would I win for a 2am start ? – I only heat the top half of the tank to 50*c for a hot morning shower- the thermostat range is 50 to75* – low pressure solar Beasley –
Perhaps ETSA’s “J” meters should have a built-in 3hr random number generator [1/180] – we live in a digital world now – – cheers.
Thomas
Thanks for this – a useful addition to the market intelligence (or is that an oxymoron?)
One aspect you did not seem to investigate is that the demand in NSW has been falling for the past few years and this would have had a significant impact on prices.
The fewer incidences of exercise of market power would also be related to the fact there have been fewer demand spikes due to the milder weather conditions in the past couple of years
Regards
David Headberry
Hi David
Great comments.
Market intelligence is a big beast, we aim to provide some insight.
The demand will have a factor. Paul has already written a couple of blogs here on the falling demand (linked here).
A quick recap shows that the average metered demand decreased by 410 MW over the period (from 1 March 2009 – 28 Feb 2011 compared to 1 March 2011 – 28 Feb 2013) and maximum demand fell by 828 MW. This reduction in demand will have an effect on the opportunities for price spikes to occur and we agree that the NEM is a complex place.
There are always many reasons for prices being different. The Analytical Report that we will send out with our QLD market review is going to be over 100 pages and only covers four months!
The reason I am highlighting the change in ownership is that surplus capacity is not a lot higher than have seen in the past but the lack of price spikes is new. Since 1999, NSW has never had fewer than 53 prices above $300/MWh in any two year period. A lot of things have happened since then but the timing of the reduction in spikes points to the asset sale.
In summery: Agree market intelligence is a noble but difficult goal, and there are many temporary factors such as demand which plays into actual outcomes in the short term.
Regards
Thomas
Paul, what do you attribute the price spikes in SA to?
Thanks for the question, Peter
That’s something I might answer, in person, on Thursday when we meet at the EUAA conference.
Cheers
Paul
I wonder how much of this afternoon peak is caused by a fall off of rooftop solar in winter afternoons. Even in far north NSW, you don’t get much solar after 2 pm in winter. The fact that it occurs even on a Sunday might suggest it has more to do with domestic demand rather than commercial.
Hi Martin
Not sure I understand this question – the evening peak in demand has existed for many years. Perhaps the input of PV during daylight hours (prior to the peak) has increased the rate at which net demand (to be met by the wholesale market) ramps up to the peak has increased in recent years – is this what you were meaning?
Paul
Martin, unless the sun goes down at 3PM PV is still going to have a reasonable output. Like Paul said PV isn’t going to impact evening demand. There is only ~500 MW of PV installed in NSW so I don’t see it drastically effecting price. Check out the examples of the German multi-GW PV capacity cutting the middle out of the daily demand …and price.
http://reneweconomy.com.au/2012/why-generators-are-terrified-of-solar-44279
Yes, sort of. PV in summer would reduce grid demand during the day exaggerating the peak in the summer evenings as the sun sinks increasing grid demand. I suppose a test of this theory would be whether the evening demand has remand flat over the last 5 years but the evening “peaks” (relative to midday) are becoming more prominant as winter approaches. Something like that 🙂
Guys,
I think it was the Germans who used to monitor short term changes in sewerage flows at night and related that to TV audiences and commercial breaks – do you want to co-ordinate with the appropriate authorities in NSW and Qld to investigate the correlation?
Good article – having the concept of surplus capacity helps provide greater context to high prices. A couple of questions:
1. Did wind generation reflect wind quality or were turbines disengaged?
2. Why is the Murraylink Interconnector down?
Michael Williams
Hi Michael,
1. The low levels of wind generation reflected wind speed. It is possible that some individual turbines (around 2MW) may be out of service but no entire wind farms were offline. It is more difficult to see this graphically as wind availabilty always matches the output unlike fully scheduled generation where it is clear to see reductions in availability due to maintenance, system constraints etc in comparison to registered capacity.
2. The outage on the Murraylink interconnector was a preplanned outage in Victoria preventing the flow into SA. At this point I dont have more detail on whether maintenance or upgrades were taking place, but it could be obtained by contacting Electranet SA or SPAusnet in Vic.
Does this help answer your question?
Regards,
John Bartlett
Hi Thomas,
Can you help me to find the Tariff 11 costs in QLD for the last few years? 10 years preferably? I sure would be grateful.
Thanks and cheers!
Alex Wilson
PS – it’s been highlighted to me that the outage at Yallourn comes coincident with Industrial Relations action onsite, and with a fire that is under investigation at present.
Here’s some further news at the ABC, and in the Australian.
If you know more, please do add this in comments below (or let us know offline).
I am more inclined to relate these demand increases to the domestic controlled load by the distributor. As 21:30 & 22:00 are common T33 switch ON times.
Thanks for the great alternative interpretation, Paul
Under that paradigm, what do you think could explain the more rapid declines in demand just before those two small jumps?
Will be curious to watch more tonight to see what happens …. (ok, will be watching the game, not so much the market)!
Paul
Not sure on the declines before the rises, but there is a decreasing trend in demand as the evening progresses most nights but it not linear. But back to sport and the NEM.. If you are looking for a sporting event that impacts the NEM then the winner in my mind is the Melbourne Cup the race that stops the nation. Demand decreases in QLD as manufacturing shuts down, office PCs go to standby etc… which is not offset by the thousands of TVs being switched on during the day.
Yes the audio does help with understanding. Even with the sound turned right up, it is still much quieter than most audio feeds and sometimes is difficult to hear. Can you boost the sound from your end?
Thanks for the comment, Martin
Being a first attempt at adding audio, we used the tools we had at hand.
Based on the belief that this kind of thing does add value to clients (by helping them more easily understand the complex interactions at work in the NEM) we are having a look at some technology that will give better sound quality.
Cheers
Paul
Hi Paul,
Pacific Hydro may have been referring to an analysis that was done earlier that I am responsible for, that utilised an erroneous demand figure (ie, non-scheduled generation had been subtracted from the denominator and thus, the demand figure was lower than it should have been in the formulas).
I’m working with the publisher to get it corrected, and I’ll let you know once it’s all fixed up. Apologies to you and Pacific Hydro for the error!
Cheers,
Ketan
I agree with your thought process, but the other factor, which you may have considered is cost. I think that a far larger part of the population are being more careful about energy costs. I believe the cost of energy to moms and dads is now becoming a significant part of the household budget and therefore things are being done to keep the costs down.
We, for example, have dropped our ducted heating system by 1 degree to 19 deg C and on weekends be actually switch it off and just wear more clothes, where we previously left it on for the weekends.
I am also meeting more and more people who are looking for alternatives. Took a cab to the airport last month and the cabbie told me that using homemade vertical axis wind turbines using 44 gallon drums and old car batteries, he had so far converted his entire shed to an off grid system and was starting to expand into his house.
Thanks, Barend
Increased awareness of energy usage, and costs (and their increasing share of budgets), both at work and at home were one of the factors noted here a couple years ago.
Your own example is not dissimilar to many, I suspect.
Regards
Paul
Ketan,
How on earth did you let that get through your own sense checker?
** Additional comments removed by the Editor as it seemed ad hominem **
Given that the June-July 2013 profiles are lower for the whole day – not just when PV is generating – your conclusion is badly overstated. PV has no effect on the breakfast and evening meal peaks or overnight values, and they are all lower too, by the same margins.
There is some sign that the minimum is shifting to earlier in the day – which can perhaps be attributed to PV generation peaking at around noon, or just prior to noon for Brisbane (solar noon is at close to 12:00 for Rockhampton, a little earlier for Brisbane). But the effect is pretty marginal, and could also be attributed to the increasing use of air conditioners in the afternoon (it’s most likely a combination of both).
Thanks for the comments, Phil
In the post I linked to a range of factors that would be contributing to various degrees, and noted that the 4am demand was also lowest in July – hence, there’s more than just PV contributing.
Not sure how I have overstated a conclusion?
Paul
Paul, I noted your other factors, which are correct in the context of reducing demand generally, as we see over the family of curves. [Other readers, I can reccomend reading that list].
It just seemed to me that your article was inferring that the demand profile had reduced at midday, and that that was probably due to widespread PV installations (“Undoubtedly, the rush of people … getting their 5kW solar PV systems installed … has had some effect”). I did not feel that the midday reduction was significantly different to the reduction over the rest of the day, so to attribute it mainly to PV was overstatement of the PV effect.
From your response it seems that I had misunderstood your intended emphasis, and that you meant that PV contributed to the reduction, rather than being the primary cause of it.
I certainly agree PV is a factor, but it seems pretty minor from the evidence of these curves.
Hi Phil
Thanks for the added details. Part of the challenge in terms of interpreting PV (for us, and for everyone) is that it is behind the meter, so not as visible at the level of the NEM.
Cheers
Paul
Thanks for this.
A common response by advocates of renewables to the criticism of “unreliables” is to propose wider diversity of the renewables. As an example, when SA wind may be becalmed, wind farms in Vic and NSW can contribute instead. Apart from the question of power transfer over long distances, and the question of whether the customers in Vic and NSW might also want the power, the assumption that Vic and NSW won’t be becalmed at the same time seems somewhat wishful – as your analysis suggests. While it’s not that common, a large high pressure system centred somewhere over the SA/Vic/NSW common border would seem to generate the right conditions for that.
I would be interested to see a graph similar to your second graph, but using half hour (or shorter) demand data (MW), rather than energy (MWh). This will give a better indication of the time-matched energy production in both areas.
As I see it, the main problem with renewables is that the energy they produce isn’t time-matched to the energy demand of consumers. In many residential houses without air conditioning, a reasonably sized PV installation will produce more kWh’s than the house uses; but you’ll get nothing – zero kW – from the PV during the evening meal.
When (not if) we get some practical large capacity energy storage systems, renewables will really come into their own.
Hi Phil
Thanks for the detailed comments.
I will add these suggestion to our growing list of potential future analysis – though keep in mind you’d be able to do this, yourself, with NEM-Review.
Cheers
Paul
Phil,
You don’t have to go to the expense or trouble of signing up to NEM-Review if you just go to the link below where its all done for you based on 5 minute SCADA data from AEMO, with historical data as far back as 2009
Leith Elder
http://windfarmperformance.info/
Phil Spencer,
This web site: “Windfarm Performance” may answer some or all of the questions you asked.
Sorry, I forgot to include the URL, but I see Leith Elder has provided it.
Phil,
You might also like to have a look at a paper I have been fortunate to have had published
recently. Properly, anonymously, peer-reviewed, by real engineers in charge of grid-control.
It looks in detail at wind farm performance for the full calendar year 2010.
The reference is: Miskelly P 2012, “Wind Farms in eastern Australia – Recent Lessons”,
Energy & Environment Vol 23 No 8 December 2012. It’s available at:
http://multi-science.metapress.com/content/f1734hj8j458n4j7/?p=3367b7f146d24db99c4ffb0a5361ab47&pi=4 .
As contributors Leith and Peter say above, you can do it yourself. I provide the methodology
with all links to relevant AEMO data in the paper.
Also, there is an earlier paper by two of my colleagues:
Miskelly A & Quirk T 2010, ” Wind Farming in South East Australia”, Energy & Environment Vol 21, Vol 20 Number 8 – Vol 21 Number 1 / December 2009 – January 2010. Available at:
http://multi-science.metapress.com/content/j2l218v143733536/?p=5b117ed08c1e43a7a3ae567df515f5f2&pi=13 .
Note that http://windfarmperformance.info/ was designed and is maintained by Andrew Miskelly,
co-author of this second paper.
Hi Paul,
I absolutely agree with your findings here, simple probability suggests that there must be occasions where total wind generation falls to zero. However no one is suggesting that we run a country with a single energy source. It wouldn’t be a great idea to run solely on coal/gas/hydro/nukes either.
Like Phil pointed out we won’t be able to have a very high proportion of our energy coming from renewable without storage. In the meantime we can aim for a lower proportion safe in the knowledge that we have an enormous volume of fossil capacity ready to ramp up when required.
Phil,
Tom Quirk and Andrew Miskelly published a largely ignored paper in Energy & Environment Vol. 20, No. 8, 2009/Vol. 21, No. 1, 2010 based on 5 minute data.
http://aefweb.info/data/Wind%20farming%20in%20SE%20Australia.pdf
Paul Miskelly has followed up with a more recent paper based on a much longer data set in Energy and Environment Volume 23, Number 8 / December 2012
http://multi-science.metapress.com/content/f1734hj8j458n4j7/
Someone might be able to help me, perhaps I have a misconception. The renewable power percentage is currently about half the 2020 target value, which should mean that about 10% of our electricity should currently be covered by RECs. The data in this blog suggest less than 5% on average coming from wind.
Is the RET under delivering?
James,
This Robert Gottliebsen article may provide some insight as to what is in store for us to try to achieve the RET, especially the cost implications:
“Prepare for an expensive energy future”
http://www.businessspectator.com.au/article/2013/7/24/resources-and-energy/prepare-expensive-energy-future
Hi James,
Most of our renewable energy comes from hydro. The RET has been more or less on track to the fixed GWh target so far. However the large retailers (at least Origin and EA) are refusing to buy into new renewable projects which will make the target hard to meet. They are hoping they can inflate the cost of the scheme by reducing supply of RECs vs demand and then ask the government to reduce the target to protect consumers. Oh and their profits. Don’t forget the profits.
The chart demonstrates the volatility of wind power. Most reading this web site would be aware of some of the consequence.
The purpose of the RET and the many other policies intended to give advantage to wind and solar generation is to reduce Australia’s CO2 emissions. But seldom do people consider the cost of these policies in terms of “what is the CO2 abatement cost?”.
A rough estimate is that the CO2 abatement cost is about $120/tonne CO2 – based on the following assumptions:
– cost difference between coal generation and wind generation = $60/MWh (perhaps higher)
– Average CO2 intensity of NEM coal plants = 1 tonne/MWh (perhaps higher)
– Wind generation effectiveness at reducing emissions = 50% (has not be determined for Australia). This means that 1 MWh of wind generation avoids just 50% of the emissions from 1 MWh of NEM generation. This is based on the Joe Wheatley (2012) study [1] which demonstrated that the wind generation in EirGrid in 2011 was just 53% effective at abating CO2 emissions. EirGrid and SA’s grid share many important similarities, including: high wind energy penetration, similar generation mix, small interconnectors and small imports and exports.
The volatility shown in the WattClarity chart is a cause – not the only one – of the reduced effectiveness of wind power (and solar power) for CO2 abatement. However, it is seldom discussed and as far as I know there has been no real attempt to determine the CO2 abatement effectiveness for Australia. Given the cost of the CO2 abatement polices it seems amazing that we have no well justified estimate of the abatement cost.
[1] Wheatley, J (2012) (submitted) “Quantifying CO2 savings from wind power: Ireland”
http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf
Hi Peter,
The chart shows volatility in wind generation over a significant period of time. You would be hard pressed to find many instances where short term volatility in wind power caused upwards market volatility. However it is a lot more common for large fossil plant to trip offline during elevated prices and cause prices to hit $12k+. Even so our current generation base is such that the loss of 500 MW is no big deal in term of system security. My point here is that it will require a hell of a lot more wind installations before we have to get concerned about system security.
Your CO2 abatement cost is also pretty inaccurate. The Irish grid has nothing to do with ours. The effectiveness of emissions reductions would be much closer to 1 given the generators that are usually limited to make way for wind; namely black coal, brown coal, and lower efficiency gas (Torrens). Also the actual cost borne is $35 for a REC minus a reduction in the pool price. I reckon the cost of abatement would be more like $40/ton.
Regards,
Tom
Tom,
What is the market mechanism by which brown coal generation is displaced by wind? I thought brown coal generation bids into the market bid stack at a lower price than gas generation. Wind will displace the most expensive alternative in the bid stack, which is most likely open cycle gas turbines.
The marginal carbon abatement due to variable wind generation does not appear to be measurable in Australia, hence conflicting claims from pro and anti wind lobbyists abound on the various blogs. Why don’t we all admit that nobody knows, reframe the problem in this context, and hold the wind industry accountable for providing evidence of displaced generation (gas or coal) due to wind generation.
Another misconception surrounds the ratio of wind energy to total supply. Retailers are accounting for about 10% via surrendered RECS, however only about 4% is being generated. This results in the cost of one MWhr of renewable energy being somewhat higher than the cost of one REC. The market would still appear to be awash with phony RECs created to support solar credits for phony rooftop solar energy.
Hi James,
It seems a bit unfair to tar wind with the solar multiplier brush doesn’t it? 🙂
Roughly speaking brown coal is displaced by wind when gas units are off or at their minimum capacity which is common during mediocre demand and/or high wind. If brown coal generators bid strategically above their marginal cost they may also be displaced at any other time. The interconnectors make it a bit tricky to pin down exactly what’s happening as one generator can be displacing another one 3 states over.
If we re-ran AEMO’s dispatch engine, with and without contribution from wind it would paint an accurate picture of the actual abatement.
For a more rough idea just look at the fact that gas and wind generation are rising, and coal generation is falling. Of course demand is a major driver but if wind was mostly displacing gas it would be taking a huge chunk out of gas generation.
Haha, you guys are such nerds!
Have a good week-end!
Thanks, Reinhard
Geez Paul – thought I would watch the video for fun…totally forgetting I had entered the competition, and low and behold my name was drawn out in the “nerdiest” way possible. Long live the Nerds!
By the way – I fly out to Paris in 10 days for a holiday (seriously) and was about to purchase a camera….thanks Paul, you saved me a head ache or two and quite a few dollars there.
Cheers Craig J.
This is interesting analysis. Thank you. What would also be interesting to see is the minimum aggregate average power output (MW) from wind generaton in a 5 min trading interval for the year and some sort of frequency analysis, such as aggregate wind generation was X MW or above 90% of the time and y MW or above 50% of the time for the year.
Thanks for the suggestion, Tagent
I will have to save that one for another post!
Regards
Paul
Hi, So what you are essentially saying is that the variability of the weather patterns over the last 4 years that the increasing number of wind farms in the NEM are operating in has not changed much.
Hi Muzz
It seems that weather patterns in the expanding area over which wind farms have been installed, thus far have not been sufficiently asynchronous to reduce the small but significant percentage of time when the entire portfolio of wind assets have produced a negligible amount.
Does this help?
Paul
Of course, the vast majority of the energy is coming still coming from SA and Vic, which have relatively strongly-coupled weather patterns, and most of the wind generation in NSW is concentrated in one area close to the Vic border.
What might be more useful to say here is that wind power is generated by “useful” weather patterns, and as long as wind farms are installed based on economic gain, the most useful wind patterns will continue to be exploited and improved reliability resulting from the geographical diversity of generation will not be realised.
The story becomes interesting when network limitations prevent wind farms from being installed in areas with the best weather patterns – we are approaching that situation now, and it is likely that as long as network augmentations designed specifically to “unlock” wind resources do not go ahead, that we will observe an increase in geographical diversity of wind generation (because the economic gain to the wind farm owner will be reduced under conditions of network congestion).
Thanks for the comment, Quadranum
I’m sure the economists have a technical name for the behaviour you observe – wind farm developers maximising the benefits to themselves (in terms of locating where they will get maximum MWh generation, and so also RECs) but not having an incentive (yet?) to consider where diversity would be promoted.
With new planning restrictions in Victoria we’ll be interested to observe performance patterns as new wind farms sites further afield are opened up.
Paul
Interesting analysis. An interesting next step would be to estimate what would be the average cost of electricity for the 365 days.
Since the purpose of the RET, and for installing wind power, is to reduce CO2 emissions from electricity generation, the next step would be to calculate how much CO2 is actually being avoided. To do this, we need to know the effectiveness of wind power at cutting CO2 emissions. Many studies using empirical evidence show it is much less that is normally claimed based on modelling studies. The study by Joe Wheatley [1] suggest wind was only 53% effective at cutting CO2 emissions from the fossil fuel generators in Ireland in 2011; i.e. 1 MWh of wind generation reduces the emissions from the 1 MWh of displaced generation by just 53%.
Having calculated the average cost of electricity and the CO2 emissions from the NEM with and without wind generation it will be an easy step to calculate the CO2 abatement cost (in $ per tonne CO2 emissions abated). Expect the CO2 abatement cost to be around a hundred of dollars per tonne CO2 avoided.
[1] Joeseph Wheatley (submitted 2012) “Quantifying CO2 savings from wind power: Ireland”
http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf
Thanks Peter
You raise 3 points:
1) Regarding the cost of electricity under the hypothetical case, each analyst will have their own point of view – with a wide diversity almost guaranteed. Suffice to say that this would be impossible to forecast what the actual price would be under such a scenario, as to do so would require predicting human behaviour in terms of trading decisions. It seems likely that the wind farm generators, were they to provide a more significant stake of energy in the NEM, would not continue to follow their “price taking” behaviour at present.
2) Regarding the MRET, my understanding of the original genesis of the scheme was that it was partly to do with carbon reduction, but also partly to do with support for a nascent renewable energy industry in Australia (for reasons broader than just carbon).
3) Regarding avoided carbon emissions, frequent readers here (and elsewhere) will see a tiring pattern in the fact that you frequently add this in, citing a single reference from a different market that aligns with your own particular perspectives. It’s not something we will be focusing on through WattClarity (the focus of which is energy supply, whatever the source).
Paul
Hi Paul
Not sure if you have analysed the AEMO 100% RE study. The 2030 scenario 2 has a demand of 290 TWh/yr = 800,000 MWh per day. Of this wind provides 111 TWh/yr or 38% – around your maximum. Maximum demand in the network is 46,000 MW.
Total capacity installed in 2030 (all RE) is 98,000 MW of which wind is 30,000 MW (a generous capacity factor of 42% – must include off-shore wind as well). They admit that they would need >200% more capacity than maximum demand.
Regards Martin
Thanks for pointing that out, Martin,
The 21GW figure above for idle capacity would probably be larger than that if we looked on a 5-minute by 5-minute basis, rather than the day-by-day basis of the above – hence maybe it would be somewhere near the “equal to peak demand” figure you note.
Paul
Great analysis Paul. Much more realistic than the usual overly optimistic/pessimistic reports I read.
With regards to your assumptions: I think it likely that geographic site diversity will be become more important than capacity factor over time. If your wind farm has the same output profile as everyone else you can expect to get lower spot prices and vice versa. Wind turbines designed for low wind conditions will be enable the move away from the southern coast. We haven’t even touched the wind resources that aren’t world class. Anyway it is likely that there will be a smoother profile than what you have extrapolated but it would not remove the need for back up supply.
I did some research on lithium batteries a couple of years back, and since then it seems that project costs have dropped by 50%. If the battery technologies keep moving at that rate it won’t be too long before they stop being horribly expensive and start playing some role in the energy market.
The Germans and Danes have put some effort into turning an oversupply of wind power into hydrogen. This may also become viable in the future with an international gas price on the east coast and occasional electricity gluts.
Thanks,
Tom
Thanks Tom
Can you point to some numbers on the cost of energy storage technologies to expand on what you’ve written?
Paul
Hi Paul,
Elec to hydrogen costs $2/W. Pretty expensive for a reverse peaker 🙂
http://fuelfix.com/blog/2013/06/14/hydrogen-plant-starts-storing-wind-energy-in-germany/
Lithium was ~$6/W and is now ~$3/W if I recall correctly. I didn’t find any decent links.
According to the ESAA’s latest Annual report, there is well over 45GW of non-wind generation capacity installed in the NEM. In addition, there are periods overnight on almost any evening, when over 21GW of electricity generation is “on stand-by”. Therefore, it is not clear what the significance is of your conclusion that 21GW of non-wind generation capacity will be requrired “on standby” should wind energy achieve a 37% market share (more than twice the current LRET target).
If you are making the point that no additional “back-up” generation is required to achieve much higher levels of wind generation then I would certainly agree with that. AEMO’s latest SOO states that no new thermal generation (“back-up” or otherwise) will be required in conjunction with the current LRET target, so AEMO would apparently agree with such a conclusion as well.
Thanks Jonathan
My main motive in have a look at this case was just in seeing how much wind would theoretically fit in the NEM under the assumptions listed (i.e. not related to current MRET target level).
In terms of the energy (and hence capacity) gap that would arise between windy and non-windy days, it would follow that this would also be a hypothetical case.
It’s probable that the challenges to be addressed are both:
1) Technical, in terms of balancing supply and demand over various time-frames (above is just day-by-day transitions)
2) Commercial, in terms of some form of changed compensation structure to provide the residual plant some incentive to remain operational for lower energy volumes (hence revenues in an energy only market), whilst still remaining available for much the same level of capacity reserve requirements. One option might be just to continue with the current structure, but understand that there’d probably be a lot more volatility present.
Paul
There is currently plenty of peaking capacity that rarely gets used. In the AEMO 5-minute production data for stations on the network, many of the Queensland gas power stations rarely get used. I have never seen power output listed from Snuggery (near Mt Gambier), nor Bell Bay gas. The hydro stations seem to be well below their potential for balancing the peakiness of wind. Tumut 3, for example, has 1500 MW of generation capacity and about 200 MW of pumped storage. If we were approaching the limits of wind into the network, it would be operating at close to capacity at times of low wind and high demand, while under the converse conditions its pumped storage would be operating at full capacity, and the 200 MW of pump from Lake Jyndabye to Eucumbene, After looking at many days of the AEMO data, I have never seen this occur. With wind currently supplying such a small proportion of the Vic-NSW demand, there is as yet little need for pumped storage to balance out the vagaries of wind.
The gas backup capacity will become much more expensive once the LNG plants at Gladstone start competing on the Eastern Australian gas market. Additional wind capacity will mean we won’t need to use as much gas in power stations. The price relativity between wind and gas will be less, but we still have the capacity built at a time when gas was cheaper.
My personal preference is that as wind is now a relatively mature technology, our public sector by incentives should be moving to geothermal, which is scheduled capacity. For example can we decrease the cost of power produced by this technology by the horizontal drilling now used by the fracking industry ? Instead the erratic wind power is making the price of renewable energy certificates too cheap, and hindering the development of geothermal.
Hi Paul,
I’m not sure I fully understand your analysis. Am I correct in saying that you’ve scaled up wind production until it achieves 100% of daily energy on a single day, and using that as the maximum capacity/energy that can be installed?
If so, I don’t think that’s a particularly meaningful number – it’s quite feasible you’d install more wind, and spill some energy on that day. After all, we currently install much more capacity than the peak demand day.
Ultimately, the penetration of wind comes down to two factors: economic and technical. There’s no technical reason why you couldn’t install much more wind than that, and curtail it on the highest wind/demand ratio days. Indeed, some of ROAM Consulting’s integrated resource planning modelling (slide 15, although the full detail isn’t on this slide) shows that a high penetration of partially curtailed wind can be least cost, particularly under moderate carbon prices and high gas prices. (E.g., if the LCOE of gas was $120/MWh and of wind was $100/MWh, then wind could be curtailed by 15% and still be least cost.) Longer term, this seems the sort of world we’re approaching. If costs are higher however, then your argument is probably more accurate, but you still need to consider economics.
On the technical side, this could potentially act as a limiting factor on penetration of renewables; AEMO suggested that the power system is operable with at least 15% of energy in each period coming from synchronous generation. More sophisticated studies are needed to understand exactly how much inertia, spinning reserve, etc., is required online.
This is something we’re thinking lots about, and would be very happy to discuss further.
Thank you. Interesting analysis. It makes sense.
An outcome I see is that, when we get over our infatuation with solar and wind, peak and intermediate gas generation will fit perfectly with baseload nuclear power. It is easy to envisage a future with 75% of electricity generated by nuclear (as in France), 6% hydro, minor renewables and the remainder by gas.
This would be the least cost way to generate electricity with low GHG emissions, as the CSIRO eFuture scenario analyser shows: http://efuture.csiro.au/#aboutsite
CSIRO reference BREE who state a LCOE of $100/MWh without considering disposal or decommisioning costs. They also assume it can be built in 6 years rather than the 12 that is the norm overseas. Pretty unlikely for a first of a kind build. They also assume a 60 year life span which certainly reduces the LCOE. What if it only makes 50 years? or 45?
Nuclear is great if you have cheap labour or no local resources. Both wind and coal-CCS are cheaper and more socially palatable.
It seems like it man be time to level the playing field for generators by removing the RET and other distortions favouring renewable energy. It makes no sense to encourage residential solar PV given it costs $600/tonne CO2 abated and wind costs over $100/tonne CO2 abated.
We need new distortions that favour nuclear! Or why doesn’t the taxpayer just fork out $180 billion for 30 GW of nuclear reactors. Are you with me Peter?
Thanks for another excellent example of the real costs of political interference (i.e. with carbon taxes, etc) in energy markets. What a blunder (net cost of ETS would be $1,345 billion to 2050 according to Treasury figures)! Will we ever learn?
Mortlake probably has the cheapest gas of any gas-fired power station on the NEM. Its owner Origin owns the gas pipeline from the Port Campbell gas plant, so it doesn’t need to pay third-party pipeline owners. While it doesn’t own the Iona gas plant at Port Campbell, it does have a majority share in two offshore gas wells. So it’s strongly vertically integrated, and thus not subject to different financial pressures to other operators. Compare that with Tamar Valley, which requires the use of a premium 3-rd party asset to transport gas under Bass Strait, then another premium third-party asset to convey the power back to the mainland. Furthermore, the spot price of gas in Victoria is usually lower than in Adelaide or Brisbane. So Origin have done their sums well in choosing the location.
In recent weeks, Mortlake has been producing power right through the night, which is unusual for a single cycle gas turbine station. Several NSW generation units at Liddell and Eraring have been out of action, and at night net power import into NSW has been over 1000 MW. So is the cost of coal + carbon tax off-lining some coal-fired generation ?
Hi Malcolm
Thanks for your clear and precise comments – perhaps there’s a role for you in posting on WattClarity 🙂
The shoulder seasons are normally maintenance periods – but with the overhang of surplus capacity, the incentive to work around the clock to bring units back is not what it once was.
Regards
Paul
Paul, it is refreshing to see someone such as your self question the myth that because renewables effectively bid in at zero that it always results in lower prices. If generators that are dispatchable are relying on pool and contract revenues rather than REC payments, then they will endeavour to “game” prices to maximise total revenues. I think it’s also important to note the effect of less gas contracting by peaking OCGT’s that will mean higher SRMC’s, shorter run times and consequently higher prices.
Thanks Owen
For those particularly interested in where Electric Vehicles might be headed, these reports written by Don McPhail as part of his E.S. Cornwall scholarship-sponsored world travels will be of some additional value.
Paul
There are multiple proponents of “EV’s as energy storage”. While I concede the possibility, I’m not convinced of the actuality.
* The EV owner wants to drive home with a charged car. Peak lopping of commercial demand from EV V2grid could leave him with barely sufficient charge to get home. Yes this could be managed, – my point is that it needs to be managed, and that the “free lunch” is a pretty meagre feed.
* Charge cycling of conventional batteries decreases battery life. While this is apparently less of a problem for Li-Ion and some other new technologies, the EV owner still needs to be convinced of this
On the other hand, EV’s at parking stations throughout the day might well be a convenient sink for renewables generation. With some smarts in the charging systems, varying charge rates could significantly reduce the need for alternative back-up generation for renewables. For example, sudden reductions in PV in a large regional city as clouds passed over could be offset by suspension of EV charging, rather than requiring a step-up of gas turbine generation.
I agree Phil, until batteries are really cheap no-one would be crazy enough to cycle them for a measly dollar or two. And if offpeak flattens right out with all the new dispatchable demand there might be even less financial incentive to so on a day-to-day basis.
It was a good presentation from Owen, should this scenario eventuate I cannot wait for the relevant network and retail tariff for the new peak period between 11pm and 2am.
Thanks Nathan.
EV charging certainly does highlight a few issues which would need to be thought through before uptake becomes large enough to cause headaches. The AEMC did some work on the implications of current market arrangements for EV and natural gas vehicles in late 2012. See: http://www.aemc.gov.au/Market-Reviews/Completed/energy-market-barriers-for-electric-and-natural-gas-vehicles.html
Why is the SA price close to $200/MWh, with a much higher spike into the thousands, when the Vic interconnector is running at only half capacity ? Surely the interconnector would curtail the price spikes unless it is at capacity.
It is also surprising that despite the forecasts of hot weather, there was absolutely no output from the Dry Creek gas peaking plant. Most of the peak was supplied from Torrens Island. Surely on days like this the Dry Creek plant would be made ready for production, and be placing bids below the $200/MWh spot price.
Hi Thomas,
Can you please tell me in simple terms why a private sector retailer had to be placed in the supply chain.
If a responsible government was going to keep prices to a minimum, would it not make sense to not have yet another layer of skimming in the supply chain.
I believe that in some instances the government has had to pay these retailers some compensation for lost revenue or something like that.
I simply can’t understand how anyone could take such a simple issue and complicate it.
My electricity bill has gone through the roof in recent years.
Regards,
Alan
Nicely summarised article Ric thank you.
Your final comment is a bit timid on reasonable trend line analysis and extrapolation though Ric; to paraphrase “In the final analysis, it may not matter that much to the solar industry whether $40 is obtained through the Clearing House or slightly less than $40 is obtained through the market a little earlier.”
You are wrong I predict, and it will be a very important point when the possibility arises.
Consider this scenario; currently agents trading STC’s on behalf of liable entities still have a very slim business case that they can put to their customers, that they can continue to provide them a service and exist on the slim differential between current STC price and Registry floor price [$40], which the liable entities know they will have to pay once the clearing house comes into play.
So up to about $39.50 some traders may be inclined to still trade, but anything over $39.50 their ability to continue effectively and profitably evaporates. They will not volunteer to continue to act as a buyer for liable entities for no return. So the buyer will start to purchase from the clearing house direct without the services of an agent.
The agents will exit the business and move in another direction entirely. Once the clearing house comes into play it is likely to continue as the sole agent for STC sales unchallenged. Trend lines are very important tools and they tell us a lot about future likelihoods for PV installations in Australia.
Pardon my ignorance…but is this high, near record demand happening in spite of the high solar pv penetration in QLD? I had thought the solar influence would have considerably flattened the peak load curve which I presume is largely caused by air-conditioning.
Hi Neil
Thanks for the question.
How demand grows (or declines) is a result of the interaction of many factors – some supporting growth, some promoting decline.
1) A couple years ago we posted about these factors that were coming together to deliver the decline in (average and peak) demand seen in the past few years;
2) More recently, Hugh Saddler has put together this detailed analysis that highlights the magnitude of the effect these factors are each having on demand suppression.
Interestingly, the effect of solar PV is currently, according to Hugh’s numbers, five times less significant than the most significant factors (i.e. 7% contribution compared with 37% to a decrease from Business as Usual). I’d stress that this is “currently”, as it may be that the effect of solar might rapidly increase, if the take-up rate continues as it has in the past couple years – though, as Ric noted here, there has been a cooling in the take-up rate despite decreasing installed costs.
So, Neil, in answer to your question – scheduled demand (which AEMO has to meet in dispatching the market) would have been higher without the PV that’s operational now, but the amount of PV that’s currently installed is too small to prevent such peaks ever* occurring.
* as to whether or not that would be a good thing – economically or environmentally – each reader will have their own personal point of view
Does this answer your question?
Paul
PS – the news reports this morning indicating that the Boyne Island aluminium smelter will be running at less than full load this quarter, due to high Q1 power prices, makes this new record Saturday demand even more remarkable.
thanks Paul. Yes it does. I had obviously thought that solar pv was having a more significant impact. I will read Hugh Saddler’s report and others you’ve referred to.
Seem like old times with the Qld Government owned generators gaming the market!
Have been watching this with interest today Paul.
Can you offer any insight into why Tas has been trading at -$79/MWh all day? Struggling to understand the bid strategy that would cause this outcome…
Thanks Clint
That’s the sort of question that might be better answered offline! I will call.
Paul
The strategy is to ensure maximum Basslink exports and inter-regional revenues. Bid negative to get dispatched.
Any chance of calling the summer over today, I think I’m sitting pretty for the BBQ. Probably not much chance of the demand being lower tomorrow!
Chris
The various emergency departments in southern Australia would probably also wish that the heatwave, at least, would be over.
I see the predispatch forecast is for over 33,000MW today – which would make it, I think, the highest the demand has been since this hot week in February 2011, nearly 3 years ago.
Maybe we should allow another round of entries, for a consolation, consolation prize?!
Paul
Thanks for the up to date coverage of the heatwave and its effects on the NEM, Paul! It’s going to be very interesting on Thursday…
It will be even more interesting this time next year when the QLD LNG plants start operating and there is no fuel available for baseload CCGT power stations and a further 1900MW of generation in the NEM will not be running
Plus 1000MW of coal fired generation in NSW to be mothballed as just announced
While I loathe the heat, watching the NEM on these days is lots of fun. While South Oz is hot, you cannot underestimate how the wind generation has changed the SA market, compared to the similar temperatures 5-6 years ago.
I wonder if the heat impacts on solar pv units on households?
These are great posts.
It is interesting that these high network demands are occurring despite over 2 GW of solar PV capacity across Australia, and that market penetration is particularly high in Adelaide. A South Australian assessment concluded that PV had a capacity factor of 40% at the time of peak annual demand. As solar panel technology improves, we should expect this to increase as panels are developed that are less affected by high temperatures. Panels that perform well under high temperature conditions should be a much more cost-effective solution to heatwave demand that additional poles, wires and generators.
Another interesting feature is that MurrayLink (connecting South Australia to Mildura) has had minimal capacity to import electricity into SA, and has in some of your previous posts has been exporting from SA. Under normal conditions MurrayLink can import 200 MW into SA. However there appear to be constraints on the distribution system to the east of Mildura. There is one 220V line from Geelong, which serves Ballarat and Horsham before reaching Mildura. Another starts in NE Vic and serves Shepparton and Bendigo before Mildura. An third line runs from Wagga Wagga to Broken Hill with a cross-border connection to Mildura. The first 2 of these lines are no doubt close to capacity serving other cities along the route, particularly constraints on the Geelong-Ballarat sector. Extra capacity from Geelong to Ballarat could free up more capacity further up the line.
all eight at Pelican Point?
Apologies, Interested Observer – must be too much going on at present!
Error #1 – meant to type Torrens Island, which actually does have eight units. Have noted above.
Error #2 – one of our other readers has pointed out that the peak demand on 29th january 2009 was actually 10,496MW as we showed back then in this post. Will update image above also.
The arrow to Anglesea as having a transmission constraint is surely in error. It has its own 220 kV line to join the grid at Geelong. Surely the arrow is meant to go to Morton’s Lane. This wind farm shares the same 66 kV line to Terang, where it joins the 220 kV network. However today there would be plenty of local demand from the 66 kV line, with air temperatures of 42 C in the Western District. I work in the area, and our poor air-conditioner at work was blowing out air at 41 C !
It is interesting to note that Morton’s Lane Wind Farm consistently has a higher capacity factor than either Oaklands Hill or MacArthur, even though it shares essentially the same wind resource. Morton’s Lane is not located on a noticeable prominence – perhaps it’s simply better design and operation.
PS. Morton’s Lane shares the same 66 kV line as Oaklands Hill, and is subject to constraints, whereas the MacArthur Wind Farm feeds the 500 kV line to Portland, and does not have transmission constraints.
Hi Malcolm
Thanks for all the comments the past couple days (and the compliment).
I have made a couple errors today, but the bit about Anglesea being subject to constraint is now (oops, not) one of them. The “+” symbol on the display for that station (and the “-” for the AGL Hydro plant) means that they are subject to a bound constraint – at this time it was the one AEMO calls “V>>SML_NIL_1”.
AEMO provides the following as the reason for the constraint “Avoid overloading Ballarat to Moorabool No.1 220 kV line for loss of Ballarat to Elaine to Moorabool 220 kV line and Mt Mercer Wind Farm” – the user can drill into the constraint in ez2view to get this detail.
The “+” symbol for Anglesea means that the constraint is (slightly, because of a low positive factor) trying to wind the output of the station down.
Does this help?
Paul
Hi Paul. Great stuff. I’ve been reading lots about how much solar pv has contributed to keeping the peak from going higher. Any thoughts on this?
Thanks James,
I did reference Kerry’s analysis in this post on Saturday, which was more directly to do with Demand Response.
Paul
A user with a steady usual load doesn’t even need to participate in the NEM pool in order to benefit from negawatts. It is a two-step process: (1) They write their PPA with their retailer at pool pass-through prices (plus network and other charges). (2) They enter a hedge with a retailer (can be same or different retailer) or other financial intermediary for their usual consumption profile. They will collect the high spot price from the hedge counterparty when they reduce their consumption below the hedge quantity. And if they don’t reduce, they are still hedged against the pool price. That is, you don’t need a Rule amendment, and you don’t need to play the pool, in order to get value for your negawatts.
Hi Graeme
Thanks for the added insights – some of our clients do a variety of hedging (all sorts of permutations, we understand) to back up their physical ability to curtail.
This also varies by season, and with their order book, etc…
Paul
Hi
I’m enjoying your analyses – thank you. Given that Qld has more than double the PV capacity of SA, and around the same rate of penetration, I’d be keen to see your view some time as to the impact that PV has on summer demand in Qld, especially at critical peak times. You may or may not have seen this – http://reneweconomy.com.au/2014/solar-saved-southern-states-from-new-and-costly-demand-peaks-92609
Regards
Dean
Thanks, Dean
Yes, solar is promising to become increasingly relevant (and of interest) in a number of ways, so we’re looking at what we can do.
Coincidentally Kerry’s article you referenced we linked here as a preamble to discussion about Demand Response, as another contributor.
Regards
Paul
Paul
Thanks for the article. I have looked at the SA demand and price curves for the hot days last week and they do show the same “jagged” peak you refer to in the article.
I know that some users that take spot risk have a staggered load shedding program with less essential demand being load shed at much lower spot prices than MPC, with trigger prices for least needed demand occurring below $1000, with other demand being shed at increasing price triggers – essential services stay on regardless. The challenge is to get the “essential; demand as low as possible – and some have found that this essential load can be less than 10% of normal demand.
Others install their own generation, still load shed and sell back into the grid and not only avoid the high rice events but contribute to the capital cost of their generation plant through the revenue earned from the market.
So there are number of ways to skin this cat!
Regards
David
Thanks for the added comments, Dave
Yes, we also know of people doing this – and there are, no doubt, others as well – some of whom probably keep their cards close to their chest as a piece of competitive advantage in their sector.
Paul
>”5) As others have noted, we see that the wind farms continued to produce with the expected intermittent pattern. In particular we see an inverse correlation between wind farm output and spot price, which is entirely expected (after all, as others have pointed out the wind farms help to lower spot prices when the wind blows – so the inverse is to be expected).”
Is this an impartial statement? I would have said the most obvious conclusion regarding wind in this chart is that when demand is high wind power drops.
We could draw the following conclusions about wind generation from this chart: Wind does not show up to work when most need and capacity credit should probably be lower than AEMO says.
It is likely wind is causing a lot more increase in the cost of electricity than simply the spot prices might suggest.
Importantly, these issues will become much more of a problem, and electricity prices will be forced disproportionately higher the more wind capacity is added to the NEM.
Hi Peter,
Wind IS proportionally lower during peaks, about 20% less on average in SA. It can seem like more because the older wind farms are non-scheduled and their production is subtracted from demand. In NSW wind production is more closely correlated with demand. This can be seen by the dispatch weighted prices that wind farms receive. In SA a WF gets about 20% lower spot prices than the average, in NSW it’s only a few %. Seems fair given my first statement?
There are no capacity credits in the NEM. In WA where there are capacity credits WFs receive a much smaller credit. The credit was originally equal and the IMO reduced it after review.
In fact WFs are lowering the wholesale price of electricity. This is necessarily true of any new generator that produces energy. In SA wholesale prices have dropped over the last 6 years while they have risen in all other NEM states, during which time the only new generators installed have been WFs. Also of interest the frequency of very high prices has fallen more than the frequency of very high demand days. Sometimes the wind does blow at the right time, however this makes for a boring headline.
This whole discussion about who did what is rather silly. The only conclusion I take from this is that a diverse mix of generators is ideal.
“Gas power does what it is supposed to”
“Wind power doesn’t do what it wasn’t built for”
Unfortunately all of this partizan bickering is making the average Australian even more confused about energy. Thanks for the wrap up Paul.
No problems, Tom
We’re keen to do what we can to help making energy markets more understandable.
Regards
Paul
I was curious to see if it would become so hot that many wind farms would lose generating capacity due to high temperature shutdowns (which often occur around 40 degrees celsius for many models of turbine).
Though it’s hard to discern from generation data exactly what caused a shutdown, this didn’t seem to impact the fleet as a whole, which is good to see.
Cheers for the post, Paul.
And almost every other CCGT power station in the NEM will go the same way in the next 12 months. Most operators will not formally announce mothballing and will say that the units are available for peaking duty, but as most of the plants do not have bypass stacks they will be useless as peakers.
Thanks for another comment, Interested Observer
Each gas-fired plant will have their own local considerations to work through, with factors including type of plant (e.g. cogen/not, combined/open/other, etc…) , contracted gas supply arrangements, location, portfolio dynamics and drivers etc…
We previously commented on Tamar Valley as just one example.
However in an environment where gas prices are rising (if current trends continue), and (again, if current trends continue) wholesale electricity prices remain capped by declining demand and increased non-scheduled or semi-scheduled generation, then the changing spark spread could be expected to have other impacts.
Paul
Is the ACT too small to be included ?
Thanks for the question, Keith
What’s shown in the NEM-Watch snapshot above above are boundaries between NEM Regions, as distinct from state boundaries (though in practical terms it’s pretty much the same thing at present).
Hence, it’s more a matter that Canberra is tightly meshed in the NSW region so does not need its own pricing region – moreso than the fact that its electricity demand is relatively small.
For many years at the start of the NEM there was a Snowy Region between NSW and Victoria – as captured in this snapshot on 7th December 2005. Snowy was abolished from 1st July 2008 with part allocated to NSW and part in VIC.
Does this answer your question?
Paul
Hey Tom,
Dr Mark Diesendorf told us for years that a fleet of geographically dispersed wind farms will
displace base load coal-fired powerstations. Policymakers are still proceeding on the basis of his
now-famous “The Base Load Fallacy”. Remember it? So, let me correct your statement above here
to now read: “Wind power doesn’t do what it was built for, which was to generate base load
power in place of coal”. And to add, “What’s more, it doesn’t matter how many wind farms are
built, it never will.”
Nobody is “bickering” here, Tom. Peter Lang is merely stating the facts as they are. He is
absolutely correct when he says that wind does not show up to work when it is most needed.
Wind is an abject, futile, hideously expensive, and dangerous-to-grid-security, failure.
May I too add: Thank you for the analysis, Paul. It is most informative.
Regards,
Paul
Renewable energy primary justification is that it is expected to cut CO2 emissions. But it is ineffective at doing so and hugely expensive.
Renewable energy proponents like to make claims about how much capacity has been built, and the rate of growth of capacity. But these claims are highly misleading. They should talk only about how much firm capacity has been added; the cost per kW of firm capacity; the proportion of firm capacity compared to the total firm capacity of the grid; and the cost per tonne CO2 avoided.
About 5%-10% of wind capacity is firm capacity and that will decrease as the proportion of wind capacity increases in the grid.
The proportion of solar PV that is firm capacity is also very low, and will decrease as the proportion of PV in the grid increases.
Wind and solar are about half as effective at avoiding CO2 emissions as is claimed by their proponents. And they become less effective as their proportion of the electricity generation increases.
Because of the low abatement effectiveness, the abatement cost ($/toinne CO2 abated) is about double what is claimed by the proponents.
I agree with Tom. “Unfortunately all of this partizan bickering is making the average Australian even more confused about energy.” The ridiculous claims and twaddle by the renewable energy proponents should stop.
I think you are missing the other option that will be relevant in many locations. Community ownership can be considered a mix of private and utility ownership where the term “distributed” really comes into its own – and local smart-grids can implement the integration of all relevant options in the most economical way.
Local smart-grids (community or utility owned) will need to operate like the internet, where various levels of operation will occur to meet current conditions. As far as end-users are concerned, their power supplies should be seamlessly selected from the available supplies – via the transmission system or from local sources depending on availability, security and cost considerations.
Hi Ian
Thanks for adding that one.
There could be others evolve, as well – in the NEM there are a couple of Demand Response Aggregators operating in a variety of forms. It may be them that take the step of owning and dispatching the resources.
Paul
Hi Ian,
I think you raise a good point, local mini grids have many advantages and I think there is definite potential for these, particularly in rural areas. The business model will be much more of a complete energy provider, supplying both generation and distribution (gentility?). There is the potential for a large cost saving with this model, and I think it will become more common as the price of storage drops below distribution infrastructure costs.
Jono.
Hi
just had to say that a co-op and that is what you are getting at will not work (sorry) no co-op works that good.
Stand alone is the only way to go. Generate your own power,store it in battery’s and use it when you need it.
Now generating power and have it on hand is the hard part so it seems I must admit that it could not be done however I have seen a genorator that works off a 12 volt system and genorates it’s own power running an inverter that runs this persons home. He made it in his shed.
It was something I had never seen before and I have been round for 1/2 a century. It really did work how I do not know! Will he make another one not sure! Will he sell it I can give the answer to that NO I tried to buy it but no way and I had to make sure I did not tell anyone his name or where he lives.
Thanks
David.
No co-op works that well?
http://reic.uwcc.wisc.edu/electric/
http://www.mondragon-corporation.com/eng/co-operative-experience/history/
“No co-op works that good” – Have you heard of Fonterra? It is the largest company in New Zealand and one of the largest dairy companies in the world which is a cooperative owned by dairy farmers. WIth $18B in annual revenue and $700M in profit I would say Fonterra “works good”
I would have thought coops for electrons would suffer seriously with small scale operation due to the problem of peak demand and all electrons not being of the same value at varying times. The coop would clearly need to engage itself with demand management- ie smart meters and peak pricing and the investment and admin of that on a small scale may well prove the greatest hurdle, absent the economies of scale of Big Gen now.
As well we live in a mobile world and attaching a lien to my home via a street, block or whole suburb power coop would be one I’d have to carefully consider buying into. After all if I sell up and move, there’s no guarantee the next buyer wants to be shackled with my home’s power coop obligation. That would leave most to consider stand alone system economics only and their risk return should there be a change of plans and any forced move in future. Again Big Gen has the edge over any small scale coop there.
This is great news. I can now replace my very dated and carbon intensive charcoal BBQ with a lower carbon emission alternative. Coincidentally, I had just run out of coal anyway.
Interestingly, my entry was based on AEMO’s 2013 NEFR forecast, but as we all know that doesn’t forecast NEM wide maximum demand, and provide probability of exceedance numbers rather than a single number. So a lot of crafting went into this, including finding reasonable regional diversity numbers (we were in a ENSO neutral period, so I went back and looked for similar years as guide) and I assumed it would be a fairly hot year based on the temperatures up till around Christmas and picked a number closer to the 10% POE than the 50% POE. Finally I then subtracted the forecast level of demand side participation during high price events.
Thanks for the details, Magnus
There was a diverse range of entries – ranging from W.A.G. up to very scientific approaches (such as yours).
Well done – and thanks for the entry, and for sharing your (formerly) secret methodology. We’ll look forward to seeing a greater number of close entries next summer.
Paul
Woo! Thanks Paul. I was pretty pessimistic about this Summer, it’s always nice to see some volatility in SA though 🙂
Thanks Tom
It was not just you, it was the majority that were bearish about demand over summer.
Paul
Hi Paul,
I assume those prices aren’t adjusted for inflation?
$35 in 2002 would be equivalent to $47 in 2012 (3%). I guess it’s still not a huge difference.
I would be interested to see the increase in demand over that time period. Did the dry conditions also drive higher consumption?
Cheers,
Tom
Hi Tom
Prices shown are not adjusted in any way – they are published spot prices from NEMMCO/AEMO.
As many have noted, wholesale spot prices have not really escalated over that period, but instead can be understood by supply/demand dynamics (i.e. higher prices when short generation, lower prices when long).
With respect to your other question about demand levels, you’d be able to easily perform your own analysis with a licence to NEM-Review.
Does this help?
Paul
This is not impartial. It is clearly pushing for renewable energy. There is not a mention of the much cheaper and batter fit for purpose technology to cut emissions, i.e. nuclear energy. Why not?
Nuclear power is better than renewable energy for meeting all the important criteria. Renewable energy cannot be justified, on a rational basis, to be a major component of the electricity system. Here are some reasons why:
1. Nuclear power has proven it can supply over 75% of the electricity in a large modern industrial economy, i.e. France, and has been doing so for over 30 years.
2. Nuclear power is substantially cheaper than renewables
3. Nuclear power is the safest way to generate electricity; it causes the least fatalities per unit of electricity supplied.
4. Nuclear power is more environmentally benign than renewables.
5. Material requirements per unit of electricity supplied through life for nuclear power are about 1/10th those of renewables
6. Land area required for nuclear power is very much smaller than renewables per unit of electricity supplied through life
7. Nuclear power requires far less expensive transmission (much shorter distances and much smaller capacity in total because the capacity needs to be sufficient for maximum output but intermittent renewables average around 10% to 40% capacity factor whereas nuclear averages around 80% to 90%.
8. Nuclear fuel is effectively unlimited.
9. Nuclear fuel requires a minimal amount of space for storage. Many years of nuclear fuel supply can be stored in a warehouse. This has two major benefits:
• Energy security – it means that countries can store many years or decades of fuel at little cost, so it gives independence from fuel imports. This gives energy security from trade wars and military conflicts
• Reduced transport – nuclear fuel requires 20,000 to 2 million times less ships, trains etc per unit of energy transported. This reduces shipping costs, the quantities of oil used for the transport, and the environmental impacts of the shipping and the fuel used for transport by 4 to 6 orders of magnitude.
There is no rational justification for renewable energy to be mandated and favoured by legislation and regulations.
Peter,
Nukes don’t make sense for Australia. Capital costs are too high, water consumption is high, peak consumption in Aus. is high. By all means it’s great for the eurozone and the asian power economies.
Addressing your points:
1. France has less than half the peak load per capita of Aus. and significant import/export capability with neighbours which helps counter nuclear inflexibility. Their baseload is also higher per person. Nukes are perfect for them but the comparison doesn’t work for us.
2. Nukes are only cheaper where labour is cheap. The Brits tendered for a new nuke plant recently and the best price they were quoted was the equivalent of $170/MWh. Significantly more than solar or wind in Aus. and they have better nuke experience and easier procurement. BREE reckon $100/MWh with no decommisioning costs and a 60 year life. 60 years is significantly longer than the typical 40. Either way it’s still more expensive than $80 for wind. You’ve probably also noticed that PV is cheaper than retail, can’t see anyone getting residential nukes…
3. True but this includes hydro which has had some disasters. Renewables and nukes are both much, much safer than coal. The difference between them is not significant.
4. True but again we are comparing to coal which is much worse. Slight differences between the clean techs is nothing compared with getting rid of fossils.
5. Sounds right. We are running out of uranium faster than steel though…
6. True, even though mine area is large. This is why nukes are so great for Asia/Euro. Land is not scarce here and wind, solar and all other industry is much more productive per hectare than agriculture (by a few orders of magnitude).
7. True but this is only a small part of customer’s costs. The reduced cost of renewables compared with nukes offsets this.
8. No one is building breeder reactors. They cost too much. There’s no point turning cheap fuel into very cheap fuel if it increases capital costs. Also people get worried about plutonium.
9. True but again this is an advantage for countries with no resources. We are not threatened by a lack of energy supply.
Why am I making a big deal about this? Because advocating for nukes distracts from actually reducing emissions from the electricity sector in Australia. Australia is too small to build nukes at scale, or profit from fuel processing. Let’s continue supplying the U to the rest of the world, they need it more than us.
Paul
Thanks for putting up such a carefully considered and insightful set of comments, including those in relation to the role and uses of market modelling. I don’t have time to respond in any detail, but I will correct you on your recollection about the original purpose of the RET Act. Here’s an excerpt from the 2003 Tambling Review of the operation of the original Renewable Energy (Electricity) Act 2000:
1.1 In November 1997, the Prime Minister, the Hon. John Howard MP, announced
a package of national greenhouse response measures, Safeguarding the Future: Australia’s Response to Climate Change, which included the following initiative:
Targets will be set for the inclusion of renewable energy in electricity generation by the year 2010. Electricity retailers and other large electricity buyers will be legally required to source an additional 2 percent of their electricity from renewable or specified waste-product energy sources by 2010 (included through direct investment in alternative renewable energy sources such as solar hot water heaters). This will accelerate the uptake of renewable energy in grid- based power applications, and provide an ongoing base for commercially competitive renewable energy. The program will also contribute to the development of internationally competitive industries which could participate effectively in the burgeoning Asian energy market.
1.2 This initiative, now known as the Mandatory Renewable Energy Target (MRET), is one of a number of Australian Government greenhouse gas abatement measures and has a strong focus on renewable energy generation and industry development. This focus recognises that emissions from fossil fuels used in electricity generation comprise around one third of Australia’s total greenhouse emissions, and that increases in market penetration of renewable energy can help displace fossil fuel emissions and improve Australia’s greenhouse and renewable energy industry performance.
(retrieved via http://pandora.nla.gov.au/pan/121641/20101007-1302/www.mretreview.gov.au/report/index.html)
I think that makes it clear that a core initial policy objective of the RET and its successors was reduction in GHG emissions from the electricity sector.
Thanks Allan
Thanks, especially, for pointing out 1 of the mistakes I made in the post (it’s almost certain there is more than just that one, so will look forward to other comments helping me see other slip-ups).
Curiously the one you note illustrates the error in just relying on my memory. Obviously my memory is not what (I remember that) it once was!
Paul
On “the age of entitlement is over” I dont agree that the RET can be classified as an entitlement. When the MRET was introduced, and through subsequent amendments, successive governments invited companies to invest in renewable energy infrastructure. Whatever the objectives behind the scheme, it was an invitation to commercial entities to invest and help achieve a national goal. It should not be confused with the car industry for example, which effectively would give an annual ultimatum to governments – “would you like to subsidise us or should we go?”
Thanks Lane
“Entitlement” is such a loaded & divisive word – bit like an obsessive focus on “Rights”. Certainly achieves an effect for Treasurer Hockey in its use, but what are the long term consequences?
Regarding the RET, it seems that there’s at least 2 categories:
1) Capacity already on the ground and that has been put there on an expectation the RET would continue into the future; and
2) People who’ve invested in projects not yet on the ground (shovel ready, or still on the drawing board).
If the RET is changed (which seems likely) what are the implications for (and compensation options for?) each.
Just one more example of the policy mess the NEM seems to be increasingly burdened with …
{but it’s not specific to energy – being a software company we’re also reading commentary about changes in the Budget about changes to funding/grant arrangements there, and the effect that this will have on sunk investment}
Paul
As a quick PS to my article above, this description of the RET on the Government’s website is doing the rounds on Twitter and would have been useful for me in helping me avoid the mistake I Allan pointed out above.
Perhaps also useful for you?
The “Notes of LETAG meeting” which were leaked and posted on the Australia Institute website in 2004 clarified some of the thinking around the RET. It was ostensibly intended as a response to Kyoto, but was supposed to provide a kick start for innovation and Australian industry – it was never intended as a subsidy for imported wind power. The difficulty is that, once enacted, it created a whole industry of corporations, consultants and lobby groups who relied on the RET since the self-supporting industry development simply didn’t occur. As I recall, in 1997 Robert Hill took a proposal for a $10 carbon tax to cabinet but was rejected. Somewhat ironically, this would have been a far more economically efficient policy and the NEM would have determined the most efficient level of wind and solar, whatever that would have been.
http://www.tai.org.au/node/1192
Hi Graham
Thanks for your interesting addition to this article.
Paul
Paul –
To add to your list of recent RET modelling I would add RepuTex’s analysis from April 2014 that looked at the implications of a downgraded LRET on the shape of the electricity mix in the NEM.
RepuTex used a bottom-up least cost dispatch model that mimics the bid and dispatch process of the NEM with a resolution of 200 periods per year. Important assumptions included in this modelling exercise were the continuation of a fixed penalty price for the RET, the ending of carbon pricing from July 2014, and inclusion of existing and committed new entrants. Scenarios considered included the existing RET, an updated 20% RET, and no RET from 2015.
One of the main findings was that once the CPM is assumed to be repealed, changing the LRET has much less impact on the resulting renewable capacity that is built. This suggests the combination of carbon pricing (even at EUA equivalent levels) and the LRET are important polices that work together to transition the NEM to a cleaner energy mix.
As for longer term implications, this study also found that new capacity will continue to favor renewables over coal and gas on a financial basis. This suggests that if NEM baseload demand continues to grow to exceed the capacity of the existing mothballed coal plants, developers will choose large-scale renewables as the most cost-effective capacity to build in the future.
Public summary slides of the report can be accessed at – http://www.reputex.com/publications/market-update-renewable-generation-powering-down/
Thanks Paul – your questions provide a salient reminder of the context behind how we got to the current situation on RET in the first place.
If there is mainstream political intent to retain RET as it was intended to work after the expansion in 2008/9, it does seem that a floating target, based on a % of actual load, is going to be a necessary, regardless of the level or the profile to get there. The uncertainties on load forecasts, through PV take up, home efficiency gains, customer price responses, gas pricing, etc, are much greater than they were back then. The idea of predicting those outcomes this time around with a fixed volume target that gets a ‘good’ outcome seem highly optimistic. A ‘good’ outcome means avoiding unintended costs of things like 1. over build of new renewables that don’t assist the demand supply balance and create unintended value transfers eg. between solar PV owners and others, or 2. under building renewables through too low a target that leads to more high emission plant being built and creates emissions levels above the Government’s (or the community’s) policy objectives.
It will be interesting to watch it play out………
The justification for policies to encourage PV is to reduce CO2 emissions. Therefore, arguably the most important questions that need to be asked are:
Q5) How much CO2 is being avoided per MWh of PV energy generated? (To work that out we need to know the amount of PV generated at about 10-15 minute intervals in the NEM and the amount of CO2 emitted by each fossil fuel generator unit – as is the case in Ireland, France and other EU countries)
Q6) What is the CO2 abatement cost with PV? (To calculate this correctly all costs that should be attributed to having PV connected to the grid must be properly attributed to the various generators)
Hi Paul,
thanks again for another interesting article.
Let me declare an interest: I have a 5Kw system and I am on the 44c rate for the rebate.
The following is a series of ‘post project review’ comments on planning and installing panels, off topic I know but bear with me please.
1. There are enough houses with panels on them now (is it in excess of 1 million?) that the roofing suppliers and the panel installers should be talking to each other about providing wind-rated fixtures so that panels can easily be installed at the optimum angle.
2. Some installations of panels are just a plain eye-sore, not if’s or but’s allowed. Not only that, they are an inefficient eye-sore, from the generation point of view. The panels industry needs to step up and ensure that people get value for money from properly oriented and installed panels.
3. Just as all diesel engines should run turbos, all domestic panels should be controlled at the micro inverter level, ie panel by panel, again so people get an output which approximates the nameplate figure. Large commercial panels don’t need this as much.
4. People on the low Fit’s, 8c and the like, will really appreciate the storage systems which are not really far away, say 5 years max, with the feverish work that is going into improved storage. The widespread use of efficient storage will be another major upset for the generators and the retailers. Storage will certainly decrease the evening peak; it will depend on the efficiency of the storage as to whether the morning peak is flattened as well.
5. The political response to domestic storage of electricity will be interesting to watch. With conservative governments in place in most jurisdictions, the response to rooftop panels has not demonstrated even an acknowledgement of the facts, let alone an enlightened approach in any sense. Hasn’t electricity reticulated from central generation been a political weapon/football over quite a number of decades? Now, in terms of metropolitan domestic generation, is it coming to an end? What will its death throes be like? To mix even more metaphors, is the ‘domestic panel generation’ tail is going to wag ‘large coal-powered generator’ dog?
6. There was a beer ad where the blokes built a combined shed where their properties met, and made it a ‘man cave’. What if the boys decided to pool their beer money, get a solar system spread across part of each of their properties, and then each take some of the output. I wonder what (being a Brisbane resident with Energex the local ‘poles and wires’ authority) Energex would say about the boys wanting that connected to the grid?
Whenever Mike Swanston from Energex contributes to this debate, he is always worth listening to. Thanks for this contribution, Mike.
Many of the people who can’t install solar would live in strata-title properties – even if they own the unit / villa / townhouse they don’t own the a roof. The individual property owner has to negotiate with their strata committee to install any solar panels, making things that much harder.
Can we have a winter demand forecast competition for Tasmania?
Hi Derek
If winter ever arrived, we might think about it!
Paul
You deserve a gong from the Brits’ dear Queen for the effort above, Paul.
Much to chew over for many people. Thanks.
Hi Steve
I’d be surprised if old Mrs Windsor had an interest in something as arcane as Demand Response – though I had heard that she takes a keen interest in switching off lights in that big old house of hers.
Paul
Grrat discussion on DM Paul.
There are many parties in the NEM who dont want to see DM succeed that’s why we have had such a long process regarding the Power of Choice review.
The other improtant role that DM plays in electricity netowrks and its good to see the AER incentivising DB’s for DM solutions; howver there should be a lot more of it.
Thanks Peter
Some methods of Demand Response are already succeeding in NEM (for instance, spot exposure for large industrial energy users – a method we have experience with).
The proposed new “Self-Scheduled Negawatt Mechanism” (bit of a mouthful – maybe the “Self-Scheduled Demand Sellback Mechanism”?) will be another method available, in the energy market, should it be introduced.
You note about the network side – this has not really been our focus, to date. Hence we’re less able to comment on how much of that is being used.
Without some form of dynamic network pricing, I suspect this will be problematic to introduce equitably – but with the debate going on about the see-saw of network flows at a household level, perhaps there will be other reasons to move towards network charges that are:
1) More related to kW than kWh; and
2) Change reflecting the loading on the network in that particular part of the world.
Mike Swanston talked about part of this in his “size of pipe” analogy in the forum I posted about here. Click through to the podcast.
There are technology offerings that could be developed, if this would be the case, to automate the response. Economic storage would help, as well, with a couple different models in terms of how it could be rolled out.
Paul
Paul
A late comment: you may want to check on how the “Mr Blue” and “Ms Green” data series were compiled by Energex.
Obviously there has been very strong uptake of solar PV across the period shown in the second chart. So lots of Ms Greens transformed themselves into Mr Blues over the period by installing PV.
If the average annual usage data at each time point on the chart is based on all those in each segment (ie PV vs no PV) at that time, then the changing averages could in theory be driven solely by the diffusion of customers from one segment to the other over time, not by changes in the actual usage of individual customers across time.
If (hypothetically) the average usage of those who left the no-PV segment was:
a) higher than average for that segment – quite plausible as higher users with larger bills may be more motivated to install PV, and
b) roughly equal to the average of those already on PV (less clear why this would be the case)
then that could explain both the (apparently) falling average usage in the green segment and the static average usage in the blue segment, without necessarily implying any changes in the “underlying” usage patterns of individual customers. If the customers installing PV were tending over time to be increasingly large users, then that could even mask a declining usage trend for customers already within the segment.
I guess my main point is that it is very difficult to hypothesise about what charts like this one really show – let alone any underlying reasons – without fully understanding how they are constructed.
Thanks Allan
Another great question.
I inferred, from what Mike had said on the night, that there was a change in the underlying consumption – but will see if I can find out more.
Cheers
Paul
Hi Paul,
Thanks for a thought-provoking post! I have a couple of clarifications (as requested), then some comments.
Significance of the capacity market to DR
In part 4a, you asked where EnerNOC gets its revenues from if only 2% is subject to Order 745. All the other sources you mention play a part, as does our growing Energy Intelligence Software business, but capacity payments are the most significant. However, I don’t agree with your inference that this means that we’ll see correspondingly low uptake of the DRM in the NEM.
Where there’s a capacity market as well as an energy market, peaking resources (both conventional generation and DR) derive almost all their revenue from the capacity market. If it’s not mandatory to participate in the energy market (in PJM, it’s not for DR), then they many not bother with it.
If you don’t have a capacity market, then peaking resources have to make do with the energy market and associated derivatives markets. So long as the market is functioning correctly (I realise that’s a big caveat), at equilibrium they should get the same amount of revenue this way. The NEM isn’t quite a perfect market, so I wouldn’t expect participation to reach as high as it ideally would, based on fundamentals, but the potential is still significant.
Huge consumer savings from DR in PJM
In part 4e, you asked whether the reference to PJM saving $11.8b, or 40% of total costs, through DR participation was a typo or a calculation error. It is neither. The independent market monitor calculated this by rerunning the 2013/14 auction without the bids from DR and EE suppliers — i.e. procuring the capacity needed from generation alone:
This is from p.52 of Monitoring Analytics’ Analysis of the 2013/2014 RPM Base Residual Auction Revised and Updated. PJM’s auctions are run three years ahead, to give time for proposed resources to be built if they clear. Hence we know the corresponding figures for the next three years: $9.6b, $13.7b, $10.1b. Pretty soon, you’re talking real money. These savings show how much cheaper it is maintain reliability if you involve the demand side, rather than relying purely on the supply side.
Comments
As I hope I made clear in my talk last week, the kind of price-responsive demand with which you assist your customers is by far the best option for the small number of very large customers who are sophisticated enough to cope with the associated risks. By buying at wholesale prices, avoiding paying a margin to the retailer for risk management, they can come out a long way head.
The purpose of the proposed DRM is to allow many more customers to participate by introducing an alternative approach that doesn’t require them to take market price or volume risks 24x7x365 just because they want to be able to provide some demand response during extreme events.
I don’t think that introducing the confusing “negawatt” term, or suggesting that it involves the creation of a separate market for a separate commodity, really helps. There’s only energy involved. As you noted in part 3b, there’s no incentive payment; it’s not an incentive scheme.
One thing I would add: ideally, demand response should be a scheduled resource. This makes it much more useful to the system operator than price-responsive load. I haven’t seen the final version of AEMO’s rule change proposal. However, in the working group discussions before AEMO went away and wrote it, it was clear that the DRM was proposed to be non-scheduled only because AEMO considered it too hard to alter the NEM’s dispatch systems to support load participation in the near term. The intention was that this would be revisited once they had some initial practical experience with the DRM.
AEMO’s phased approach makes sense. It takes a while to get the implementation details right, just as it does for the market to adjust to broader participation, and for customers to take advantage of their new opportunities. This is why it is important that we start soon, rather than waiting until the next time it’s obvious that we’re paying over the odds due to the low level of customer participation.
Cheers,
Paul.
Excellent, thanks PaulT
Will think through the rest of what you say later, but want to ensure I understand the savings noted in the PJM capacity market, as that’s what really caught my attention!
Using the numbers you have quoted, and an installed capacity in PJM of about 180,000MW (think that’s about right – should be right enough to see that I understand the orders of magnitude):
Under the Reliability Pricing Model (RPM) we have two scenarios:
Scenario 1) Real = Total payment of $6.7b for a year across 180GW is $37,000/MW p.a.
Scenario 2) A re-calculated “no DR or EE” scenario = Total payment of $18.5b for a year across 180GW is $103,000MW p.a.
These numbers are reasonable – in the ballpark of what I’ve seen of the annual price for capacity in Western Australia, for instance.
Can I assume that there’s about 20,000MW of Demand Response and Energy Efficiency in PJM? I think it’s less than that but the number should be good enough to allow me to get my head around this.
Assuming 20,000MW of physical generation pushed out at the top of the bid stack by the DR and EE, this means that there was about that much that bid the difference between the two total revenues.
This means 20,000MW bidding, collectively, $11,827,280,831 – or $591,000/MW p.a.
Given it’s a capacity market, I assume this can be peaking capacity, but even if we assume capacity is built for $1,000,000/MW (expensive for a peaker, I would think, even with price escalation) this would mean that there are generation developers out there pitching for a 2 year payback on investment.
What am I missing?
Thanks
PaulMc
Hi Paul.
The maths works out a bit differently as it is not a pay as bid auction, everyone is paid at the clearing price.
The additional cost is spread over all 180 GW of generation, not just the 20 GW in question.
There will have been 20 GW of generation bidding between $37,000 and $103,000 /MW p.a., which changes the price paid to all cleared bidders, results in increased costs of 180,000 MW x (103,000 – 37,000) $/MW p.a. which is $11.8 billion p.a., making the 20 GW of DR pretty close to the mark.
The payback period would then be close to 20 years, depending on discount rates etc.
If those same 20 GW of generators were bidding at up to $591,000, then that would be the clearing rate, and total costs would be ~$120 billion
Thanks Scott
That makes more sense – apologies for not figuring it out myself.
Are bid prices made public, after the fact? I would be interested to look at the bid stack, to understand more …
160,000MW of physical plant to bid at or below $37,000/MW p.a. – at 100% utilisation this is only $4.22/MWh capital cost (at most), unless I have made another mistake? That seems way too low to cover fixed costs.
If this is true, then the approach seems to be to under-recover in the capacity market and recover more in energy (unless the entire fleet is depreciated away)?
{please excuse the questions, if they are naive – we don’t have a capacity market to work with in the NEM}
Paul
Hello Paul, generators don’t run at 100% utilization, but the capacity revenues are still a small portion of their revenues compared to the energy market revenues. For DR, however, which is only called a few hours a year, capacity is the vast majority of its payment stream.
Here is a blog I wrote on the FERC 745 ruling and the recent PJM auction: http://www.navigantresearch.com/blog/a-dark-day-for-demand-response
Hi Paul.
Unfortunately, PJM doesn’t release bid data, even after the fact. I believe this may have something to do with the fact that their auctions take place over several rounds, with participants able to adjust their bids at several fixed times.
In relation to your comments about Capacity revenue matching up with fixed costs and energy revenue matching up with variable costs, I don’t think many generators will approach it from that perspective.
A peaking generator will earn exactly as much capacity revenue as a baseload generator rated at the same wattage, but will actually earn less energy revenue, because it is only on and earning energy revenue a fraction of the time, whereas baseload is earning energy revenue during low price periods as well as high price periods.
Peaking generators tend to have lower fixed costs than variable generators, but higher running costs.
So a peaking generator, with its low fixed costs, high variable costs and low energy revenue may end up approximately covering its fixed costs with its capacity revenue, and then the energy revenue is profit on top.
A baseload generator however, has higher fixed costs but the same capacity revenue – they likely won’t come close to break even there, and they don’t particularly intend or need to. The energy market is (as Brett points out) where they make the big money.
DR, which earns energy revenue a lower fraction of the year than a peaker is even more reliant on capacity revenue (again, Brett got to this point before I did)
Fascinating to cross-reference your post with the estimated contribution from solar (http://pv-map.apvi.org.au/), and the demand charts on AEMO’s website. In Queensland on the 17th solar contributed about 500 MW at midday, or about 50% of the solar installed capacity, and there was a strong slump in demand co-incident with the solar contribution. However in Victoria a cloudy day caused a maximum solar contribution of 15%, and there was little midday slump in power demand.
On Sunday and Monday night this week there were strong winds overnight at a time of low power demand, and data from AEMO showed a South Australian price of $20/MWh compared with about $30/MWh in Victoria. About 100 MW were being exported from SA. Thermal power was still being generated in SA from Osborne (gas cogeneration, with heat sales to a nearby factory), Pelican Point (gas combined cycle) and Northern (coal). Each of these were generating about 160 MW. Some questions
1. Why is there such a large price differential of about $10/MWh between SA and Vic ? (The same occurs when SA is importing from Vic, except the SA price is $10/MWh higher than Vic). Yet much larger flows of energy occur between Vic, NSW and Qld on much smaller price differentials. Is SA’s Electranet using its market power (and wind power generators who have no-one else to sell to) to extract larger price differentials than in the eastern States ?
2. Is this pricing behaviour what was intended in the NEM ?
3. Why would the gas power station at Pelican Point continue to generate into such a low-priced market ? If it curtailed its production to below 60 MW, so there was net import into SA, then the SA market price would rise to parity with Victoria plus the $10/MWh margin Electranet build in. So by reducing production from 160 to 60 MW the owners could increase the SA price from $20/MWh to $40/MWh, and only pay for one third as much gas.
Hi Malcolm
Thanks for the questions.
Some quick answers at the end of the day:
Q1) You ask “Why is there such a large price differential of about $10/MWh between SA and Vic ?”
Q2) You ask “Is this pricing behaviour what was intended in the NEM ?”
Q3) You ask “Why would the gas power station at Pelican Point continue to generate into such a low-priced market ?”
By the way, ElectraNet is the transmission company operating in SA (see our Power Supply Schematic) so does not have the involvement in the market you’re thinking of.
If this does not fully answer your questions, maybe next week I can come back to this…
Paul
Paul, have you taken into account that for most of the first week of July there has been a transmission outage constraining Qld-NSW flows on QNI?
Hi James
Thanks for highlighting one more factor. As noted above, there’s enough there for another post at some stage.
Paul
PS James
Have posted this chart here to illustrate your point.
Thanks
Paul
Most of the volatility was in the South Australian market, which has the highest penetration of wind capacity, and has seen the greatest off-lining of thermal capacity. We could expect this volatility to reduce once the Heywood inter-connector is upgraded, and provide another 190 MW of transfer capacity to a much larger market. But this depends on whether the extra capacity is used to provide a relatively constant import or 190 MW, or whether the capacity is kept to smooth out the price peaks. It can’t do both at the same time.
To me the surprises in these data are
– why do the thermal generators (such as Torrens Island) not seem to be using the forecasts of wind power production to set the number of generation units kept on the boil ?
– do the generator operators not trust the forecasts ?
– spikes in the Tasmanian market should be easy to smooth out with some experience in managing quick-response stations such as Reece, unless the bidders want to intentionally generate price spikes
The next point of interest in volatility will be when the LNG exporters start bidding at the end of the year and gas prices start to climb. It would make it much more expensive to keep generation units at Torrens Island boiling ready for price spikes.
Some numbers I have seen indicate there is a lot of dumping of gas, with generation from the Darling Downs up significant despite warmer winter. I wonder if the other combined cycle plants are lifting production in Queensland. Is this not short term at least until the LNG plants clear the surplus.
Thanks, Ian
Yesterday there was certainly a sizeable volume of gas generation in Queensland – as I noted.
What I noticed, but did not include in the post, was that the spot price for gas in the Brisbane hub was below $2/GJ – and again today ($1.88/GJ) which provides one reason why it’s being burnt in volume. See GasWatch for more.
Paul
It would be interesting to see this look at teh bid data repeated for the 17 July repeal date. My hypothesis would be the muted response from 1 July could be indicative of mixed sentiment around likelihood of the repeal bills passing the senate. However, once made certain there should be a “second wave” of bidding in the lower SRMC band.
Hi Wayne
Thanks for the comment/question. Others have asked questions along similar lines, offline. It’s on my “to do” list.
Paul
Thanks Paul,
Interesting. How much money did Infigen make on their dispatched plant? I assume they didn’t curtail any capacity after the 5 minute spike?
Will have to have a look sometime, Ariel
Why was there a price difference of $3/GJ between Victoria and Queensland prices, when there was adequate pipeline capacity to move gas between the 2 markets ? Could it be that the pipeline owners charge this to shift gas south over this distance ? Once the LNG export plants start operation, could we expect a similar $3/GJ difference to move gas north from Victoria to Queensland, so perhaps we might see a $9/GJ price in Queensland and $6/GJ in Victoria ?
‘This article has been written in an environment where (it seems to us) two opposing views of the future of the electricity market have been verbally jousting from ever-more-entrenched positions. As I have noted before, we strive to remain technology agnostic – focusing, rather, of the market impacts of generation, whatever the source ‘.
It’s rather quaint that you don’t take a stand on the use of coal for electricity generation. It flies in the face of science as we know it surely.At least 2/3 of it has to stay in the ground for goodness sake.
Thanks Neil
I appreciate that that’s your perspective – as it is for many others as well. Our focus is on how the market is operating, under whatever the rules are at present.
Paul
“How much of this cost should be paid now, and how much later – based on assessments that technology costs will decline, and calculations about net present value using a variety of discount rates?”
If you take a discount rate larger than your rate of growth the right time to solve a problem will always be right before it’s about to happen. This is the problem with economists like Bjorn Lomborg saying that we should be dealing with the problem later. The size of the issue is too large for any one generation to deal with alone, the costs have to be spread out over many decades.
Also note that technology costs are declining because of scaling not research. Solar and wind technology is virtually unchanged since the 80s. Some improvements have been made with research, but most of the improvements are due to building a bigger factory, getting normal financing rates, and experience with installation.
Thanks, Tom
At the high end of the scale some might want to use commercial target rates of return – which would tend to favour deferral. At the other end of the scale, my vague recollection is the Stern report used a discount rate of 0?
Just one part of “a diabolical problem”.
Paul
Paul
Outstanding work. I commend the approach of always looking for an agnostic position and (although I am inherently biased) I have strived to remind readers that the energy industry is not evil per se. We need each other and your point on “what happens if” is most valid; the merit order effect is not perpetual! Good on you for sharing your incredible depth of knowledge to this important debate.
The issue at hand (IMHO) is quite simply are the non renewable sector (and consumers) simply being “gamed” by the incumbents because they can?
Clearly, the entire energy sector needs a tweak and a tune but in the meantime, its hard to know which mechanic to believe.
Thanks Nigel
Nice to hear that you find it useful.
Paul
Great outline of the issues. I would just note that the NEO is not adequate to define what Australia needs from its electricity sector. That’s not to say the NEO should necessarily be expanded to include carbon reduction goals but that we need to recognise that carbon reduction is central to electricity policy and therefore can’t be ignored if we want long-term policy stability. The question that needs to be answered, or at least discussed more fully, is – how much carbon reduction do we need from electricity? And following from that – when, how, and who pays?
Some useful references on the “social cost of carbon” (SCC) are:
– The US government’s interagency working group, which ended up recommending four different price paths in recognition of the debates around discount rates, quantification of impacts and scientific uncertainty. “Three values are based on the average SCC from three integrated assessment models (IAMs), at discount rates of 2.5, 3, and 5 percent. The fourth value, which represents the 95th percentile SCC estimate across all three models at a 3 percent discount rate, is included to represent higher-than-expected impacts from temperature change further out in the tails of the SCC distribution.” The average 3% price is generally used as a central estimate. See also the US EPA’s page that looks at how the SCC is incorporated into regulatory impact analysis. http://www.epa.gov/climatechange/EPAactivities/economics/scc.html
– “Omitted Damages: What’s missing from the social cost of carbon” which provides more accessible information on what’s in and what’s out of the integrated assessment models used to calculate the SCC. http://costofcarbon.org/files/Omitted_Damages_Whats_Missing_From_the_Social_Cost_of_Carbon.pdf
– closer to home, the Climate Change Authority’s report on Australia’s carbon budget and emission reduction targets. This gives a thorough explanation of the long term emission reductions needed internationally and nationally, the role of the power sector in contributing to this, and a carbon price path that can deliver appropriately. http://climatechangeauthority.gov.au/reviews/targets-and-progress-review-3
Thanks Olivia
In case you were not aware, Keith Orchison has picked up on your comments here.
I’d add that these 6 enduring challenges still remain, though perhaps the first one now looks like “generation capacity retirement” instead of development, especially if the RET continues in force.
Paul
Great piece. Just one correction the Earthquake and Tsunami in Japan was 11 March 2011 not in 2004. 2004 was the Indian Ocean Tsunami which occurred 26th December 2004.
Paul
Thanks for the pick-up, Paul. Have updated above to avoid further confusion.
Hi Paul,
A couple of points. There is much discussion in this piece about spot prices but what is the financial impact of forward contracts? The discussion on the Merit Order Effect and the “very polished” animation seems to ignore them.
The “2030 onwards” discussion about the AEMO report, where indeed they suggested that 100% renewables was technically possible, but it came at considerable cost – between $219 to $332 billion plus land cost. I published a piece on this which might be of interest to your readers:
http://www.energyinachangingclimate.info/100 Per Cent Renewables Study Needs a Makeover.pdf
Sorry try this URL
http://www.energyinachangingclimate.info/100%20Per%20Cent%20Renewables%20Study%20Needs%20a%20Makeover.pdf
Thanks Martin
Some believe that the spot price will lead the contract price – and some believe the opposite. Reality is somewhere in between. Oversupply of capacity, and continued decline in demand, will drive both down – until the supply side responds (or new methods of demand – like EVs – are introduced).
I listed the AEMO 2030 study because someone else would have. Did not get into costs – thanks for your link.
Paul
Hi Martin,
I notice many have made mention of the potentially large additional land cost of the 100% renewable study from AEMO. Most of the land requirements from this study is related to wind generation. The leasing cost of the land for wind generation is tiny compared to capex costs for the turbine. ie, ~$10k per turbine per year leasing cost, compared to the ~$5 million capex cost of the turbine. It is pretty much in the noise band in terms of cost.
I suspect this will be true of most of the renewable technologies.
Further to David’s point:
The cost of leasing is part of the LRMC, it’s not an externalised cost. Also the turbines and roads take up a tiny fraction of the total land area.
Even for large scale solar the cost of land is minor.
Paul,
Note that AGL are the participant for, but not necessarily the owner of those wind farms.
Cheers,
Tom
Thanks Tom
There are a number of different ways of assigning companies to generation assets – NEM-Review has three:
1) Owner – in many cases not the developer, or the company trading in the market. Some are just infrastructure investment funds. Our “Power Supply Schematic” Market Map also highlights these.
2) Bidder – which I used yesterday – takes account of the entity interacting with AEMO, and in most cases is the owner of the energy generated. But it excludes Non-Scheduled.
3) Trader – which I used here.
I understand that slicing the numbers different ways provides different results. The core finding today is that AGL is not insignificant in contributing to growing wind’s market share.
Paul
Paul, this is a result of government legislation not goodwill on the part of AGL. Their current opposition to the RET highlights this. No RET = no “AGL” windfarms. Development of them and on-sale also provided a great short term earnings hole plug via development fees. Maybe you could do an analysis of just how far out of the money those deals were – with the PPA prices published so inflated, seemingly to create up front development fees that they are now suffering “losses” on their green products – refer this weeks profit waterfall diagram….
Just for your interest; AGL is the third largest equity investor in Australian renewable energy. Infigen is top with 1,500m., Meridian is a smidgen above AGL both with about 600m. and there are another dozen players with 100m. to 500m. invested. Note that none of the other vertically integrated companies are on the list. (Bloomberg collated data)
Tom is absolutely right, most of their portfolio they developed and built but sold and signed a PPA with the current owner. Their suggested portfolio is nothing other than “greenwashing”.
There are plenty of operators out there who own more wind farms than AGL, this is just a nonsense look at what AGL has contracted. AGL sold out of its wind farms at huge profit, and now they’re trying to destroy the RET. What a joke of an article – it really reduces the credibility of Global Roam as a modelling house. Maybe disclose how much money AGL have paid to you in the last 12 months and how much you hope to get out of them in the next 12 months next time you post your next sycophantic article that attempts to gloss over AGL’s terrible renewable credentials in recent months.
Hi Stirling
Thanks for the comments. You seem to make 4 claims:
1) That asset ownership, not contracted position, is a truer measure of greenness
I understand this perspective. There are a number of different ways at looking at who’s greener than others – each has some value, so long as the writer, and the reader, understands the measures used, as I noted here. In this post, I just chose another measure.
2) That there are larger portfolios (in terms of ownership)
Some assets, for which AGL still “owns” the market risk (for both electricity and RECs) have been sold to infrastructure funds. This is not uncommon – for wind farms, or other forms of generation as well.
We’re currently finalising the 2014 issue “Power Supply Schematic” Market Map, which will enable us to collate up totals owned. Perhaps this can be another post at some stage in the future.
Happy to have a look at your data, if you can share with us (tel 07 3368 4064)?
3) That GLOBAL-ROAM Pty Ltd is a modelling house
We’re primarily a software company, not a modelling shop. Our website explains more. I think you have us confused with ROAM Consulting Pty Ltd, a completely separate company (see here for some background).
4) That this article was “sponsored” in some way by AGL Energy
This is not the case. In total over the past 12 months, AGL has paid us the royal sum of Aud$1,600 (exGST). I do not expect the following 12 months to be any different.
Regardless of what they (or any of the other 100+ clients we serve – green, black & indistinct) pay us, that will not influence what we post about on WattClarity.
Does this help?
Paul
Yeah OK I got owned.
Interesting article. Yes those ownership stats are based on Bloomberg New Energy Finance data, published in a recent note we released to clients on exposure to assets that could potentially be stranded by changes to the RET. We track ownership and investment in renewable assets worldwide. For more details you can have a look at http://www.bnef.com or contact us at sales.bnef@bloomberg.net if you’d like to know more.
Thanks Kobad & Tom
It’s been pointed out that the BNEF charts on equity & debt providers are here on RenewEconomy, for those who want to access them
Paul
Hi Paul
Chris, Ben and Waldo from Powershop and Meridian here.
As you know we are committed to protecting the current level of the RET. You can see many of our comments here.
GetUp and its members have a similar commitment to the RET, and went through a process to identify retailers with an aligned point of view. They didn’t find many.
GreenPeace’s guide to which you refer, and Giles Parkinson’s RenewEconomy article , show that there are a number of retailers out there who (regardless of what their marketing teams say about their renewable credentials) work hard at undermining the RET and the renewables industry. Even the more “renewable” amongst the cohort give at most faint praise to the RET, with none coming out hard and unequivocally calling for the RET to be left untouched.
We have a commercial arrangement with GetUp!, where we pay a stipulated fee to GetUp! for every customer that signs up to Powershop under this campaign, and this arrangement is explicitly disclosed by us at http://www.powershop.com.au/p/betterpower and by GetUp! at https://www.getup.org.au/campaigns/renewable-energy/switch/victorians-switch-to-save-renewables.
We, and GetUp!, are encouraging customers who favour renewable energy to switch away from Big Three to send an unequivocal message of support for the RET. Whether they switch to us, Diamond Energy, or one of the other renewable retailers highlighted in the Greenpeace guide is not the point.
We understand that these payments made to GetUp! are going exclusively towards growing the Better Power campaign to save the Renewable Energy Target. The “owners” of GetUp! are the thousands of members who are participating in the campaign, and it’s a not for profit organisation that campaigns to support causes that its members believe in.
It’s no good to buy renewable energy (GreenPower for example) from a company who is looking to undermine or weaken the renewable energy industry in Australia. It may make you feel better, but it won’t help the future of renewables – because GreenPower only works to support renewables in an environment where renewables are supported by policies like the RET.
Hi Chris, Ben and Waldo
Thanks for the detailed comment. Helps me to understand the nature of the commercial relationship between Powershop and GetUp.
I think I understand the reasoning behind your comment:
i.e. buying Greenpower now from someone achieves a 1x benefit (i.e. you buy a kWh you get a kWh) however you’re saying that you buy Greenpower from someone who advocates for the RET as well and there’s a multiplier in there (in terms of supporting growth in renewables beyond just the kWh you pay for).
My sense is that everyone will have their own personal perspective there – cost per kWh of green would be one consideration for many I suspect.
Regards
Paul
Not to mention that the big three retailers charge $60 for an LGC when the going rate is under $30…
We don’t know how much CO2 is emitted because NEM generators don’t measure or estimate with sufficient accuracy and precision at sufficiently fine time intervals (e.g 15 minute intervals as in Europe) the emissions data from the fossil fuel power stations. The AEMO figures are based on long term averages and applied to all units in a power station. They make no allowance for the effects of cycling, spinning reserve, start up and shut down, or operating at less than optimum efficiency.
This is how Ireland does it for comparison: http://www.eirgrid.com/operations/systemperformancedata/co2intensity/
[The methodology that used to be described in a succinct, clear paragraph has been removed. I’ve asked for it to be reinstated.]
Thanks Peter
It’s my understanding that the data submitted by the generators to the Department of Environment as part of their reporting requirements is indeed accurate (using international best practice standard).
This information is not published by AEMO – which, as you know, is a separate organisation focused on electricity as distinct from emissions.
Data provided in NEM-Review is an approximation based on emissions intensity by station – which, as you say, does not take account of variation by load factor, etc… However it is useful enough for these high-level comparison.
Paul
Paul,
Thank you for your reply. But my understanding is your understanding is not correct in that Australia does not measure emissions per unit (as the USA EPA requirements [1]. [2]), nor estimate them as is done in EU and as per the example I gave with the EirGrid web site I linked in previous comment. You can see similar web sites for other EU countries too.
This is why we have only a limited understanding about how much CO2 emissions the intermittent renewable energy generators are achieving. On grids where they have reliable information, such as Ireland which has many important similarities to South Australia, peer reviewed analyses show that wind generation was only about 53% effective at abating CO2 emissions per MWh of wind energy supplied to the grid. That is far less than is claimed by the wind energy advocates.
[1] US EPA “ECMPS Reporting Instructions Emissions” http://www.epa.gov/airmarkets/business/ecmps/docs/ECMPSEMRI2009Q2.pdf
[2] US EPA “Plain English Guide to the Part 75 Rule” http://www.epa.gov/airmarkt/emissions/docs/plain_english_guide_par75_final_rule.pdf
I received this by email from EirGrid overnight:
“Thank you for drawing out attention to this. We will ensure that the website is updated in due course with the missing text, which is:
EirGrid, with the support of the Sustainable Energy Authority of Ireland, has together developed the following methodology for calculating CO2 emissions. The rate of carbon emissions is calculated in real time by using the generators MW output, the individual heat rate curves for each power station and the calorific values for each type of fuel used. The heat rate curves are used to determine the efficiency at which a generator burns fuel at any given time. The fuel calorific values are then used to calculate the rate of carbon emissions for the fuel being burned by the generator. ”
Australia has no processes that go anywhere near world best practice for calculating or measuring (as the USA EPA requires of US generators), CO2 emissions.
Peter Lang, in the end this all matters very little because the National Greenhouse and Energy Reporting System requires measurement of actual coal consumed and sampling analysis of the chemical composition of the coal coming from the field in which it was taken. This provides us with a quite accurate measure of the carbon burnt in total over the period of the year and therefore CO2. At the same time AEMO metering measures the amount of electricity sent out to the grid by the power station. So on an annual basis even if not on a fifteen minute by fifteen minute basis we have a very good understanding of the total annual CO2 emitted and total GWh generated and the average emissions intensity. This then gives us a very good hint as to whether there might be a material deterioration of efficiency at work over time. If this were the case then we might then want to investigate what the cause might be whether it be increased variability of output due to say wind farms (I know what you’re getting at Peter) or for example a change in water content of the coal or some problem with spoiling or a range of other factors.
Oh, I see, so we’ve called in the renewable energy advocates to help out on this so called technology site. Confirms what has been evident for a long time.
Tristan Ellis (Editor of ‘ClimateSpectator’ and a passionate renewable energy advocate), you are wrong on this as has been pointed out many times in the past. Measuring electricity output from each individual generating unit at 15 minute interval and getting a very rough estimate (contrary to your claim) of carbon burnt in power stations over years does not allow the emissions avoided by intermittent renewable energy generators to be determined.
If such rough estimates are good enough why to the US EPA require measurements of emissions at 15 minute intervals along with calibration of the instruments measuring the emissions and reporting of these to the EPA? Read the links I provided above.
And why does Ireland (an example of EU method) calculate emissions at 15 minute intervals as per the EirGrid methodology quoted above.
Also as stated above, EirGrid’s emissions data allow calculation of the emissions avoided by wind generation as just 53% of what they would be if 1 MWh of wind energy avoided the emissions of 1 MWh of the energy displaced from the other generators. It is likely that South Australia’s grid is even less effective than EirGrid.
And this information is not new or rare. The analyses that have been done based on empirical data (as distinct from modelling studies) show similar results.
They also show that the higher the penetration of wind energy the lower is the abatement effectiveness.
What all this means is the CO2 abatement cost with wind energy (and all intermittent renewable energy generators, is much higher (e.g. a factor of ~2 higher) than the renewable energy proponents would have us believe.
Renewable energy is high cost for no benefit – i.e. no benefit in terms of avoiding climate damages, which is the primary justification for subsidising and mandating them.
I’m baffled by your logic, Peter
Because Tristan has decided to post a comment on WattClarity, it seems that you have inferred (without any other evidence) I asked him to post and ‘gang up on you’.
Yet you, yourself, have commented a number of times on this post – and you know that I did not ask you to do that. Given I did not ask you, and yet you still posted, I can only surmise that you believe you made your comments because you are my Aunty?
Furthermore I would conclude, following the same reasoning, that because Giles has chosen not to comment he must indeed be my Grandfather
Can I remind you of your prior comments posted only last night – that you later requested I delete when the truth became self-evident? This was another example of you jumping to conclusions in attributing motive
As I have noted before, we have a broad range of readers (including yourself, and Tristan) who choose to comment (or not) of their own volition. As you have seen, most choose not to – we don’t mind, either way.
Once again I remind you that we are carbon (and technology) agnostic. We post about it occasionally as it is of interest to some of our readers – including, it would seem, you.
In terms of my article, above, it is not clear to mewhat value estimations of emissions at 15-minute granularity would achieve in this case?
Leaving aside the inherent difficulties of actually trying to measure emissions on any basis (and especially on a short-term basis), I am unsure how such measurements, when added together across each of the years shown above, would actually change the main conclusion in the article (that AGL’s larger portfolio produces significantly more emissions than any other portfolio)?
I have learnt from experience that you have your particular hobby horses that you like to repeat (on WattClarity and other sites) ad infinitum, but it does seem a little off the topic – don’t you think?
Yes, the emissions calculations uses in NEM-Review are approximations, as we clearly note here.
Any power station performance engineer will be able to explain why any emissions metric will not be 100% accurate. Calculating estimates on a 15 minute basis such as what you suggest would offer some advantages, but also carry some disadvantages.
As noted above, NEM-Review is fit for purpose:
(a) it helps a range of clients compare trajectories of estimated emissions over time, and works with information in the public domain;
(b) it helps draw out some key points for readers of WattClarity, such as in this post.
If you want to suggest to Minister Hunt that there are better methods for measurement & reporting of emissions under NGERS, I’m sure he’ll be all ears. You can reach him here.
If you want to suggest to AEMO that the estimates they publish could be improved, you can find them here.
I know that some people at AEMO do read WattClarity (though they are probably not the same people dealing with their emissions approximations). I could almost guarantee that Greg Hunt does not read WattClarity. Hence, it’s probably the wrong forum for you to be promoting such suggestions.
If we ever did come to analyse & post about the emissions avoided by any particular generation source then we might continue this discussion.
… but for tonight I’ll leave it here and (instead) think about how I would have otherwise worked out the winning lottery numbers if I had not been investing the same time reviewing off-topic comments on WattClarity.
Paul,
I don’t understand the purpose of that long rant. If you didn’t call up Tristan or ‘Anon NGERS worker’ to reply why didn’t you just say so? Instead you went into a long, mostly irrelevant, rant, as has happened before in response to my comments.
You say WattClarity is technology neutral. Yet it seems pretty clear, from you’re responses and the balance of posts over time, that you are an advocate and supporter of renewable energy. The fact you frequently jump in quickly to attempt to dismiss comments and papers that point out the poor economics, high abatement cost and lack of demonstrable ability to achieve the main justification for subsidising and mandating them also suggests pro renewable energy bias – i.e. not technology neutral after all.
Your comments in the past responding to the low emissions abatement effectiveness of wind generation (and therefore high abatement cost) suggest you tend to jump in quickly on this issue and make comments that show this is not your rea of expertise. Such comments require a response.
“it is not clear to me what value estimations of emissions at 15-minute granularity would achieve”
In that case I’d urge you take some time to read and study the literature on the empirical studies of abatement effectiveness of intermittent renewable energy technologies – or don’t comment on it at all on the subject. You might like to start by studying the Joe Wheatley paper “Quantifying CO2 savings from wind power” because there are relevant comparisons with the South Australian grid http://www.sciencedirect.com/science/article/pii/S0301421513007829 )
Stating you are carbon and technology neutral may be your intention but you cannot prevent your own biases showing through, just as Tristan and Giles Parkinson can’t help their biases showing through.
PS, Peter and others,
If you wanted to comment to AEMO about their processes for publishing their estimated CO2 emissons figures, then this opportunity has arisen that would might want to contribute to.
Submissions due 9th October, so would be ideal if we could help you through the power station course before then!
Paul
Gents, I’ve done some work on the marginal abatement I’d like to share with you. In a nutshell I found that due to interconnection, most of the abatement from wind farms tends to be on NSW black coal generators. This is due to the market size and the fact that NSW coal has a higher marginal cost.
Renewable deployment actually displaced the least amount of CO2 in QLD due to cleaner fuel and modern plant (0.92 kg/kWh). TAS and VIC had the highest amount of abatement at 0.98 kg/kWh due to the market power of brown coal in those regions. SA and NSW were in between.
This may not seem intuitive but it does line up with some of the fuel consumption info that AEMO or Reputex put out.
In response to comments above here’s how NGERS works:
Power stations report annually to Government under the National Greenhouse and Energy Reporting Scheme (NGERS).
Power stations must estimates emissions and energy consumption and production in accordance with the NGER (Measurement) Determination. All medium and large coal and gas power stations estimate CO2 emissions using “method 2 or 3” which involves sampling and analysis of their fuel inputs to determine emission factors. This method results in an accurate assessment of emissions with relatively low uncertainty. Reporting is subject to the NGER framework which includes a system of audit and the publication of certain data (including additional information for designated generation facilities).
Data reported by power stations are incorporated into Australia’s National Greenhouse Accounts and are used to compile scope 2 emission factors which reflect the average emissions intensity of electricity consumed in each state and territory (published annually in the National Greenhouse Accounts Factors).
“which involves sampling and analysis of their fuel inputs to determine emission factors. This method results in an accurate assessment of emissions with relatively low uncertainty.”
What does “relatively low uncertainty” mean? Please quantify the uncertainty and state how it is calculated.
The CER figures do not take into account the variations in:
– C content of fuel being consumed at a point in time
– thermal efficiency of the different units operating
– thermal efficiency of the units depending on its output
– thermal efficiency of the units as their output is ramped up and down
Perhaps you could state whether all coal fired power stations are using method 1.4 (d) and at what interval are the emissions measured (if not all, what proportion by annual energy generated?):
” (d) method 4 provides for facility specific measurement of emissions by continuous or periodic emissions monitoring.”
http://www.comlaw.gov.au/Details/F2012C00472/Html/Text#_Toc330214226
The USA EPA requires the emissions to be measured at 15 minute intervals and the measurement instrumentation to be calibrated and all this to be reported to the EPA and publicly available. We have nothing like that.
Without emissions intensities from each unit, either measured as required in the USA or estimated based on measurements as listed above, assertions about how much CO2 is abated by wind generation are so uncertain as to be meaningless.
I look forward to your quantification of the uncertainty and explanation of the methodology for estimating the uncertainty. I trust you have reliable estimates for the uncertainties on the assay values of the coal consumed by each unit at 15 minute intervals.
Can you provide a link to where I can download the generation and CO2 emissions intensity figures for the NEM generators per 15 minutes (or per 1/2 h or 1 h) to enable the type of analysis reported in this paper:
http://www.sciencedirect.com/science/article/pii/S0301421513007829
A pre-submission version is available free here:
http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf
Peter,
The power station operators will have to take regular samples of coal quality so they know what the boiler will be experiencing. I don’t know about 15 minute frequency or data availability, but it wouldn’t be hard to report if you are already collecting the data.
p.s. I asked a friend working at coal power plant who said they sample pretty much constantly and at different points along the delivery path.
I have read with some interest and some amusement.
I would like to make a few comments.
Firstly nearly all the reports I have read regarding coal fired power stations and emissions are calculating or estimating their emissions based on the power sold (as shown in the AEMO 5 minute generation data) The fuel used and emissions calculated are estimations reverse calculated, using predetermined efficiencies and moisture content of the fuel. These emissions do not take into account the inefficiencies caused by idling generators to be available to back up wind farms or generator failures. Power stations refuse to publish actual fuel consumption in the same 5 minute time slots that the AEMO publish the generation data for every generator. They have this data but refuse to publish it due to the exposure it would bring on how inefficient they are being forced to run, not only due to backing up wind farms, but also covering line losses pushing power across the countyry. I am told the losses now from Adelaide to Melbourne exceed 15%. AEMO site shows on 4 July 2013 at 14:55, a fault dropped 336 Mw of Macarthur wind farm off line. It was instantly picked up by 334Mw of Black coal fired power station in NSW and Qld. Not a flicker of a light bulb and no blackouts. The coal was being burnt anyway and the load was picked up thousands of km away. Macarthur was not needed so in a net effect was not abating any GHG at all, and was in fact increasing the country’s emissions due to its maintenance and construction cost, and the losses in the power lines across the country to back it up. This backup of wind farm stoppages happens repeatedly, and the AEMO have confirmed that 500Mw and greater backups are needed many times a year.
If comparison is made from tonnes mined at power station coal mines, where this can be seen on annual reports, and compare it to Mw sold from the same power station, then the numbers are not even close to what they are reported in the theoretical reverse calculations used to report emissions. This discrepancy has been widening steadily over the past 6 years as more wind farm capacity is installed.
And finally, I talk to power station engineers and to contractors working on the sites. Engineers tell me that their coal feed does not reduce and increase at the same rate as MW sold, in some cases they can be burning 20% more coal than what the Mw being sold. A contractor working on a Black coal infeed conveyor told me how they were under pressure to keep the conveyors running during a maintenance job, the coal was continually fed into the generator boiler for the 4 or 5 hours of the job. Even though the coal feed remained constant, comparing AEMO generation data, the output nearly doubled then again halved over the period. A lot of coal was not being converted in to saleable electricity, it was wasted.
Opinions, thoughts and estimations are wasting so much time and money, we need the power generators to openly publish real time 5 minute coal feed data for every generator, so this can be matched against the AEMO 5 minute generation data, then we can all see what is happening in an open honest and truthful manner. Then we can clearly see how much coal is really being burnt, and how small the net emission savings by wind farm are.
Ah, Hamish (and Peter Lang)
Everyone loves a good conspiracy theory.
Problem is, most of them do turn out to be myths (though I am sure that Elvis is still out there, somewhere).
Whilst we could go to-and-fro electronically for a long time, and we might achieve a raised level of understanding eventually, it would not be a very efficient use of my time, or yours. Instead, I have another suggestion:
For a number of years, QUT has run an excellent Introduction to Power Plant Course that has helped to lift the veil on the magic that happens behind the closed doors of a power station.
Both you, and Peter Lang, do seem to have made considerable effort to start from nothing and learn what you can about how thermal coal-fired power stations. Such a course would build on this base to help you to understand how one actually works.
In the interests of saving my time on WattClarity, and helping you both up the learning curve, I would be happy for our company (GLOBAL-ROAM Pty Ltd) to contribute $1000 each to the cost of you attending this course. Perhaps some of our other readers would like to chip in as well?
However note that:
Please advise if you will accept this offer to assist in your education?
Paul
Hamish,
Whoever you are talking to is bullshitting you.
1. There are no formal reserves as backup for wind. Trips on thermal generators are far more common (see Liddell, Eraring) and yet still no problem with security. Also note that SA has no new fossil generators despite massive WF capacity. WF output can reach 90% of load served.
2. Transmission losses are usually 2-5%. If they were 15% it would be economically attractive to build power plants in the CBD.
3. Fossil plant burn less fuel at lower capacity. From thermo law 1: energy in = energy out for a system with constant internal energy. If the plant was burning the same amount of fuel for a lower output they would need to actively cool the boiler to prevent damage (i.e. increasing internal energy). It is much easier to feed less fuel in (not to mention cheaper). The efficiency at min load is similar to that at 100%. Peak efficiency is at ~80%. Note that even if efficiency drops 1% when you halve your output, you are still burning about half as much fuel. Thus if a WF’s output forces a coal plant to turn down (nearly always the case) you still get 98-99% of the savings.
Hi Tom
Regarding your points:
1) About reserves
2) About losses
3) About heat rate
Finally, Hamish, one of the main reasons some generators may not want to have such commercial details as coal burn & heat rate published in real time is because of the highly sensitive commercial in confidence nature of the information. If their competitors knew this type of information, they might be able to utilise it for commercial gain (and their detriment) in the market (in our software development business, we’d certainly be looking to further enhance ez2view to help them do this, if possible!)
Hope this helps?
Paul
Paul,
Wow! What a pathetic, arrogant, gratuitous response. It demonstrates many of the signs of intellectual dishonesty (more below)
So Paul McArble is resorted to offering ‘shut up and go away money’ and for it to be paid by ROAM Consulting.
This smacks of similarities to corrupt union behaviour. What’s next, banned from any further comments? That’s how his evident allies, Tristan Edis and Giles Parkinson (name first mentioned here by Paul McArdle, not me), prevent comments they don’t want rather than admit they are wrong on an issue. Anonymous commenters also appear on Tristan Edis’s politically partisan, renewable advocacy web site to defend WattClarity. So there is an apparent allegiance between these passionate (but irrational) renewable energy activists.
No 4 of the “10 signs of intellectual dishonesty” is:
“4. Avoiding/Ignoring the question”
http://judithcurry.com/2013/04/20/10-signs-of-intellectual-honesty/
Instead of admitting you made baseless assertions and you cannot support them you try to bluff your way around it. You’ve demonstrated this type of response on previous occasions too when you made comments that showed you don’t know what you are talking about on the issue of how Australia’s CO2 emissions are estimated and the uncertainties in the estimates.
You also admitted, on this thread, you don’t understand why measurements at time intervals of 15 minutes (or 30 minutes maximum) are relevant for estimating the CO2 emissions avoided by intermittent renewable energy generation. [I pointed you, again, one of many papers which shows why very clearly; apparently you never read it or didn’t understand it, or any of the others]. And you seem to go into a state of denial whenever it is pointed out that the empirical evidence from other grids where such information is available (on CO2 emissions per individual unit per 15 minutes taking into account the thermal efficiency at the level of output for that particular unit at that time) shows that wind generation is much less effective than the proponents claim at abating CO2 emissions.
One such study shows that Ireland’s EirGrid was just 53% effective at abating CO2 emissions in 2011 (Wind generation supplied 17% of electricity and many relevant features of EirGrid are similar to South Australia’s grid which is held up by the wind energy advocates and activists as the poster child for renewable energy in Australia.
That means that CO2 abatement using wind energy is about twice as expensive as the proponents would have us believe. That is relevant. Why are Paul McArdle, Tristan Edis and Giles Parkinson trying t avoid the issue. Why are they trying to bury it? Why is Paul McArdle offering ‘shut up and go away’ money – which is to be paid by ROAM Consulting.
Why don’t you just answer the questions and provide links to the 15 minute data, the actual consumption and the assay values of the coal entering the burners at 15 or 30 minute intervals? We know the reason – they don’t exist. You know it too and are trying to doge admitting you are wrong and made baseless assertions up thread.
Avoiding/Ignoring the question”
http://judithcurry.com/2013/04/20/10-signs-of-intellectual-honesty/
Is just one of the 10 signs of intellectual dishonesty displayed by Paul McArdle.
Peter
I’m sorry to see that (for the 3rd time in a few days) you have jumped to conclusions – this time in assuming some nefarious ulterior motive in a genuine offer to help.
It’s unfortunate that we cannot help you.
Paul
PS Hamish, we would still happily help you.
Point one,
Your “offer” is not a genuine offer at all. It is an attempted diversion from answering the question to get you off the hook and appeal to your followers. If it had been genuine it would have been done quietly by email. Your comments are arrogant, insulting and rude. You haven’t a clue what I’ve been working on for the 40 years before retirement (including having been on and in the dams, tunnels and power stations of the Snowy Mountains Scheme since about 1952, aged about 6). Whereas you comment, and try to make out you speak with authority, on subjects you clearly know little about. They suggest you are young, inexperienced and gullible.
You still avoided the questions. They are relevant to the debate which followed from your post and my first comment “We don’t know how much CO2 is emitted because NEM generators don’t measure or estimate with sufficient accuracy and precision at sufficiently fine time intervals (e.g 15 minute intervals as in Europe) the emissions data from the fossil fuel power stations”
Which you really didn’t have to answer at all since you seem to know next to nothing about the subject.
As I pointed out, your comments and avoidance demonstrate intellectual dishonesty – e.g. “10 signs of intellectual dishonesty”, No 4 here: http://judithcurry.com/2013/04/20/10-signs-of-intellectual-honesty/
I don’t know why you, as the blog owner, just drop it – don’t respond at all on subjects you know next to nothing about.
At this point I am persuaded your personal, professional and intellectual integrity is not very high.
As I suspected, my response has gone straight to moderation.
Well, what can I say. You want to glide over facts, first hand from people on the ground (with no vested interests) and assume that no one but you has any knowledge or contacts that know more than you are prepared to admit. Some comments to correct your assumptions. The conveyor was feeding the actual boiler, not a stockpile, if you read my original comment closely. PowerStation engineers with 35 years service as I understand, with considerable first hand experience running the stations tell me excess stream is vented, to put it simply, so that capacity is available with live hot available steam at any time. You are probably unaware that I personally worked in facilities that provided standby and peaking power to the grid, and are probably unaware that I worked under contract for state and federal ministerial advisers on a couple of high level energy projects for their American, Japanese and Chinese friends. I could probably run a course that could teach you a thing or two.
As for “commercial in confidence” being the reason why they won’t supply the coal feed figures, you must be joking. Anyone in the industry would have a very accurate idea of what all other powers stations in the grid are capable of, and their efficiencies and capabilities are published in annual reports. There is no real excuse for not publishing live coal feed data, unless you are trying to hide reality.
And as for SA, they more often than not are backed up by Victorian Brown coal generators. With industrial and manufacturing downturn, Victoria has an enormous unused capacity available to be sent anywhere in the Eastern states. Even allowing for considerable losses.
Tom interesting that you said “Transmission losses are usually 2-5%. If they were 15% it would be economically attractive to build power plants in the CBD.” Firstly if you seriously believe transmission loss from Qld to SA is less than 5%, then I have little faith in your understanding of the NEM or the impact wind farms are having. The Singaporean company who own the 500Kva line that runs through to Portland, currently are looking at the maintenance or upgrade that is required, because it has hit the 15% l.oss figure. They are also costing and in early stages of planning permissions, to construct a line beside the existing one with twice the capacity. The drawings have already been done for the power station to be built at Portland as well. And it is not a wind farm and it is not fossil fuel powered. If you do a search on the speech made a couple of years ago by the then federal Labour energy minister, you will see that the Federal Government stated then, that when the country finally realises that wind will not work, then they can build the new power station and all the Green voters will accept it, and we will have built the grid and interconnectors on the pretence of wind saving the world, to enable the power station to run the country. However today I was told that it may not be built in Portland now that there is a mothballed desalination plant in Gippsland that just happens to be a good size fit for the desalination plant required for the proposed power station. This significantly reduces the cost required to build the power station. Also for Melbourne, prevailing winds and weather would be safer having the power station next to the desalination plant, rather than at Portland. I am afraid that wind farms are just a distraction for the masses, so the government can implement the goal with less fuss. Not a conspiracy, not a work of fiction, the drawings have been complete for some time now, and both state and federal planners are fully briefed on the plan. If you personally know people in these departments as I do, then they can confirm this to you. If you do not know people in these departments, and have not seen the drawings, then you have no right to comment on it nor cast doubt on the credibility of those of us that have. I do not plan to make any further comments on this matter.
Cycle steam is not vented except in emergencies. Demin water costs too much.
I agree, SA WFs are often backed up by VIC brown coal. While the coal plant is run at a lower capacity it is consuming significantly less coal. That’s the whole point.
Wow, normally there are harldy any comments on here but for some reason this post has released not one but two nutters!
Thanks for a really helpful explanation of the issues underpinning some of the warnings we’ve been hearing about supply security. The slow build-up of a systemic security risk resulting from skimpy maintenance is pretty convincing. And to me, as a representative of the “people focused on broader decarbonisation” it suggests that there’s a strong security argument for a clear policy driver of orderly plant retirement. Not that I’m clear on what that should look like, yet. But would be interested in your thoughts on how such a policy could be designed to best address security risks.
Thanks Olivia
Will return to your question, as time permits
Paul
Hi Paul…pointing out more of a typo than anything. In the first graphic, it is probably more accurate to say wind is suppressing scheduled generation not scheduled demand. If only demand was scheduled, it would be so much easier to supply!
Hi Liam
Thanks for the comment.
I was focusing more on Scheduled Demand (which NEM-Watch highlights) more than Scheduled Generation, in this instance – though the wording could have been clearer.
We see South Australia is coloured a deep blue, being down towards the bottom of the historical range of Scheduled Demand experienced in the region’s recent history.
Paul
Okay it’s back to the MMS refresher course for me. It’s not a typo, it’s
Got it now thanks Paul. Even so, “Scheduled Demand” (other than demand response)does border on oxymoron territory…
Why is there such a large difference in gas prices between Brisbane and Adelaide ? As your gas chart shows, it’s clearly not limited by pipeline capacity. Is this price differential what the pipeline owners charge for the use of their asset ?
Thank you for this. Interesting and much appreciated. I’d point out for the benefit of other readers that
1. US EPA requires power stations monitor GHG emissions from the exhaust/stack at 15 minute intervals and sum hourly for reporting (from memory). The instruments are required to be automatically calibrated at frequent intervals too (either every 15 minutes or hourly or similar, from memory).
http://www.epa.gov/airmarkets/business/ecmps/docs/ECMPSEMRI2009Q2.pdf
http://www.epa.gov/airmarkt/emissions/docs/plain_english_guide_par75_final_rule.pdf
2. EirGrid calculates CO2 emissions at 15 minute intervals. The total emissions per year reported to the EU and IEA by other government departments based on total fuel consumption and properties reconcile closely with the EirGrid calculations. The calculation method is described here: http://www.eirgrid.com/operations/systemperformancedata/co2emissions/ :
“EirGrid, with the support of the Sustainable Energy Authority of Ireland, has together developed the following methodology for calculating CO2 emissions. The rate of carbon emissions is calculated in real time by using the generators MW output, the individual heat rate curves for each power station and the calorific values for each type of fuel used. The heat rate curves are used to determine the efficiency at which a generator burns fuel at any given time. The fuel calorific values are then used to calculate the rate of carbon emissions for the fuel being burned by the generator. ”
This paper is excellent and shows how to estimate emissions saved by wind generation for grids that have reasonably reliable emissions estimates at 15 minute intervals: http://www.sciencedirect.com/science/article/pii/S0301421513007829
Here is a (free) pre-submission version: http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf
3. As far as I am aware, neither AEMO nor CER have estimates that are even close to the time spacing (e.g. 15 minutes), precision and accuracy needed to make a reliable estimate of the emissions saved by intermittent generators such as wind and solar power. Therefore, we don’t know how much emissions are saved by wind and solar power in the NEM. Our best guess is that the effectiveness of wind power in the NEM would be roughly equivalent to wind power in EirGrid – i.e. about 53% effective at 17% energy penetration (proportion of electricity generated by wind power). That means 1 MWh of wind power avoids just 53% of the emissions from 1 MWh of electricity generated by EirGrid without wind power. This is important because it means the CO2 abatement cost with wind power is around twice what is normally claimed by the wind power proponents.
4. Of course, if we took a different approach to reducing GHG emissions, there’d be no need for rigorous, precise, accurate emissions monitoring to the standards needed for commerce in CO2. The compliance cost of emisisons monitoring will become huge if we go down this route. There is a better way. Just fix the problem. The USA and IAEA could do that by removing the impediments that are making nuclear power far more expensive than it could and should be. As the costs come down nuclear would replace fossil fuels world wide with no need for carbon taxes, binding international agreements, police forces to force compliance, international trade disputes and emission monitoring.
Readers may be interested in next week’s Energy Users Association of Australia Annual Conference- New Energy Paradigm- Better Energy, Better Business. The program features an impressive range of speakers from the regulatory, research, retailing, networking, generating, policy and political arenas, with the focus on the new energy paradigm and how best to manage it, including the pivotal issues critical to Australia’s large energy users as you describe in your article.
New Energy Paradigm- Better Energy, Better Business will be held at Crown Melbourne on 13 and 14 October. For further information or to secure your space, please visit http://www.euaaevents.com.au
I agree entirely with what Michael says about needing to know how your company uses energy is critical to reducing the costs of energy and finding the various tools to be better able to identify ways to limit the costs incurred – so much of the energy cost paid by users is for someone else to provide the resources for what they think your firm needs and to manage the risks involved in sourcing and delivering the energy your firm needs.
What is just as important is making sure that the tools are available to allow each energy user to implement the various approaches noted by Michael. The work and expertise provided by Michael is possible because of the work that has gone on many years before to make these opportunities possible.
But this, and similar work, on the electricity market still continues and needs the specialised expertise that the MEU provides.
For example, the establishment of the gas STTM was a long term project that involved many years of work but it was essential for energy users that throughout this lengthy process for the needs of users to be always top of mind as the development proceeded. This is a role for groups of users, such as the Major Energy Users (MEU). Throughout the development of the STTM, the MEU was involved at all levels of the process and ensured that the STTM delivers the maximum benefit to gas users.
The MEU continues to be heavily involved in making sure the electricity and gas rules are revised and rewritten so that energy users have greater flexibility in how they access and use the resources of the energy market so that the interests of energy users are maximised.
Energy efficiency may not be all it’s made out to be. The rebound effects need to be taken into account.
http://thebreakthrough.org/index.php/issues/energy-efficiency/lighting-electricity-steel
Worth a read.
Great post.
It’s good to see a website produced by an enquiring mind.
Same idea has been used by Peter Martin in The Age.
http://www.theage.com.au/comment/power-down-what-the-melbourne-cup-can-teach-us-about-fighting-climate-change-20141103-11fv8q.html
Another story of note recently occurred on Saturday afternoon (1 Nov) when there was a long period of sustained negative prices in South Australia. There was high wind power production (~1000 MW), low demand (~700 MW), probably because of a high solar contribution behind the meter, low thermal power production from coal (Northern ~250MW) and gas (Osborne ~150 MW, Torrens Island 50 MW), and maximal export to Victoria (~400 MW). Victorian prices were not much better at about $7/MWhr, with a high wind contribution (~900 MW) and maximal exports to Tasmania and NSW.
Tasmania is another interesting story. Prior to the abolition of the carbon tax, it consistently exported about 400 MW to Victoria. This was a deliberate strategy, and was achieved by running down its dam levels. Since the end of the carbon tax, Tasmania has instead been importing about 400 MW from Victoria. The strategy made $140m profit in 2012/13. Its dams are now at only one third of capacity, and winter inflows have been poor. I expect it will take several years for the dams to return to their normal supply levels, and I hope they have budgetted for the costs of importing power over that period.
http://www.themercury.com.au/news/tasmania/lower-rainfall-increased-power-generation-lead-to-fouryear-low-for-hydro-dam/story-fnj4f7k1-1226872669585
Thanks Malcolm
You bring up an interesting point – because of what some might see as a windfall gain for existing hydro plant coming from the carbon tax, hydro plant in Australia ran harder during the 24 months of the carbon tax – an entirely rational behaviour. We highlighted the first week in 2012 here.
That’s not the only opportunity for borrowing from the future to arbitrage time – another way has been to run hard one year and lean the next in order to generate above baseline production volumes and hence create renewable energy certificates (LGCs/RECs). I thought I had posted some analysis on this online, but could not find in a quick scan.
Paul
You got quoted today in the Melbourne Age. http://www.theage.com.au/comment/power-down-what-the-melbourne-cup-can-teach-us-about-fighting-climate-change-20141103-11fv8q.html
Existing hydro investments could be used for much more generation balancing to complement the volatility of wind than it currently is, particularly the Tasmanian hydro stations. The Snowy stations and Dartmouth are used in this way, but while the carbon tax was applied Tasmanian hydro did little more than provide baseload power across Bass Strait, selling into what was at the time a higher priced market. In the process it has run down the hydro dams to one third of capacity, and now imports brown-coal power from Victoria, which is now cheaper without the carbon tax.
The NEM covers two quite different markets with different economic drivers. There is the SA-Vic-Tas cluster, which has the majority of the NEM’s wind and hydro capacity, and all its brown coal capacity. There is also the NSW-Qld cluster, which is driven by black coal, gas, and a relatively small amount of wind and hydro. The linkage between these markets has a capacity of little more than 1000 MW, so when it is maxed out, there is no further capacity for generation balancing. To accommodate the intermitancy of wind in the SA and Vic markets, better use could be made of Basslink and existing Tasmanian hydro stations. Basslink has a capacity of ~500 MW, but this could go from -500 MW (Vic to Tas) to + 500 MW (Tas to Vic), thus providing a potential buffer of 1000 MW. Unfortunately now that Tasmania have run down their dams they may have little capacity to export into Vic at times of high power prices, so until their dams recover we are unlikely to see this desirable market behaviour. There are also peaking gas plants in Victoria and SA that can also cover the volatility, but as gas prices increase they will need a higher power price before dispatching power from their plants.
The Labour government tried to purchase some of the older coal-fired power stations such as Hazelwood in Vic and Playford in SA, but the asset owners appeared to be holding out for a higher price than the government was prepared to offer, expecting that power demand would continue to increase. I wonder how many coal-fired asset owners now regret them not having taken up the government’s offer. This would have provided a more orderly path to reducing coal-fired capacity than waiting for receivers to step in.
Re security of supply with volatile wind and solar renewable energy, existing hydro investments could be used for much more generation balancing to complement the volatility of wind than it currently is, particularly the Tasmanian hydro stations. The Snowy stations and Dartmouth are used in this way, but while the carbon tax was applied Tasmanian hydro did little more than provide baseload power across Bass Strait, selling into what was at the time a higher priced market. In the process it has run down the hydro dams to one third of capacity, and now imports brown-coal power from Victoria, which is now cheaper without the carbon tax.
The NEM covers two quite different markets with different economic drivers. There is the SA-Vic-Tas cluster, which has the majority of the NEM’s wind and hydro capacity, and all its brown coal capacity. There is also the NSW-Qld cluster, which is driven by black coal, gas, and a relatively small amount of wind and hydro (but that hydro has important capacity for security of supply – Tumut 3 has a 1500 MW capacity but only enough water for a 4% capacity factor). The linkage between these markets has a capacity of little more than 1000 MW, so when it is maxed out, there is no further capacity for generation balancing. To accommodate the intermitancy of wind in the SA and Vic markets, better use could be made of Basslink and existing Tasmanian hydro stations. Basslink has a capacity of ~500 MW, but this could go from -500 MW (Vic to Tas) to + 500 MW (Tas to Vic), thus providing a potential buffer of 1000 MW. Unfortunately now that Tasmania have run down their dams they may have little capacity to export into Vic at times of high power prices, so until their dams recover we are unlikely to see this desirable market behaviour. There are also peaking gas plants in Victoria and SA that can also cover the volatility, but as gas prices increase they will need a higher power price before dispatching power from their plants.
You should have a running tipping contest for which station is to be closed or mothballed next similar to your annual peak load contest – however you would probably have to give away too many BBQ’s to keep up over the next12 months. Or is everyone still pretending that “bid unavailable” and “peaking duty” are different to mothballing?
In both of the extreme cases wind is being punished; when it’s windy everywhere and price is low wind farms get paid less, when it’s not windy and price is high wind farms miss out on the super lucrative times.
Tom, “punished” is an emotive word that evokes thoughts of unfair treatment, but I wonder if that’s actually the case?
In the case of “low” prices
When it’s windy all generators who contributed to that low price (wind and thermal) are paid the low spot price. This means the wind farms that generate plus the thermal plant that still bid very low prices to run at least at minimum load.
It’s a separate commercial decision for each portfolio and it seems that, at least till now, portfolios continue to be “happy” to run at very low prices for transient periods rather than back off further, or shut down completely. This decision applies to wind farms, just like any other.
In the case of “high” prices
Why is it the case that a wind farm is “penalised” for being unable to generate when the market requires it (i.e. which, in this particular case in point, would be when every other wind farm is still)?
At times when the ability of groups of generators to produce is restricted, there is added opportunity to extract higher prices from the market – no matter whether this is due to lulls in the wind, or major technical problems at power stations, or unavailability of fuel for thermal stations.
One point that I notice on a weekly basis is that the limitation of wind, systemically, occurs far more often than it does for other groups of assets (no matter which other way I can think to group them) – this is the main point I make in the article (i.e. once a week, perhaps, for 2000MW wind to not perform, as compared with once every few years for a 2000MW station to not perform)
In this article 15 months ago, I was surprised to see a much stronger correlation between wind farm output at individual stations than perhaps I’d been led to believe, and noted that this highlighted an opportunity for:
(a) Wind farms in different parts of the NEM where the wind profile was very dissimilar (ideally inverse). Perhaps this would be in North Queensland – I am yet to see data to verify; and
(b) Of course, there is the argument for storage to level out the injections from individual plant – and to level out “market” demand, perhaps as well. However such storage would involve investing more money in a market that’s already oversupplied, with demand declining.
Another point to note is that the (relatively) high prices experienced when wind is low (say $35/MWh in South Australia) is commonly held to be still below a long-run marginal cost of operation for a large thermal plant with – and this seems to be an overlooked point – that long run marginal cost defined on an assumption that it achieves a capacity factor that might be higher than that which is being achieved at present. A lower actual capacity factor further increases LRMC required – hence widening the gap between what’s being achieved and what’s required.
A more neutral interpretation might be that the wind farms that have been constructed, to date, have not taken sufficient account of locational factors as they relate to the NEM (as distinct from how they relate to wind production volumes) and so are themselves achieving sub-optimal returns, and being part of the puzzle that’s driving returns lower for everyone else.
More generally
Finally, I recognise that your choice of “punished” might have been a spare-of-the-moment decision and that a more correct word might have been chosen to convey what you actually meant.
However in my limited use of social media there does seem to be a fair degree of victim mentality pervading the discussion on all “sides” of the debate. It seems to me that that’s not going to get us to a truly sustainable solution.
Yep, I didn’t mean wind is a victim, just that the value of production is self discounted. Given that capacity availability is rewarded through high spot prices, wind is not being overpaid for its capacity.
I definitely agree about the suboptimal geographical placement. Unfortunately SA average prices were and still are much better than VIC or TAS. And you’d be crazy to compete with Tas Hydro. No one wants their revenue to be defined by a competitor.
We are already getting close in SA where wind generation meets the regional demand. So one day I can see that wind farms will have to bid a realistic price for their output (rather than budding the minimum) in order for them to get their targeted return. At the moment they are depending on thermal generation to bid so that the spot price + LREC provides the return on capital the wind farms need. What would wind farms do if there is so much wind generation that the spot price is set by wind farms for many hours – presumably consumers get paid by the wind farms to use their output. Oh joy! Come the day
David, those would be strange days indeed – but somehow I don’t think that energy users will actually ever see that respite.
Trying to ascertain how wind farms should bid to maximise returns in an oversupplied market with declining demand (assuming no major closure of other capacity) would be a tricky challenge indeed, for reasons including:
(a) A fuel source that is uncontrollable, and very much aligned with the fuel source a sizeable number of your own competitors use (especially in South Australia); and
(b) The fact that you lose REC revenue, as well, if you choose not to generate.
A “Death Spiral” for wind generators, if you will?
It does seem that it will be necessary that strategies mature over time, unless we collectively invent other ways of socialising costs whilst privatising profits (e.g. by progressively raising the cap price for the REC to make up the shortfall + layering in some other support for storage).
The “saving grace” for wind farms in South Australia is the interconnector to VIC (or this problem would have been met years ago), so measures of % production from wind for any given region (or other sub-set of the interconnected NEM) are a bit of an academic nonsense, most times.
Two markets where the wind generators don’t sell into a low market are Tasmania and NSW. In Tasmania the load balancing is from hydro (which has its load factor limited by river flows), and the 2 wind power stations are not highly correlated, one being exposed to westerly winds, the other to easterlies. In NSW wind is a such a small proportion of the market that these problems haven’t yet become apparent.
The upgrade of the SA-Vic interconnector will reduce the problems of low prices during high wind periods in SA, but could have the effect of passing these problems on to Victoria. The upgrade is only about 200 MW, so if an additional 200 MW of wind capacity were added in SA this would nullify its effect.
So where are the next wind farms likely to go ? My guess is Tasmania and NSW.
Thanks for this comment, Malcolm (and a couple others I have not had time to respond to either).
The surprising thing I found in this analysis was that all wind farms that existed at the time (across SA, VIC, NSW and TAS) were still reasonably in sync, meaning that there were still quite a number of whole days when production was close to zero .
It hasn’t hurt the generators outside of SA as much, in terms of driving spot prices significantly lower, but this would suggest that it’s only a matter of time, as new capacity continues to be added.
Personally, I’d really like to understand how similar (or dissimilar) a wind production profile would be as far away as possible in the NEM – say in North Queensland?
Paul, Cup Eve Monday is a very commonly scheduled RDO in many industries, as well as commonly taken as annual leave; the Monday reduced demand pattern is therefore not unexpected,and clearly predictable as you have shown.
I am amazed that the proponents of wind project haven’t taken into consideration the effects of correlated wind speed of multiple sites on market prices in their business case. In the last few minutes I have tested a correlation between hourly wind speeds at Hamilton (close to the epicentre of Australia’s wind capacity, with MacArthur, Oakland and Portland wind farms in close proximity), and Armidale on the Northern Tablelands of NSW. This is about as far north in NSW as I have seen wind farms proposed. The two stations are positively correlated and the correlation is statistically significant, but the R-squared is only 4.5%. So for practical purposes they are not correlated. Armidale Airport has a mean wind speed at a height of 10 m of 4.6 m/s, and Hamilton is windier at 5.2 m/s, so it is easy to see why the Hamilton area was chosen first. The most favoured location near Armidale would be at the top of the Tablelands between Armidale and Glenn Innes, which coincides with the interconnector between NSW and Queensland.
Nearly all Australia’s wind capacity is within 100 km of a line between Port Augusta and the north-west tip of Tasmania. Most cold fronts almost exactly match the angle of the line, so nearly all the stations experience high winds together. Cathedral Rocks wind farm near Port Lincoln is a little earlier than the majority, and the NSW wind farms and Musselroe in Tasmania a little later.
Another consideration is the complementarity between wind and solar. Wind tends to predominate during winter and as lows pass over. Solar predominates in summer and when there are highs. While the right wing bloggers might chide wind for not producing at times, these are often the times when solar is producing extremely well. So an analysis of the volatility of renewables needs to include both wind and solar, and include consideration of the times of low demand at night when neither produce as much as during the daytime.
Hi Malcolm,
Market prices are definitely taken into account for the business case for a wind farm. Unfortunately production is the major driver as market prices are are not much better for sites with different profiles. For example if the wind speed at hub height scales for the Armidale and Hamilton sites you mentioned the Hamilton site will have 44% more energy available. Even the worst correlating wind profiles (for price) experience only a ~20% discount to average market price.
The wind speeds I quoted are from Bureau of Meteorology weather stations and Armidale and Hamilton Airports, both of which are located in exposed areas. The standard measurement height is 10 m. While the airport wind speeds are lower at Armidale, there are more local prominences along the top of the range between Armidale and Glenn Innes than there are near Hamilton, which could mean a well-located wind farm could be exposed to higher overall wind speeds than in the Hamilton area. For example, there is a range just west of Glenn Innes that is about 200m above the surrounding landscape. In the Northern Tablelands, the transmission line approximately follows the ridge line and therefore close to these promenances.
In calculating wind power, what consideration is there of (i) altitude and (ii) temperature? Some of the Northern Tableland sites are up to 1400m, so there could be some loss of power because of a thinner atmosphere. Frost and snow are also considerations for the mechanical operation of the wind turbines at this location. By contrast in the tropics, wind doesn’t feel it has to have the same “penetrating power” beyond the coast. Apart from cyclones I don’t recall the Queensland coast being particularly windy when I lived there. Is this anything to do with temperature thinning the atmosphere ?
I have lived in Armidale, Townsville and Hamilton, and Hamilton definitely feels windy more often.
Density does drop with altitude and humidity but not a whole lot. Usually the increase in wind speed with altitude is well worth the small drop in density. Colder temperatures at altitude also offset the pressure drop a little.
I also wouldn’t worry about frost. Armidale temperature goes positive every day [I used to live there too 🙂 ] so buildup of ice is not possible. Ice buildup is a problem with temperatures constantly very negative for weeks.
The Armidale region could be a decent location due to the excellent transmission but you’d have to find a good ridge. Hopefully the New England Wind Coop gets off the ground.
Paul,
Good post. I get the distinct feeling that your general references to ‘social media’ are really a specific reference to a tweet I posted recently that compared a single instance of wholesale pricing and wind energy penetration in the NEM:
https://twitter.com/KetanJ0/status/526871486481846274
Given the constant assertion that wind energy is driving up prices, a single instance of high wind penetration and lower-than-average prices is a demonstration that that’s not a rule you can apply, in general, to the genesis of wholesale electricity price.
I agree with the anonymous comment on that tweet that broader conclusions can’t be drawn from a single instance (I didn’t draw those conclusions, but that seems irrelevant to the critics). I’ve also included a comparison, over two years, of spot price vs wind output.
As I’m sure you share the same antagonism towards the presentation of single instances, surely one could asses the ‘left’ and ‘right’ corners of your post above together, and look at the net effect of both instances?
IE: Wind is either high, or low. The net effect is that it’s there, sometimes. Compared to a situation where it’s never there, ever, there’s a net decrease in wholesale electricity price.
Is this the middle ground you’re looking for, when both potentialities are considered?
Ketan
No, I don’t think that is the middle ground. At least not in the longer term. Adding more supply to a market has an impact on future investment (and perhaps disinvestment). So it’s not reasonable to assume that everything else stays the same. Over the long term, if there is more wind, there will be less of something else. This can then be expected to result in higher prices when there is little to no wind generation. Whether it will average out as lower prices overall is dependent on assumptions around demand and the timing of entry/exit of other generation. At high levels of intermittent generation, it becomes more likely that reliability settings such as the Market Price Cap will need to change to maintain reliability.
Other costs that may increase and feed through to end user prices include interconnector upgrades and ancillary services, let alone the certificate costs.
Higher end user costs is not necessarily the wrong outcome if the community wants more renewable energy. But this tradeoff is rarely acknowledged.
But I agree that cherry picking individual instances of high wind/low price or low wind/high price and extrapolating from them is not meaningful.
If you think it is relevant to state that adding wind, as long as nothing else changes, will decrease the wholesale price, then fair enough. But one could say the same about brown coal.
Hi Ketan
Thanks for the compliment.
Re cycles of wind in sync with social media:
No, I had not seen that particular tweet – there are plenty of others that celebrate the high wind times and (it seems) a number on the other side that note when wind is at its lowest. To me it just seems illustrative that there are two “sides” shouting at each other with neither really listening in return.
Re drawing a linkage between wind and low (or high) spot prices
You comment “Compared to a situation where it’s never there, ever, there’s a net decrease in wholesale electricity price.” is true on the assumption that all else remains the same, both in terms of:
(a) How all the portfolios bid; and
(b) How assets are made available in the market.
for that particular dispatch interval.
I think this is the point Kieran is making with his “one could say the same about brown coal.” comment – i.e. what would have happened if you added more brown coal production, instead of adding wind.
What I have been grappling with is the effect to which we might see changes in future – either
1) as a result of conscious decisions by portfolios to withdraw capacity from the market, or reprice it higher to earn a decent return – or
2) in a “doomsday scenario” if we have the same type of response thrust on us more randomly as a result of collective/systemic “run it into the ground” decisions taken with whole groups of assets not earning a return. Kieran notes the possibility of increasing the Market Price Cap further to provide a return for capital-intensive plant running lower capacity factors than designed, whilst others have talked of capacity payments. They won’t be the only two possible responses.
These types of inter-temporal questions were part of my comments here with respect to one of the main limitations of modelling done for the RET – including ACIL Allen’s official study. I’m not aware of anyone who’s really looked at this in detail? Even more-so, I am surprised to see that not many people even acknowledge it
This type of market-promoted feedback mechanism is also sorely lacking in descriptions of what’s become commonly quoted as “The Merit Order Effect”, despite the impressive graphics – i.e. it assumes that new capacity can continue to eat “someone else’s” lunch (socialising costs whilst privatising profits) with no competitive response in the market.
Paul
Thanks Paul and Kieran,
This is all really good stuff. I work as an advocate for clean tech like wind farms and solar so it’s pretty valuable to get a more nuanced and technical side of things from those in the know.
I’m having some difficulty grappling with one particular thing – Kieran’s comment that adding (say) 1 GW of brown coal capacity would have the same wholesale-price-depression impact as adding 1 GW of wind energy capacity (for instance).
My understanding is that wholesale price is lower due to the presence of energy in the lower end of the bid stack – pushing the ‘price-setter’ lower. Wind energy sits at the lower end because:
– High installation costs are offset by the LGC scheme
– Fuel can be extracted with near-zero costs
Am I wrong in thinking that these are the mechanics of what’s known as the ‘merit order effect’? If, in the above scenario, we simply had extra brown coal, wouldn’t that leave price at the same level, as we simply have more generator bidding in the same price band (as opposed to greater volume in lower price bands)?
Or is the point that brown coal is cheap, and as such, also serves the purpose of filling out the lower price bands (and therefore triggering a reduction in wholesale price)?
The question really comes down to: Are wholesale prices depressed due to some characteristic of wind energy, or the simple addition of megawatts of new generation of any kind?
If the second is true, would the creation of 1,000 megawatts of diesel generation decrease wholesale price, regardless of the frequency or magnitude of its dispatch? Or, the creation of 1,000 megawatts of gas?
Again, thanks for the replies – this is very informative.
cheers,
Ketan
(@KetanJ0 on Twitter)
Ketan
Reality is, oftentimes, more complex than any pretty animation (or any market model) can reflect.
A drop in price (i.e. Price A – Price B, comparing two scenarios for a dispatch interval with all else the same) occurs when added capacity in Scenario B is added below what the regional price is for that region in Scenario A. It mostly* does not matter how much below Price A the added capacity bids.
Hence for that hypothetical dispatch interval Price B would be identical, no matter whether the added 1MW of capacity was brown coal or wind.
However this is further complicated by thinking about:
1) What price the additional capacity bids.
2) The availability factor (capacity factor for wind) of the asset will also play a role (remembering that we’ve started to only look at a single hypothetical dispatch interval):
3) Timing of the availability:
Hope this helps?
Paul
480MW may not be much in the AGL portfolio, but combined with a couple of other gas fired generators in SA that are rumoured to be planned to be mothballed next year it could get interesting on a hot day with no wind in SA
Good point. We might get to the situation where the wind lovers realise how much they need fossil fuels today.
Note that it’s not an immediate closure. It would coincide with an extra 190 MW of capacity from Victoria when the Heywood inter-connector capacity is increased, a drop in demand with the closure of Holden and other larger users, and an increase in gas prices as the Gladstone LNG plants compete for Moomba supplies. The Torrens Island plant is not a rapid-response gas turbine like Dry Creek, but needs to be kept hot. This makes it less competitive as a true peaking station. The Dry Creek plant has rarely been used recently even on days of high demand.
How effective is wind energy at saving CO2 emissions?
What’s the abatement cost ($/tonne)?
A topic for another post perhaps one day, Peter
Hi Peter,
It depends on the state. The worst case is QLD with 0.92 t/MWh (due to modern black coal plant). At the current LGC price of $33 that’s $35.87/t.
QLD doesn’t have much wind though so a better representation would be SA or VIC with abatement costs of $34.73/t and $33.67/t respectively.
The displacement calculations assume that wind displaces the marginal unit (but not hydro units as they can hold their water until later).
Tom,
Thanks you. Can you give me a link to where the methodology for those calculations is documented?
Hi Peter,
I took the marginal unit per 5 minute from AEMO’s NEMPriceSetter.xml files.
My assumption is that if you added 1 MW of wind at any point in time that you displace some typical unit. The abatement effectiveness is based on the t/MWh from ACIL Tasman’s 2009 paper, this is weighted by the frequency that each unit sets price. This method isn’t great but it is more representative than using local generation as the smaller states are strongly affected by VIC and/or NSW generation. The result is that if you add renewable generation anywhere in the NEM you are most likely going to displace a NSW black coal generator.
I reweighted the values to remove hydro as that energy can be used later to displace fossil generation. Must-run hydro doesn’t set price and can’t be displaced. If wind was the marginal unit then the addition of more wind power abates nothing; that energy isn’t stored.
OK, so how do you know I’m not completely full of **it? Since 2010 gas generation dropped only 2% (0.378 TWh). Black and brown coal generation both dropped 20% (31.7 TWh total); mostly because of demand. Wind is up 74% (3.3 TWh). Hydro is ballpark after their run on the carbon tax. The weighted average emissions intensity from the change in coal generation is roughly 1.04 t/MWh. So either my method underestimates wind abatement effectiveness or energy efficiency is better than wind at abating.
I did this because Grant King kept going ON about how wind displaces CCGT. You’d think the guy would know something about gas generation wouldn’t you?
Cheers,
Tom
Tom, Thank you for further explanation. Is the method written up somewhere where I can access it.
I suspect your assumption about the effectiveness of wind at abating CO2-e is optimistic.
ACIL Allen came up with a similar cost of abatement for wind as referenced in pg. 20 of the CCA’s review http://www.climatechangeauthority.gov.au/files/files/reviews/ret/2014/review.pdf
They had a cost of $32 which would suggest that the abatement effectiveness is above 1t/MWh, as the LGC cost has consistently been above this level.
pg. 166 of the actual report https://retreview.dpmc.gov.au/sites/default/files/files/ACIL_Report.pdf
Please let me know why you think my estimations are optimistic and what wind would be abating if not coal generation.
I tried searching for an email address for you but all I got was fashion and publishing businesses. Let me know and I’ll send you an Excel file.
Tom,
“Please let me know why you think my estimations are optimistic and what wind would be abating if not coal generation.”
Joseph Wheatley (2013) ‘Quantifying CO2 savings from wind power’: http://joewheatley.net/how-much-co2-does-wind-power-save/
See the full paper (linked in the post)
Daniel T. Kaffine, Brannin J. McBee, Jozef Lieskovsky (2013). ‘Emissions Savings from Wind Power Generation in Texas’ http://www.iaee.org/en/publications/ejarticle.aspx?id=2509
Daniel T. Kaffine, Brannin J. McBee, Jozef Lieskovsky (2012). ‘Emissions savings from wind power generation: Evidence from Texas, California and the Upper Midwest’: http://econbus.mines.edu/working-papers/wp201203.pdf
Herbert Inhaber (2011) ‘Why wind power does not deliver the expected emissions reductions’: http://www.sciencedirect.com/science/journal/13640321
This paper references many studies into the effectiveness of wind power at reducing GHG emission. All these studies used empirical data. Many studies were not used if they had used data that is not publicly available.
Could you ask Paul McArdle if he’d mind forwarding your spreadsheet to me.
Peter,
You keep using that Irish source. I get it, if you displace gas the emissions abatement is small. Same thing for CA and TX. We hardly use gas in Australia though.
Check out the abatement for MISO in the paper you linked – 0.92t/MWh (pg. 7). MISO has a lower proportion of coal generation than the NEM, but like the NEM coal is most commonly the marginal unit. Unlike Ireland, CA, and TX they don’t get half their electricity from gas/nuke. A much closer representation of the NEM.
http://energy.gov/sites/prod/files/2014/09/f18/Remarks%20of%20Kurt%20W.%20Bilas%20-%20Exec.%20Director%20Government%20Relations%20-%20MISO_0.PDF
“I get it, if you displace gas the emissions abatement is small. Same thing for CA and TX. We hardly use gas in Australia though.”
You need to read ans understand the papers.
Sorry, but I Am not interested in your comments any more. You have no credibility.
Seriously, do look at pg. 35 of the econobus link you have above. It has a nice graph of emissions saved by wind vs. coal generation fraction. This would be helpful when comparing between regions.
Have a nice day,
Tom
Regarding Inhaber’s paper, his abatement curve uses Germany as the example. Germany’s main priority is prematurely switching off their nuclear fleet, thus emissions from the electricity sector have remained flat as renewables are only keeping up with the reductions in nuclear energy. I would phrase this as keeping emissions from rising while removing nuclear.
However you phrase it the example does not apply elsewhere as other nuclear fleets are not retiring by force. You should not use his fictional equation for other regions.
Inhaber also cites de Groot on part load efficiency.
Total BS.
80% load, peak efficiency is 35% for subcrit coal.
50% load, lowest efficiency is 33%.
100% load is a bit below peak efficiency.
At worst that’s 5.7% more emissions per MWh but with 37.5% fewer MWh sent out.
“Regarding Inhaber’s paper, his abatement curve uses Germany as the example.”
Wrong!
You should read and understand. You have no credibility.
Back up generation just has to be available. It’s not actually generating.
If a coal plant running at 100% has a SRMC of $35 and sells into NSW at $35 they are breaking even. If the same plant ramps down to 50% but consumes the same amount of fuel they will lose $35 per MWh. This ramping happen to the black coal generators every night. They would go bankrupt very quickly if you were right about backup.
OR you could suspend disbelief and realise that they just feed less coal in.
The Boauhinia rail electrification project has just been completed, so rather than diesel trains carrying coal from the Rolleston mine all the way to the coast it will be electrics, which will no doubt further increase electricity demand. Queensland now has a total 4,000 KW of electric locomotives, so if all locomotives were at 100% load factor it would be 4 GW of demand !
http://www.aurizon.com.au/projects/bauhinia-electrification-project
4,000 kilowatts is only 4 megawatts. Each locomotive might be 3MW or 4MW and each coal train might need 3 of these
Another Queensland story of interest is the end increase in gas prices – today Brisbane trades are $4.21/GJ, which is higher than any other capital city. The arrival of the first LNG tanker in Gladstone Sydney prices to $4.00/GJ, but no obvious impact yet on Vic prices at $3.50/GJ.
So with higher Queensland gas prices, power prices from peaking gas plants would need to be higher. How much higher Queensland gas prices will go will be interesting, and probably not what the investors want to hear. The high prices for LNG landed in Japan now appear to be a thing of the past as new low-cost LNG producers such as PNG are joining the supply side, and there appears to be only modest growth in Japanese LNG demand.
“With the three projects each opting for electric compression of gas delivery to Gladstone, we’re looking to see a sizeable increase in demand for electricity as a result”.
This is not quite correct – althout QCLNG and APLNG are electrified using electricity from the NEM, the GLNG upstream project has gas turbine driven hub compressors and electric driven nodal compressors supplied by open cycle gas turbine generators so does not increase electricity demand on the NEM.
Thanks for the added point of clarification, Interested Observer
Paul
The excellent data available through NEM-Review tells me that a large new continuous load of around 400 MW came on line in Queensland on 27 October last. This suggests that the 2014-15 summer peak is likely to be considerably higher, for equivalent weather, than in recent years; this has already happened on 9 and 17 December. However, without knowing what the load is, it is difficult to be certain with such a prediction.
Thanks Hugh
We have begun to investigate – see in this new post
Paul
Oh and yes, it’s too late for anyone to take account of this in their forecasts for peak QLD demand in summer 14-15 (to win a portable BBQ) as entries closed on Fri 9th Jan.
A very good initiative has been taken to save our electricity bills. Usage of solar power also helps in keeping the environment clean.
Hey Paul,
So what does this mean for gas powered generators? My understanding is gas can change output much much faster than goal which means some companies must be making a bucket load.
Do you think with more renewables gas will play a much bigger role in the future to supply peak demand when the renewables wane? I’ve been quite skeptical of the many green oriented articles stating that gas electricity will play a tiny role in the future of Australia’s energy market.
Could it be the gas liquefaction plant near Gladstone?
That’s one possibility, Craig – especially when thinking about the broader LNG supply chain, including upstream electric compression (as noted here). We’ll wait to see what our further investigations reveal, in terms of size and timing.
Several of the process diagrams for LNG trains show that the compressors are driven by gas turbines (costly?). Also, if they were consistently running the liquefaction plant could they store all the LNG? Surely someone can confirm or deny the rumours!
My personal opinion is that the unique G20 public holiday in November caused a step change in southeast Queensland customer behaviour. The pristine weather before the supecell hailstorm was just the right time to head to the Gold/Sunshine coasts and drive up the tourism load. Also the early holiday could have driven consumers into tropical summer holiday mode earlier than usual, turning on the air conditioning earlier than would otherwise have been the case.
Thanks Peter – that’s certainly a different theory 🙂 – I, for one, can’t recall what I was doing on the G20 holiday!
An AEMO Market Notice on 26 September 2014 said that the second 275/132 kV transformer at PowerLink’s new Wandoan South substation was commissioned.
Then there’s APLNG Columboola and the PowerLink Orana 275 kV substation that feeds APLNG there.
See also http://www.originenergy.com.au/files/APLNG_Project_Upstream_Delivery.pdf
So my guess is that the mystery customer in APLNG.
PS: I wouldn’t burn my valuable natural gas to make electricity for my compressors if I can buy cheaper electricity from a coal fired power station 🙂
On the other hand http://www.originenergy.com.au/files/APLNG_Site_Tour_Combined_Presentations.pdf says that the APLNG plants near PowerLink’s Columboola and Orana substations are only good for less than 100 MW 🙁
Thanks for the great links Mike. They also built two 30 MW gas power stations in the area (pg. 17). It’s probably not too expensive if you run them on unprocessed gas and it would give them a bit more flexibility in managing gas supply vs. demand.
Pretty sure its just the start of summer, it has been hot and humidity has been up. There aren’t any regional loads that are up significantly apart from the gradual growth in the Surat. The real demand in QLD still has a while until it hits.
10 years management consulting? How impressive!
How was the estimate of solar PV generation done?
What is the assumed fleet-average capacity factor of PV installations that are actually still operating? How do you know?
What proportion of the originally installed PV installations are still operating? How do you know?
Why would they not be operating? The panels have a 20 year warranty and last much longer than that. If your inverter breaks it’s pretty cheap to replace.
[I answered this previously but the comment has been deleted.]
Tom,
Re-read and try again. Not the words “fleet average” are significant.
If Green Markets ws genuinely concerned about reducing global GHG emissions, or Australia contributing to global reduction – as distinct from just pushing jobs, income, economic growth off shore – you’d be advocating for nuclear power not renewable energy.
1. Nuclear power is a much cheaper way to substantially reduce global GHG emissions than renewable energy.
2. Nuclear power has the capacity to provide all humans energy needs effectively indefinitely.
3. RE cannot sustain modern society, let alone in the future as per capita energy consumption continues to increase as it has been doing since humans learnt to control fire.
4. There is much greater capacity to reduce the cost of nuclear energy than renewable energy.
5. The issue with nuclear is political, not technical. The anti-nukes and RE advocates are the block to progress and they have been for the past 50 years.
You are a joke.
Hinkley C tender price was $170/MWh. That’s at close to 100% CF. If we went 100% nuclear we would have 60% CF and closer to $230/MWh.
If you wanted to use breeder designs it would cost even more again. Uranium from the ocean? Fuel costs go up 6 times.
“Much greater capacity to reduce cost” Just because it’s so expensive doesn’t mean it is going to get cheaper. Nuclear power has increased in cost over the last 50 years. Renewables have plummeted in the last couple of decades.
I can’t wait until all the old pro-nuke dinosaurs die out of the energy industry. You are a waste of oxygen.
Tom,
Re-read and try again. Not the words “fleet average” are significant.
[Ignore repeat comment.]
The CSIRO ‘MyPower’ calculator shows that, even in Australia where we have cheap, high quality coal close to the main population centres and where nuclear power is strongly opposed, nuclear power would be the cheapest way to reduce emissions: http://www.csiro.au/Outcomes/Energy/MyPower.aspx
Below is a comparison of options with different proportions of electricity generation technologies (move the sliders to change the proportions of each technology). The results below show the change in real electricity prices and CO2 emissions in 2050 compared with now.
Change to 2050 in electricity price and emissions by technology mix:
1. 80% coal, 10% gas, 10% renewables, 0% nuclear:
electricity bills increase = 15% and emissions increase = 21%
2. 0% coal, 50% gas, 50% renewables, 0% nuclear:
electricity bills increase = 19% and emissions decrease = 62%.
3. 0% coal, 30% gas, 10% renewables, 60% nuclear:
electricity bills increase = 15% and emissions decrease = 77%.
4. 0% coal, 20% gas, 10% renewables, 70% nuclear:
electricity bills increase = 17% and emissions decrease = 84%.
5. 0% coal, 10% gas, 10% renewables, 80% nuclear:
electricity bills increase = 20% and emissions decrease = 91%.
Source: CSIRO ‘MyPower’ calculator
Points to note:
• For the same real cost increase to 2050 (i.e. 15%), BAU gives a 21% increase in emissions c.f. the nuclear option a 77% decrease in emissions (compare scenarios 1 and 3)
• For a ~20% real cost increase, the renewables option gives 62% decrease c.f. nuclear 91% decrease in emissions (compare scenarios 2 and 5).
• These costs do not include the additional transmission and grid costs. If they did, the cost of renewables would be substantially higher.
That site uses old BREE data. They have since updated.
For 2020:
nuclear – $100 to $170 (assumes plant lifetime longer than ever achieved, no decommissioning cost, free fuel infrastucture)
wind – $60 to $110
PV – $70 to $190
I agree that if you want to go 100% clean power you need nuclear, batteries are still in their infancy. We haven’t got the political will to even get 25% clean power. May as well choose the cheaper option which is clearly large scale wind and rooftop PV.
Tom,
As usual you haven’t a clue what you are talking about. The BREE are projections for 2030 and 2050, and they are the updated data. You didn’t even bother to read any of it did you. And the plant lives are generous for renewables and penalise nuclear. you really haven’t a clue.
And we haven’t even included the cost of transmission and the risk (almost certain) that renewables cannot do the job the advocates claim
Furthermore, the differences between the 2012 and 2013 are trivial compared with the very large superiority of nuclear over renewables.
You are basically ignorant of any of this
The MyPower links to information are broken. They used to say they used the 2103 BREE AETA report data. However, the linkt to “what are the numbers based on” now takes you to eFutrure and that data is based on 201
Correction: “and they are the updated data”
This statement may be incorrect. The links to the data source are broken. My recollection is that they were updated some time last year to the 2013 AETA report and the information links said so. There were discussions about it at the time. Now the only link (perhaps out of date) points to eFurture assumptions and that says it is based on 2012 AETA report. My recollection is that MyPower was updated to 2013 data, but I could be wrong. But the differences are still insignificant in context.
Right, so we should stop reducing emissions now and wait until 2050 when nuclear MIGHT be cheaper. Got it.
All transmission costs are borne by the generator. It’s the law.
This includes connection, frequency control, and reactive power support.
Maybe you’d know that if you worked in the power sector.
Again,, you demonstrate you don’t know what you are talking about. Transmission cost is not borne by the generator; it’s borne by the consumer. Wind farms pay only for the connection to the grid. But the cost of transmissions for a largely renewable energy generation system is huge.
Nuclear is far cheaper than renewables now. It can be far cheaper than renewables when the impediments on it – imposed as a result of 50 years of anti nuke propaganda and disinformation – are wound back.
Sorry, Tom, you haven’t got any understanding of any of this and your confirmation bias prevents you from doing any unbiased research. You don’t even Accept BREE and CSIRO figures when they don’t say what you want to hear.
You are an example of the people who have been blocking progress for the past 50 years or so.
The connection cost includes any transmission line additions required to connect to the grid. Any new customer or generator has to pay for line additions and they do not own them either. WFs also pay transmission charges like all generators. You are imagining that large new transmission lines are being built on the public dollar for the sole purpose of connecting a WF. In that case it would be expensive. But it doesn’t happen.
I’m not trying to stand in the way of nukes. I’d be happy for them to be included in the RET. I’m annoyed with you for spreading misinformation about RE.
Tom,
You are misunderstanding. you clearly have not read the link I gave you. To greatly reduce GHG emissions we need a large proportion of electricity generated by low emissions technologies – such as nuclear and/or RE. The additional transmission cost for say 50% nuclear is very much smaller than for 50% of electricity supplied by renewables – e.g. rough estimate $37/MWh for renewables v $2/MWh for nuclear (this is addition to the existing grid). Very rough but gives you some idea of scale.
Correction:
The additional transmission cost for say 50% elecgtricity supplied by nuclear is very much less than for 50% of electricity supplied by renewables – e.g. rough estimate $18/MWh for renewables v $2/MWh for nuclear (this is addition to the existing grid). Very rough but gives you some idea of scale.
Tom,
You said: “I’m annoyed with you for spreading misinformation about RE.”
I don’t believe I am. I believe you don’t understand the relevant facts. You’ve made many unsubstantiated, and pedantic and irrelevant, assertions but you have not shown that the important, relevant facts are incorrect, let alone shown they are wrong by a sufficient margin to change the conclusions.
The three main components of the total system LCOE for a system that can substantially reduce GHG emissions are:
– LCOE of the technology (e.g. as per CSIRO eFuture and MyPower using data from BREE AETA)
– LCOE of additional transmission required (much greater capacity and much longer lines to widely dispersed areas)
– Expected value of risk that RE cannot do the job (meet the requirements of the electricity system economically).
You need to show that the numbers I’ve given for these three items are so far out that the conclusion is wrong.
If you can’t show your alternative figures to substitute for mine, and provide authoritative sources (relevant to NEM) for all of them, then you have no valid argument.
Nuclear power cheapest electricity with largest emissions cuts by 2050
The lowest cost way to generate our electricity and reduce emissions by 2050 is with a large proportion from nuclear power.
Here I use the CSIRO ‘eFuture’ calculator to compare two scenarios to supply electricity to meet the projected electricity demand on the National Electricity Market (NEM) in 2050 as well as cut CO2 emissions. The two scenarios are: 1) nuclear power not permitted and 2) nuclear power permitted. eFuture determines the generation mix that gives the least cost electricity for that scenario using the selected inputs. The scenarios compared here use the default (central estimate) for each user selectable input. The two scenarios are compared on the basis of CO2 emissions intensity and wholesale cost of electricity.
CO2 emissions for nuclear not permitted are 80 t/MWh versus 25 t/MWh with nuclear permitted. That is, if nuclear is not permitted emissions would be 3.2 times higher than if it is permitted.
The table below lists the LCOE (wholesale price) with nuclear not permitted and with nuclear permitted; the third column shows the ratio ‘No/Yes’ (nuclear not permitted/permitted)’. Cost items of common interest are itemised. Costs are in $/MWh.
Item No nuclear With Nuclear No/Yes Ref.
LCOE from ‘eFuture’ 130 85 1.5 1
Accident insurance 0 0.1 2, 3
Decommissioning 0.15 0.01 4
Waste management 0 1 5
Transmission, 50% penetration 18.5 2 6
Total LCOE 149 90 1.8
Without nuclear is 80% higher cost than with nuclear permitted. With a higher proportion of renewables in the default scenario, the cost of electricity would be even higher, probably more than double the cost with nuclear permitted.
Policy analysts also need to include in policy options analysis an estimate of the risk that renewables will not be able to do the job. We know nuclear can provide around 75% of electricity in an advanced industrial economy because France has been doing it for over 30 years. But renewables have not demonstrated they can or will be able to. Many practitioners think they will not. An estimate, in LCOE equivalent terms, of the risk that renewable technologies do not meet the hopes of the proponents is $54/MWh.
The risk that renewables will not be able to do the job is the major risk that should be questioned, not the costs of waste disposal, decommissioning, accident insurance etc. of nuclear all of which are negligible compared with LCOE and the risk that renewables do not deliver the benefits claimed by their proponents.
References
1. CSIRO ‘eFuture’ scenario options calculator http://efuture.csiro.au/#scenarios
2. US NRC, 2014, ‘Backgrounder on Nuclear Insurance and Disaster Relief’ http://www.nrc.gov/reading-rm/doc-collections/fact-sheets/nuclear-insurance.html
3. EIA, ‘Nuclear electricity net generation’ for year 2012 http://www.eia.gov/cfapps/ipdbproject/iedindex3.cfm?tid=2&pid=27&aid=12&cid=regions&syid=1980&eyid=2012&unit=BKWH
4. IEA, 2010, Projected Costs of Generating Electricity, 2010 Edition http://www.iea.org/publications/freepublications/publication/projected_costs.pdf
5. OECD, 2013, ‘The Economics of the Back End of the Nuclear Fuel Cycle’, Figure ES.1 http://www.oecd-nea.org/ndd/pubs/2013/7061-ebenfc-execsum.pdf
6. Peter Lang, 2013, ‘Renewables or nuclear electricity for Australia – the costs’, Figure 7 http://oznucforum.customer.netspace.net.au/TP4PLang.pdf
7. WNA, 2014, ‘Decommissioning Nuclear Facilities’ (“0.1 to 0.2 cents/kWh”) http://www.world-nuclear.org/info/Nuclear-Fuel-Cycle/Nuclear-Wastes/Decommissioning-Nuclear-Facilities/
8. DECC ‘Offshore Renewable Energy Installation Decommissioning’ https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/47955/900-offshore-renewable-installation-decom.pdf
[deleted at request of commenter]
Please delete 6:47 pm comment. It was posted in the wrong place. I’ve reposted it where it was intended to be.
Paul
Just on the question of why AEMO’s MTPASA and STPASA forecasts seemed so different for the week you examined, the reason is that the MTPASA takes no account of the actual weather forecast – it is simply based on long term climatic statistics. The STPASA forecast does take into account the actual weather outlook.
The MTPASA demand forecast will almost always be significantly higher than the STPASA forecast unless the near-term weather forecast is predicting conditions (broadly) corresponding to a maximum demand day for an entire season, which is the the demand metric that the MTPASA forecast is attempting to represent (at 90%/50%/10% probability of exceedence levels).
Cheers
Allan
Thanks Allan
That helps me – and probably some of our readers as well.
Paul
The sole justification for wind power is to avoid GHG emissions. Therefore, I’d suggest displaying the emissions in real times (as is done in other countries). More importantly show the emissions avoided by wind in real time as well as:
CO2 emissions by technology (t CO2/h) and t CO2/MWh
CO2 abatement cost ($/t CO2 avoided)
See EirGrid for example: http://www.eirgrid.com/operations/systemperformancedata/co2emissions/
And
France: Demand, Generation, Trading, CO2 emissions in real time: http://www.rte-france.com/en/eco2mix/eco2mix-co2-en
I should have said “the principal justification for wind power is to avoid GHG emissions.”
Great write up, just a couple of thoughts
I think the Pitt and Sherry “mysterious jump on Queensland demand” is answered in AEMO’s last two editions of Energy Update February 2015 edition page 2 and December 2014 edition page 3.
Quoting “Queensland experienced record breaking temperatures in late October and November last year. The November 2014 mean maximum temperature was the highest on record for the state, leading to some very warm nights for Queenslanders,” said Ms Pimentel.
Secondly, In addition to PV behind the meter there is also the price triggered demand side diesel generation embed in the distribution networks to consider. Some of these embedded generation sites can be quite large >1MW and there surprising number of of them out there. Looking at yesterday’s spot price some these embedded generators may have potentially run well into the evening. So if we were adjust for both solar PV and embedded demand side generation what is the underlying demand?
And finally is anyone tracking the number of failed and “abandoned in place” solar PV systems that the owner cannot afford or be bothered to repair. There are plenty of anecdotal stories of failed PV systems where the installer has gone out of business (no warranty) and home owner is faced with the cost of replacing the now ~5 year old inverter, if you haven’t yet recovered the install cost on the original PV system, then paying for a major component is bitter pill to swallow for many.
Cheers
Thanks Paul
In terms of the “mysterious jump” I think what AEMO has noted is part of the answer, but have not had the time to really dig further.
In terms of embedded generation behind the meter, solar PV has been (as much as I understand) the biggest growth factor in between 2010 and now. That’s one reason why we have focused on it, in particular – to provide a more integrated view in NEM-Watch v10 complete with that data.
In terms of price-responsive demand changes, either by diesel gensets starting or by actual demand reductions, there was some of that happening yesterday we know (some are our clients).
In terms of anecdotal evidence of solar faults, I have yet to delve into that – but I know that Tristan and others have commented. If you know of other sources of published data, in addition to the CER, please list them here so I can check them out as time permits?
Paul
My concern is that moves in this way might be described as “jumping from frying pan into the fire”…”
I agree 100%. And I agree with your comments about the RET. Simplifying enormously (for clarity in a blog comment), the rational answer seems simple. It’s to dump the RET and allow Direct Action to work to buy abatement at the least cost (based on reverse auction processes)..
However, as ACIL-Allen pointed out in their commissioned report to the RET Review Expert Pannel, simply dumping the RET and repealing the legislation sends a message of heightened regulatory risk in Australia (no worse that what some advocate to shut down coal power stations of course, and probably less economically damaging because all repealing the RET would do is to remove the guaranteed sales of all the RE generators produce and the subsidy of the RECs. UIf they are cheaper than fossil fuels, as some claim, they don’t need any production subsidies. Let’s see.
ACIL-Allen suggested that closing the RET to new projects would be preferably and gandfather the exisiting projects. I am persuaded this would be the best approach. Let’s give the Direct Action approach a chance: Reasons to support Direct Action (and close the RET to future projects) are:
1. It’s much lower cost than the Labor-Greens CO2-e emissions restraint policies
2. It’s highly flexible – so it can be adjusted to fit with whatever policies the world agrees to
3. Carbon pricing is highly unlikely to succeed [1], [2], so its a totall waste of time energy and resources continuing to advocate for it
4. The US is implementing direct action, and has no intention and not realistic prospects of implementing CO2-e emissions pricing
5. Direct action has demonstrated it is by far the best way to reduce emissions – e.g. France’s CO2-e intensity of electricity is 10% of Australia’s and it has near the cheapest electricity in EU.. That’s not due to CO2-e pricing. Its due to direct action policies – as are virtually all the countries that have nuclear power substituting for fossil fuels.
6. Countries with the highest proportion of renewables have the highest cost electricity and high emissions intensity from electricity. The countries with the most nuclear have the lowsest cost electricity and lowest emissions intensity of electricity [3]
References:
[1] ‘Why carbon pricing will not succeed Part I’ : http://catallaxyfiles.com/2014/10/26/cross-post-peter-lang-why-carbon-pricing-will-not-succeed-part-i/
[2] ‘Why The World Will Not Agree to Pricing Carbon II’ ; http://catallaxyfiles.com/2014/10/27/cross-post-peter-lang-why-the-world-will-not-agree-to-pricing-carbon-ii/
[3] ‘How much does it cost to reduce carbon emissions? A primer on electricity infrastructure planning in the Age of Climate Change’ (see especially slide 10 and note the irony in slide 14) ; http://canadianenergyissues.com/2014/01/29/how-much-does-it-cost-to-reduce-carbon-emissions-a-primer-on-electricity-infrastructure-planning-in-the-age-of-climate-change/
On your post [1], you are assuming that the cheapest options are excluded first which isn’t necessarily so. Electricity is a sector with low abatement costs and is easy to measure.
Both direct policies and taxes have their benefits. I previously worked on an energy efficiency scheme where businesses wouldn’t accept money to upgrade their equipment. As soon as the threat of the carbon tax came along those same businesses were desperate to upgrade even though the rate of return was lower without rebates. People are more afraid of loss than they are excited for gain 🙂
Direct action nuclear takes capital costs out of the electricity sector and puts them on the tax payer. You could just as easily do that with renewables and have free energy.
Interestingly, the generally higher/more volatile wholesale electricity prices are also coinciding with a generally rising domestic gas price.
So I wonder if the volatility is purely demand side driven or perhaps due to some supply side constraints from the gas turbines
Your point 2d) is why I’m against capacity credits. The benefit of a high market cap is that the rewards for actually delivering are high and the rewards for not delivering are zero.
You could probably justify building a OCGT if you could capture 10 hours of $13k per year. If supply was an issue people would knock up some peakers in no time.
Alternatively, if the Market Price Cap actually reflects the value of lost load than there will be a greater incentive to bring capacity to the market, make available additional ramping capacity and extend asset life. Unfortunately, there are strong political reasons not to allow the MPC to reflect VOLL that are counter to the markets and consumers long term interests.
>”Gas also faces competition from renewable energy, with increasing renewable capacity, mainly wind power that services the base load demand. The penetration of solar PV systems with energy storage pose as a potential competitor to GPG in the peak electricity segment in the medium to long term. The ability to store electricity would enable households to change consumption patterns by charging batteries during off-peak and using stored electricity throughout peak period. This would see falling demand for grid-electricity during peak times.”
Does Core Energy’s projection of solar PV systems assume the massive subsidies (and hidden cross subsidies) will continue indefinitely? What is Core Energy’s projection for the scenario where the subsidies ceased (as clearly they should be if policy was based on rational analysis)?
Peter this is an area of material uncertainty. If solar becomes competitive without support mechanisms and storage becomes a cost competitive reality then gas faces a major challenge in the water heating and room heating space. It is simply too early to make an informed call. It is an area Core is analysing closely as it is has major implications across the energy chain.
If? PV is the cheapest domestic fuel, subsidies or not. You can get a system for $2.3/W before STCs these days. With 5% interest you’re laughing as long as you consume at least a third of what you produce.
If gas stoves weren’t nice to cook on domestic gas would be dead.
Paul,
I don’t know what figures you are using, but I cannot understand why people advocate solar power. It doesn’t seem rational to do so. not only is it many times more costly than fossil fuel or nuclear generated electricity, it is also not sustainable – that is it requires fossil fuel of nuclear to support it. If you add the cost and energy required for energy storage it is even further from being sustainable. The fact it is not sustainable is a slam dunk.
If the issue is GHG emissions reduction then why not advocate fro nuclear. It is far superior against all the important requirements: energy security, reliability of supply, cost, health and safety and environmental impacts?
Peter,
A typical offgrid system has an EROI well over 10. How is that unsustainable?
Solar is now cheaper than nuclear*. Rooftop is much cheaper as you save on distribution costs. You can’t just go out and buy a 10 kW reactor.
*Hinkley C $180/MWh, Chinese new reactors $100/MWh, Saudi Solar $60/MWh, undiscounted unsubsidised Aussie rooftop solar $100/MWh
Thanks for the insightful article and analysis. However I think your conversions in table 1.1 might be out. Total gas of 171 PJ equaling 18.2 GWh doesn’t feel right. Should it be 18.2 TWh?
Thanks for picking that up, Tim
We corrected the transcription error above for ease of readability.
Paul McArdle
One area where more analysis is required is whether solar will be effective in meeting winter heating needs in the residential and commercial sectors. Has anyone seen quality research/analysis addressing this topic?
It depends on your system, climate, and insulation. Offgrid systems often tilt their panels up towards the winter sun to give a more uniform annual profile.
With respect to gas, does it matter? A reverse cycle aircon at peak rates is still cheaper than gas.
Related fun fact: German domestic gas is half the price of Australian domestic gas. Their wholesale price is about twice ours.
If only the government didn’t mandate peak/offpeak hours. Great article how do we spur more debate on this issue.
thanks for the commment, we are working with economists to put a good case to government.
Great article!
You could also charge for capacity and energy separately like for large customers. Many homes already have interval meters or smart meters. Such tariffs have encouraged offpeak heating and cooling systems to be built in the US.
most interval meter systems are poorly matched to real network congestion and thus do not make the best case for alternatives
Interval meters don’t have to mean TOU. You can measure customer load over an interval and charge based on average kW. It’s not as good as peak kVA but it’s a start.
A part of the ‘big issue’ is that networks in some States have been built much larger than the largest load. Better pricing is important, but the AER putting the clampers on high costs and over-investment will deliver the first and biggest wins for all consumers.
A certain margin is sometimes necessary, in the CBD in Brisbane a few years ago it was fed by a few old cables whose status was not fully known. We were running a risk like Auckland which was not fully restored for up to several weeks. Thus instant capacity compared to peak load (particularly for CBD) is not a good measure of a gold plated network, it is likely to be driven by reliability standards.
That is much less true for suburban areas and congestion pricing is much more necessary.
When ever I talk to networks companies about this issue they almost inevitably tell me the problem is “lack of granular data”, IPART said the same thing a few years ago. They either don’t know or can’t manage the data fast enough to offer genuine or dynamic pricing signals; so they smear things in a crude way.
In fairness, they are under huge pressure to reduce costs and so new data systems and analysis doesn’t seem to be on their agenda and of course, the current regulations don’t really incentivise or encourage innovation.
Seems to me (happy to be corrected) that rethinking the regulations and incentivising granular data collection is fundamental to changing the status quo?
It seems to me we do need to move to a new regeim , it may be that it will take new players to find the business case.
I’ll be very interested to know the average wholesale price of power from the solar farm. Royalla has contracted to sell electricity at $183/MWh (or there ablouts from memory). That’s about six times the average LCOE of existing coal plants (according to BREE AETA, 2012). and over three times the average spot price.
Of course all that is because the very large hidden subsidies and cross subsidies are included. What an unbelievable boondoggle solar and wind power are. And for what purpose? What effect will they have on the climate?
Talk about being short-sighted. By very definition we’ll have to phase over to renewables eventually, the sooner we start the better.
Clearly you don’t live anywhere near a coal fired power plant. The number of environmental hazards they pose is extensive, that’s excluding their massive CO2 output. Coal is cheap because there are unpriced externalities. It’s that simple.
As for solar subsidies, a recent (Nov. 2014) ABC News article stated that: “Australian coal, oil and gas companies receive $4b in subsidies”.
There’s no hidden subsidy or cross-subsidy here, it’s pretty overt.
Also, that contract price is the same as the cost of nukes, except it’s a fixed cost not rising with inflation.
Why does solar total 10 MW on the REnewconomy site? The column has only one entry for 2 MW.
Peter
Next time you see something like that please take a screenshot so we know what to investigate – it’s a bit hard for the customer service team to know what they are investigating otherwise.
Paul
Interesting analysis Paul.
A couple of points I’d like to make.
Looking at the lack of correlation between wind and solar on a 30 minute basis, it is worth mentioning that a lack of correlation is a lot better than a positive correlation. It certainly means that there are going to be much fewer instances of both wind and solar being simultaneously at near record lows (or highs) than just wind or solar alone. This will likely have implications for any estimates of storage requirements (or spillage from over-generation).
I’ve noticed that when looking at monthly averages of wind and solar, that there is a very small negative correlation. This is because wind tends to perform a bit better during winter than summer (and obviously solar performs better during summer than winter).
Regarding your questions about solar’s capacity factor.
The APVI website indicates that the overall average CF achieved at the PV peak on March 3 is a little over 50% (ie, a long way short of your preliminary estimate of 79%).
http://pv-map.apvi.org.au/live#2015-03-03
This would suggest that they are using an installed capacity of a about 3.3 GW, which would give an average capacity factor of 13% for this 7 week period. This seems a little on the low side to me, most reasonable systems I’ve seen are more like like 16%, however I have no doubt there may be many systems suffering from shading etc that would bring the average down.
Having said that, they are estimating 4.0 GW of installed capacity as of Nov 2014 (or reported installed capacity of 3.9 GW), so I’m a little confused with that discrepancy.
Given that it is rare for a single PV system to ever achieve close to 90% of it’s capacity (unless the system has an greatly undersized inverter relative to the panel capacity), having every system in Australia simultaneously averaging anywhere near 80% of capacity would be highly unrealistic.
Thanks for the comments, David – will review the information you’ve supplied.
In terms of degree of correlation, I would like to be able to understand how a big wind farm in north Queensland would match against the existing stock in the south-east of the NEM. Can you help me with that one?
Paul
I don’t have that info to hand at the moment, though I can tell you that when looking at 2014 wind output on a daily basis, the following average correlation coefficients between states apply (R value)
SA:SA 0.69
SA:Vic 0.60
SA:NSW 0.28
SA:Tas 0.16
SA: WA 0.08
Vic:Vic 0.77
Vic:NSW 0.42
Vic:TAS 0.36
Vic:WA 0.01
NSW:NSW 0.76
NSW:TAS 0.36
NSW:WA -0.01
TAS:TAS 0.57
TAS:WA -0.04
WA:WA 0.39
So you can see that (on a daily basis) there is essentially no correlation between wind farms in WA and the other states. And given the relatively poor correlation between wind farms in SA and those in TAS or NSW, I think we could guess that the correlation between SA and QLD is going to be pretty close to zero on a daily basis.
Total national PV capacity (includes ~50MW of large generators) as used in the map is 3,896MW, based on monthly CER installation data. That data is ~3 months old at present so it’s an underestimate: we’re certainly >4GW now.
Capacity per state is shown in charts at http://pv-map.apvi.org.au/historical
Thanks Michael,
however I remain a little confused. As 12:20 pm today (March 30), the live data page indicates that QLD solar is generating at 43% of max capacity, and also that it’s generating 497 MW. This implies a max capacity of 1156 MW. But this is in disagreement with what it has for QLD’s state total of 1249.7 MW of PV. Can you shed any light on this disagreement? Is it assuming a max capacity that is less than the generation capacity?
Clearly an extreme case – but it does raise serious questions about the realty of relying on electric vehicle battery storage for NEM backup. It also brings into question some of these “100% renewable energy” analyses that have circulated over the last few years.
Thanks Martin
As a scenario assuming onlywind supply, with EV battery backup, it’s clearly not a plausible scenario.
I’m not sure that it makes sense to leap to it brings into question some of these “100% renewable energy” analyses as I’d believe that those studies would have included more than just those 2 technologies.
Clearly there needs to be more than just those two – and a number of people have suggested others to me – online and offline (such as Peter in his comment on this copy of the article on Climate Spectator.
This kind of analysis I’m doing as time permits, but next steps might include:
Step 1) Overlaying solar over the same period and scaling to see how the “unmet demand” gap might reduce (this is perhaps the most obvious next step as solar, along with wind, are the two renewable technologies being deployed at scale in the NEM already);
Step 2) Following this, I could look into how existing hydro capacity might be able to alter its operation (in this hypothetical scenario) to fill the gaps. Keeping in mind that most of this is not pumped-storage, but does have existing storage, it would seem to be a less expensive form of storage than anything new.
Step 3) Looking into other technologies that others are working to try to bring to commerciality to see how it would work to fill in gaps.
I suspect, at the end of the day, that a Pareto principle will apply here as well – in that the cost per greenhouse abatement will escalate quickly as we move up the “% Renewable” scale. Hence absolutely “100% Renewable”, under all scenarios (e.g. even in drought years), might not ever be an economically desirable outcome for the majority – unless some new technology comes along.
Paul
1271 GWh. That’s a lot of batteries certainly but it’s very doable by other means such as hydro. Whilst they are not pumped storage as such, we’ve already got *individual* hydro schemes storing far more than that so it can certainly be done.
Just as we won’t likely see all energy into the grid coming from wind, so too we won’t likely see all storage in the form of batteries.
Dual axis tracking systems look amazing, but the cost increase far outweighs the improved energy harvest. The improvement in energy harvest is also affected by latitude. It only becomes appreciable at higher latitudes – talking >35N or S. There is a significant increase in BOS for dual axis tracking – motors, controllers, comms within the plant, not to mention the foundations. When the modules were expensive it was a great idea, now that modules are cheaper, the most cost effective way to increase energy production is to put in more fixed modules. Similar arguments can be used to justify preferencing flush mounting vs tilt brackets in residential installs.
Single axis is a lot simpler mechanically so the economics are subsequently a bit softer. The increase in energy harvest is reduced too – I’ve not had much to do with single axis trackers, so I’m not sure where the line falls.
The graphs are from the 19th – I’m not sure what the tilt angles of the fixed and tracking systems are, but if they’re close to latitude then these graphs will be of almost the ideal day – close to the solstice. Comparison of the peak output / installed capacity of each system will be interesting at the end of June. The capacity factor of the dual axis should remain relatively stable over the year, the other two will vary quite a bit.
The arguments for or against west vs north orientation are heavily dependent on the tariff and the site – unless afternoon tariffs are much higher or the west facing roof is the only suitable roof and its a choice between some solar and barely any, west facing will always result in a fairly large reduction in energy harvest. At typical residential roof slope it’s around 25% reduction from ideal. This reduction is less noticeable/important at low tilt angles (say 15 degrees or less), which does allow for a lot more freedom on large commercial/industrial roofs, and can allow for very high coverage ratios on space constrained ground systems.
The relative cheapness of modules may allow to offset this drop in output through increased capacity, but with no time of use tariff or a flat load profile, the timing of the solar peak is a bit irrelevant so you go for the best $/kWh.
Also, while orienting to the west (or east) does shift the timing of the peak, it’s not really noticeable until you get to really silly tilt angles and start to lose a lot of output due to atmospheric absorption and a lot of space due to mutual shading.
Sounds a good idea. Any idea when the costs will come down enough so that this technology will
be on peoples houses in Queensland? Or will the cost always be too high do you think?
Any idea what the following is (approximately)
Dual Axis Total Energy Over typical March day
——————————————————————————- = 120%, 110% ????????
Optimal Fixed Total Energy Over typical March day
Keep up the good work!
Dear Paul
Very good response from Nathan. Its true the best PV outputs for fixed tilt are often recorded around the equinox – in Melbourne at least, this is often in October/November or April, when the elevation of the panels allows the highest levels of radiation captured by the plane of the PV array.
The calculations for tracking are pretty simple – is the additional capex of the installation of tracking exceeded by the additional value of energy on an NPV basis? You would add to this cost the maintenance and risk of downtime involved with tracking. If you follow some of the bigger installations you may see that the developers have already sone these sums and decided not to track.
On the residential PV side, the depressingly small feed-in tariffs in Victoria have led installers to so something weird – orienting half of the modules east and the other half west. This is an export minimisation strategy as it pushes the generation into the morning and evening peaks and minimises export typically seen in the middle of the day. It doesn’t give you the best kwh harvest but rather the best $ harvest. So PV policy can give us some odd PV architecture. Note don’t try this by connecting these arrays together – they should have separate inverters!
As mentioned, no-one is expecting to go 100% wind, and I doubt many people think we can go 100% wind + solar.
The UNSW 100% renewable study indicated that solar and wind can meet about 88% of our electricity needs on an hourly basis throughout the year. The remaining 12% was obtained almost equally from gas turbines and existing hydro. The study required about 200 GWh of storage, primarily from molten salt associated with CST, though I suspect this is more likely to come from battery storage (and with the CST replaced with PV). Not insignificant, but achievable (10 million homes x 20 kWh, though I suspect a 3-way split between residential, business and distribution is more likely and sensible)
http://www.ies.unsw.edu.au/about-us/news-activities/2013/04/least-cost-100-renewable-electricity
As a PS to this post, we’ve received the following suggestions (thus far) of what might have contributed to the reduction in the all-hours demand:
1) Mothballing of Port Stanvac desalination plant
2) Operation of industrial generation plant behind the meter
3) Closure of Penrice Soda
4) Reduced energy usage at mining sites, including Olympic Dam
5) Car manufacturing closure.
Note that I have not checked any of this, but just record as a starting list for further digging, when time permits.
With respect to other claims made, this reduction in all-hours demand is distinct from (layered underneath) the reduction of daytime demand due to embedded solar.
Could a simplified tracking system work with floating arrays, such as the recently-opened unit a Jamestown in South Australia ? A whole chain of floating panels mounted at the latitude angle could be moved between NE, N and NW during the day by a relatively simple automated pulley system. The UQ tracking systems would be expensive because they are holding the entire weight of the panels. If instead that weight is floating, a much simpler system could be used.
There’s quite a few unique things electrically in Tas.
First is that at the far off-peak times, heavy industry accounts for 80% of the total load and just one factory is one third of the entire load in the whole state in the middle of the night.
Second is the dominance of peak rate water heating, with a market share around 88% of residential hot water. Typically these are 160 or 250 litre storage systems (no such thing as instant electric hot water in Tas) with 2.4 or 3.6kW element. Hence the big jump in demand in the morning as people take showers. Off-peak has only 8% or so of the market, and virtually all of those are on the off-peak rate with afternoon boost – there’s only about 1000 or so customers using the off-peak night only tariff in total so it’s very minor. That said, gas, heat pumps and solar are slowly gaining increasing market share at the expense of the traditional electric storage HWS.
Third is that electric heating dominates (67% market share) with most of the rest being wood. Gas is sufficiently rare that if you have a house with gas and are selling it, the real estate agent really will mention the gas as a definite feature of the house.
Then there’s the means of heating itself. About 40 – 45% of homes use what are universally known as “heat pumps” in Tas – that’s a single split system air-conditioner in the terminology used in other states. They’re always installed for heating, with a heating calculation (but no cooling calculation) to determine the size and spend 99% of their operating time in heating mode. Typically, a 7kW or so output system in the main living area only and electric resistive heating in bedrooms etc is very common.
Then there’s those with resistive heating systems, and typically that’s a 6 – 10kW fixed heater in the living room. Yes, a 10kW single phase 240V electric heater – and they literally do glow red hot in operation (as they’re supposed to). Others have just a fixed fan heater, but still 3.5 – 7kW generally.
So there’s a lot of heating equipment installed versus relatively little cooling equipment. Hence cold weather can drive demand far higher than hot weather ever will. But heating is becoming more efficient as such, with heat pumps going from basically nothing 20 years ago to their current usage and a move away from resistive heating.
Also notable is that the traditional use of off-peak electricity in Tas is space, not water, heating via storage heaters. These are decidedly out of fashion these days and in terminal decline. Common a generation ago, before too long they won’t really exist at all.
Comparing past years, also of significance is the closure of the once-massive paper factory at Burnie. APPM (Associated Pulp and Paper Mills), or better known locally as simply “the pulp”, that place dominated the town for decades but it’s completely gone now and so too is its’ power consumption. At its’ peak it had 10 production lines producing paper, a pulp mill to supply the paper machines and another production line producing boards, slowly declining over 30 years before final closure in 2010. The separate Wesley Vale paper mill, with one machine, also closed at that time.
The remaining paper mill, Norske Skog at Boyer (near Hobart) has also reduced electricity consumption due to process and feedstock changes. Many still call that place ANM – Australian Newsprint Mills – and that’s what it used to make. These days one of the two machines has been converted to produce higher grade paper and the feedstock changed to rely solely on softwood which requires less energy (electricity) to process via the mill’s mechanical (not chemical) pulping process.
Then there’s a few small industrial co-generation plants, there’s a significant one at a vegetable processing plant and another one at a public hospital, installed in recent years which have also reduced grid demand a bit.
So overall, quite a few things have individually reduced load by small amounts but collectively it becomes a bit more significant. Also a lot of LED traffic lights are around now and street lighting is starting to change to LED’s too. And of course the move away from incandescent lighting in homes is another one.
All that said, so long as the “big 4” factories continue operating then overall demand can’t possibly change too much. The big 4 use roughly half the state’s power on average and two thirds at the far off-peak times.
Hi Shaun
Thanks for adding some excellent local context – very useful for me, and for our readers as well.
Paul
I presume you mean Thursday 5 March 2015 (not the 6 March), when half-hourly operational demand reached a maximum of 8,831 MW at 17:00, 66 MW off the all-time record.
The maximum temperature on 5 March 2015 was 36.9 degrees and the mean temperature was 28.3 degrees (at Archerfield). While not doubting this was a hot day, it ranks around the 10th percentile in historic weather terms, which may be loosely associated with a 90 POE demand. This would imply that the actual recorded demand was significantly above AEMO’s forecast.
Thanks for the correction, Richard – I see I had it right in the image, but wrong in the transcription. Noted this above.
Thanks also for the clarification of where temperature was at on that day – but not sure I understand the which was around the 10th percentile (i.e. average or peak) and how this translates to a 90% POE. Apologies that I’m slow on that translation.
As per the chart above, that day was certainly different than the days either side!
Paul
One of our readers has emailed in with these additions:
1) Another one that you may want to include is the role of aggregators for green and white certificate creation – eg STC, LGC, VEET and ESC schemes.
2) And it would seem with the recent ERF auctions that aggregators were the prodominat suppliers of projects into the ERF for abatement…
UChoose is the only Unbiased Comparison Service(aggregator) – the only one in Australia which receives the same commission irrespective of which offer the Customer chooses. Plus we do not have any preferred partners or sponsored offers. True choice and unbiased.
Great post. The high correlation between wind output in SA and Vic pointed out in a previous post means that prices in both markets are low or negative at times of high wind output and low demand. Increasing the SA-Vic interconnector capacity through the transformer upgrade at Heywood will do little to alleviate this problem. However, correlations in daily wind output between Vic and NSW are much lower. Surely it’s time to re-rate the Vic-NSW interconnector limits. There are 4 high-voltage power lines, but they seem to have very conservative ratings. (The coal generators in NSW would hate this because it would increase the low-cost brown coal power coming from Victoria at times of low demand in the early morning.) The power lines are
1. Khancoban-Tumut 3 (330 kV)
2. Khancoban-Tumut 1 (330 kV)
3. Dedarang-Wagga (330 kV)
4. Red Cliffs-NSW (220 kV)
Khancoban and the Murray power stations are regarded as part of the Victorian system, even though they are physically in NSW..
Looking at the Google Streetview image of power lines just north of the Khancoban switch yard (36°13’0.55″S, 148° 8’29.84″E), there are 4 sets of power lines, 2 of which go from Khancoban to Victoria and can carry 1500 MW of power from the Murray power station into the Victorian market. The westernmost 2 power lines look identical to those going into Victoria, and should be able to take 1500 MW from Khancoban into the New South Wales grid. Yet the maximum Vic-NSW interchange at 4 am this morning was only about 1200 MW, when there was a price of -$10/MWh in Victoria and $23.96 in NSW. The Dedaran-Wagga line should also have a similar capacity of about 750 MW, and the Redcliffs line perhaps another 200 MW. There could be further rating increases with dynamic ratings, because much of the transfer would be required at night at times of high wind and cool temperatures that would keep the wires below their threshold of about 50 degrees C. So there could be a total capacity of well over 2450 MW.
The Qld-NSW interconnector has been progressively up-rated, and looks much less substantial than those at Khancoban, yet I believe carries up to 1000 MW, while the 275 kV SA-Vic interconnector via Heywood looks similar and will soon have a capacity of 680 MW. So the Vic-NSW combined limit of about 1200 MW for 4 major power lines indeed looks on the low side.
A correction to your comment about Tasmania having to conserve hydro supplies . There have been excellent inflows over the last week into the Pieman, Mersey-Forth and Derwent catchments, to the extent that 2 of their hydro dams are spilling. However the catchments with the greatest storage – the Great Lakes (for Poatina) and Gordon have received minimal inflows. So the catchments with good inflows will allow water in the Great Lakes and Gordon to be conserved.
http://www.hydro.com.au/water/lake-levels
There have been periods of negative power prices in Tasmania today – perhaps the combination of high wind, low demand and run-of-river hydro power stations.
To quote hitchikers’ “Don’t Panic”. As you pointed out, possible LOR are common but actual LOR are not.
The key issue here is less wind and more the scheduled outage of an interconnector.
The issue of interconnectors to SA and Victoria was a key part extreme weather impact on the NEM study conducted as part of National Energy Security Assessment (NESA). The study identified SE Australian heat waves as the type of extreme weather event likely to stress the NEM. Interconnector failures are among the most significant power system risks during such events, with South Australia at risk for both being at the end of the NEM chain and heat driving demand.
The most recent january 2014 heatwaves pushed SE Australia to the limit, but while Reliability and Emergency Reserve Trader (RERT) contracts were formed they were not invoked.
A key issue is network congestion in Victoria affects what the interconnectors can carry to/from SA, and whether it is frequency synchronised. Each jurisdiction has its own preferences for responses to such challenges (ie how much political flak for large blackouts).
The rise of renewables has changed the paradigm of ‘spinning reserve’ – but the rise is gradual, and it is not the only possible response: already the NEM has a range of special schemes in both Tasmania and South Australia. As noted, demand response and storage are increasingly becoming economic options.
The short answer is that the future is likely to be a more complex mix of generation and demand tools than the historical ‘more of the same’, but ultimately it is very unlikely the lights will ever go out.
Thanks James
For everyone’s benefit you have referred to the National Energy Security Assessment, which is here:
http://industry.gov.au/Energy/EnergySecurity/nesa/Pages/default.aspx
Paul
What is missing in the market now is pumped storage hydro to level out the market, particularly the 1 am to 5 am periods of negative power prices when there is high wind. According to Wikipedia, the Murray power station has all of its generators capable of pumping, so if true it should have the potential to soak up to 1500 MW from the Victorian system at less than $10/MWh and sell it back to the grid a few hours later at over $30/MWh.
In the NSW system the Tumut 3 power station is pumped storage, and there is also a pumped storage system operated by Sydney Water, so a higher Vic-NSW rating capacity would provide access to these systems.
The NEM report files do show a “Snowy pump” of 200 MW operating during some of this time. Not sure if this is on Tumut3 (NSW system) or Murray (Vic system). Now that intermitancy is becoming a problem, surely one of the best power station investments would now be in these pumped storage systems.
Thanks Martin
The role pump storage hydro might play in balancing increased intermittent supplies is one of the intriguing questions in this all. I wonder what type of consistent price spread there needs to be (between pumping and generation) to make it work, given that (I assume) it would have to be running much more frequently than any of the 3 existing schemes do now.
Paul
Paul, my concerns with reporting that “1/4 of electricity was produced by wind” based on a 5 min DI is just the kind of story that the “greens” run with to suggest this is now the norm. I see it’s down to 8% in the current interval.
Martin
You’re a frequent reader of WattClarity, so will understand that:
1) I do try to remain objective; and
2) I have posted about the highs of various types of generation, along with the lows.
Paul
Paul,
As I mentioned on Twitter, I’m perplexed by the choice to focus on wind energy as a causal factor in this event, despite it playing a minor role in its occurrence. As you may know, a lot of people are regularly seeking to find reasons to lay the blame on wind farms for a very large range of adverse impacts – health, grid reliability, electricity price increases, mobile phone reception, etc.
Most of these are predicated on an inability (or an unwillingness) to consider causality – if something bad happens, and it happens at the same time that a wind turbine was generating electricity, that doesn’t mean wind turbine operation was a necessary or sufficient (or both) factor in that thing occurring.
I understand you’ve included AEMO’s correction at the bottom of this article, but someone motivated to implicate wind energy in problems that would have happened anyway would be well served by the headline in this article.
I might append a further example – Tony Abbott said in 2013 that ““renewable energy targets are also significantly driving up power prices right now”
The RET scheme contributes ~2% to 4% to residential bills at the moment. So whilst his statement of contribution is true, it’s not a reflection of the major cause of price increases – network infrastructure spending.
Numerically (or, qualitatively) accurate remarks about contribution can still misinform, without the inclusion of prominent and unmissable context. Headlines are easily the most striking and ‘stickiest’ component of any blog post or article – see: http://www.niemanlab.org/2014/02/4-headlines-that-will-restore-your-flagging-faith-in-journalism/
Thanks for the thoughts, Ketan
I will follow up further the comments made on Twitter and LinkedIn and investigate further. As my understanding increases, I will post more commentary here.
I do understand that there are different people who view wind farms very differently, just as there are people who view coal seam gas extraction very differently, just as there were many inquiries in years-gone-by about the effects of living in close proximity to high voltage power lines. Controversy is not a new thing in the energy supply industry.
Paul
Another contributing factor to the negative power prices in Vic at times of low demand and high wind production is that the brown coal generators don’t appear to adjust to the situation with lower output. To them, brown coal is “free” because the operation of the coal mine is a fixed cost, whereas in NSW it is a variable cost (because black coal can be sold on the world market). This market behavior differs from when there was a carbon price was in place and there were variable costs associated with brown coal use.
But their profitability is now so low that another coal-fired power station (150 MW) will soon close.
http://www.theage.com.au/victoria/alcoa-to-shut-anglesea-coalmine-and-power-plant-20150512-ggzgw9.html
Goodness me, there’s my name in lights!
I can certainly tell you, it’s been a windy week down here. Gloomy too, unfortunately.
As for what the market could do if the improbable occurred… I take it the abrupt loss of transmitted wattage from SA would potentially jeopardise the rest of the NEM? But apart from that, isn’t there a whole heap of Victorian and NSW hydro effectively on standby, judging by those generation numbers?
Paul,
You mentioned credible contingency. In cases like this the NIL_MAXG constraint will be active limiting wind farms and the interconnector and requiring other generators to increase generation. If the interconnector trips (or Northern/Pelican Point) both wind and gas can ramp up.
To extend on this theme, some readers would be interested in the work published by the IMF in May 2015 – and linked here:
http://www.imf.org/external/np/fad/subsidies/?hootPostID=05f97b4cc6736d8be1eeea546bef6422
With some blog posts:
(a) http://blog-imfdirect.imf.org/2015/05/18/act-local-solve-global-the-5-3-trillion-energy-subsidy-problem/
(b) http://blog-imfdirect.imf.org/2015/05/18/global-energy-subsidies-are-big-about-us5-trillion-big/
(c) and a presentation http://www.imf.org/external/np/fad/subsidies/pdf/presentation.pdf
We’ve been subsidising solar for 45 years at least. The subsidies for production should cease. It cannot be justified.
I recall Dr Mark Diesendorf, Dr David Mills and Dr Stephen Kaneff in about 1991 arguing that “solar is baseload capable and cheaper than nuclear now, if the stupid government would just give us more money to demonstrate it.
That was 24 years ago. But still solar is not viable. Solar is still hopelessly uncompetitive and all investment would stop if not for the subsidies and regulatory incentives to prop it up.
Don’t make me laugh. We’ve been subsidising nuclear for even longer, and at higher cost, with worse results.
Residential solar is the cheapest consumer fuel -without certificates. That should be your argument to cease subsidies.
Nuclear costs more than solar even when run at 95% capacity factor. When nuclear does load following it can’t compete with diesel let alone anything else. Leave nukes to those without any other resources.
Tom,
As usual you don’t have a clue what you are talking about. if you want to compare subsidies for different technologies you have to normalise – i.e. show the subsidy per unit of energy supplied through the the plants life. The subsidies for solar are an order of magnitude higher than for nuclear.
We got a return on investment with solar, look at the second graph. Nuclear subsidies per MWh are rising.
Nuclear $180/MWh, Solar $100/MWh. What don’t you get here? It costs more to subsidise nuclear.
Having a hard-on for nuclear power doesn’t constitute some form of expertise.
“Having a hard-on for nuclear power doesn’t constitute some form of expertise.”
Word, Tom.
There’s been a fascinating sequel to this high Tasmanian demand. On Thursday I recall the Victorian spot gas price hit $10/GJ during a cold snap, and peak power prices up to about $90/MWhr. There was an almost complete absence of gas peaking generation across SA, Tas, Vic and NSW. Wind was a strong contributor, providing up to 1000 MW each in SA and Vic, and peaking being provided almost entirely by hydro, despite very low dam levels at 2 of Tasmania’s largest power stations. The only gas units operating over this period were Osborne and a couple of units simmering at Torrens Island.
Then on Sunday a fire shut down SA’s Northern power station, but its contribution wasn’t missed because there was about 1000 MW of wind power in SA. This was more than the market could absorb, because there were extended periods of negative power prices in SA on Sunday afternoon and much of the Monday public holiday. Surprisingly, during this period of negative power prices there was minimal export of power to Victoria. Is a failure of the market bidding process ?
Thanks Malcolm
Yes, it has been interesting to watch (but unfortunately no time to post more, this time).
Cheers
Paul
Point 2 needs further development, “The physical limitations of the transmission network and how it is represented in the NEM algorithms to allow competition across locations”.
There are proposals for a few transmission lines to move to a dynamic rating (eg Ballarat-Bendigo), but not the main interconnectors. I suspect the ratings on many lines are set conservatively low. However Basslink can exceed its static rating by 25% for up to 5 hours. Why only Basslink ? A dynamic rating would reduce the spikes that occur in only one State.
The temperature of a transmission line is a function of its amperage, plus weather parameters of wind speed, air temperature and solar radiation. I work in agricultural science, where we can predict the temperature of apples using these weather parameters to within 3 degrees C. Surely it’s not too difficult to work toward a dynamic rating system updated hourly from the BoM forecast grid.
Wartsila has the gas power generation technology (smart power generation: start to full load in less then 5 minutes and high open cycle efficiency) that can give operators the possibility to suppress price spikes better than any other technology current on the market.
We have a rigorous study made by a consultant company that prove the point about our technology. If you are interested in the report, please send me an email, I will be happy to,share it with you.
Luca Febbraio/ GM power Plants Wartsila Australia
Paul,
You should probably mention that gas generators are withholding capacity to drive up price. Considering that wind is a smaller player it seems strange to blame them for the opportunism of others. I noticed that brown coal did not increase production at all in response to demand… 😉
Cheers,
Tom
Thanks Tom
I have not looked in this instance, but it would be a natural part of the energy-only market design if generators are “repricing capacity” at times of tight supply/demand balance (a topical activity given this debate going on at present).
On other occasions other readers have noted that generators don’t physically withhold capacity – so in the past I have used the term “economic withholding”.
Paul
PS in terms of brown coal in South Australia not increasing production, I think it has other things keeping it occupied at present…
Paul,
A Combined Cycle Gas Turbine can be quite attractive in most cases.
Regards
Paul
Thanks Paul
As I think I said above, beauty is in the eye of the beholder (or, in the case of CCGT perhaps, beerholder).
Paul
Hi Paul,
Much is said about the merit order effect reducing average wholesale electricity prices. It occurs to me that prices must be lower when the wind blows, hence wind turbines must receive below average market price. Conversely, as you suggest in your article, scheduled generation can achieve above average price by withholding output until prices are high.
Have you thought to use NEM Watch to determine a wind output weighted average price in SA? My intuitive feeling is that if the time weighted average is around $35 per MWh then the wind weighted average might be closer to $25 and with increasing share of wind this might drop further. This bodes badly for wind investment in SA as the $65 penalty will be cheaper than a PPA with a wind farm.
Thanks James
1) Your hunch is correct – wind farms do tend to show lower diversity than I was initially envisaging (even across regions) and hence collectively “shoot themselves in the foot” in terms of spot price outcomes. Other posts on WattClarity that look in this area include the following:
2) You (or other readers) can crunch the numbers yourself with the help of the NEM-Review package, which (as the name suggests) is specifically aimed at this sort of historical analysis:
http://www.NEM-Review.info
Paul
Whatever the effects on spot price the effect of wind farms is essentially to reduce the amount of fossil fuel burnt per MWHr consumed which might be important, don’t you think, in a few years. Our children and grandchildren may even thank us for our efforts, at least in SA.
P.S. if you really want to see ugly power infrastructure check out the Dry Creek Power Station at the corner of Grand Junction Rd and Churchill Rd Dry Creek in Adelaide. These roads carry thousands of commuters each day and the stacks of DCPS have been waiting for a paint job since at least 1992. Perhaps we should start a brickbats & bouquets award for infrastructure as they do for architecture. DCPS Coords: 34°50’52.65S, 138°34’56.59E
Paul,
the electricity supply industry has never been one to consider the visual impact of its infrastructure on the landscape or the suburbs, largely because no one in government ever insisted that they do so. Admittedly, there is some improvement round the ‘burbs (Brisbane) where the Energex substations are now somewhat more pleasant to the eye.
Unfortunately, much of LV distribution network has been put in above ground and its remaining useful life is probably 40+ years. So there is not much change possible there.
Other owners of infrastructure in SEQ (Queensland Rail, Brisbane City Council) are making significance steps in decorating what they own, eg Milton Railway Station, Coopers Railway Station in Brisbane, plus the traffic signal boxes in Brisbane. They are even involving the community in achieving this!
So there is some infrastructure which can be made pleasant to the eye, if the owners can step out of their old ways.
Is that enough for Joe Hockey and his leader? Who knows? No holding your breath now.
Steve
Rotor angle stability. Google it.
Thanks Pete
… but I’m missing something – not sure what this has to do with why Northern is being closed?
Paul
There are precious few power system stabilizers in South Australia. If Northern closes there will be even less. What happens on a high demand, windy day, and a few big lines trip?
Thanks Pete
You, like a number of others, are concerned about what the stability of the SA grid might look like (minus Northern, and maybe Torrens A as well if AGL’s “might. maybe” closure becomes more certain).
Paul
A colleague (leaving the utilitysector for greener pastures) once said “everyone wants the product, but no one wants the packaging”. That pretty much sums up the above.
Thanks Alastair
I understand the sentiment, and (it seems to me that) there is some truth to this.
However it also seems to me that there is also other layers – in that we’ve seen there are things a large number of people do want, but some have not wanted to pay their “fair” share for these. Some obvious cases in point being:
1) Historically, the mass roll-out of residential solar PV to hundreds of thousands of people, supercharged on the back of (initial) overly generous feed-in tariffs (of which I am one of 400,000 Queensland beneficiaries, for instance).
2) In the current day, a rocky road ahead in the move to more cost-reflective network tariff structures (that’s not being helped by the understandable claims that the self-interested network companies are only making moves now that demand is declining, whereas this should have happened 10 years ago when the take-up of air-conditioning was accelerating – i.e. that they are only concerned because its affecting them now, not when it was “only” a matter of one group cross-subsidising another).
Paul
Re stability of the SA grid, there is a relatively new gas combined cycle power station that has hardly produced any power for the last few months. Pelican Point has a capacity of 475 MW, and I’m sure its owners would be watching the latest developments with Northern and Torrens Island A with optimism.
http://www.adelaidenow.com.au/news/south-australia/pelican-point-power-station-will-cut-more-than-half-its-generation-capacity-early-next-year-threatening-jobs/story-fni6uo1m-1226978458743
Unfortunately Pelican Point runs on natural gas and once the Queensland LNG plants are commissioned there will be no gas to run base load generation anywhere in the NEM
Here is one answer, and an obvious one. I am unable to buy roof-top mounted solar panels as I live in a flat, currently rented, by the way. (Refer the correct comment by Edward Boyce, above. ) Strata title buildings must secure unanimous agreement to make a move. It strikes me as inequitable, indeed, a failure of democracy, that government provides incentives not equally available to the entire population. These solar rooftop incentive schemes represent a transfer of wealth to the better-off. There are other reasons, some architectural, that inhibit the take-up. You do not broach some other complex issues about sunshine accessibility in built-up areas,. e.g. If a high-rise building or tree overshadows someone’s’ collection area, is there just cause for complaint? While the installation of roof-top solar circumvents network distribution costs – the larger part of a consumer bill – the more equitable alternative is large-scale solar (e.g. AGL’s NSW power plants) providing the asset owner is prepared to accept the commercial risks of wholesale competition. Your comments make the assumption that the network utilities remain dangerously asleep and are unable or unwilling to improve their efficiency. Maybe, union-dominated bureaucracies will prove too lazy to save themselves, but I do not buy the death spiral story yet.
The renewable energy advocates have blocked substantial progress on cutting CO2 for the past 20 years and are continuing to do so. Locking in the RET will delay substantial progress for another 20 or 30 years.
Renewables cannot have much impact on cutting global GHG emissions: http://www.onlineopinion.com.au/view.asp?article=17447
http://euanmearns.com/the-difficulties-of-powering-the-modern-world-with-renewables/
http://bravenewclimate.com/2015/06/05/less-than-the-sum-of-its-parts-rethinking-all-of-the-above-clean-energy/
http://bravenewclimate.com/2014/08/22/catch-22-of-energy-storage/.
Only nuclear can have a substantial on cutting the emissions intensity of electricity. Anyone who has crunched the numbers knows this. Rational analysts have know this and have been showing it for at least the past 25 years. But the RET is forcing renewable capacity into our market and it will block the possibility of substantial emissions reductions from electricity for 20 or 30 years. Even once the public and voters see through the renewable energy ideologue’s dogma, it will take another 20 to 40 years for the impact of the irrational policies that have incentivised renewable energy to wash out of the system. That’s how much damage the green ideologues are doing. They either don’t realise or don’t care.
“How much does it cost to reduce carbon emissions? A primer on electricity infrastructure planning in the Age of Climate Change” (see Slide 10, and note the irony in slide 14 (this is what the RE advocates are unwittingly advocating for):
http://canadianenergyissues.com/2014/01/29/how-much-does-it-cost-to-reduce-carbon-emissions-a-primer-on-electricity-infrastructure-planning-in-the-age-of-climate-change/
Miles Prosser,
Thanks you. Excellent post. Very informative. From my perspective, the sooner governments sot trying to pick winners (e.g. pro-renewables, anti-nuclear, market interventionists) the better for everyone (other than who like to live off government imposed subsidies).
Would you be happy if nuclear was included in the RET?
How is Nyngan plant going now ? I haven’t noticed any production from it in the NEM reports lately.
Malcolm,
Sounds like you really need a copy of NEM-Watch (www.NEM-Watch.info) or NEM-Review (www.NEM-Review.info) so you can explore your question, at your own leisure.
There are ups, and there are downs (or is there a question behind your question about specific downs?)
Paul
Perhaps I’m missing something, but there seems to be a problem with the final figure.
Since 2005 (but excluding 2015), Australia has installed over 4 GW of PV, and just under 4 GW of wind. That’s an annual average installment rate of about 400 MW each, or 800 MW in total.
The columns in that chart are clearly much less than this (perhaps 300 MW average?). Even it it excludes PV, it still seems to be wrong.
The columns in the chart are only for large scale and add up to just short of 4 GW through to 2015 so the numbers tally fairly well to your figures.
Do you really think people want power companies putting extra cycles on their expensive EV/hybrid batteries?
Hi Paul,
thanks for your informative comments.
I’m interested in the Adelaide session. Could you please provide registration details?
Cheers
Hang on, so SA’s decrepit thermal plant is unavailable and the state has a dearth of fuel reserves but wind is to blame for higher prices?
The fact is that if none of the wind plant had been built prices in SA would be even higher. The reason so much wind capacity was rolled out there rather than in VIC or TAS is precisely because of the high average prices.
Hi Paul,
Interesting reading as always.
Can you please send me details of the Melbourne session?
Thanks
Hello Paul,
Can you please send details of the Brisbane session.
Thanks
Sam
I am in Ireland. I’ve estimated the cost to power Ireland entirely with wind turbines and pumped hydro energy storage. The estimated capital is 400-1500 billion Euro Their electricity consumption has been 26 TWh per year for the past 3 years.
Thanks Paul, very interesting.
On wind farm location: unfortunately SA price is still very attractive. Weathering the occasional bout of low price is still worth it.
Last month dispatch weighted price for some SA farms; Snowtown $40, Lake Bonney2 $49, North Brown Hill $43. Average SA price was $73. “A rising tide floats all boats” etc.
In NSW Boco Rocks got $38 and NSW avg. was $39. Portland got $32 and VIC was $34. Musselroe got $31 and TAS avg. was $35.
One of our readers has let us know the following:
With respect to Comparator Websites, the ACCC has released this informative guide:
It is going to be interesting to see how the combination of PVT’s and Tesla Powerwalls are going to influence energy useage. I came to your site here to see if there was a relationship between the daily timing of PV and wind, and as expected wind lags solar with the daily wind peak being in the late afternoon. I’d be interested to know if the APVI solar data includes self use energy or is it the total rootop energy production? The powerwall is going to blur the data as it takes the dc from the panels and the inverter only sees the energy as it is drawn from the system. It is at the inverter that the wifi link reports the energy output from the rooftop system. So the powerwall will serve to better represent energy consumption timing. However as thermal energy from PVT’s can be estimated from the electrical output from the panels, the Powerwall will mask that energy throughput. As the thermal energy from PVT’s is more than double that of the electrical output of rooftop systems this is a significant influence which will then only be visible in the drop in offpeak energy draw from the grid.
Disclosure: BilB has been banned from commenting at websites such as BraveNewClimate and Catallaxy
I think David Osmond has covered much of the territory.
But. Remeber that we are only at the beginning of the energy innovation revolution. It was just a few months ago that Elon Musk announced his Powerwall, and although there are none in Australia yet we are already counting the impact that this new technology will have. Similarly PVT’s are on in their infancy, and practical affordable backup power generation has not yet emerged. I am championing the Liquid Piston 70cc rotary engine out of Connecticut as being the disruptive technology that will fill this gap for residential and small business. This engine fueled by natural gas (methane) will produce a maximum 3.5 kilowatts, but is ideally suited for a steady 2.5 kilowatt output. So for low solar periods this engine will bridge the electrical energy gap feeding DC energy directly into the Powerwall and its consequent waste heat at 5 kilowatts directly into water heating filling in for the thermal output from the PVT’s. Where energy can be used in this balanced way the Liquid Piston engine is operating with 80% and above efficiency from methane gas. My estimation is that such backup will be used up to 600 hours per year in the average residential or small business premises.
The question naturally arises, “where does the methane come from if we are to have a 100% decarbonised economy?”. Good question. as it turns out, according to Victorian government figures, each Victorian disposes of 2 tonnes of waste material per year, more than 50% of which is waste paper and cardboard ie cellulosic material that comes from trees. This amount of material, if gassified and recombined into methane gas produces enough gas to generate 15,000 kwhrs per year per household of 4 people. One can derate that figure in as many ways as one likes and it still provides the complete answer to the low solar period problem, and from a renewable sustainable source.
It is my expectation that by 2030 rooftop solar will be predominately 4.5/9.0 kilowatt PVT’s Powerwall and Liquid Piston engine backed up to the extent that there will be sufficient over capacity to smooth out all but catastrophic climate events. There is one other variable that is yet to come into play, and that is the EV, or from my expectation the plug in hybrid PHEV. Such vehicles fit well into the 4.5/9.0 PVT household as such a system has the capacity to charge 2 such 8.5kwhr batteried vehicles 365 days a year. How ever in the event of a major low solar energy period there will be the option to not charge the vehicles and run on their fossil fuel or even their more sustainable bio diesel fuel.
The short and the long answer really is that zero wind for a number of days is a problem for now, but not in the long term. Perhaps the real issue is how will industry, high rise residential, and high rise commercial resolve their energy problems?
Hi Paul, great article.
Just a guess – but perhaps it is due to how the projects are hedged. For example, if hedging using the exchange, the price at contract expiry is what matters (which is both unrelated to when you produce, and is sort of ‘balanced’ by the non-windy prices). Having said that, I understand that the liquidity in the South Australian futures market is pretty low – and wind farms are pretty unlikely to hedge with the exchange (for other reasons).
Related to this – a poorly designed / implemented CfD (and/or PPA) might have a similar effect. Simply going for the lowest price CfD (rather than lowest cost) would likely encourage wind farms in regions with good wind resource / high capacity factor and favorable planning laws (i.e. South Australia). The price might be the lowest here – but that says nothing about the difference between this price and the value in the market (and hence cost). For example, a higher priced CfD in NSW that was perhaps anti-correlated with wind in SA might end up costing less much, due to a price premium in the wholesale market.
This is all speculation of course – and presumably the people who design such things have already thought of this!
You’re still a young fellow Paul!
Moving through my fourth decade in the industry…, I see new energy technologies and energy market uncertainty compelling both residential and business consumers to respond by taking increasing control of their own destiny. In doing this these consumers become, what we call ”Proactive-Consumers” or “Prosumers”.
We see this evolution every day. Over the past 8 years our company focus has been on helping businesses enter contracts for grid supplied energy and arranging the best available network tariff, but of late our business is increasingly about helping businesses run the maths over alternative energy supplies, solar PV, battery storage, negotiating energy buy back agreements and in some cases designing and costing complete grid disconnection.
Even in Tassie roof top solar now amounts to over 80MW installed capacity, so consumers have invested probably $150M over the past 10 years. This tells us consumers are taking increasing interest in, and control of, their energy choices. Therefore overtime the traditional energy supply business model and political intervention will increasingly loose, or at least drastically change, in its ability to influence consumers.
When we discuss Network tariff reforms or investment in more centralised generation, we also can’t help but wonder if some incumbent businesses are planning for future scenarios which may not ever materialise. The challenge we think is rather to describe likely future scenarios and then plan around these alternative futures.
Cheers
Marc
Only in my second decade in and around the NEM, but have also seen the same cycles. The basic problem is that the NEW capex is set at the start of the process based on forecast demand… and forecasts are always wrong. Having agreed that an investment is “required” the regulator can’t very well turn around and say “oops sorry, we were all wrong and now you don’t get to make a return on that capital”. On the other side, regulated prices are intended to prevent networks exercising monopoly power in setting prices. I think it is a fair trade off to allow networks to recover a a “fair” return on those “good faith” investments that all sides have agreed are likely to be needed. Yes if it was a “real” market the investment would be stranded and not earn a return, but on the other side if left to that same market, network providers would address that stranding risk by shortening the payback period on that capital expenditure – resulting in higher network charges anyway. You essentially can’t argue for regulated cap to protect consumers on one side and then argue against that same regulatory system protecting the return on capital for the investor on the other. Benefit sharing is as good a mechanism to “correct” these forecast errors as they become evident, as the operator is incentivised to not spend capex/opex they don’t require as that (lack of) need becomes apparent.
Thanks Mervyn.
What’s the process of engineering variation into switching times?
How would you propose that SA’s controlled load peak (sometimes the daily peak load), be flattened?
Great article Paul! I think the ROE diagram you were looking for is here: http://www.businessspectator.com.au/article/2015/8/6/solar-energy/power-industry-proposes-new-age-entitlement
Sorry I didn’t realise that link was subscriber-only. Here’s a link to the image itself: http://www.businessspectator.com.au/sites/default/files/network%20returns.png
Thanks Rob
A very similar pattern will emerge in Qld with solar PV (farms) and as they start to ramp up and build more capacity towards their 50% target. Spot prices here are already feeling effect of PV, and as growth in PV continues (both small and eventually large-scale), spot prices will continue playing less of a contribution to revenue-earning side of the project.
Re Alan’s comment, for PV + battery there is one recharge-discharge cycle per day to recoupe a battery or pumped hydro investment, and combined with wind or coal for night charging there are 2 cycles. Periods of high wind occur on approximately a 3-day cycle, so it would take much longer to recoup a battery investment.
Queensland already has one pumped-storage hydro station that could reduce some of this volatility.
Re the article, some of the South Australian discounts will spill across to Victoria when the Heywood interconnector upgrade is completed (190 MW of additional capacity). Since the main Vic and SA wind farms have highly correlated wind speeds, this will increase the discounts for Vic wind farms.
Paul McArdle,
You might find the summary post of interest:
“Renewables future: A Summary of findings”
http://euanmearns.com/the-renewables-future-a-summary-of-findings/
This post begins:
“Since February 2014 I find that I’ve published 24 posts on renewable energy here on Energy Matters (linked to in order of appearance at the end of the post) . In them I’ve written about wind, solar and tidal power, hydro, biogas, hydrogen and methane, CO2 emissions, interconnectors, exports and imports, energy storage, load management, backup capacity and ramp rates, the UK, France, Germany, Norway, California, remote islands like Eigg and El Hierro, the world as a whole and even mythical places like Atlantis and the Island of Denmark. I’ve reviewed “energy future” plans formulated by others, such as DECC, the National Grid and France’s ADEME and even come up with some plans of my own. I can’t think of any stone I’ve left unturned, or for that matter anything more I can write about for the time being.
So, time to sum up.
The ongoing transition to renewables – is it leading us into a clean, green, sustainable energy future, or will it leave us freezing in the dark? Or will it do neither?”
This post, and some of the other more recent ones listed in the 24 previous posts may prompt so thoughts you may not have otherwise had.
I found posts 18, 19 and 23 especially interesting. No 23 is:
“Decarbonizing UK Electricity Generation – Five Options That Will Work”
http://euanmearns.com/decarbonizing-uk-electricity-generation-five-options-that-will-work/
Thanks Peter, there’s certainly a few linked there.
Paul, what value do you use for our present (2014-15) energy output for wind (1W) and solar (1S), in GWh?
Hi John,
Numbers requested are as follows:
1) Wind in the W1 (i.e. current for 2014-15) totals 9,406GWh
2) Solar in the S1 (i.e. current for 2014-15) totals 5,068GWh. Note the correction I have added above to clarify that it’s actually just the small-scale series I used as “solar” to avoid the bump late in the FY when large-scale begins.
Paul
Nice one Paul. It seems to me that we are likely to start having significant amounts of battery storage in the coming decade, particularly in homes with PV. As a first guess, say 1 million homes x 10kWh = 10 GWh.
It would be interesting to see how the flow duration curves would change if you model this storage. A simple battery algorithm such as store whenever energy would otherwise get spilled, and release as soon as generation reduces below consumption. Since spillage only starts around 6W+6S, you’d only have to model this scenario upwards.
Perhaps repeat with 20 & 50 GWh of storage.
Of course, some of this could also be done with existing pumped hydro instead.
cheers
Thanks David
Yes, this type of “next step” is already underway.
There are virtually an infinite number of permutations for how storage might be used (especially considering all the different actors who will have a finger in the pie, and their various incentive). I had already opted for a simplistic approach, such as what you suggest, but realise that it will be quite different than how it will play out in reality.
Paul
Another simpler suggestion is that when looking at larger wind capacities, you use a much larger scale factor for NSW than for SA. It is certainly unlikely that SA will increase capacity by 10x, but that may be quite possible for NSW (and you could also use NSW as a proxy for QLD additions too).
This will help increase geographic diversity, which is currently dominated by SA & Vic
Because my focus is on “NEM-wide” in this analysis, I had already decided not to do this kind of thing. Fundamentally it would open up another degree of freedom in the analysis that would make it even harder to produce some useful results in the time available.
Thanks for the suggestion,though, David
Paul
If you scaled the renewable generation as a function of state demand, then you wouldn’t add an extra degree of freedom. ie, if NSW has 35% of annual NEM demand, then scale NSW wind so it provides 35% of wind generation and likewise for solar. Unfortunately that still leaves a problem of how to model QLD wind, perhaps using NSW wind as a proxy may the best you can do.
Hi Paul, you’ve mentioned what the spillage is in terms of MW, and % of time. I’d also be keen to see what it is in terms of GWh, or even better, % of renewable generation (GWh spilled / GWh of renewable generation).
This is the big question. If you spill 10% of potential production that only increases cost 11%.
Love how you tied solar & wind to the baseload label, well done.
Seems to me that the accuracy in weather forecasting will be the key factor in not just the future electricity market, but also in industry (if they adjust to the peakiness of the market).
I would add another factor for the Alinta decisions: a rather poor (and unjustified) design choice. Specifically 50 MW capacity with 15 hours of storage. That is terrible configuration to asses at this point in time. The same size field / plant (albeit with a large power block and smaller amount of storage) – say 150 MW with 5 hours of storage, would be much more sensible (and would cost a similar amount).
As a side note – the recently constructed solar reserve plant (U.S.) has a 110MW/10hr configuration, which makes more sense in that market, (it was able to secure a PPA for $135/MWh as well).
Perhaps this would still be non-viable – due to the landscape considerations Paul mentioned. But given the price premium for being for being able to sell in the evening peak (or at times of low wind, and/or sell more on demand energy relative to wind) – it would be quite close.
Paul
How right you are about the need for better ramp rates in the “new electricity supply world”. In its response to the AEMC review of the AER ramp rate rule change, the Major Energy Users commented to the AEMC that the AEMC should be looking to the minimum ramp rates needed for the future market rather than setting ramp rates that could be provided by the existing fleet. By mandating a higher ramp rate requirement on generation plant, the AEMC would recognise the need for more agile plant in the future but could also provide a mechanism for allowing exemptions for existing low ramp rate plant.
But the AEMC persisted with a low ramp rate minimum in its preferred rule change
Some of our rapid-response fleet has come from hydro, and some from gas. An ideal gas plant would have its own storage facility nearby so it is not at the mercy of the spot gas market. The only one like this I am aware of like this is Mortlake, which has a nearby gasfield as its storage facility a start-up time of less than 5-minutes.
An interesting scenario is emerging for this coming season for drought-affected Tasmania. Their 2 main storages – Great Lake and Lake Gordon – have shown virtually no net replenishment since their power stations were run hard until the end of the carbon tax in July last year. Has the owner of Hydro Tasmania (the Tasmanian government) been asking too much in dividends, rather than sparing these dams by more imports from Victoria ? If water levels get close to the low supply levels on the intake structures, they may be in some trouble which they may blame on the drought, but is really because of their own management. Meanwhile Snowy Hydro has managed its resources much more conservatively.
Re: Malcolm. I think line-packing is pretty common for a lot of gas generators in Australia – which is really just storage by another name.
Off topic, but an issue of the Portland Observer newspaper last week reported the transfer of a 250 tonne transformer from the port of Portland to the Heywood switching yard. This is a key part of the Heywood Interconnector upgrade, which will add 190 MW of capacity. According to the AEMO website, it is due to go live in July 2016, but if the transformer is almost in place, perhaps it will go live early.
This upgrade is a key reason for March closure of the coal-fired Northern power station, because it will enable additional cheap brown coal power from Victoria into the South Australian market. It will also enable additional wind power for SA into the Victorian market at times of excess generation.
Okay, I’m in suspense. So what did cause the significant drop in demand in NSW?
…it was Tomago Aluminium.
Thanks Tom
That’s my understanding also – however I did not know, for sure, if that information was public knowledge hence was not going to be the first to state so.
Paul
Th e story portrayed by the aluminium industry is that they couldn’t possibly participate in any load shedding at any time, because then all their pot lines would need to be emptied out of solid aluminium. Yet when when the problem is their side of the meter it doesn’t seem to be a problem.
Paul,
I think that you have incorrectly characterised the issue that Snowy are flagging in their Rule change.
You have suggested that the the delta between CLEAREDSUPPLY and INITIALSUPPLY represents how much demand has varied from what AEMO expected over that 5-minute period. This is not the case – in fact the difference between these two variables is exactly equal to the amount of demand change that AEMO have forecast for the 5-minute period.
The problem that Snowy have highlighted is that after AEMO issue the 5-minute price and dispatch targets, non-scheduled loads (and generators) can suddenly change their consumption/supply, which is seen by the NEM as an unexpected change in demand. The scheduled generators however are required to continue to ramp to their 5-minute dispatch target.
As an example, suppose that AEMO forecast that the demand in a region is expected to increase by 20MW in the next 5-minutes, and calculate 5-minute price of $300. AEMO issue a dispatch instruction to the marginal generator in that region to increase its output over the 5-minutes period by 20MW to meet the expected demand increase. The marginal generator then commences increasing its output to achieve a 20MW increase by the end of the 5-minute period.
However, other non-scheduled entities note the high 5-minute price ($300), and choose to reduce their consumption in aggregate by 50MW. This demand reduction was not anticipated by AEMO, and despite the fact that the demand is now 50MW below what was expected, the marginal generator needs to continue increasing according to its dispatch target (or risk being declared as non-conforming).
AEMO need to balance the power system supply and demand, and they do this by reducing the output of the generators assigned to frequency control.
At the end of the 5-minute period, AEMO then “capture” the fact that the demand reduction of 50MW has occurred, and then ask the marginal generator to ramp back down again.
The intend of the Snowy rule change is enable AEMO to anticipate the 50MW demand change, and prevent the unnecessary and inefficient ramping up and down of marginal generators and frequency control services, and the resultant inaccurate pricing forecasts.
Thanks for the input, Chris
I’ve added a PS to the post to explain my method:
https://wattclarity.com.au//2015/11/puzzling-through-three-rule-change-requests/#method
I’m genuinely interested to understand what I have done wrong?!
Paul
PS Unfortunately I’m heading on leave today for 3 weeks so will be slower to respond than would normally be the case…
Do these calculations include the impact of reduced Tasmanian hydro inflows on the supply of certificates ? We have now had 2 years of low inflows, and over the next 5 years or so the system will need to be run much more conservatively to allow storages to build up from their current 26%. It may mean that an additional ~500 MW of wind capacity (over and above the forecasts above) would be required simply to replace the missing hydro certificates.
Thank you Malcolm, that is a good point you’ve raised. The basis of the future requirement is calendar year 2014 data, which includes half a year in which carbon pricing encouraged strong hydro outputs.
It would be reasonable to assume lower future hydro outputs in the near term as you note; how it varies as we get nearer to the higher LGC targets in 2020 is unclear. If the 2014 performance is higher than long term hydro outputs, that would make the target more challenging and need further new capacity as you point out.
Regards
Andrew
What you say makes it even more challenging than I first thought. While the carbon tax was in operation, Tasmania typically exported about 400 MW of hydro power to Victoria on close to a 24/7 basis. This was achieved by running down their storages, particularly the Gordon Dam. I can’t remember what the net flows were between the end of the carbon tax in July 2014 and the end of the 2014 calendar year, but for simplicity of argument, lets assume the net flow was zero. So for the whole of 2014, this would be an average of 200 MW. Now Tasmania is importing about 400 MW from Victoria on a 24/7 basis. We therefore have 200 + 400 = 600 MW less hydro generation than in 2014. To make up for the reduced certificates from this hydro using wind with a 33% capacity factor would require 600 / 0.33 = 1800 MW of wind capacity over and above the new capacity foreshadowed in the article.
Snowy Hydro has also been generating less than normal because of low inflows, which may require further wind capacity.
I’m sure there are arguments with the HEC as to whether the reduced inflows are cyclic or a more permanent reduction through climate change, as has occurred for rainfall in south-west WA. There would now be an imperative of security of supply, guarding against the possibility of a further year of low inflows in 2016. Tasmania can only import about 40% of its electrical energy requirements on Basslink, 10% is from wind, and there needs to be enough water in storage to supply the remaining 50% for perhaps a 2-year buffer period. So I expect the high level of import from Victoria to continue for at least the next 2-3 years, resulting in a further a shortfall in certificates.
The ACT target of 90% renewables is only going to make it more difficult for everyone else. LGCs from the ACT’s PPA sponsored generators are to be passed to the government and then cancelled in the same way that green energy commitments are met. ACT retailers will still need to purchase a number of LGCs. Renewable energy activists in the ACT have made sure that the target was additional to the amount of renewable energy generated through the RET scheme. If for some reason the ACT doesn’t end up with sufficient sponsored generation the government intends to buy LGCs on the market to make up any shortfall and thus achieve its target. ACT electricity consumption will be about 2.8 million megawatthours in 2020. The target also includes that PV electricity generated within Canberra.
Paul interesting analysis.
I would only add that everyone is paid on the half-hourly price and so may respond after a high price event. How do the graphs look if the trigger price uses the settled price rather than the dispatch price?
Ben
Thanks Ben
Been away on leave so no time to respond. The 5/30 issue does complicate this consideration in a number of different ways:
1) You note that generation (or negawatts) might respond after a 5-minute price, so long as it’s in the 1/2 hour. This is one permutation.
2) Another is that those participating in this way might try to pre-empt spikes by dispatching based on predispatch prices, gut feeling, or some other method.
I don’t have time to delve more presently, but will see if I can return to this in future…
Paul
Not sure the LNG sector will consume 20% of QLD’s electricity given that the three downstream plants on Curtis Island and GLNG upstream have their own gas turbines and are not connected to the grid
The LNG plants will certainly have their own off grid power for liquefaction process. I think the is expectation that some of the power required for upstream gas production and transmission may be taken from the grid.
What caused the spike in price? Was it a fall in wind production?
Demand spiked, Angaston (diesel) dropped 37 MW, and Ladbroke (gas, price setter) was constrained down due to network limitations. Wind dropped 5 MW the first event and 18 MW the second time. Small changes compared to diesel and demand.
And if it stays there now then I’m a happy man with regards to your competition!
But alas, it’s very early in the summer, and who knows what January and February still hold for us.
In an El Nino summer it seems fairly likely we’ll see demand higher than that one before too long…
Hi Paul,
Note that Tomago (800MW) was taken off just after the first $13800 5 minute price, just before the weather change. AGL ability to do this for a couple of hours is a significant tool to reduce VOLL in the market. Also may affect who wins a barbie!
Great web site.
Regards,
Rod
My understanding was the wind bids at the bottom of the stack (essentially -$1000/MWh) due to it being non-despatchable as per market design. That is to say that wind is a price taker, not a price maker. Its low operating costs have little to do with its position in the merit order for generation in the setting of the price in the market.
Feel free to correct me if I’m wrong.
Paul – it may be worth a commentary on Tasmanian electricity pricing. With Basslink down for the next three months there has been significant RRP volatility.
Thanks Farhad
It’s on the list – but a few too many balls in the air at present unfortunately.
In the meantime, not sure if you read Marc’s updates here:
http://www.goannaenergy.com.au/newslettersAndUpdates.htm
Paul
Thanks for sharing your thoughts, David.
You refer to the uptake of rooftop solar being a transfer of wealth from the lower to the middle and upper classes. It needs to be noted that it is the lower income households that are installing PV at a greater rate. Multiple reports/studies/actual statics analysing and comparing regions of PV uptake have found that it is the locations with lower income ratios that have the higher uptake of rooftop solar PV. Especially in recent years where there is low or no FiT for PV owners exporting surplus energy back to the grid, the transfer of wealth is actually the other way. These households pay $1,000s of their money for PV that has long payback in most cases, but has multiple positive externalities, e.g.social, environmental and economic benefits.
The increased supply of energy in the grid is contributing to maintaining a stable cost of energy, through negative pricing events and when aggregated and with large-scale RE generators have a depressing effect on wholesale prices -merit-order effect.
It is too simplistic to claim they are free-riding and having a (negative) wealth transfer.
I agree with the mandatory introduction of dynamic tariffs. How about making the transition attractive to home owners?
This could be done by mandating dynamic tariffs based upon coincident peak demand, instead of maximum site demand. Residential peak demand is less correlated to network coincident peak demand.
Or, by mandating that demand charges are uniform for all customers, e.g. 10 c/kW daily charge, irrespective of the site (residential, commercial, industrial)? The rate could be adjusted each year to ensure the networks make their regulated returns.
Or, by mandating that dynamic tariffs only be introduced if fixed (daily) charges are eliminated? This way, end-customers would be incentivised to install solar + storage systems that reduce their impact upon the grid. A reasonably efficient single family residence with a correctly sized solar + storage system could draw zero load from the grid during summer, and only 1 kW during winter (limited to off-peak times). The home owners could have a zero bill during summer, and an almost zero bill during winter. In the meantime, the contributors to coincident peak demand would pay for the grid, as they are the ones who determine how expensive the grid needs to be.
All good Mike, as long as you can swim.
Floaties ain’t gonna help ya in this pool.
Ha, thanks Peter.
That’s why I’ve built an ark and grown a beard to save a select few from the flood.
The effect you seem to have missed as a discrete element in the equation, I believe, is the increase in the rate of change of dispatched generation required late in summer afternoons as the peak demand arrives and the solar input decreases meaning that scheduled generation is required to respond more quickly. I think this would be more apparent if you looked just at SA demand where there is less industrial electrical load as a percentage, due to gas substitution, and the high impact in SA of air-conditioning demand which rises trailing the temperature increases. This means that late in the afternoon the demand rises more quickly as solar output falls away. Fortunately in SA we have relatively flexible generators due to the high penetration of CCGTs, OCGTs and even our gas fired steam generators are relatively flexible compared to large coal fired base-load generators. The Vic-SA interconnector also has a major effect but fortunately, due to the relatively small import required, does not tax the rate of change of interstate coal-fired generators.
In winter of course any solar generation will be all gone by late afternoon having reduced the output of the scheduled generation to a minimum or even to shut-down in early afternoon and then the evening peak will be entirely filled by scheduled generation and possibly wind. This will really test responsiveness and once NPS is OOS show the value of CCGTs.
Thanks for pointing this out, Shane
The increased rate-of-change was something that I looked at here:
https://wattclarity.com.au//2015/10/ramp-rates-for-unserved-demand-much-higher-in-this-future-world/
I understand it’s causing some head scratching for a number of people – such as what it might mean for frequency change and hence FCAS requirements.
Paul
The Catagunya diesel “power station” has been registered with AEMO so production data will be available via the normal means from AEMO.
Thanks Shaun
Will have a look – does not necessarily mean there will be data published – but, if it is, will be fed through to:
1) The NEM-watch Widget (http://reneweconomy.com.au/nem-watch)
2) The new (internationally focused) LiveGen Widget (http://data.reneweconomy.com/LiveGen)
3) The full NEM-Watch product (http://www.NEM-Watch.info)
4) NEM-Review, for historical analysis (http://www.NEM-Review.info); and
5) Other products we have
Regards
Paul
Data is being published for the Catagunya site now “CATAGUN1” is what you’re looking for.
Thanks Shaun
Data is being shown under “Liquid” in the NEM-Watch Widget located here, and a growing number of other places:
http://reneweconomy.com.au/nem-watch
Paul
Hydro Tasmania must have been under a lot of pressure from the Tasmanian government to produce short-term dividends rather than let the dams fill up. With the Gordon Dam so far below its maximum head, each unit of flow is generating much less power than normal. To operate at its maximum rather than its current head would be equivalent of another 120 MW wind farm, with no additional capital or operating costs.
It is understandable that the big users who have been asked to cut demand would take a hard-nosed bargaining position. Hydro was willing to take short term profits at the expense of the longer term security of supply, so why should the large consumers forgo profits too ? Perhaps they have agreed to short term reductions in return for extension of their low power purchase prices.
Thanks Malcolm
Whilst there are anecdotal claims by a number of people along the lines of:
I’m not sure there is anything on the record we can point to about that?
Paul
Hi Paul, thanks for a well-researched summary of the current Tas energy situation.
It may be worth noting the environmental consequences of the requirement to draw the storages down so low and keep them there – many are now below the environmental risk thresholds developed by HT itself and may be sufficient to cause long-term ecological changes. Do you know if the environmental flow for the Gordon River (world heritage area) is still being delivered with Lake Gordon storage levels so low?
Lower output per cumec at low levels increases the costs, as well as potentially damaging turbines as sediment loads increase. Nail-biting times for HT, the Government, industry, the Tas population and some of Australia’s most beautiful places.
Hi Paul,
Thanks for the article, i have been looking at the Moyle connector between scotland and northern island. Installed in 2002, it has been running at half power since 2012, due to repeated problems in the polyethylene insulator for the return layer, see here for example:
https://www.sell2wales.gov.wales/search/show/search_view.aspx?ID=JUN037964
http://www.bbc.com/news/uk-northern-ireland-16434031
http://www.bbc.com/news/uk-northern-ireland-21277249
http://www.4coffshore.com/windfarms/interconnector-moyle-icid20.html
It looks like now they have had to lay a new separate return layer now to try and get cable working again to full capacity.
Any comment for above relative to basslink? ie is it same technology, both have a fault date 10 years after entering service and a similar looking fault type.
I think the work of Amory Lovins and RMI would be of use here; I’m thinking of “Small Is Profitable” and “Natural Capital”.
Small is Profitable covers off how distributed resources can add value to an existing network so maybe a bit off the end game. “Natural Capital” spends some time discussing the concept of energy as a service, which us how I would see a massively atomised electricity market playing out – even if the energy is free there is still a cost to managing it yourself and there will always be a capital cost. Its not realistic to think even the engaged prosumer will be across the technical details of managing and maintaining their energy source and storage system… there will be a big role for those who can provide the service, trouble free. Whether thats on a unit basis or flat rate like rent will depend on how these things scale.
Thank you for this interesting exploration, Paul.
Given that concerns about climate change were mentioned in 3a, I would propose a Q6 section: “Would it address the need for strong climate action?”
I’ve only been doing energy/climate for a few years but I’ve noticed a severe lack of analysis around the prevailing and near-totally unspoken assumption that intermittent renewable sources + storage = effective climate action. Emphasis is always on the liberation of the consumer or breaking the stranglehold of utilities, with perhaps mention of how clean it would be. Well, how clean exactly?
Earlier this year I finally found some work from Stanford. http://gcep.stanford.edu/pdfs/events/workshops/Barnhart_NEA%20Stanford%20April%202015.pdf As you can see on slide 27 the projected lifetime emissions intensity of solar power supported by the front-runner of distributed storage technologies – lithium ion – is pushing 200 gCO2/kWh (I think “kg” is a typo). There are examples of conventional national and regional electric grids which operate at far lower than this intensity.
So a “rapid decreased in lifetime emissions intensity of Distributed Solar and Storage” would ideally be added to the list of assumptions here.
An interesting question. I expect electricity won’t be “free” for many years, apart from short periods when there is excess from renewables, because there is too much market inertia in the system. A fully off-grid house needs access to sufficient sunlight for house power and probably to charge an electric car. Pay for an up-front capital cost, and the rest is “free”. Suitable for rural and some suburban areas, but there simply isn’t enough sunlight to do this for most apartments. There is little incentive for these sorts of investments in rental properties. Furthermore, residential demand is only about 20% of Australia’s total, so the real action is in the commercial, retail and industrial sectors. These have their own paradigms, and while there is already a good return on investment for solar panels within such businesses, most organisations aren’t set up with the internal investment structure and empowered internal proponents to make this happen. So I would expect a slow gradual decline in grid power demand.
Paul, on your comment re the possibility of corporate subsidies, here’s a link to how the arguments around that are playing out in Kenya:
https://www.greentechmedia.com/articles/read/Could-a-Solar-Giveaway-in-Kenya-Destroy-the-Countrys-Burgeoning-Off-Grid-S
Some of our readers will find this article by Blad Plumer on Vox Energy & Environment provides an excellent discussion of some of the factors that need to be kept in mind relating to forecasts for how much, and how quickly, solar PV costs will decline, as it gains market share.
Great to see the report on the closure but its a shame the Port Augusta solar project hasn’t got up yet. Fingers crossed for those workers to jobs in the new economy.
Prof. Ledwich, you forgot to mention the role of Energy Efficiency.
Much better Building Energy Efficiency Standards than the prevalent 6-star standard – which is not checked after the build by the way – would help a lot to bring down the peaks. 8-star homes should be mandated and insulating homes should be incentivised again.
There are two main factors contributing to the peak:,
The first is aircon use to bring down the temperature in these inefficient ‘glorified tents’ as they were also called. An efficient home like a 9-star home needs minimal cooling, meaning a much lower peak.
The second culprit is cooking. Unfortunately with cooking nothing much can be done to reduce the peak, except for mandating efficient appliances. An efficient fridge is using 100 Watts whereas an old inefficient one can use up to 400 Watts. The same goes for ovens etc.
I agree that control of hot water and pool pumps during peak times (or mandating efficient pool pumps) is also necessary. Hot Water could mandatorily be heated during the day which the ‘duck shaped demand curve’ above shows is reasonable. Unfortunately the Newman Government also got rid of the heat pump or solar hot water requirement for new housing.
I disagree that lighting and TV use is contributing much to the peak
They would possibly contribute a maximum 600 Watts to the peak, depending on the house of course (except for plasma TV’s – not very common nowadays). LED downlights bring that value down quite a bit as well.
Cooking and aircon use however would draw 5000 W and on average around 2000 Watts.
Another factor is diminishing the willingness of people to save at peak times.
Because daily network charges are so high now – at least in Qld – saving electricity doesn’t really save the customers a lot anymore. Going back to low daily charges would be a lot like TOU charges because every kWh used would in this case signal that it costs them more to use more without having TOU rates. This isn’t the case anymore with high daily connection charges. My bill is over 50% made up of these charges.
Economists normally value the impact of such an event from the “resource cost” implications, i.e. the additional input costs to the industry (such as fuel and hired plant), plus the forgone value of any lost sales. For an event like this one, the resource costs will be much lower than simply multiplying the marginal spot price inflation by demand.
Thanks Ben
Your comment reminds me of the distinction between “Market Benefits” and “Customer Benefits” that have perplexed many who’ve looked to justify an upgrade of QNI over the years through the Regulatory Test.
I appreciate the distinction you make, however believe that the broader community will think of the cost of the Tassie crisis in terms of increase in costs paid collectively – independent of whether an economist might class some as “wealth transfers”.
Paul
A few key main suspects for me Paul:
– Liddell issues early April had quite an signicant, extended impact on supply, combined with some other unreliable baseloaders (Eraring in particular) and a heavy outage season (majors at Kogan, Darling Downs and Vales). Note the PDC for VIC reflecting separation overnight from northern states, overnight was particularly affected by low baseload availability (particularly with some participants bringing down plant at their lowest position eg Eraring clinkering overnight, similarly of late Loy Yangs with coal issues conserving during the day/overnight such that they can run morning/evening peak)
– Repricing of TIPS/SA gas. The Brooklyn compressor issue earlier in the year limited participant ability to fill storage and the impact is most notable in the gas market. Whereas LNG has capped the market in the past we’ve seen a number of prices into the $20 and even $30 range. The TIPS pricing in the $400 range is more than the spark spread but suspect they might be attempting to conserve gas to last out winter and a combination of transport considerations
– And as everyone has been talking of, the impact of LNG trains in QLD. There has been an increase in load from upstream compression, combined with intermediate gas generation largely limited to running only morning/evening peaks
A very important question and I wonder who has the most to lose and who has the most to gain?
I wonder if this is fair statement
Traditional suppliers/ Distributors = Big loss
retailers/customers/country = Big Win
I don’t know much about the energy markets but this seems like a likely equation. The question is what will the retail market look like, clearly it will need to be more IT savey.
Hi Paul
The information in these graphs is really interesting particularly the Distribution of Prices for each State. Have you put together similar graphs for the Q1 prices or Annual prices. I looked through the Archives but could not spot anything.
TB
Ben’s comment about resource costs is well made (as is Paul’s about the opportunity cost of consumption foregone by customers who reduced their usage). The other thing to remember about simple spot price (uplift) x volume calculations is that they miss the impact of financial contracts between buyers and sellers (predominantly Hydro Tas) which would substantially mitigate the net purchase payments of buyers. In looking at the effective additional financial cost to buyers (as a whole) one would really need to know their netspot electricity exposures and multiply those significantly smaller volumes by the assessed spot price increase. Of course there is no publicly available data on the size of these net exposures but it would be safe to assume that they are much smaller than gross demand volumes.
Interesting comparison would be WA, which has a portion of its gas allocated to domestic users (as have all other LNG exporters in the world except on the Australian east coast). The WA spot power prices on the AEMO website are reminiscent of a year ago on the east coast.
Well, since Stages 1 and 2 are a total of 64 turbines
http://corporate.siemens.com.au/en/home/news-centre/press-releases/hornsdale-wind-farm-has-secured-financing-for-stage-2.html
And the official project is 100 turbines
http://hornsdalewindfarm.com.au/
I’d say we can look forward to another 36 turbine 3rd Stage, and more charts and serious commentary from WattClarity.
There’s something interesting I’ve noticed – someone will pick a few large-sounding SA price intervals, then present the penetration of wind in that state, and then conclude that the high prices are caused by wind.
Considering this strange phenomenon is occurring to a similar degree in all NEM regions, is it fair to assign the penetration of wind as a necessary factor in this phenomenon?
IE: if the penetration of this specific fuel type was lower or zero, would this phenomenon cease to exist? A hypothetical worth thinking about, considering how this has now taken the form of an unquestioned assumption, in some media outlets.
Cheers,
Ketan
With regards to wind correlation across the distances you cite, I think Dr Morgan’s analysis is as independent and thorough as we might get right now.
https://bravenewclimate.com/2015/11/08/the-capacity-factor-of-wind/
Surprisingly good analysis.
He states the cost of curtailment as fact though, without any supporting information.
Curtailing a few % of potential production makes very little difference to lifetime cost of energy. See the maximum generation periods in winter and how little duration there is.
‘Innocence’ and ‘guilt’ are such odd concepts, when you think about it – but they’ve been created for this issue due entirely for the need for a collection of media outlets to campaign in a way that protects incumbent fossil fuel interests.
I try to think about the logic of the claims. If we had a system of all gas, and no wind, the problem would still exist. If we had a system of some wind and some gas, the problem exists.
So, wind is necessary in the emergence of this phenomenon – as you point out, periods of low wind result in the dominance of gas, which is expensive, and so price goes up. But gas’s contribution to the issue – namely, high cost, is both necessary and sufficient – if the fantasies came true and we took an axe to any generation facility that doesn’t rely solely on pollution-intensive fuels, we wouldn’t be living in a wonderful land of low prices and bountiful supply (and emissions would be high, which is a bad thing).
Hence, my tweet pointing out that laying the blame solely on a [necessary but insufficient] factor, and entirely ignoring the [necessary and sufficient] factor is wrong.
As you say, there’s no getting around the fact that there are times when wind output is low across the entire South Australian fleet. As you also say, gas prices are high. So, the most constructive way forward would be to openly and honestly deal with *both* of these components, in a way that doesn’t mean a return to carbon intensive fuels, simply because we can’t be bothered solving this problem.
I guess it’s worth pointing something else out – spreading misinformation about the national electricity market is a direct contributor to problems that can arise out of the system, as decision-makers are influenced by this information.
In old media, statements are declarative, no evidence is ever included, and challenges occur over the course of days, long after the original declarations are assumed to be correct. In new media, even if misinformation springs up, at least these can be challenged or criticised (as you’ve done with my tweets above).
A major factor in the flurry of falsehoods that have emerged is traditional media outlets failing to understand the nuances of the system they’re talking about. The end result will almost certainly be a worsening of the problems they’re concerned about, with the added effect of destroying a new Australian industry. I feel that this impact is under-discussed – everyone just seems to shrug and accept it.
Actually, I don’t think Hornsdale will help very much. The problem is not when Hornsdale can produce, all the other wind farms can produce then and there is plenty of electricty. The problem is that when Hornsdale is not producing all the other wind farms don’t produce either and there’s not enough electricity in SA.
It’s the fundamental problem with intermittentancy that people have been carping on about for a long time. Putting more wind farms in SA won’t fix it.
I see this as a self-inflicted problem, with the best solution simply being to not inflict the problem, i.e. distributed generators should simply reduce local demand, they should not be allowed to feed any surplus into the grid. Any surplus can be stored in the batteries that renewable enthusiasts insist are not merely myths.
Energy price remains in the negative for SA, probably due to all the wind. Pelican point is running, my guess is as it can make money on the FCAS support? Osborne is off which is the first time for awhile but QPS5 (also Origin) is now running. I have a vague recollection that QPS5 is registered for FCAS but not Osborne, but not 100% sure
As a quick PS to my article this morning, I’ve included the following comments from an anonymous (and more learned) reader.
The reader emailed these comments in to help me, but I thought they would be more broadly useful. Thankfully they have agreed to let me post:
It’s even better/worse than that: SA Power Networks and ElectraNet are aware that during the day, tripping certain feeders with high PV penetrations might exacerbate the frequency imbalance, as some of these feeders have ‘reverse’ power flows on them. The problem is they don’t know which load points this is happening on, and it changes during the day!
One note – a shortage of FCAS-contingency services in SA (or any state) could be lessened by the AEMC approving and furthering the ‘ancillary services unbundling’ rule change, which they are due to comment on by 1-Sep.
Today loads (demand side) are technically able to offer the three FCAS Raise contingency services, but must be offered to market by the load’s retailer. Due to the complexity of a retailer administrating such a service and short retail contract durations, almost no demand side loads participate in the FCAS market today. AEMC notes that “with the exception of an aluminium smelter and two pumped hydro loads, all of the 20 registered participants to provide FCAS services in the NEM are generators.”
The A/S unbundling rule change would allow demand-side specialists (“DR aggregators”) to recruit and offer aggregated FCAS (raise), which should democratise the FCAS market, increasing demand-side participation and supply of FCAS, theoretically putting downward pressure on FCAS prices. In the context of SA, provision of increased FCAS from the demand side may lessen incidence of load shedding (UFLS).
I guess something worth not skirting around is the fact that challenges will be presented as insurmountable, catastrophic and civilisation-ending by a range of parties keen to ensure that decarbonisation does not occur.
In that context, it becomes extremely important to channel the inspirational and genuinely wonderful spirit of engineering and science, which is legitimately technology neutral and looks to correctly diagnose and remedy these quirks in a fairly optimistic and logical way. Which is why what you’ve highlighted above is so important – the difference between a ‘challenge’ and a ‘problem’.
Thanks very much for the analysis. My intuitive guess is that this correlation penalty is going to get worse as more Wind is added in Victoria as well since there will be less opportunities to export. My other intuition is that wind generators will increasingly suffer as residential solar + batteries takes off.
Fossil fuel generators have the capacity to tailor their production, at least to some degree, in accordance with predictive weather patterns but Wind generators will just keep competing with each other. Their only option will be substantial investment in their own storage technology (whatever that may be) but it’s not clear the CAPEX would be worth it.
It seems that the ACT electricity consumers are taking on all the risk of the PPAs signed with Horesdale Wind Farm, set to provide LGCs to the ACT government as an offset to the non-renewable supply fed into the ACT. The 1200GWh will represent about 40% of the ACTs electricity use. In order to lift the nominal renewable use to 100% the ACT government will surrender these LGCs.
John,
Your analysis of the ACT wind subsidy is spot on (if you can excuse the pun). Minister Corbell is spinning the line that ACT consumers are protected by a rising average spot price which closes the gap in the contract for difference. However the wind outputed weighted average spot price is declining (if it wasn’t then merit order effect would be another fantasy). ACT consumers will pay a much higher subsidy to SA wind farms than the government is predicting.
The media in the ACT is accepting the Government’s propaganda. Several op/eds have praised the canny Minister’s scheme.
I suggest it’s a combination of high labour costs and the increasing time and cost in bringing the aging generation units back into production after failures that are killing Hazelwood. For the last 3 weeks of September it produced little more than 1000 MW, or a 55% capacity factor. According to The Age article, there are 550 direct employees and a about 300 contractors, which is a lot of people to support from just 1000 MW. For comparison, AGL’s Loy Yang A has 500 direct employees, but in September it produced an average of about 1600 MW. Furthermore, its output has ranged from 1000 to 2000 MW, tending to be lower when total Victorian wind output was higher. AGL also operate the MacArthur wind farm (plus others), so they would not want to depress spot prices in a way that affects their other generation assets. For this reason, Loy Yang A would tend to be achieving a higher price for its output than Hazelwood, but also has the benefit of lower labour costs and lower maintenance costs.
The owners of Hazelwood (GDF Suez and Mitsui) also own the newer Loy Yang B, and would be acutely aware of the lower operating costs of the newer power station. They would be beneficiaries of less competition in the market for base load power after the closure of Hazelwood.
Federal and State Government can help the #LatrobeValley community in #Gippsland through a “Just Transition”
We are yet to meet with Josh Frydenberg (Federal Minister for the Enviroment and Energy).
We have already met with;
Daniel Andrews (Victorian Premier),
Lily D’Ambrosio (Minister for Energy, Environment & Climate Change, Suburban Development),
Wade Noonan (Minister for Industry & Employment, Minister for Resources),
Jaala Pulford (Minister for Agriculture),
Harriet Shing (MLC for Eastern Victoria), local Councilors and other Community leaders.
This has new in an attempt to help save this region from the uncertainty of the BrownCoal Energy industry transition specifically relating to Hazelwood and Yallourn Power Stations and Mines.
The initial feedback is good but the pressure will need to be maintained to ensure commitment of funds and resources to our area of Latrobe Valley (Morwell, Traralgon, Moe, Churchill, Yinnar, Glengarry, Tyers, Trafalgar, Yarragon and Warragul).
1. Transition Framework is needed to facilitate existing power workers moving from a closing power station or mine to an active one.
2. Create a BrownCoal Transition Group in Latrobe Valley to co-ordinate the closures in a planned way. The group membership needs to consist of Federal, State, local Council, Unions and Industry representatives.
3. Build a Genuine World Class, High Speed Rail (HSR) that operates between Sale – Traralgon – Warragul – Pakenham. Create a transfer Hub at Pakenham to enable integration and lower costs.
(Not Melbourne to Sydney that draws the negative involvement of the Airline Industry lobbyists and adds costs.)
3. Ensure local residents are given priority for jobs created by this process.
This will help maintain local housing prices and also council revenue through Rates which will be needed to help our community.
4. Build a new Gas pipeline from Sale’s Longford Gas plant next to the old one, through to Hazelwood and Yallourn Power Stations to allow conversion from Coal to Gas generation. This was fully researched in 2008.
This helps employment at both ends of the pipe and prevents the need to do coal seam gas or fracking.
5. Develop a Green Factory Centre dealing in Wind Turbines, Blades and other technology in Morwell behind MidValley where an existing rail siding exists and integrate it with the high speed rail (HSR) to be used at night and during off peak periods.
This has been done successfully elsewhere by speaking with OEM’s and including in the signed deals.
Twitter @MrCyberdude
Mark – for our readers you might want to clarify who you mean by “we” in terms of who you have met?
This can’t be due to insufficient capacity at this time of year, maybe loss of a frequency control unit, not enough synchronous capacity available to deal with that, maybe due to some being offline at a time of plentiful wind power?
Paul,
Great site by the way , there is very little out on the web that is not extremely biased.
I do enjoy reneweconomy, they are slanted way too green and hence not an accurate resource but are a good source of news.
Their green assumptions slant their data but I still enjoy their mostly accurate direction most of the time and their stories are relevant.
While I may work in the coal industry I also have 40 solar panels on my roof of 10kw running on APS microinvertors. The reality is that with the 5c Feed in Tariff it’s a joke and is disappointing so my next step is to maybe connect an EnPhase battery system for night use and store my daily generation. Price point and are still an issue.
Cheers
I may
I did reply to the we comment but it looks like you skipped it?
thanks for sharing this Paul, looking forward to your update when more is known about the transmission failure.
It was fascinating here in Adelaide watching from the hills some suburbs in darkness while some had just had their power restored.
My main question is why it took so long to restore power, from about 4 pm until 7:30 pm at the earliest? There is enough inter-connector capacity to power the entire southern half of Adelaide, and there had been no damage to transmission infrastructure in the southern part of the State. Indeed, this is what occurred with the first restoration of power to my area, that all but about 100 MW was from Victoria. They apparently had a meeting – was to make up for not having pre-agreed contingency plans for black starts? Were there contradictory positions that the various market players were arguing about ?
Is the apparent slow staged restoration of power about maintaining stability and control of frequency? Not that I fully understand the complexities of electricity generation, but I thought it interesting that the ElectraNet boss on radio this morning mentioned that wind generated electricity cannot be used for ‘black start’ processes, that they need to get full control of load and frequency before introducing wind..
andrew w
black start with windmills? – that’d be “interesting”
I hope that an honest and thorough engineering led evaluation of the failure is performed – then we’ll see. The moment you see pols pronouncing on the matter – you’re being managed…..
As Paul is aware, Bruce Mountain has an article up over at RenewEconomy based on Paul’s above animation (http://reneweconomy.com.au/2016/how-the-south-australia-blackout-occurred-what-the-data-tells-us-65806). I’ll repeat the gist of the comments I made over there based on some research and discussion with a few people in the industry:
Good clear article Bruce and thanks also to Paul McArdle for getting his analysis up so promptly.
A couple of very minor corrections to Bruce’s numbers – there was ~300MW of wind generation in the South East at 16:15 AEST – the other 80 MW was from Ladbroke Grove (gas turbines). Torrens Island was producing 274 MW from three (not two) B units (200 MW capacity each), I’m not clear how Bruce knows the 4th Torrens B unit (TORRB2) was synchronised. According to bid data it was not offering any capacity to the market.
ElectraNet says that “3 of the 4 transmission powerlines between Adelaide and the north of South Australia” were damaged by the storm – this would refer to the four 275kV lines running from Davenport / Robertstown to Para and Mannum respectively (see Figure 2-2 in https://www.electranet.com.au/wp-content/uploads/report/2016/06/20160630-Report-SouthAustralianTransmissionAnnualPlanningReport.pdf) These lines run on 3 separate corridors almost all the way, so losing three of them in a single event is extraordinary. That said it’s not yet clear exactly where the failures occurred and why that would necessarily lead to the loss of all supply from generation north of Adelaide – we’ll have to wait for fuller information to understand why.
Given that loss of supply though, and with ~700+ MW already flowing from the South East there is no way the Heywood interconnector and transmission to the South East of Adelaide could have securely picked up the slack (I understand that Heywood interconnector flows briefly shot up to over 1000 MW at which point protection mechanisms tripped it off), nor could TIPS – or any other plant – ramp anywhere near fast enough to meaningfully address the supply-demand imbalance. The question then is why the South Australian Underfrequency Load Shedding scheme did not trigger and reduce load in a more controlled way before the system went black – again we have to wait for a detailed analysis of events before jumping to conclusions.
So nothing to do with SA’s second largest subsidy farm, Snowtown North stage 2 reaching 96% capacity, then auto-furling and 140MW of supply disappearing in an instant, just before total grid failure?
http://i65.tinypic.com/24onc3o.jpg
(Actually 2670MW lost from two Snowtown subsidy farms combined at the same time).
That moments later, the destabilised grid suffered a cascade collapse must be just a coincidence, or something to do with some transmission towers, or something. I’m so glad paid Big Wind shill Bruce Mountain-of-it cleared all that confusion up.
Meh.
should have read –
Actually 260MW lost from two Snowtown subsidy farms combined at the same time.
(Unlike Bruce Mountain, I believe accuracy is important. Even if it means admitting error).
Electranet claim to design for a windspeed of 41 m/s. That’s 147.6 kph. 70 kph and even 90 kph should be just a breeze to a transmission line 😉
Bear in mind too that doubling the wind speed quadruples the wind pressure.
Massive power black outs happen all over the world. New York City completely blacked out. Central London where Scotland yard was completely lost. The power grid between Canada and USA where power line easements were ignored and trees shorted out the whole system. Risk management in the power industry looks a 85% reliability. When the 15% failure occurs all hell breaks out. My belief is learn from North Queensland and buy a domestic generator from Bunning’s. North Queensland has always had blackouts from natural disasters. For the academics.
V=IR.
V stands for power stations. Adding pressure to the system.
R stands for transmission lines a pathway for the energy.
I stands for current or energy flow. lose any one and lose the lot. Vectors, Engineering maths, unreal numbers etc etc. Queensland had a similar event in 1986 when Joh ruled and coal was the bees knees. Australians have become an ignorant population. Commercial TV and the owners should be classed as those who broke an ideal way of life for the sake of “”Who is the Greatest”.
The 70 km/hr wind speed at Whyalla Airport could well be magnified several fold on the hills that the transmission line crosses. There may also be some harmonics that set up by strong winds on the line.
BOM reported 100+ kph winds at Snowtown
28/03:37pm W 50 104
28/03:34pm WNW 54 104
28/03:30pm WNW 46 104
28/03:28pm NW 35 104
That is windspeed followed by gust.
AFAIK those BOM values are ground level air speeds. What sort of speeds would you expect at 30m above ground (where the towers are)?
From BOM: Sep 28 UPDATE Severe Weather Warning, locally DESTRUCTIVE wind gusts about 130km/h are possible KI, York Pen and Mid North.
So on the face of it, I could believe that the towers were hit by local gusts in excess of design load.
The question is – is a design goal of 41m/s in fact adequate?
Forget about the design wind speed and recorded Met wind speeds when you have localised tornados touching down in the area around those transmission lines-
http://www.adelaidenow.com.au/news/opinion/tory-shepherd-federal-government-using-sa-blackout-to-prod-debate-over-renewable-energy/news-story/fb756bf99f54d7beba2c5212888122e8
I find this quite interesting. I live in Casper, Wyoming, USA. Today we have a high wind warning—meaning approximately 50 to 65 kph sustained winds and gusts to 95 kph. At times, we will have such winds for 2 or 3 days at time. Transmission lines sometimes go down, though usually it takes more than the 95 kph to bring them down. Some towns have had 160 kph wind gusts repeatedly. I realize that the high winds are not usual to South Australia.
Wyoming does have some wind plants, but most of our power comes from coal plants. The wind turbines often shut down due to the high winds, but no one seems to notice. I have to tell people to watch and see that the turbines don’t turn. I’ve seen them shut down in wind—turning one minute, still the next. People just don’t understand how wind turbines work and how variable and limited the output is.
“The ultimate outcome of this would be that thousands of battery storage, solar, community wind and CHP units would be incentivised and optimised for grid stability – aligning DG owners and the grids’ interests while at the same time lessening volatility on the wholesale market.”
Nice powerpoint. I sat through many of these types of presentations in a long IT career. Everything would be integrated and optimised into a smoothly working environment. Never really happened. Now looking at the evolving power industry and the desire to have vastly distributed, non standardised equipment purchased by individuals, community groups and companies all interconnected, playing well together and optimised for grid stability. I do believe the extent of challenge is understood, or perhaps its not acknowledged because it doesn’t fit the desired narrative.
Small example, we have had our house with smart meter for 4 years. Even though one organisation controls the equipment, the standards, the deployment etc and gets penalised for non delivery, they still have not deployed remote metering (coming real soon they say). That is a relatively simple project , albeit with large scale. Integration of 1000s of disparate systems, with fine control to deliver grid stability, when you don’t control all the levers seems to be a courageous goal Minister.
Remember too that at higher windspeed the wind turbines will shut down ( and are braked) to prevent damage.
Hi Paul
I am still a Learner so I am chasing information rather than voicing my undeveloped opinion.
Can you explain the AEMO spot price for electricity in more depth please.
Alternatively point me to some resources.
I have been informed that Wind Farms have negotiated a set price per KW for 20 years.
Does Snowtown Wind Farm really get paid $112 per KW for all the electricity they produce?
Thus the spot price does not affect them at all.
Is it not true that large industrial electricity users, eg ALCOA, have negotiated a fixed rate (tariff) and so are not included in the spot price.
Is it not true that large power stations bid for and get a locked in amount and price 24 hours earlier
So the spot price only reflects the small amount of shortfall or surplus electricity requirements for that 30 minute time frame.
Sometimes, not often, the AEMO price is negative. As in -$2.34.
I am to believe that all electricity providers are paying some one to take their power.
A very poor business model I would think.
Please set me straight
regards
Stephen
Thanks Stephen,
That’s a few questions there. I’ll try to respond in more detail in future posts (weaving into other commentary) but here are some starters:
Q1) Explain the AEMO spot price?
The AEMO has run educational courses for many years. I’ve actually never been on one myself, but we have sent other team members on some from time-to-time. More recently they have created this “General Public elearning – National Electricity Market (NEM) Overview” which, at only $220, is probably high value for money.
Q2) Do wind farms (and other generators) have long-term contracts in place?
The NEM is what’s known as a gross pool, meaning that all big generators actually receive the spot price for the energy they generate. The spot market also (largely) dictates how they are dispatched. Hence they all care about what happens in the spot market.
That said, most generators have a range of hedges in place to provide (partial) protection from the volatile revenues that come from spot. Some hedges are in the form of long-term PPAs (power purchase agreements) like you mention.
These hedges are not complete protection, so all big generators should still be concerned about what happens in the spot – though in reality some don’t realise this until they get burnt one way or another.
Q3) Do large loads have long-term contracts in place?
Because the NEM is a gross pool, ultimately all energy users are buying with reference to the spot price. As is the case for generators there are (most times) hedges in place that have the effect of smoothing the pain (or the benefit). These hedges can be of various durations – though long-term ones (e.g. 20, 30 years) are legacies of arrangements in place prior to the current market starting in 1998.
Q4) How does bidding work?
Generator offer prices (they are called “bids” in the NEM) are fixed the day before for the current trading day, but volumes in price band “buckets” can vary up to real time. Hence, dispatch volumes are only decided by AEMO in real time.
Hence the spot price really, really does matter.
Q5) Can prices go negative?
Yes, they can drop to a floor price of -$1,000/MWh – and sometimes do (like yesterday in QLD for a dispatch (5 minute) interval). When this occurs, yes, suppliers are paying the AEMO (on behalf of wholesale customers) to take their power.
Does this help?
Much appreciated Paul
One more. Do wind farms have a guaranteed price for their electricity
I have two themes in life
Don’t hurt anyone and try to understand things
I’m someone who actually follows a few utility market blogs like your own, so I’m reasonably more informed than most people out there. This is literally the first time I have ever heard of this program. I’m now going to have to go back to the quarterly novel that my power company sends me and see if they ever mention these plans.
But here’s what I see when considering this:
1) First of all, I never knew about it, and I just asked a few co-workers none of whom have ever heard about this.
2) The charge for 10% is $1.10 per week and the charge for 20% is $1.80 per week. Those are pretty damn affordable, but at such low percentages they make me feel like they’re not really making a difference. This also makes me suspicious that this is just like other green schemes where you’re paying money for the promise that someone has done something, while I wouldn’t feel confident that they actually did. What is their overhead on this plan, how much goes to admin, versus just paying for the electricity?
3) The charge for 100% is 5.5 cents per kWh. That’s a great percentage, though it would literally mean roughly $2 per day for me which is $182 per quarter. Which is a fucking huge jump. That’s $14.30 per week.
4) If I’m the only person doing it and take up rates are that low, then they definitely aren’t spending this money on alternative power sources. In a hypothetical situation imagine no one in my state actually took up this plan and I was the only person who did. Do they turn on my own special wind turbine for an hour to supply my yearly needs? Of course not, so it’s not like I’m actually paying for something physical. I don’t even feel certain that by voluntarily paying more for my power that these companies would then invest in that technology more. I see these more as placebos.
5) Why does it step from tiny amounts (10% and 20%) to 100%? Why can’t I pay for 50%? If you can pick percentages at all, why is there such a disconnect between almost nothing and absolutely everything? Things like this make me uncomfortable with why it’s structured like this.
6) I don’t like to think about my power bill. You’re forcing a complex choice which requires some extra thought on top of picking my provider, plans, billing period, and similar. All of this just adds to further anxiety / analysis paralysis when picking these things. If you really want to see who doesn’t want this, make it the default setting that everyone pays for 100%, then have people opt out. You’d probably see these percentages reversed.
This is why I believe there is a disconnect, it’s not because people don’t value it. It’s because the way in which you’re marketing it and selling it to me doesn’t make me feel like you’re actually doing anything different. This isn’t a vote with your wallet situation because it’s the same person promising me that they won’t just take the extra money.
But then again, maybe I think differently than everybody else.
Thanks for your perspective, Uriah,
Just to clarify – when you say “… the way in which you’re marketing it and selling it to me doesn’t make me feel like you’re actually doing anything different…” that I am not one of the ones offering to sell you the GreenPower – just someone on the outside, looking in and trying to understand.
Think I understand your dual concerns about “additionality”, and “auditability”. My understanding is that the Clean Energy Regulator address both of these in how they audit the production of LGCs, and allow for voluntary surrender (for GreenPower, and other schemes like the ACT PPAs).
Paul
Hi Paul,
Sorry, wasn’t saying you were selling it, I should have written the way they’re marketing and selling it.
Thanks,
Uriah
The reason I don’t pay for green power is because the electricity companies have to buy green power whether or not I pay extra for it. I thought about this for 10 seconds a couple of years ago and stopped paying for 20% green power because I feel like I’m ust subsidisng everyone else and the power company.
Thanks for the explanation of your situation, Mike
The way I understand it, if you had purchased GreenPower, it would have meant additional LGCs purchased from the REC Registry (hence more green power supplied, over and above the MRET target).
I understand the way it would feel like subsidising others, though.
Paul
I think a lot of people are willing to pay a little extra but they don’t feel it is fair if they are the only ones paying. That and the previously mentioned points about general awareness of the scheme.
The answer can be found in Psychology literature. See for example http://www.sciencedirect.com/science/article/pii/S2306774815000034
Government Accredited GreenPower is actually a highly regulated mechanism. When you read the actual rules on how it works, you will find that both Mike and Uriah’s concerns are adequately dealt with. The certificate schemes means that the premium is definately spent on green power and that they only have to buy the green power if their customers are signed up.
The real problem is that people don’t know about GreenPower. At sustainability forums, you can ask who supports renewables, doesn’t have solar but has GreenPower, many people don’t know about it.
One way the disconnect between the many who support renewables and the few who purchase green power could be decreased is by rolling out a default for GreenPower for new customers who then have to opt out if they don’t wish to pay the premium.
Paul there was some analysis a couple of years ago that when times are tough people stop paying for the green power option. I don’t think that it has ever recovered from that, It had a good take up then when it was shown on the bills how much it was people simply said no. It costs extra and most won’t pay for it so agree words over actions.
I think there is a difference between supporting renewable energy and deciding to pay for additional green power. Most people agree that emissions must be reduced, and that renewable energy is a means to this end. However, few people feel that their individual contribution will make any difference. Contrast this to roof top solar wherein a tangible individual benefit is perceived.
The ACT government has recently cancelled all their green power purchases for government business because they thought they could do more good for ACT ratepayers through investment in energy efficiency.
Climate change mitigation requires national and united national effort. Worthwhile climate change adaption can be achieved by individuals through choices that affect their built environment.
Don’t know about you but I’ve had a number of Indian reps cold call offering to change my light globes, shower heads and install chimney balloons, door seals and power boards-
http://easybeinggreen.com/energy-efficiency/
so many consumers would have taken up the free offer and as such believe they’ve done their bit, particularly with dramatically increased power prices. We all know there’s no such thing as a free lunch and so the cost of such programs is in our power bills.
Now I’m in Adelaide with an older Tudor home and I can tell you within a few weeks 2 of the 3 chimney balloons deflated, the TV power boards crapped themselves, the family moaned so much about the shower heads I changed them back and I already had mostly CFL globes but the Indian guy was happy to give me a heap of them and sign his form to say he’d installed them (they have someone from head office ring you on his mobile there and then and you nod and agree all is kosher, like lots did with the pink batts scheme when they already had insulation.
Most folks I know have experienced all this and figure they’ve done their bit, particularly as they know they’re getting screwed for all this dodgy rubbish via their power bills.
The reasons I don’t buy Green Power are:
1. I’d be encouraging bad policy, making Australia less economically competitive and less efficient. Renewables reduce our productivity.
2. Renewables are hugely expensive, requiring massive subsidies *(well over 100% subsidy) which is forcing competitive power suppliers out of the market
3. Intermittent renewables are a very expensive way to reduce emissions. At 20% penetration, wind is about 50% effective at reducing emissions. Nuclear on the other hand is greater than 100% effective (because it tends to replace baseload coal).
4. Wind and solar are not sustainable. Their ERoEI is insufficient to power modern society as well as replace themselves. They are entirely dependent on reliable power sources to support them.
5. They are supported by a purely ideological agenda. This has a short life.
6. After 60 years of solar PV and and 100 years of solar thermal engines, solar supplies just 1% of global electricity, wind 4%. And there’d be almost none if not for the ideological agenda forcing governments to massively subsidise them.
These are my reasons. I expect many others have made a rational decision and decided the risks and possible rewards are not worth the costs.
I hope this helps.
My widower father was in his dotage living alone in a unit and I increasingly looked after his financial affairs, a major problem being doorknockers and charity collectors. When I opened an electricity account to deal with I was surprised to find he’d switched to dearer Green Power with Origin and I enquired as to how that came about. It was then he produced the ‘presents’ he’d got posted later via the nice young fellow knocking on his door. A nice blurb from Origin, a sower head and some CF globes and a miniature green Origin footy to go with the dearer power bill.
So I’m straight on the phone to Origin to complain about sales staff preying on pensioners like that only to be told not to worry as the Green Power deal was off and all customers would be refunded anyway. Seems the Govt had knocked back their claim to Snowy Hydro being flogged as Green Power after Origin had been the winning tenderer in its privatisation.
Somehow the dills in power think we don’t talk about these dodgy practices and scams among family and friends and how quickly the word gets around.
If you read my submission http://www.greenpower.gov.au/Business-Centre/Program-Review/~/media/4488FFC5C5B04BACAEA881E393F33BB8.pdf or summary of recommendations http://www.greenpower.gov.au/Business-Centre/Program-Review/~/media/6DDD9A53908E49AA9BE6A0AE098154AA.pdf … on the Review of the GreenPower Program undertaken last year by UTS, you may appreciate the technical and legal issues associated with GreenPower and the simple fact that there is no legal framework that allocates the ‘renewable use’ attribute or the ‘reduced emissions’ attribute to those customers paying for GreenPower. Neither the Renewable Energy (Electricity) Act or the NGER framework transfers these attributes to a paying GreenPower customer. These attributes are shared across all grid customers. Consequently, the pricing structures for GreenPower are not fair. More customers understand this and it contributes to why the scheme is failing. GreenPower Marketing guidelines guide the customers to claim reduced emissions whilst those emissions are already allocated across all other customers under the NGER Framework and National Greenhouse Accounts Factors..
So “GreenPower pricing should be based around the cost of producing and delivering renewable electricity rather than as a penalty above standard electricity.”
Well it’s like this. If you pay homeowners 54c/Kwhr for any unreliable leftover solar power they’re not using, when reliable baseload costs maybe 8c max, we’ll fall all over you to be green and squeal like stuck pigs should you want to take away all our compassion for saving the planet.
What’s the real problem with green power pricing? It needs to be costed on the reliable base load it can always provide, just like thermal always was, but has now been bastardised by fickle renewables taking the cream off the top of all the milk. The only way to redress that is to enforce renewable generators to tender their power on the basis of the maximum base load power they can guarantee to the marketplace at all times. Naturally they’d have to partner deals with thermal power providers in order to do that and price accordingly.
Encouraging “clean” electricity without capacity payments for “dirty” electricity is a recipe for disaster, because the “clean” variety tends to be AWOL during heatwaves. Who would build a new conventional generator when the govt is intent on shrinking its market?
Paul, AEMO media release can be found here with a little more detail:
http://www.aemo.com.au/Media-Centre/Media-statement—South-Australia
As a PS to my early morning article, I realise I made a mistake in commenting that no load was lost.
Here’s a few other articles that provide more details:
1) Loss of load at (at least one) major C&I customer:
http://www.afr.com/business/energy/bhp-billiton-ceo-andrew-mackenzie-raises-job-fears-after-new-power-blackout-20161130-gt1gtn
2) Another perspective:
http://reneweconomy.com.au/vic-network-fault-causes-outages-in-south-australia-conservatives-blame-renewables-84808/
3) Of course the AEMO report, noted by Kieran above.
Australia needs to take care of and replace its conventional generators, the only component of supply that can deal reliably with broken equipment, and the low-wind days of heatwaves, but is it possible to change the tide of public opinion and the adverse economics before widespread power outages does that the hard way?
A question that arises with the Victoria and SA being short energy over the 17/18 summer. Does this result in the SA-Vic inter-connectors being under utilised as NSW/Tas can not import enough to flow through Victoria onto SA to meet demand?
Check out the wind turbine State’s output for Dec 6th by knocking out all check boxes except SA and subtotal to see the obvious problem-
http://anero.id/energy/wind-energy/2016/december/6
Seems the regulators are going to have to guarantee via the market, the returns of Torrens Island and Pelican Point by all accounts-
“The COAG meeting discussed an interim report from Chief Scientist Dr Alan Finkel on energy security, which was commissioned after the September 29 blackout.
Dr Finkel told The Advertiser that measures had been put in place to safeguard SA’s power supply, including new procedures adopted by the Australian Energy Market Operator. “Within the last week, AEMO has put forth a requirement that there has to be a minimum of two synchronous generators operating at all times within SA,” Dr Finkel said. “That provides voltage support and system strength.”
The synchronous generators are gas-fired turbines, which are more reliable than solar or wind power and stabilise the network.”
http://www.adelaidenow.com.au/news/south-australia/new-rules-to-reduce-blackout-risk-in-sa-as-jay-weatherill-and-malcolm-turnbull-brawl-over-energy-security/news-story/6bf5edaf8944068f2b2e3d6d3d80b078
So much for Australia’s phony national electricity market where all electrons were supposed to be equal but anyone with any technical brains knew some are more equal than others.
When you create a phony national electricity market then naturally there’s only one answer when you realise you’ve stuffed it. More of the same-
http://www.adelaidenow.com.au/news/south-australia/31-million-to-help-sa-businesses-struggling-under-high-electricity-bills-announced-in-midyear-budgt-review/news-story/3bad76be1b38c0507610fa6eb70a4218
“What game are these union heavies playing, one does wonder…”
Gotta get the redundancy payouts fully bedded down with all these Labor/Greens about. Sort it now or we’ll give the punters a bigger taste than SA of what you’re up to with all your unreliables.
The problem is you’re trying to model an extremely variable and complex system when it’s obvious you should stick to computer modelling a simple system like climastrologists do.
As someone responsible for a fair few modelling exercises I concur with a lot of what you say Paul.
Two additional observations:
1) Modelling is not the same as forecasting. Non-modellers in particular too frequently seem to believe that anything coming out of a “complex” model must be a forecast, and far too often go on to treat the results as holy writ.
You’ve clearly pointed out the plethora of challenging assumptions needed to feed any model, and the real world uncertainties and unknowables that models cannot adequately capture. I’d argue that these difficulties have increased in uncertainty and complexity in recent years (consider the future price / availability of gas for power generation as just one example).
Given this, treating any single set of model outputs as a prediction of the future, rather than just a scenario reflecting a particular set of inputs and modelling method, is naïve at best and quite disingenuous if done deliberately.
2) Modelling and model results are too often employed to advance particular policy positions. This is related to your point about modellers talking their own book, but I guess my observation is that once every policy position is advanced and buttressed with the support of sophisticated (but often opaque) modelling, presumably with the aim of rendering it immune from debate and criticism, it actually devalues the modelling process in general.
I think it might be a mistake to totally discount solar. One of the behaviour changes that will improve the benefit to solar owners is if they can load shift to times when power is available to them from their solar panels; eg rather than overnight water heating instead heat to full capacity in the middle of the day.
The other impacts that are likely to be significant are: –
Shift to LED lighting, including by local councils which must substantially reduce overnight lighting loads,
Greater awareness of parasitic loads and a greater portion of consumers actually turning off devices overnight, and
General improvement in the efficiencies of appliances, fridges and air conditioners
Mick
Paul McArdle,
I’m rather disappointed at the one sided comments from you about workers when it was actually AGL that threatened Victoria’s electricity supply by stating they would lock out all workers, even those not involved in the 14+ month dispute.
The real threat to the Energy Market and the Government’s quick response came from AGL’s escalation of a dispute and a lack of reserve in the NEM to cover an individual company decision to not just lock out the Power Station workers, but also the Mine Workers which are not even involved in the dispute.
AGL’s decision to lock all workers out was simply done to cause an AGL induced coal supply shortage to ENGIE’s Loy Yang B station. This was a legal manoeuvre to affect a 3rd party company, so as to force an arbitrated outcome rather than negotiate in good faith bargaining.
I’ll repeat the workers view, that it has been 18 months, where ALG would not even meet with the unions to negotiate. For this to be the first threat of industrial action from the CFMEU after the 14+ months lack of good faith bargaining by AGL. It’s very rich of you to side with the Boss.
Maybe you should stick to the facts and not speculation, like you normally do. Basking the workers and insinuating they are smoking something and want a longer Christmas break is hardly impartial considering that it wasn’t balanced with comments about expected multi million dollar profiteering by the remaining generators. I would have thought by now that you especially would know that the electricity generation sector is a 24/7 365 day a year operation and that each and every shift worker has spent many a Christmas Day, Easter and public holidays at work and away from their families.
By the way, the AGL workers have all voted for “Protected Action” as they are entitled to do under the FairWork Act, in a secret ballot run by the Australian Electrol Commission (AEC). The protected action bargaining order (PABO) was a request from the power station workers and the CFMEU represents its workers as requested, especially when there appears to be no end in sight.
http://cfmeu.com.au/loy-yang-workers-to-take-protected-industrial-action-following-agl%E2%80%99s-refusal-to-bargain-in-good-fait
Contrary to the AGL propaganda the dispute is not over pay raises, it is all about job security, as AGL wants to cut worker numbers as evidenced by their forced redundancy clauses.
You should really be reporting about how all the Generators are simply looking to game the market and maximize their profits at the expense of all consumers.
ENGIE owns Pelican Point Gas, Hazelwood and Loy Yang B, so when they close Hazelwood they are looking to make a killing from the market. Some of the Hazelwood Units could run until 2020,2019,2019 and 2018 without any extra expenditure but are being shut down for profiteering and ideological reasons, without a second thought for the Workers, Latrobe Valley Community and consumers.
To put the announced closure into perspective, ENGIE hired young workers 2 years ago and promised the a job until 2025 with a potential of 2033.
ENGIE’s previous announcement of redundancy payments were a negotiated requirement of the EBA they signed 2 years ago and not a cent more was offered.
I’m one of those directly employed 500-750 Hazelwood Workers that are about to be made redundant in March because of this corporate profiteering.
How about you stop the union bashing and look at the problem for what it is, a political football between State and Federal Governments all based around the Generators manipulating the AEMO/AEMC rules for maximum benefit (To the shareholders of course).
I for one appreciate your more impartial and fact driven articles, that’s why I read and often refer people to http://www.wattclarity.com.au rather that the one sided reneweconomy bullshit.
I hope you take my criticism constructively and aim to deliver more facts than speculation.
There was a storm and blackout in Melbourne on the 1st May:
http://www.theage.com.au/victoria/melbourne-weather-wild-storms-lash-city-20160430-goj4ya.html
AND there has been a decided move away from off-peak water heating (less heaters replaced like for like) by a combination of policy makers encouraging gas heaters as less emissions intensive and higher less competitive rates for controlled off peak tariffs. As mentioned by Michael, LED replacements for Street Lighting would be significant. This would also be the case in NSW so not sure if a similar impact would be evident there also?
Re shift to gas HWS we should be able to observe that increase in gas usage so are the figures available?
It surely can’t simply be a coincidence that 28 September 2014 happened to be among the warmest for that time of year (including an all-time September maximum record being set at Avalon Airport). http://www.bom.gov.au/climate/current/month/vic/archive/201409.melbourne.shtml Could this be indicating that some sort of sweet spot is being hit between space heating and cooling demand? Unusual pre-heating of water going into electric-powered hot water systems may also be a contributor?
Hi Shane Cremin,
As you completely fail to mention Australian Standard AS4755: Demand response capabilities and
supporting technologies for electrical products, it appears that you are unaware there are technical solutions to the problem you introduce in your first line of your Joke Punchline article.
If you are aware of AS4755 then it’s disappointing that you have not informed the readers of your article the following information. WattClarity readers are not your average dumbed down blog and deserve more detailed information.
The solutions are not without consumer backlash though.
It’s called Demand Response Enabling Devices or (DRED) and there are various states of operation from DRM1, DRM2, DRM3 and DRM4.
DRM1: shut off, or operate at minimal load, for the duration of the event. In an air conditioner, this means the compressor must not operate, but fans and controls may continue to do so. During normal peak load events the utility (or a pre-programmed DRED) would ‘cycle’ air conditioners to operate for, say, 15 minutes and then shut off for the next 15. In emergencies, a constant DRM1 signal could switch appliances off for the duration of the emergency.
A pool pump controller responding to DRM1 must switch the pump off.
An electric, solar-electric or heat pump water heater must cease the use of electricity for heating water (whether by resistance elements or compressor operation) but water circulation pumps and controllers may continue to operate. This makes DRM1 a technically superior method of control for solar and heat pump water heaters than traditional off-peak, which de-energises the entire unit.
DRM2: continue operation at not more than 50% load for the duration of DR event.
DRM3: continue operation at not more than 75% load. This was differentiated from DRM2 at the request of electricity utilities, on the grounds that it will be easier to enrol customers in DR programs if they can be assured of having at least 75% cooling during DR events.
DRM4: switch on even if timers or user settings would not require operation at that time. This is intended to allow appliances to come on at times when electricity price or CO2-intensity is low, eg when renewable generation availability is high. It would apply to swimming pool pumps – by bringing forward but not extending daily running hours – and to electric water heaters – allowing additional energy to be stored by temporarily raising thermostat settings. Utilities may also use DRM4 to pre-charge solar-electric water heaters on winter mornings, to prevent their boost elements from operating during the evening peak when space heating, lighting and cooking loads coincide.
.
Now here is the real Joke Punchline..
For an appliance to be AS4755-compliant, it must be able to respond to only DRM1(Shut off or Minimum load), which is not really consumer friendly.. An old adage used to be that the customer is always first.. This adage iis lost when you prevent a customer from using the air conditioner that they brought for the sole purpose of cooling themselves down on a hot day, all because it’s a hot day and Lack Of Reserve LOR exists due to insufficient reserve in the National Electricity Market (NEM)
The real problem!
Who is managing this whole debacle that is centred around closing existing power plants like Hazelwood Power Station without cheap reliable alternate energy sources ready to take up the Lack Of Reserve LOR created… NOBODY!
Well allegedly it’s the Australian Energy Market Operator (AEMO), under the control of the Australian Energy Market Commission (AEMC), which was actually created by the Council of Australian Governments(COAG).
Currently the entities of AEMO, AEMC and COAG are the equivalent of a rudderless ship being dragged by the undercurrents of the energy Generators.
Can you guess the names of these engie companies?
Mark,
The availability of demand response options does not negate Shane’s main point – the cost of funding infrastructure needed to satisfy the increasing peak demand due to air conditioning.
There is a limit to how demand response can be used to manage this. Turning off lights, swimming pool pumps, and other miscellaneous power uses can shave a bit off the peak. You can’t really turn off your fridge and freezer on the hottest Summer days. In order to make a substantial reduction you would need to turn off or ramp down air conditioning. How would consumers perceive this? They spend money on a home air conditioning system only to turn it off during the hours they need it most. If forced into such action, they may start to wonder why they purchased air conditioning in the first place.
There are solutions in the form of better building designs however anyone involved in the building industry will understand how difficult it is to promote these sorts of design improvements. Construction standards in Australia are poor by comparison with other OECD countries. The industry is geared up to churn out cheap lightweight structures at very low prices. Any sustainability initiatives struggle to gain acceptance in this environment. When consumers are prepared to spend more money on their homes it tends to go to amenities such as granite kitchen bench tops rather than sustainability features and initiatives.
Note also that the Building Code does not help. It contains energy efficiency provisions based around minimizing annual energy consumption. This does not lead to the same outcomes that would be achieved with a focus on minimizing peak electricity demand. There are some perverse outcomes from this when NatHERS star rating software encourages designs that do not minimize peak demand. For instance:
– there is limited credit for deep overhangs and other external shading features
– dark coloured roofs are sometimes favoured over light coloured roofs
For most of the 200 or so years of European settlement in this country people have survived without air conditioning. For locations in the Southern part of the country, it is still possible to provide a comfortable home without the need for air conditioning. The critical factor is thermal mass. Lightweight homes can never be adequately comfortable without air conditioning.
If the retail electricity sector were properly deregulated so that consumers paid the true price of power at the time they use it, uptake of air conditioning would be greatly reduced. This should include time based transmission and distribution fees – not just for energy consumption costs. Such deregulation would then expand the market for well designed homes. Unfortunately, those who had already invested in poorly designed lightweight homes would suffer (financially, thermally or both).
Note also that cost increases resulting from the closure of coal fired power stations are a separate issue. Even without these closures, the increasing uptake of air conditioning was always going to drive cost increases – in generation, transmission and distribution. These issues may be compounding however it is important not to conflate them.
Cheers
Hugo
Hi Mark,
You are correct – AS4755 provides some mitigation of the issue, but as the article pointed out, I don’t think this standard really came into effect (with any real impact) until well into the air-con roll out. The point of the very brief article (not initially written for the readers of Watt Clarity) was to highlight the slow, reactive nature of regulators in an era of fast-paced change. Whatever your view of the rapid uptake of solar (and likely uptake of storage), it will have consequences to how energy markets behave. I didn’t make a big issue of it, but over here in the West, we have been bogged down by a 3 year “Energy Market Review” which has not really achieved much and hasn’t even addressed the looming issue of solar + storage. I can’t help feeling that in a couple of years, we will look back and wish we had the foresight to implement some market reform that might mitigate some of this impact – where I specifically, albeit briefly, focused on the capacity market and its inadequacies.
Shane.
“Who is managing this whole debacle that is centred around closing existing power plants like Hazelwood Power Station without cheap reliable alternate energy sources ready to take up the Lack Of Reserve LOR created… NOBODY!”
Govt is managing it alright by intervening in the marketplace with subsidies and legislative fiat to create a whole bunch of suppliers to engage in the purist form of dumping one could imagine. The only problem is the nature of supply via the dumpers when they’ve driven the dumpees out of the market. Still, there’s no doubt blackouts and brownouts will reduce the emissions they want to and that’s the payoff apparently.
OTOH to stop dumping in its tracks Govt would only have to legislate for a level playing field (well aside from the subsidies and other specific legislative fiat which would be small beer under anti-dumping legislation) All it needs is for legislation that only permits suppliers to tender to the grid up to a maximum they can supply 24/7 all year round and bingo! naturally they’d have to cut deals with thermal providers to do that but bang goes their desired emissions payoff.
I should add the obvious caveat that it assumes our historical and ongoing inability to store energy which is rather pitiful, apart from in the form of calories and pumping water uphill. Yes I know the cheer squads place great faith in electochemical storage, but all I’d point out to them is we’re still using essentially the same lead acid battery in our cars Henry was plonking in the Model T.
Roger S,
I know there is a real potential to store electricity via a molten salt energy system.
Specifically by initially using off peak or hydro grid electricity as well as other heat sources to heat a molten salt energy storage system. This stored energy would later be used to heat steam for a turbine. Ideally this would be used in a NEM price peaking scenario and would likely be competing against gas generators. In some states the option of a solar heating array could be implemented.
http://www.solarreserve.com/en/technology/molten-salt-energy-storage
Mark,
This type of system is ideal for managing peak electricity costs in the wholesale market.
It does not however deal with transmission and distribution costs. These typically represent between 40 and 60% of an electricity bill.
I find it odd that the Office of the Chief Scientist should declare something to be “unstoppable”, which sounds much more like politics than science, and deciding the outcome in advance is the antithesis of good science. Predictably the marketing departments of the “Transition” industry are in raptures about “unstoppable”. The notion of consumers eagerly embracing “The Transition” is also somewhat less than convincing to me, eagerly embracing free money is probably much closer to the truth.
Both solar and wind will never have much direct impact on peak demand, but why does the Chief Scientist fail to mention this fact? It is already widely known that solar PV in South Australia has pushed peak demand times towards dusk, so adding more capacity will have very little effect. Similarly the large wind capacity in SA has pushed peak demand on non-wind generators to times of very low wind power, which hobbles any further wind capacity.
Moree has single axle tracking while the Broken Hill and Nyngan plants have fixed panels. This might be the reason why Moree gets up and running before Nyngan does. It is also several hundred km north east of Nyngan
Can I suggest that a good step can be taken towards a More Mature Discussion by dropping the ubiquitous word “challenges”, and calling it for what it really is, problems, problems, problems.
Ironically Tasmania, only recently beset by a major problem, now looks like the best place to be for electricity consumers, because of its relative immunity to governments setting targets.
PS one of our shyer/shier readers has emailed in the following explanation:
Our thanks to them for that!
Paul
Presumably the 2940 MW is a 5 minute scheduled demand figure – from what I can see the lowest half-hourly scheduled demand on 2 Jan was just over 3050 MW.
Interestingly if you use scheduled + non-scheduled (intermittent) generation as the demand measure then the minimum value was just over 3300 MW, the Dec 2016 minimum was not much higher at just over 3400 MW, and way back on New Year’s Day 2000 the minimum was 3377 MW – although back in 2000 AEMO didn’t report total non-scheduled generation, however this would have been a pretty small number before Victoria gained any non-scheduled windfarms.
Correct, Allan,
Noted “Scheduled Demand” on the image, but perhaps not in the text.
Paul
Interestingly, in a research report published by Morgan Stanley yesterday and titled “Inside Industrials: Electricity: Demand -1% but Supply -23% (03 Jan 2017)”, the following passage was included:
Will help to provide some context to the above.
Postulate a distribution of eleven household energy consumption groups around the mean shown in the graph for years 2009 – 2013. Annual kWh :-
11250
10500
9750
…
3750
Year 2009 average 7500 kWh.
Remove the greatest household energy group and recalculate the average for 2010. Result 7125kW.
Remove the next greatest group and recalculate for 2011. Result 6750 kWh
For 2012 result 6375 kWh
For 2013 result 6000 kWh
The ‘Average Annual Domestic Consumption (kWh)’ graph is generated.
Now take the removed groups to form the ‘Domestic (with solar)’ average.
For 2009, the greatest group without solar was 11250 kWh but 7000 kWh with solar, so solar average generation for this group was 4250 kWh.
For 2010, the next greatest group was at 10500 kWh without solar but the consumption with solar was 5000 kWh. This requires a 5500 kWh solar generation.
The average consumption of the two greatest groups (7000+5000) with solar is then 6000 kWh.
Solar generation in 2011 of 3500 kWh for the third group and solar generation of the fourth group in 2012 is also 3500 kWh and the final group requires 2500 kWh solar generation.
The ‘Domestic (with solar)’ graph is generated.
The hypothesis is that the highest electricity consumption groups in the first years buy full sized (5kWh?) panels, presumably off-setting the air-conditioning load in large houses. The third and later groups buy smaller panels or fit in less favourable orientations. The falling graph for those without solar is a statistical trick caused by the largest consuming groups leaving this group. The remainers do not have to change their behaviour at all.
There was an article that briefly appeared on The Age website that the Wonthaggi desalination plant had not yet produced any of the water that had been ordered back in March 2016. The order was for 50 GL of a 150 GL capacity, to be delivered by June 2017. The website of the plant’s owner, Aquasure, still shows that no water has been produced. One would think that in order to minimise power costs, the plant would be run at 100% capacity for 33% of the time between July 2016 and June 2017, utilising the half-hour periods with the expected lowest third of power prices. According the The Age article, there has been a commissioning problem with the underground cable from the Cranbourne Terminal Station (perhaps akin to the problems with Basslink). At full capacity, the plant uses 90 MW, so we can expect a boost to Victorian power prices when they finally resolve these problems.
The desalination plant is part-owned by Engie, who also part-own Hazelwood. It would surely be in their interests for the plant to be operating before Hazelwood shuts, firstly to obtain a lower power price for the desalination plant, and secondly to boost to the price of power sold by Hazelwood.
The 50 GL of desalinated water come at a high cost, whereas about over 80 GL of water flowed down the Yarra River past the pumping station for the Sugarloaf Reservoir in September and October 2016, because there hadn’t been sufficient investment in capacity to pump this water to other reservoirs.
Despite the afternoon price spikes Wivenhoe pumped storage power station still hasn’t turned on, (but it was generating in the evening peak on Thursday). Is its ownership by a company that has some major coal-fired power stations leading to a less than optimal market outcome ? It is owned by CS Energy, a State-owned entity that also owns Kogan Ck, Callide and Gladstone power stations. If Wivenhoe were to bid during the price spikes, it would reduce prices to spot-exposed consumers, but lower the price received by the coal-fired power stations in the fleet. One would think that if owned by another entity it would offer its full capacity of 500 MW during price spikes and evening peaks, and recharge through its pumps at night when spot prices were below $50/MWh.
These crazy prices must surely help the business case for additional large-scale solar farms in Queensland, and the GenEx pumped storage station. The spot gas prices were crazy too, at $13.73/GJ at the gas supply hub.
The high demand in Qld is presumably due to the LNG activity and the high temperatures, however, I’m curious to know how big of an impact the high temps are having. In other words, due to the LNG trains the demand in Qld increased, and due to this increase will even small-to-medium increases in temperatures (and thus load) lead to significant price jumps?
seems the 50% QRET is quiet timely, for daytime loads anyway (PV).
Things are already getting interesting:
http://www.afr.com/news/policy/industrial-relations/agl-energy-alleges-covert-industrial-action-at-loy-yang-20170116-gts3e6
http://www.theage.com.au/victoria/agl-fails-at-industrial-umpire-over-shortage-of-overtime-workers-at-loy-yang-power-plant-20170116-gtsdhw.html
Might be time to make a bowl of popcorn and watch the show!
Not sure how closing Hazelwood is a win for emissions reduction, the same amount of electricity will be needed after it closes, some sources will get to increase their outputs, and there is only one type of generator that can do that useful trick.
Because no plant is more emissive on a t CO2/MWh metric than Hazelwood, so even if you are assuming its output will be replaced entirely by other coal plant, emissions will be at least marginally lower overall.
Hi Paul,
Prior to Solar was this as high? What effect has solar had today?
Lucas.
See 2nd update above, Lucas
Great to see PV heavily utilised and reducing the carbon emission through most of the day, certainly the only bright spot. Such a shame most of the generation was through coal.
Hi Paul
On a half hourly basis the scheduled demand peak looks to be 9369 MW. This is about midway between AEMO’s 50% and 10% probability of exceedance forecasts for this summer, and more than 400 MW higher than the previous high in 2010 (also on 18 Jan – how about that!).
An interesting point of comparison is that the 2010 peak occurred in the half hour 14:30 – 15:00 (trading interval 30) while yesterday’s was for half hour 18:30 – 19:00 (trading interval 38). This shows how solar has – by reducing scheduled demand during the sunlight hours – pushed the time of system peak to much later in the afternoon. In 2010 scheduled demand was over 500 MW below its maximum by TI38, so had the same underlying consumption pattern applied yesterday, in the absence of solar you might have seen a half hourly demand approaching 9900 MW!
Actually 18 Jan 2010 wasn’t the highest half hourly scheduled demand for Qld prior to yesterday – that was surpassed by 9105 MW on 1 Feb 2016, my mistake 🙁
But the points about later time of peak and maybe another 500 MW of demand in the absence of solar PV still stand.
And this is probably labouring the point, but on first inspection someone might look at the 6:30-7:00pm peak, with solar PV producing virtually nothing by that time of day, and say “well, solar is doing nothing to reduce peak demand”. However this is incorrect because in deferring the time of scheduled demand peak, solar PV has moved that peak to a time of day when underlying consumption has fallen (ie some industrial / commercial load would have reduced by then), leaving a lower maximum to be supplied from scheduled generation.
But once the scheduled peak has been consistently shifted to that time of day it is true that additional solar PV capacity will not further reduce the size of that peak.
One thing AEMO seems to neglect to do is adjust actual peak demands for weekend reductions, there are many holidays as well at this time of year, so its quite possible that previous years would have had similar demands if the previous heatwaves had fallen on working days.
The same thing happens with wind power, once it gets “substantial” it pushes times of high peak demand on non-wind sources to wind lulls, and ADDITIONAL wind farms have very little effect on those peak non-wind demands. South Australia passed this point a few years ago.
I think they call it ‘potential stability issues’ in the diplomatic corps-
http://www.abc.net.au/news/2017-01-20/blackout-investigation-puts-power-import-plan-on-hold/8195394
Hi Paul,
At some point in the future, when you may be looking for something to do, it would be interesting to see what the $ impact on the market was due to the solar output. In other words, quantifying the avoided cumulative cost by the depressing nature PV has on the spot price (merit-order). Conservatively, this may assume the existing spot price, i.e. spot price during solar output. This would be instead of trying to estimate a new marginal generator and raising spot price.
With the current availability of data it is quite impossible to determine the actual total output from small scale solar systems. The APVI data is based on such a small number of the installed systems, and even that sample shows gross distortions in system output, that expansion of the data to provide a whole state estimate is almost ludicrous.
Even the systems that are reported through PVOutput.org are more likely to be recently installed (more likely to have WiFi outputs) and from systems installed by those who are more likely to follow regular maintenance routines. So the base data is likely to be heavily skewed to the “better” performing systems.
I would be very keen to see calculated confidence intervals for the APVI data, but my guess is that they would be so wide as to make the estimate of “peak” output from solar systems very difficult.
As the AEMO data is quite similar it is likely that they follow a similar procedure as the APVI data; i.e. relying on a very small number of systems actually reporting output.
I therefore suspect that picking peak usage by relying on this data is very dangerous. But that’s not really the point. There are all sorts of methods for controlling the actual peak demand on any one day, uncontrolled solar input being just one, but there are also many methods which would provide some control over the peak.
The real issue for consumers, industrial and commercial, is the price and reliability. From your graphs it is clear that the behaviour of the small scale solar units causes a narrowing of the peak demand on conventional generators. To maintain reliability we still need conventional generators capable of covering the actual peak (in the event that solar output is low). So we have not saved any capital but have pushed down the operating hours over which this capital can be recovered and so sending up the price. We can see that in the peak pricing as the solar systems start to fade.
So small scale solar system are not really helping the peak demand issue, they may in fact be making it worse. Irrespective of whether they actually reduce the peak or not.
On the possibility of west facing solar panels – it is very interesting to see the additional output from single and dual tracking solar systems. But all of this, including batteries, just keeps making solar a more expensive, rather than a more competitive power supply option.
Michael, the diseconomy you have pointed out of running base-load power stations over a narrow peak period underlines the unsuitability of coal or nuclear baseload power as a backup or complement to renewables. What we need is not baseload but dispatchable generators, like gas, hydro or solar thermal with heat storage. From this perspective it is extraordinary that the solar thermal option is routinely ignored in discussions like this and, of course, by government..
Although an enthusiast for PV solar, I can’t see the point of installing large PV arrays over flat, agriculturally useful land. There are very large areas of commercial shopping centre, warehouse and factory roofs as well as carparks and walkways near existing powerline infrastructure where PV panelled roofs would provide shade and power at zero land cost.
All investment in land-covering solar should be focussed on solar thermal with energy storage.
Thanks for this analysis.
Hi Paul,
the spot market electricity price in Qld seems to be spiking more and more in late Jan/early Feb. I can’t recall ever seeing such extensive spikes. What is going on ? and what can we expect this to do for the next round of power price rises from the retailers ?
looks like gaming and gouging to me.
Allan
These analyses are really appreciated
They provide a clear and concise explanation of what has occurred.
What is missing (and this is not your fault) is why the generators are playing games and spike the prices to MPC and not offer prices more in keeping with their costs. There have been a number of attempts to prevent the generators using their market power (MEU – economic withholding, AER – ramp rates, SA gov’t – bidding in good faith) but they still do it. I am aware that generators will seek to maximize their profits by whatever means are available, but if there were increased competition then we would see the market more reflective of marginal cost bidding rather than gaming the rules
David
Thanks for your comment, I appreciate the feedback.
You’re correct that deriving explanations for the underlying bid structures / strategies of generators are outside the scope of these quick event reviews. But obviously they are important and of great interest.
As any number of commentators have observed, the NEM market design effectively requires the possibility of occasional “contingent market power” where generators are able to have offers priced very significantly above short run marginal costs dispatched. But the issue is how frequent these possibilities ought to be in a well-functioning market? Impacting on this will be basic supply-demand balance, industry structure, and the extent of financial hedging (both explicit via contracts and implicit via vertical integration). I think a general difficulty with rules seeking to govern bidding behaviour may be that the real issue is not behavioural but structural – eg supply side concentration.
It’s probably also the case that as we have seen increased vertical integration, the interests of retailers and generators in the prevailing level of and volatility in spot prices have converged.
All that said, an overarching control is meant to be that as the level of spot & contract prices reaches or exceeds long run marginal costs of supply (LRMC), the real or implied threat of new entry increases. For a long time in the NEM we became used to seeing prices arguably well below LRMC – entirely appropriate for an oversupplied market – so with generation retirements and increasing fuel costs, as well as strong load growth in Queensland, the recent increases in price level and volatility are not unexpected directionally – I’m not going to offer a judgement here on their level. But if they prove to be sustained, I’d expect to see even more discussion about new entry – both large and small scale.
Allan
Very interesting, Paul.
Thanks for pulling together these examples.
Can you please explain how the reserves in various LOR1 notices for NSW have been calculated and how that panned out in reality. 57241, 57238, 57234, 57232, 57222, 57219, 57217 , 57216 (on 4th Feb)
Also how did notice 57218 on reference temperatures materialize?
Why was NSW coal generation so low? It can go up to 9,500 MW but did only 8,500 MW on 6/2/2017 at the critical time. Was there a plant, transformer or transmission problem? Heat-related?
Why did imports from Victoria suddenly drop between 13:00 and 18:30? A spike of almost 2,000 MW at 19:30?
Murraylink is exporting at twice the limit the constraint is set to. Doesn’t look like any spare generation in SA at the moment. LOR2 already declared, waiting for the LOR3. Perhaps we will dodge it?
We didn’t dodge the LOR3. Up to 100MW rotational load shedding implemented. TIPS A1 unit was offline, I assume for maintenance. That would have saved us but shows the supply situation is very tight on hot days. TIPSB1 has a blip and went back to 150MW for awhile as well, it hasn’t been outputting it’s nameplate (200MW) for awhile.
Ian,
Here’s the first review:
https://wattclarity.com.au//2017/02/load-shedding-occurs-in-south-australia-in-heatwave-conditions/
Paul
Pelican Point 50% mothballed still. That would have saved us this time. Fortunately the load shedding was less than an hour but it is disappointing we have available generation offline due to essentially a market failure in my view. (The most efficient gas generator unable to run economically)
“The most efficient gas generator unable to run economically”
You mean the price per kWh wasn’t high enough to turn Pelican Point on ?
High efficiency of a combined cycle plant is only useful if the plant runs as a baseload. With high renewable penetration in SA, a combined cycle plant would have to ramp-up and ramp-down to accommodate carbon-free energy from renewables. Such ramping-up/down would not allow a CCGT plant to operate at its most optimum leading to higher operating cost for the owner of the plant. No wonder, a highly efficient CCGT plant is not despatched by the owner in a market with high renewable penetration.
A reciprocating gas engine power plant is ideally suited for numerous start-stop, ramp-up/down application without impacting the maintenance cost and such a plant can also achieve high efficiency (46~47%). Such a plant can be built in 100MW chunks and allow efficient dispatch of just 10% of its capacity due to its multi-unit configuration.
I’ll be happy to provide more information.
Hi Suraj.
Great comment, where are you based and what is it you do?
Mark Yates
Mark/Ian/Suraj – will put you in touch with each other offline during the week to avoid the need to post contact details in open comments.
Hi Ian,
Quite keen to connect with people with similar interests in this field, would be great to talk further and understand what you do.
Mark Yates
Ian
Sorry not to respond earlier to your comments, afraid I got busy digging deeper into the events of yesterday – shows how in the above post I focussed on Thursday’s outloook and completely missed the risks just around the corner. If I lived in Adelaide I suspect I would have realised that AEMO’s predispatch was well on the low side.
Should be more about Wednesday on WattClarity this afternoon, but your observations are right, Murraylink exceeding constraint limits especially.
Cheers
Allan
Great article Alan. I’m thoroughly enjoying your reviews and learning more about my ezy2view in the process.
Cheers
Today 9/2/2017 17:00 in South Australia
Demand 2,824 MW
Generation 2,765 MW
Available Generation 2,660 MW
Imports from Victoria 567 MW
So SA generation plus imports from Victoria was 3,332 MW much higher than demand
Why is that? generation plus imports must always be equal to demand. What are the transmission losses?
Great article! Amazing that PPCCGT Unit 2 has been unmothballed so quickly! One of the Port Lincoln units is running tonight as well.
I wonder if AEMO under-forecasting has anything to do with it, they don’t seem to take weekends/holidays into account. Maximum demands were quite low in SA during the last 2 summers largely by chance, no heatwaves culminated on full working days (Mon to Thu), but this year they have done so in both QLD and SA.
AEMO modelling of wind power is more subtle than stated in the article. For one purpose they use a nominal capacity factor, but for the Reliability Standard and the ESOO they use actual wind power data for the previous 6 years, adjusted to 2016/17 levels.
Given that the recall time on Pelican Point’s second unit was obviously a lot shorter than some earlier statements had indicated, I think a higher demand forecast from AEMO either on the day or a day or two ahead may have made a difference on Wednesday. It still would have come down to AEMO issuing a Direction early enough to get the unit online for the evening peak – as they did on Thursday.
I’d be very surprised if AEMO’s forecasting process ignores weekday / weekend differences but the absence of mid week extreme heat in the last two summers would have made calibrating and tuning their models more challenging.
Thanks for pointing out the actual wind-trace methodology AEMO use for ESOO and reliability standard analysis. For the purposes of this quick writeup I felt the simple capacity factor “benchmark” was an appropriate point of comparison, but yes they do adopt a more sophisticated approach in full market modelling.
Allan
Allan, you are my new Font of Knowledge on the grid. Your articles should make it into mainstream media. At least they should be on the daily clipboards of our pollies for digestion.
Some random related comments and questions….
Is AEMO running a corruptable system? Have you seen evidence of the generators being pulled back / taken off line in the face of an impending tight spot? Just to play for the $14k? Do AEMO have people who check the bonafides of any off-lines? Just wondering.
I’m staggered at the increase in demand from hot weather. There must be opportunities for controlling this excessive demand as a cost effective way to avoid load shedding.
Is it true that NSW were about 2GW short of power on the 10 Feb 17 during the worst of their power peak? And that Tomago smelter pulled consumption to avoid cutting power to mums and dads? Did NSW still have to swallow a $14k spot price during this “almost outage”? In Adelaide it was reported as one big ‘back patting exercise’ after people reduced power consumption and averted load shedding, in response to a call from their Premier.
Cheers and regards.
Your article refers to ROLLING load shedding being ordered, but the news casts in Victoria suggested those without power had no power for that whole period.
Were those reports true?
If so why was the withdrawal of power not shared sequentially over various consumer blocks as the term ROLLING implies.?
Robert
This tweet from SA Power Networks certainly describes the load shedding as rotational ie rolling. AEMO and the relevant authorities in SA (and in every other jurisdiction) would have a pre-approved list of loads / areas to be shed sequentially in this type of event. The duration of such shedding in any particular area would normally be about 45-60 minutes, then power would be restored and the next scheduled area disconnected, and so on. As this event lasted less than an hour, I imagine that only the first load block on the list was affected. AEMO’s media release above states that load reduction applied between 19:03 and 19:40.
Allan
As always there will be blame games, South Australia’s problems are home grown with renewable energy and poorly performing gas markets have taken their toll there.
But all in all, it is just too darn hot and AEMO have a tough 3 days ahead. Good luck.
Why is it a “remarkable outcome” that we are on track to break the 2011 record. How many more people are there now in NSW than in 2011? Probably half a million at least. Yes there is the penetration of Solar PV, but there are also many small and subtle power hungry developments , such as the continuing increase in average house size, the increase in size and number of appliances, huge televisions, coffee machines , computer equipment etc. Its surprising it hasn’t happened sooner.
Is the Tomago aluminium smelter back on again? With how much load? Is or was this included in the scheduled demand? A smelter should be the last to be included in load shedding. Why not disconnect the air conditioned office blocks of those decision makers who are responsible for our energy mess?
Hey Paul – thanks for the update. In regards to Solar data – am sure you though about this – do the timezones of the APVI data line up with NEM-time? It also seems a bit strange the small scale capacity seems to line come online an hour or so (ish) earlier..
Cheers, Mick
Good point, Mick
Will investigate.
This data here is all from the AEMO, so definitely all NEM time:
https://wattclarity.com.au//2017/02/about-solar-panels-predominantly-facing-north-not-west/
With respect to APVI, I can’t remember off-hand what the time-stamps are, but would think that we would have addressed this and translated to NEM time. But will check with the team.
Paul
I guess all those ‘pink bats’ that were installed a few years back have played a small part in helping save the day.
100% agree Paul! The sooner the industry infighting stops, the sooner solutions will be delivered. If it doesn’t, political solutions will override correct, ecominc and efficient outcomes.
Sorry, that should be economic
Hi Paul
really big guess here, but could the more rapid decline in small scale solar be because its predominantly roof top mounting which may be more prone to afternoon shading from neighboring homes and trees, which large scale solar may not?
I think that would certainly be a contributing factor.
Another thing to take into account is the three major large scale solar PV sites in NSW are spread out geographically east to west (Moree, Nyngan and Broken Hill) compared to small scale rooftop PV (heavily concentrated on the east coast) so generation is more evenly distributed. This also helps explain the later start time for generation for large scale PV as Moree is the furthest east yet still 400km from the coast. The Moree solar power station is also single axis tracking which has a smoother output curve over the day than fixed position installations.
with the temperatures Qld has been experiencing, I can’t see how Energex’s Peak Smart initiative would be too popular. It seems to me that if it is hot then the last thing people would want is for the network owner to throttle back their airconditioners.
The little blip at the top of Qld’s demand curve seems to indicate some load control but it is so small now relative to the MD that I wonder whether it is worthwhile messing around with load control devices at the domestic level. I wonder how much hot water load is actually switched off during these hot afternoons. Not much I would have thought.
And while we are on the subject of hot water, how many homes in Qld have solar PV and electric hot water ? Such craziness astounds me.
Most of the many large-scale solar farms soon to move into construction across the NEM are proceeding with horizontal single-axis tracking, and will be generating large volumes of power facing west during the afternoon peaks (just as the Moree Solar Farm has been collecting lately with NEXTrackers, last Thursday being a prime example). It’ll be interesting to see the extent to which just 1GW+ of these projects impact the shape of the duck curve.
Solar data on the Aneroid Energy website shows that the Moree Solar Farm was still producing at more than 80% of capacity until shortly after 6 pm (I presume this is local time, not NEM time). So all the new solar farms with tracking should be producing well right through the summer peak hour. However in the winter its decline in output commenced at 4 pm.
And as another reader noted, Nyngen and Moree large scale solar farms are tracking which extends morning and afternoon production….
So systematic market failure where the market does not provide incentives for efficient large generators to run when they are needed. Instead of trying to come up with an extremely elaborate market mechanism to balance all of these competing interests, we should nationalise them and take the profit motive out, so that we have the capacity when we need it. Natural monopolies like all utilities markets, can’t be engineered into a competitive market.
The fact that this is also happening in NSW and QLD shows it is an issue with the market structure and likely also rational self interested actors gaming the system.
I find it odd that small inefficient generators in SA can continually run making profit but large efficient generators can’t.
If we nationalised, we would just be facing government failure instead of market failure. Why do you assume a nationalised system would work better given governments’ track records? Why do you believe that government’s don’t face their own financial constraints that will influence how much they are prepared to invest in the system? How do small scale alternatives to grid electricity like solar and battery fit in? should we nationalise these too? If not, why not? And given solar is already competitive and some argue that in SA at least, batteries are too, then perhaps we are no longer in the situation of a “natural monopoly”.
Well said Kieran.
I don’t assume that you won’t have issues with government failure, nor that they won’t have their own constraints. More so, they would have LESS constraints than an ordinary business. Solar and battery are independent, like running a generator for your own home, you’re trying to increase the scope there by likening a 5KW generator to a 50,000KW generator. The market is still a natural monopoly, particularly with distribution.
Every issue you’ve brought up is a problem under a market system as well, except under a market system you’ve also got the additional issues of regulatory capture and gaming the system.
Markets aren’t magic.
“Solar and battery are independent, like running a generator for your own home..”
True but that’s a rare situation wrt the communal grid whereby there are lots of rooftop solar generators with no storage capacity bludging off the insurance of the grid and paying no insurance premiums. Quite the contrary as many are being actively subsidised to bludge off the grid like mine is.
Uriah & Kieran
Putting aside the public / private ownership question, a different issue worth considering is the generation technology mix in South Australia (and elsewhere in the NEM). While large scale combined cycle gas plants are thermally efficient, they are not nearly as flexible as some other forms of generation particularly with rising gas prices and increased amounts of variable renewable generation pushing them towards peaking / reserve operation. Here I mean “flexible” in an economic sense as much as an operational one (however CCGT plant is generally less flexible in terms of startup time, minimum run time etc than open cycle turbines).
We can see this in the decision by Engie to reduce the quantity of gas contracted for the station. Contracting gas supply and transportation for the full plant capacity would involve committing to higher fixed costs, with an uncertain likelihood of recovering these costs from a limited number of opportunities to run economically at full capacity (due to high running costs on gas which has become a much more expensive fuel than it was when Pelican Point was built, versus the very low marginal cost of renewables).
Regardless of who owns Pelican Point this underlying economic shift creates a difficulty in economically optimising the operation of a station like this and maximising its value to the system. For true peaking / reserve operation in an environment of high gas prices, the best form of “gas-fired” plant might actually be open cycle turbines running on liquid fuel – the lower fixed costs of both the generation plant and fuel supply outweighing the higher costs of relatively infrequent running. Hence “small inefficient generators” running “profitably” and Pelican Point at only half capacity!
Allan
Some people are pointing fingers at the wind power lull, but it is the wind power peaks that have killed the concept of “baseload” in South Australia. Generators like Pelican Point and Osborne operate in baseload mode, maintaining steady output, between infrequent step changes. In recent years baseload generators have been often kicked off the system by wind power surges, so its not surprising that some have closed or been mothballed. Ordering loss making generators to be on standby does not seem like a complete solution.
A major long-term problem for SA is keeping Torrens power station in operation, it is the workhorse load-follower, surely the most important electricity asset in the state.
Thankyou for an excellent and easy to read analysis of what happened with lack of supply from Pelican Point.
So, at the end of the day, if a generator can’t guarantee that it will make its money back on providing electricity to the grid, it won’t, unless they are directed to by the AEMO in which case any excess charges will be covered.
Sounds fair to me – they are not operating as charities, or even as public services.
Also sounds like a failure of the AEMO to predict usage.
How much of the planning and foundations of the NEM took the impact of secondary suppliers’ pricing into account when the national model was developed? Are there just too many links in the profit chain?
Paul
Just come across an article in which you referred to a “first of its kind” dual solar tracker being installed at the Gatton Solar Research Facilty.
Can you provide any more information?
“Under the National Electricity Market (NEM) rules, generators can not bid plant into the market if supply can’t be guaranteed”
What a laugh that is with renewables and until the AEMO has the power to restrict any tenderer of electrons to the communal grid to the maximum they can guarantee 24/7, all year round this Govt sponsored form of domestic dumping will continue without a level playing field that consumers require. If that mandate was implemented, then renewables would either have to partner with thermals to lift their average tender rates and/or invest in storage, whereas at present they’re freeloading on thermals as insurers and not paying them their just premiums. What a crock that is and the ACCC would normally have the power to jump on such dumping practices should it occur.
A simple analogy would suffice if say a Labor Govt wanted to privatise the Fire Brigade and proudly announced the CFMEU tender was the lowest priced and has been accepted. Just a wee problem in the fine print folks. There won’t be any fireys on duty when the sun goes down and/or there’s no wind or wind over 90km/hr. Yeah right!
So you’re saying small gas generators that have low start up costs by higher operating prices should not be allowed to turn on in peak periods and off in low periods? You’re saying when gas prices go up and they need to shut down, they should be forced to stay on?
I’m just drawing a comparison with your standard for wind and applying it to gas as well. Small generators have a benefit over large generators in some regards, so the large ones are “insuring” the small ones. This isn’t unique to wind.
Fluctuations happen in resources and these have flow on effects. You can debate over their size, but that’s about it.
“So you’re saying small gas generators that have low start up costs by higher operating prices should not be allowed to turn on in peak periods and off in low periods?”
No I’m not saying that at all because clearly those gas peaking plants, short of a breakdown, can produce up to their maximum output anytime demand and pricing calls them forth. OTOH they couldn’t match a base load cogen gas plant like Pelican Point that would have long term gas contracts for the whole of their plant and it would be up and running and rather than 6 hours lead time to fire up the idle generator that could easily ramp up output with both gas generators already operating. That’s because they’d either be insuring fickle wind and solar and getting their just insurance premiums, and/or the unreliables would be paying for battery, hydro or molten salt storage in order to compete on a level playing field.
We need high priced peaking gas plants too but what we don’t want is fickle generators dumping in any part of this grid equation.
“I’m just drawing a comparison with your standard for wind and applying it to gas as well.”
Actually on that point there’s no reason why it shouldn’t also apply to large base load plants like Pelican Point too. ie under their 24/7 all year round maximum they wish to guarantee to the AEMO they’d have to in turn partner with fast start gas peaking plants to insure them. Nevertheless creating a level playing field would largely impact the unreliables who are at the mercy of the elements and rightly so if we as consumers aren’t to be.
Err guys, try ruling an average 30% average line through this graph and then imagine the wind turbine industry not being able to tender any power above the line and having to make up any power below the line- http://anero.id/energy/wind-energy/2017/february
Get the picture?
Not to worry folks the CFMEU Fire Brigade quote includes an ironclad guarantee only union fireys will be employed, whereas the other nasty capitalist tenderers won’t guarantee that 😉
Here’s the big storage problem for solar-
http://breakingenergy.com/2014/10/29/at-ivanpah-solar-power-plant-energy-production-falling-well-short-of-expectations/
http://www.powermag.com/crescent-dunes-solar-energy-project-tonopah-nevada-2/?pagenum=1
and ditto wind with any mandated level playing field which is why State Govt’s like SA refuse to acknowledge what’s been going on, but with base load coal increasingly being run on sticky tape and string to extract the last bit of revenue out of them before they shut down, their day of reckoning is fast approaching.
Roger, I think you make some good points. Its a pity you use such a belligerent style — it detracts from your essential points. That we pay the same $ per MWh for dispatchable energy as we do for semi-scheduled and non-scheduled energy seems to not be a proper recognition of the value of the energy offered. Because of the variability of both sources it increases the daily range and variability that the load following plant has to account for, and its more expensive for that plant to operate in that environment. As we see here in Allan’s excellent analysis of Pelican point. It seems inappropriate to me that the generators that contribute to the creation spikes by their variability, also enjoy the high prices. Because wind and solar gets “dispatched” first, and gets paid the spot price price, there is no incentive for them to invest in storage of some sort. If there was some discount from the spot price to reflect their non-dispatchable nature it would encourage storage.
Mate, I get belligerent when the lights go out because too many technical experts have been too nice for too long, while belligerent watermelons have cowed them into shutting up and going along for the ride as they foist their theories and unreliables on us all as consumers. To put it bluntly to these leftys who turned their red coats inside out to wear the green on the outside when the Berlin wall fell and their ideology was in tatters, all electrons are equal, but some are more equal than others. Ipso facto I gave you the only rational, sensible way to make them all equal, whatever the world’s omniscient guru thinks should be the same tax price for all to price their input CO2, in order to get the world’s atmospheric CO2 back to the level the omniscient guru knows it really should be. I am not that guru but I sure as Hell know unequal electrons when I see them and the time for pleasantries is over.
Roger S:
The NEM is not a creature of the environmental movement, but of economists. The environmental movement has been advocating markedly different strategic policies in this sector for more than 40 years. Governments have preferred to take their advice from economists and the conservative press and have failed to predict or to plan for the train wreck that is now coming upon us at full steam.
A coherent policy has been made difficult if not impossible by the fragmentation forced upon the sector by vocal advocates of competitive free markets in the conservative press and conservative think tanks.
Question for you: Who is able to effect fundamental reform of the National Electricity Market? Who is in charge of the regime? It certainly ain’t the environmental movement.
Chris
Thanks for your comment.
While it’s true that within any given half hourly settlement period all generators earn the same spot price for their output (before adjustment for locational network loss factors), the average price that different generators earn across a given day, week, etc can vary widely depending on how each generator’s output moves as the spot price changes.
Taking just the 8 hour period midday – 8pm in SA on the day of load shedding, wind generators’ output fell steadily through the afternoon, while thermal generation ramped up to fill the gap and meet (not quite, unfortunately) the increasing demand (see https://wattclarity.com.au//2017/02/initial-analysis-sa-load-shedding-wed-8-feb-2017/ for a chart showing this). Prices in this period rose from a few hundred dollars per MWh up to the Market Price Cap of $14,000/MWh whilst load was being shed.
The demand-weighted average spot price over this period was $4,642/MWh. Wind generators as a group, with much more of their output occurring before the price spiked and much less during the period of extreme prices, earned an average price of $1,584/MWh, a 66% discount, while thermal generators as a group earned $5,206/MWh, a 12% premium. Within that thermal group, generation that ramped up specifically for the extreme price period would have earned a much larger premium.
This is an extreme example, but it clearly shows that the spot market – as intended – does send price signals rewarding dispatchable generation able to contribute more during periods of tight supply-demand balance. Previously on WattClarity, Paul has posted some interesting articles including this one commenting on the increasing discount to average spot prices being earned by windfarms in South Australia.
Aside from the energy price, there are other mechanisms – principally the markets for Frequency Control Ancillary Services – that can reward / penalise generators for ability or lack thereof to contribute to stabilising system frequency etc, again related to the “controllability” of those generators’ output. This is an active area of debate and development in the NEM, as the need for and nature of the services required changes within a transitioning system.
Allan
Hi Allan,
Thanks for taking the time to explain that not so obvious effect of averaging, and that it does create an incentive for dispatchable generation, and therefore for storage. Thinking about the main thrust of your article, the day ahead bidding requirement makes it quite awkward for generators that are market exposed for their fuel cost. I suppose this would be similar for pumped hydro, because their fuel, the power to fill their upper reservoir is also market exposed. For a large pumped hydro with a daily charge/recharge cycle, the day ahead bidding would require them to bid to sell energy they haven’t secured yet. To charge the upper tank they would be bidding as dispatchable load, and if their bid mean they did not get dispatched as a load, they then would not be able available to generate. That seems to place them in the same predicament as Engie, having uncertainty at both buying and selling. On the other hand, smaller unscheduled generators don’t have to commit ahead of time, and can respond the actual market price at the time, a much safer alternative, than having to commit in advance. The playing field seems quite in need of some levelling.
I have an idea, it comes from recycling something that worked.
Lets look at what the SECV did successfully for 85 years.
Take the profiteering out of the equation and treat it as an essential service
So there seems to be a lot of conflicting information.
http://www.afr.com/news/politics/wind-and-thermal-generators-failed-sa-in-blackout-says-aemo-20170215-gudbkq
http://www.joshfrydenberg.com.au/guest/mediaReleasesDetails.aspx?id=304
But the fact remains that Pelican Point could have been on in less than 15 minutes, so I’m still not sure what was at play.
G’day Paul, hoping you can help resolve a bit of confusion. The first graph is headed ‘Trended Aggregate Solar Output (small scale and large scale), while the annotation is “the aggregate numbers from the AEMO (historical small-scale solar data from AEMO…) (my emphasis). If the latter is *only* small-scale solar, why are the numbers almost identical to what would be obtained by simply reading off the graph? Shouldn’t they be appreciably smaller, if they omit the large-scale input?
I have a 3.8kW PV array on frames to reorient them N on my WNW roof slope. The reason for this is to locally power the washing machine in the morning on days when the solar dryer (clothesline) will then dry the wash that afternoon. By such load management our usually 2-person household pays nothing for electricity, only interest on investment. Energy exported at 25c/kWh for the first hour and 8c/kWhr thereafter pays not only for power drawn (at 25c/kWhr) from the grid at night and cloudy weather but pays for the line rental as well. The solar hot water helps considerably of course.
It does not matter that the peak output does not coincide with peak airconditioning demand because we don’t have an air conditioner. Our Hardiplank/brick/Colorbond house is well insulated and we open all the windows at night (in summer) and close them and curtains during the day. But even if we did have an air conditioner we, like everybody else, could effectively match our demand to the PV output peak using a timer on the air conditioner. With a sensible pricing scheme utilities could make it worthwhile for everybody to run their air conditioner during the N facing PV peak and shut it off as demand rose for the evening cooking demand peak. Heat and “coolth” represent storable energy and if the house is cool when people arrive home from work it can stay cool through the evening.
Storage batteries and pumped hydro would be useful for other load mismatches but we don’t need them for hot days. Thermal mass does the trick.
-Flexible generation already gets a price premium and wind already gets a price discount.
-“AEMO supply reliability leads to interconnection” what?
-AEMO’s capacity contribution assessments had nothing to do with supply shortfalls in SA
This guy is just trying to drum up business by chucking all the keywords in one place.
Here’s one fix: Bipartisan policy then get out of the way.
What we need is proper planning for load shedding. Cutting supplies to an aluminium smelter should be a non-no. Strict measures to reduce demand and to stop demand growth, e.g. passive solar houses instead of energy hungry residential towers. There is also need to cap power consumption of households at the meter during peak periods.
High pool price spikes suggest system has reached its limits and/or there is market manipulation.
As for the 5 min bid stacks imagine we had auctioning of petrol at our filling stations every 5 minutes!
I agree with some of your points; however there is now a need to move towards a capacity market to ensure that capacity is always available.
This could be done by AEMO invoking its Reserve Trader Powers but they seem to have been missing in action on this to date.
I do not agree with raising VoLL.
if anything VoLL should be reduced. it adds a huge risk premium to the market and it has yet to achieve its stated aim of being a signal for new generation.
It should be abandoned in favor of a capacity market which values capacity, not just energy which may or may not be there when you need it.
“It should be abandoned in favor of a capacity market which values capacity, not just energy which may or may not be there when you need it.”
There are 2 parts to that overall emergent problem. The traditional but growing one of peak demand whereby aircon use is demanding 20% more generation and distribution capacity for a few weeks of the year in summer and that’s particularly correlated with Adel and Melb weather. In that respect it makes power bill sense to price accordingly i.e. consumers with aircons can choose lower prices if they sign on to having a smart meter that can demand manage their aircon compressor for 20% of the time or stick with full user pays. The other biggie now for consumers is getting reliable electrons anytime, summer and winter, regardless of smart metering in summer, but that’s where my level playing field policy comes in. Under that regime the AEMO will know exactly what amount of reliable electrons are available in total and so will all the market players with future planning and investment in mind. Not so with the current disastrous regime.
“There should be no need to throw out the energy-only market design, but there is opportunity for market-saving innovation, perhaps by introducing a price premium for output from scheduled generation, paid for by a discount on the output from semi-scheduled generators (mostly, wind) that only ‘turn up when they can’.”
The notion that some omniscient public servant can work out the optimal price premium and discount between reliable and unreliable electrons tendered to the grid is to believe I’m the second coming, which sounds fine to me if I’m paid according to my new station in life. Much wiser for mere mortals to set the level playing field and let the multitude of players work it out between themselves. Ipso facto that means EVERY tenderer of electrons to the communal grid can only tender up to a maximum they can guarantee 24/7 all year round and let the market work it out.
I don’t know what sort of cross transfers would eventuate with the only obvious level playing field outlined, but clearly fickle wind and solar would have to pay reliable thermal partners their just insurance premia in order to up their average tender quantities (wind had an average of 29% of installed output only in 2016 but what about all their dips below that line?) and/or invest in storage to achieve it. Let them work it out over time between themselves, or else vertically integrate their generation mix. All I want is reliable, despatchable electrons, with appropriate compensation if I don’t get them and judging by the political ruckus at present, I’m not alone.
The free-riding is easily demonstrated looking at wind output as a proportion of installed maximum output- http://anero.id/energy/wind-energy/2017/february
and astute observers will notice the below zero graph lines and perhaps be interested in switching the graph to MW top right.
In 2016 the average output in Australia was 29% which marries well with an oft quoted figure around 30% of installed output for wind energy in general. Simply draw a horizontal line at 30% and look at all the dips below that line and their negatives that wind generators would have to pay a despatchable thermal insurer to cover them (fast start peaking gas perhaps?)or else install storage to be able to use some of those peaks. Notice also that to the extent they rely on thermal to backstop/lift their tender levels that reduces thermal 24/7 all year round tender capacity by the same amount, but the bonus is the AEMO and all players then know the absolute level of reliable electrons that are possible and can plan investments against any rising overall demand with confidence. Not a hope in Hell of private enterprise doing that now with the wild roller coaster they’re on, but if you think public servants can do it better with our taxes, then you’re part of the problem. The problem being you aren’t looking at the big picture, but are instead stuck within your blinkered speciality like so many-
https://wattsupwiththat.com/2017/02/20/climate-science-and-climatology-specialization-and-generalization-forest-and-trees/
Unfortunately a national electricity grid knows no such bounds.
Yeah no. The spot price you are referring to isn’t the actual price experienced by the customer. 90% is contracted at a significantly lower rate, more like 5c/kwh rather than the low 8c given to the ‘poor’ home solar people (excluding transmission and retail of course). Perhaps we should estimate the massive waste of taxpayer money incurred when home solar owners were reimbursed 60c/kwh gross (irrespective of whether they used that power or not) when the average price for generated power was 3c/kwh over the 3 or so year period the scheme was in place. Which one do you think actually hurt your wallet more?
Hi Adam,
Thanks for your comment. The analysis in this article relates to the wholesale (spot) electricity market. All retailers pay the spot price to AEMO, and conversely all generators receive the spot price from AEMO. Most market participants also hedge their exposure to spot pricing in order to firm up revenue (for generators) or cost (for retailers). Hedging instruments are settled against the spot price and hence their price is largely determined by the market’s expectation of spot pricing. The current futures pricing for NSW baseload Q1 2017 is about $120/MWh or 12c/kWh.
Residential consumers pay a fixed rate to a retailer. Of course, this fixed rate incorporates the costs to the retailer as explained above.
As for the solar bonus scheme (which has been discontinued), its purpose was to encourage the uptake of solar PV in the early stages of adoption in the Australian market and it has certainly achieved that goal. Yes, 60 c/kWh was a high feed-in tariff, but remember that solar panels are a 20 year asset. We will be reaping the rewards for a long time yet.
The owners of sparingly-used assets such as the Port Lincoln and Colongra power stations must be wondering why when those assets were required, they failed and were unable to bid into a a high-priced market. Inadequate testing prior to forecast demand peaks ? Perhaps the market operator should require twice yearly testing of their systems before the demand peaks in later summer and mid winter.
Paul
Wattclarity is an amazing source of information
While I might not agree with all of the conclusions reached, the process is clear and the information provided is accurate so we can use it to help us better understand the highly complex NEM
Please keep up the good work
Thanks for keeping it going
David
I realise this is largely a rhetorical question, but I’m going to answer it anyway. Presuming you mean ‘x works in the industry, therefore rather than x being an expert whose opinion should be respected, x is a vested interest who should be ignored’, I reckon I’ve been noticing this sort of sentiment in these circles for nigh on ten years.
Hi,
I note Jamey’s comment (below – from todays topic) that those on the $0.44c feed in tariff are still eligible for a further retailer feed in. Is this in fact the case. If so, I would really appreciate some info re the application process.
Hi Sean
In my case, I am one of the lucky 400,000 in QLD on the 44c FIT and I do receive a bit more than that (50c or 52c in total, can’t remember specifically) with a particular retailer whose name I won’t mention. I believe pretty much all of them do it, so shop around.
Paul
Apparently it wasn’t only the media getting skittish Paul – AEMO directed additional Pelican Point capacity into the market that afternoon despite no sign of forecast LOR2 on the horizon. There was an actual LOR1 declared but reserves were still over 450 MW before the direction.
Re the 44c FIT. The people who received this had a higher exposure to finance costs of construction, a higher exposure to risk (knowledge of the life of PV installed was lower confidence) and made commitments in a time where government *had* to encourage activity to build a market.
If you compare people who squeezed in ‘under the wire’ at the end of the 44c FIT to somebody who applies shortly after its cancelled, yes, it “looks” unfair. I think the government could have done a sliding scale tail-off, or a number of other things.
But this is really no different to somebody who gets first round shares, before float, and somebody who buys at market rate post float: there are winners and losers, there advantages and disadvantages, there are risks and rewards.
I don’t have solar. I value PV and wind immensely, and I am delighted that part of my bill goes to a social offset and equalization to people who have invested in PV and yes, that includes the 44c early adopters. I’m glad 44c ended: it wasn’t sustainable. I am sad 11c has been dropped to 6c and lower, I think its disincents people, from long term capital investment in PV to all our benefit.
But this really isn’t “fair” and “unfair” -it just feels like it is. If you want “fair” and “unfair” then lets talk about how people get to invest in coal and gas, and are not exposed to the longterm remediation costs because of systematic rorting of the cleanup cost.
Perhaps the emergency capacity for the Heywood inter-connector should be up-rated. It’s already had last Friday’s stress test and passed. Currently the emergency level is specified on AEMO’s website as 701 MW for summer, 717 MW for autumn and spring, and 762 MW for winter. Surely a more sophisticated set of temperature-controlled limits could be applied, as on the Vic-NSW inter-connector.
I have only just signed up to get you email which is very informative and interesting.
While i have read what you say you do it is hard to get my head around someone wanting to do this without some kind of angle – you know the old saying that there is not free lunch.
Would you like to tell me a bit more about why you do what you do, which as I said I am very interesting.
Cheers,
Ken in Adelaide.
Drop in load was probably due to “load relief” – as network voltages fall, so too does demand – its the common Np and Nq indices power engineers are familiar with
So far as the “missing” load is concerned I wonder if consumer appliances, specifically modern inverter type air-conditioners, may have tripped on account of a sudden drop in voltage and then re-started a few minutes later?
I don’t know how these systems respond to sudden voltage fluctuations but if the load did indeed go “missing” and it wasn’t due to load shedding or a large industrial load being tripped then there aren’t many other possibilities given we’re talking about 300MW or so.
Losing four operational units to one transformer fire does not seem like ‘single failure proof’ design.
If no distribution level load shedding occurred, then the shed system load may have been directly connected to the HV grid and shed either automatically or manually by AEMO?
Directly connected HV would normally be big industrial loads. Does SA have ~300MW of those? Reconnecting those loads could have been done gradually as additional generation synchronised and loaded.
This will be due to a combination of issues.
The most significant reason probably is that the other generators both synchronous and non synchronous did not drop their skirts and run away. There was a lower percentage of non synchronous generation running than for the September event and the majority of the non synchronous generation was remote from the event and probably did not see any severe voltage or frequency fluctuations anyway. So it didn’t all trip as in the September event and the synchronous generation close to the event would be more forgiving.
The second is that the measurement is probably not correct – mismatched instantaneous readings from a transient event – and you would need to examine carefully the actual voltage and frequency fluctuations at various points in the network.
The indicated drop in power is not all that great at just over 12.5% of the load immediately prior to the event. There obviously was a significant drop in voltage – reports of lights dimming – so all resistive load will have immediately stopped pulling a fair bit of power.
Likely we would probably also be very surprised by the actual load that was shed and then reconnected quickly. Every UPS in the state would have flipped to battery (and all have probably had a service/new batteries since September had just identified all the flakey ones). Every capacitor in every electronic power supply gave up a bit of it’s spare store. Quite a few sensitive large loads probably did disconnect and then immediately reclose without any real impact on their connected plant. Some loads probably stayed off but have not made the press because they regularly do that during such events. The reasons that these did not become “press events” is that users could immediately go about restarting them so there was no great damage – it just becomes one of the regular unscheduled events that operators are trained to handle
So in summary there probably was a fairly significant drop in load but it did not become an “event” because the network stayed up – probably thanks to the greater percentage of online synchronous generation than September. And the remoteness of the non synchronous generation meant it was more isolated from the event.
The more you look at the September event the more you begin to wonder if a few nutters distributed though the management systems (AEMO included) were looking to the upcoming wind event as some sort of opportunity to push the barrow of wind capability. It begins to look as if they were almost trying to break some record for the percentage contribution of wind to the SA network – perhaps trying to get to 100% of SA power from wind and took their eye off their prime roles of actually keeping the power on. Even before the event everyone would have know that too much non synchronous generation was a risk – but everyone seemed to ignore that risk. And high winds are always a network risk even if they only bring down a few isolated conductors and tree limbs.
But in September those in the know seemed to deliberately ignore the risks and plodded on blindly with a commitment to loads of wind. This recent event shows the stark contrast in the outcomes. Although they didn’t have the choice this time as the output from the wind and solar resources was relatively low at the time.
A high voltage fault at a large generating station is one of the worst due to fault infeed availability. Hence, depending of the nature of the fault, strange protection system operations can happen. Losing multiple generators during this type of event is not unheard at all. A fault becoming a 3-phase fault with possibly delayed clearing is very extreme.
All I know about this event is that some piece of equipment probably failed explosively. It happens. With a close-in unbalanced 3-phase fault, generator neutral currents can cause their protection systems to operate. If the failed component impaired protection systems, resulting in a slow clearing of the fault, the event will become quite severe over a large part of the grid. Compressors in air conditioners and fridges will see a voltage loss and kick in restart delay timers/breakers for a period of time. Other voltage sensitive equipment and synchronous motors disconnect and also have a delayed restart. Back in the day, we used to call it “load shaken off” for the lack of a technical term, so you get the idea. Unplanned load cycling.
Power systems cannot be designed for clean, controlled protection for every type of event. I am unfamiliar with the SA protection requirements but in North America, for instance, most areas are designed for proper protection for either a 3-phase fault or a single phase fault with delayed clearing. Hence you can have events that exceed design capabilities but that have a very low probability. Perhaps this was indeed a 3-phase fault with delayed clearing.
Thanks everyone for the comments, some great insights there.
Some specific responses to people’s comments:
Malcolm – I’m not quite sure which piece of data you are referring to (a reference / link would be good), but the numbers look close to the per-circuit thermal ratings (MVA) on the two 275kV Heywood (Vic) to South East (SA) transmission lines. They might represent the highest levels AEMO would be prepared to schedule Heywood interconnector flows to in an emergency (and only for a very short time, << 15 minutes) but there may be other limitations and in any case would not leave the system in a secure operating state. In general the maximum secure transfer capability of the interconnection is quoted at 600-650 MW but frequently other constraints limit imports to considerably lower levels.
Ken – yes there’s a commercial angle to this but it’s primarily about trying to demystify the workings of the market, and you can see from the other commenters’ input that I learn plenty from doing it as well.
Andrew – that form of very short term load relief on frequency / voltage dips certainly exists and assists with riding through the immediate aftermath, but doesn’t persist for the 10-20 minutes we saw last Friday. I think what happened may be more along the lines suggested by Shaun, Michael & Tom.
Shaun – I think you, Michael, & Tom are probably on the right track – it will be interesting to read what AEMO says. Your comment about inverter-connected loads also made me reflect on the large amount of inverter-connected PV generation – had there been a large scale dropout of that generation last Friday then supply-demand balance would have been that much harder to sustain over the interconnector!
PaulD – I don’t believe there is that much large, direct connected load in SA available for contingencies like this (whereas in other mainland states smelters can play this role), and in any case I’m sure we’d have heard about it if (say) Olympic Dam had tripped.
Michael – Great comments, thank you. Relative to the September event, while there was certainly a lot less non-synchronous generation running on Friday, remember that in September it was the loss of nearly half that generation that led to the collapse. Last Friday it was well over half of the running synchronous stuff that fell off. If you take the pre-event September outputs by generation type, then adjust for the loss of 465 MW of wind, you end up quantities of thermal and large-scale non-synchronous output that are actually pretty close to the mix in SA last Friday once the three TIPS units and Pelican Point had tripped – around 400 MW of online thermal, 4-500 MW of non-synchronous wind, and nearly 1000 MW over the interconnectors. On my reading of the AEMO System Black report it wasn’t then a trip of the remaining online windfarms that precipitated collapse but tripping of the Heywood interconnector on loss-of-synchronism protection. I know what happens in the immediate aftermath of a big system disturbance is nowhere near as simple as my MW output comparison here but it does point to the load response last Friday (and clearly no loss-of-synchronism event) as being pretty important. And thanks for the insights into load behaviour after a big voltage dip – which would have been much closer to the load centre last Friday than it was in September.
All I’ll say on your thoughts on operation of the system on Sep 28 is that 1) there had already been plenty of previous occasions where wind output exceeded SA system demand so I really doubt any “nutters” were going for some sort of record and 2) I remain amazed that more than 5 months after the Black System event there has been almost no sign of any independent review of how the system was operated on the day.
Tom – thank you also, some of your comments address matters far outside my range of technical competence, but again the insights into behaviour of connected loads are really helpful in understanding why the system was able to ride through.
Allan
Below is a link to the file of AEMO line ratings I referred to. My reading of it (rows 11229-11246) is that there are 2 circuits between the South-East terminal station in SA and Heywood. The summer rating for each of 2 lines is 597 MW under normal conditions, and 701 MW for emergencies.The normal rating for the Heywood transformers is 3 x 370 MW (rows 3552-3569). Both of these are ~1100 MW and much higher than the publicised target rating of 680 MW.
In the file, the ratings seem to bear little relationship to the underlying physics or appearance of the line. In some State ratings are based on seasons, while in Victoria it is temperature. Some lines with few conductor cables seem to have disproportionately high ratings (eg Ballarat-Bendigo 347 MW at 20 degrees). The ratings seem to have a stronger roots in history than physics, and managed with excessive conservatism. I suspect that the best investment in the NEM would be to re-rate the inter-connectors based on sound physics, with dynamic ratings calculated from air temperature, wind speed, and transformer temperature. Re-rating won’t necessarily be cost-free, because it may require stress tests with helicopters flying the length of the line using thermal imaging cameras.
https://www.aemo.com.au/Electricity/National-Electricity-Market-NEM/Data/Network-Data/Transmission-Equipment-Ratings
To interpret it needs to following diagram
https://aemo.com.au/-/media/Files/Electricity/NEM/Planning_and_Forecasting/Maps/Network-Diagrams-pdf.pdf
Allan, the numbers I have got for the September event has Wind providing 45% and Gas 10% of SA load prior to the event. For this event you have Wind only providing 18% and while Gas is at 42% prior to the event – quite a different generation mix. Solar at 13%/18% (last time vs this time) is pretty much the same. Imports at 529MW/613MW (last time vs this time) did give more headroom this time for imports.
I don’t think that September can be split into two separate events such as failure of wind and then separation from the NEM as separate independent events.
Michael – yep I accept that it’s artificial to split the September event into wind failure followed by Heywood trip; however the 456 MW of wind lost did occur in two main tranches – about 200 MW triggered by a voltage dip seven seconds before the interconnector opened, then another 260-odd MW less than a second beforehand, after a sixth major voltage dip. AEMO’s comment was “Voltage collapse began after clearance of the sixth voltage disturbance and sustained generation reduction of 260 MW associated with three wind farms (Groups B and C). This confirms system voltages were stable and would have remained stable if angular instability had been avoided” – see p39 of the most recent System Black report. So I think the event could be at least be split pre- and post- that 6th voltage dip (which was due to another transmission fault – see Table 6 / p26).
As the same report makes clear, the windfarm runback / disconnection was due to programmed control system actions following the number of ride-throughs exceeding pre-set limits, rather than failure to ride through the voltage dips themselves. I think that may be important in thinking about how that generation handled the primary faults on that day.
Your percentages for gas and solar generation prior the black system don’t look right at all, see p18 / Fig 1 showing 330 MW of thermal (~18%) and p24 where solar PV is estimated at 50MW (<3%).
Finally and most importantly, gas may have been 42% prior to last week's event but with more than half that tripping off, the percentage available to then respond was well under 20% of pre-event system demand. I still think that if demand had not fallen substantially, the outcome last Friday may have been quite different.
Allan
Thanks Graham for a good and concise explanation of the issues.
The hard part is to get action now so that the large gas users don’t disappear. Then again, if the large gas users do disappear, so does much of the domestic gas problem. There would be lots of spare gas and lots of spare electricity too! But if that occurs, so do future investment and jobs disappear and that increases the current account deficit and narrows the export (or import replacement) base,
Hmmm. Maybe we do need a solution
“Thanks Graham for a good and concise explanation of the issues”
With all due respect Graham is mainly pointing out the symptoms of the patient’s fatal disease-
http://anero.id/energy/wind-energy/2017/february
I would have thought somewhere along the line someone would have called out the emperor has no clothes on. That’s because it’s a fundamental axiom of engineering that you can’t build a reliable system from unreliable components and that’s why there are lemon laws around for simpler cars. A second year apprentice mechanic would understand it. No matter the emperor knows best about disproving that fundamental axiom with no countermanding precedent existing in history so here we all are with everyone looking every which way or donning dark glasses. The emperor is starkers gentle folk!
@Allan,
It looks to me like they have fitted UVLS (note: not UFLS) to the big industrial loads since the September blackout. It seems to have saved the system.
You state :-
‘SA Power Networks seeking an explanation – the latter has confirmed that there was no distribution system level general load shedding’
Perhaps the next question for SAPN is, “Did the total distribution demand vary over the timespan of the incident and by how much?’
Do we have to postulate that the solar PV stayed attached and the air-conditioners, fridges and freezers tripped because of a voltage dip? I doubt you would notice a ten minute break in air-conditioner operation; it would just look like the thermostat operating. However, …
… UFLS and UVLS:
The system frequency in SA would not vary much during the incident because it was attached to the rest of the NEM grid. It is this fact that makes frequency response in SA ineffective with the Heywood interconnector operating.
I know nothing about UVLS systems in SA but maybe large loads have them fitted? This looks relevant :-
‘Kill switch’ for mines to save electricity grid – The Australian
http://www.theaustralian.com.au/national…for…/9f99512ea64c98b7ef6975f49e050235
4 Oct 2016 – BHP Billiton’s Olympic Dam mine in South Australia. … “The under-voltage load shedding scheme … has been installed as one of the interim …
[PDF]SA Events – update report – Australian Energy Market Operator
http://www.aemo.com.au/-/media/…/AEMO_19-October-2016_SA-UPDATE-REPORT.pdf
19 Oct 2016 – Report – Black System Event in South Australia on 28 September 2016 and …… Availability of under-voltage load shedding at various locations, …
‘the grid manager has been directed to trip a ‘kill switch’ …
Looks like it works!
https://stopthesethings.com/2016/10/04/south-australias-wind-power-debacle-killing-miners-mineral-processors-scarce-jobs/comment-page-1/
Alas. The latest ‘plan’ from its vapid Premier, Jay Weatherill is aimed at keeping the lights on in Adelaide (where the voters are), while cutting power to its miners in the North of the State.
In order to prevent the grid from collapsing every time there’s a total and totally unpredictable collapse in wind power output (see above and our posts here and here), the grid manager has been directed to trip a ‘kill switch’, that will cut power to SA’s two biggest mining operations: BHP Billiton’s Olympic Dam and OZ Mineral’s Prominent Hill.
“The under-voltage load shedding scheme … has been installed as one of the interim measures that will eliminate the risk of voltage collapse spreading throughout the transmission system,” the report said. “This risk if not addressed had the potential to cause widespread power outages under certain network conditions.”
Did Whyalla spin up its gas turbine as well?
‘An update for Whyalla says they have managed to avoid disaster there, courtesy of their gas turbine but they’ll probably still be taking $30 million in losses from the blackout.
http://www.adelaidenow.com.au/news/the-whyalla-steelworks-has-averted-a-worstcase-blast-furnace-shutdown-but-arrium-expects-a-30-million-hit/news-story/a008ff502c41f01b318905b115e6f031‘
Well with today’s Advertiser article kicking off with-
“Largest power station at ‘end of life’
THE owners of South Australia’s largest power station have likened it to an elderly person sitting in a retirement home…” we can expect gas fired Torrens Island to be closing down soon like Hazelwood and Port Augusta before them. There’s only so much you can wring out of sticky tape and string as the recent fire at TI attests. We’re living on borrowed time with no reinvestment in thermal power stations as the unreliables free-ride on that insurance with their own negligible marginal cost of production. It remains to be seen how consumers react when they’re stuck with too many unreliable electrons with no more backup.
The remedy is obvious but the political fallout to administer the correct medicine now means waiting until the patient is in emergency intensive care on life support.
Although there was no distribution level shedding, I suspect there will have been automatic HV underfrequency relay trips at the transmission level.
The big difference between this event and last September was, because the interconnector had some reserve capacity and remained on, the frequency held up sufficiently to allow the underfrequency relays to operate and shed that required 300MW or so.
In September, the frequency fell 7HZ in one second, too fast for the underfrequency relays to shed load.
Graeme – I’ve seen some high resolution data on system frequency – not at all SA locations admittedly – and it doesn’t seem to show falls to anywhere near as low as 49Hz, which is the trigger point for the first blocks of UFLS. Unless you mean direct-connected transmission customers (and I believe there are < 100 MW of such load), other loads on the transmission system are distribution offtake feeders etc, and had any of those been tripped by UFLS action then SAPN would certainly have been aware of that. SAPN's comment to me was that they believe the load reduction was along the lines of what Tom referred to above as "load shaken off" by the system voltage disturbance.
But we'll have to wait for a detailed report on this to be sure.
Allan
As an experiment, I tried flicking the supply switch on my running heat pump. It stopped, unsurprisingly, and did not restart for several minutes. It did not wait for the house air temperature to drop to its normal minimum temperature, so the restart was not the normal thermostat operating.
The pump manual says :-
Frequency ON-OFF Max 4 times/h
Inching protection 10 minutes
ON and OFF interval 3 minutes
Power source voltage +/-10%
A sufficient (exact magnitude unknown) voltage dip that reached the extremities of the system would seem to be able to stop all the running air-conditioners and the automatic restarts are postponed for several minutes by design. Would this help to explain some of the ‘missing demand’ and the subsequent demand recovery? The solar PV presumably remained attached.
Perhaps it was unwise for the gas industry to fight against a domestic reservation policy. Domestic reservation is not good economics, but very good politics, because it means the gas industry would have some political allies. As it is, who are their allies ?
– the local manufacturing industry hate them because gas prices are going up so much
– taxpayers hate them because the gas industry pays no royalties
– environmentalists hate them because of high fugitive emissions of methane
– farmers hate them because of damage to their land, fed by stories of how the industry has behaved in Queensland
– the State LNP don’t like them in Vic and NSW because the farmers hate them
So for anyone not part of the the gas oligarchy the answer is … much more solar (including storage), some LNG imports, and partnering with juniors such as Strike Energy (as Incitec is doing). I’m sure that’s not how the gas industry planned it to be.
Thankyou for such an insightful article.
Where is the gas to replace coal and/or to solve problem of intermittent renewables? Apparently conventional gas on the East Coast is peaking. The whole LNG export from Gladstone was a bad idea.
In 2011 my main concern was that gas – our only alternative transport fuel – is wasted.
Australia’s natural gas squandered in LNG exports
http://crudeoilpeak.info/australias-natural-gas-squandered-in-lng-exports
US shale oil has delayed the oil problem by a few years but now the IEA warned of oil shortages if oil prices do not go up to finance new projects
HOUSTON – Global oil supply could struggle to keep pace with demand after 2020, risking a sharp increase in prices, unless new projects are approved soon, according to the latest five-year oil market forecast from the International Energy Agency.
https://www.iea.org/newsroom/news/2017/march/global-oil-supply-to-lag-demand-after-2020-unless-new-investments-are-approved-so.html
So these Queensland LNG exports have created a double whammy
We must not add energy hungry projects and adopt very energy frugal lifestyles
The APLNG and QCLNG projects developed enough new upstream gas to meet the needs of their downstream LNG plants – it is GLNG that should never have been built because their upstream fields are not capable of meeting the demand of the downstream LNG plant. Therefore they are reliant on third party gas which has resulted in existing domestic gas being sold to GLNG and the mothballing of gas fired generators.
“Therefore they are reliant on third party gas which has resulted in existing domestic gas being sold to GLNG and the mothballing of gas fired generators.”
I would have thought more a Catch22 at work here. The gas fired generators (ie particularly Pelican Point) with subsidized and mandated unreliables impacting their bottom lines are not confident enough to negotiate long term contracts, so the gas suppliers and infrastructure developers natural look elsewhere for predictable returns. Again it is all against the backdrop of unreliables free-riding on thermal producers to insure their fickleness and now we have a State Premier demanding it of them and promising to build another thermal competitor with taxpayer funds to really ice their cake.
Although as a humble layperson and power consumer I should really leave professional analysis of the situation to my betters in these matters-
http://nla.gov.au/nla.obj-158793567/view
I find the article and discussion very intriguing. Particularly intriguing is the very plausible notion that there is a high level of “invisible” load shedding provided by inverter powered compressors. Something that occurs to me is that one of the benefits that’s claimed for large battery installations is rapid voltage support and this may be contrary to this load shedding effect. I wonder if the relatively slow response of synchronous generators turns out to be an advantage here… because it allows enough time for the inverters to sense the disturbance and switch off. Of course the inverters of the battery installation can be set to whatever is needed, and it seems it might be good if they are set to match a synchronous generator response, rather than aiming for something “better” that would prevent voltage collapse. There’s probably a middle ground here of enough voltage support so that it does not collapse too deeply causing generation and transmission to drop out, and yet enough collapse that inverter powered devices can can sense the drop and switch off.
I suppose the action of the inverters is designed to protect the device, not the grid, but it seems like there would be a good opportunity to use this phenomenon deliberately to protect the grid.
@chris
I have given thought to your suggestion that inverter-driven heat pump tripping is a useable affect. IF that was demonstrated to be the case, it could be an answer to SA’s present woes that are basically driven by heat pumps (OK, wind power generation collapse matters as well) in hot weather and in particular by large numbers being turned on at much the same time of day.
Because all the houses are hot by late afternoon in summer, starting a domestic heat pump makes it run continuously at high load until the house internal temperature drops to the minimum set point. It is a crazy way to run a heat pump. Pumps should be left on all the time on thermostat control with the set point altered a bit when the the house occupants are not there.
That is the way I run my heat pump. It is switched on in September, off in April and runs under thermostat control, drawing 270 to 350 watts when running; overnight the set point is dropped 2C. Done that way a small pump suffices to heat (in the UK) a whole house, except for when outside temperatures dip to freezing. A small pump is not only cheaper than a large one but also more efficient, smaller and less noisy and critically uses far less peak time electricity.
Your suggestion would be a form of fast-acting demand management with a guarantee that the pumps do not restart for a few minutes. Nobody should be that put out that their heat pump is stopped for a few minutes, provided it is not damaged by this being done repeatedly. I suppose the awkward squad might fit surge arresters?
http://digital-library.theiet.org/content/journals/10.1049/joe.2014.0180
In this study, results from testing six heat-pumps are presented to assess their performance at various voltages and hence their impact on voltage stability.
…
Note that all units continued to operate down to a voltage of 0.7 pu or below.
Heat-pump performance: voltage dip/sag, under-voltage and over-voltage
Table 1
under-voltage cut-off, pu
good (0.7)
low (0.52)
low (0.55)
moderate (0.6/0.68)
very low (0.24)
low (0.58/0.54)
Analysis of results obtained indicates that all units have a reasonably well-defined under-voltage cut-out (and cut-in) level. However, for many of the models this is set at a surprisingly low voltage.
All five units fell to low power operation after the transient and resumed their previous operating conditions after their normal start-up delay (typically 1–3 min).
http://www.abc.net.au/news/2017-03-15/aemo-report-torrens-island-fire-almost-second-statewide-blackout/8357316
‘The market operator believes the rapid change might have occurred as rooftop solar photovoltaic systems tripped out during the disturbance but then reconnected.
AEMO recorded a 400MW drop in energy demand because of the incident.’
Surely, a trip of solar PV generation would be seen by the grid as an INCREASE in demand? Try working the problem again and suggest it was the air-conditioners that tripped and restarted later under micro-processor control?
I have checked the underlying report and the problem is there. One of the recommendations is to check the solar PV performance under grid transients but no mention of the air-conditioners.
Fault at Torrens Island switchyard – AEMO
https://www.aemo.com.au/-/media/…/2017/Report-SA-on-3-March-2017.pdf
FAULT AT TORRENS ISLAND POWER STATION SWITCHYARD AND LOSS OF MULTIPLE GENERATING UNITS,. 3 MARCH 2017. Australian …
As PaulD has picked up, AEMO released its initial report into this event a few days ago.
It provides a lot more useful detail on the nature and sequence of the faults that led to the TIPS B and Pelican Point trips. It also confirms that there was no underfrequency load shedding, but that there was a sharp drop in demand immediately after the first faults / trips, and it was probably this that prevented another trip of the Heywood interconnector which would inevitably have led to another System Black in SA.
As it was the interconnector went uncomfortably near breaching the (revised) triggers for Loss of Synchronism protection – see p12: Based on the above, AEMO believes this event came very close to tripping the Heywood Interconnector.
Strangely there is no discussion in the report of the reasons for the 400 MW loss of underlying demand – offset initially by what AEMO believes was loss of 150 MW solar PV generation yielding a net 250 MW demand drop (see Figure 3 in the AEMO report). I’ve seen subsequent comments attributed to “an AEMO spokesman” relating the underlying demand drop to the “natural response of modern electronic equipment” to low voltages caused by the faults, and stating that “AEMO will develop better modelling of load and rooftop PV response to severe voltage disturbances, to fully understand the implications for power system security.
Just as well – if the numbers had been the other way around, 150 MW loss of load and 400 MW loss of PV generation – then the lights would definitely have gone out!
Love your analysis as always Paul. One comment about your conclusion in C1. The following is a simplistic analysis assuming that all electricity is traded on spot market, but hopefully it has enough truth that it demonstrates my point. If you look at periods in the last 12 months where demand in SA exceeded 2.6 GW (43 half hour periods), price increases very steeply with demand. It’s not a great correlation, but a linear fit indicates a $23/MWh increase in spot price for every MW increase in demand. A 100 MW battery could potentially reduce prices at this extreme end of the spectrum by over $2000 per MWh. That’s a saving of over $6m per hour, or $137 million over the 43 half hour periods. That represents 10% of the total value of the SA spot market in the last year, and goes a long way towards justifying the cost of the battery.
My understanding is that the 100MW gas turbine at Quarantine failed to start under load, and that is seen by AEMO as the straw that broke the camels back
The protection system for other power plants and the interconnector often uses an I^2t rule which depends on the thermal time constant of the systems. It might be set at a number like 100 so if the over-current is say 10% the system will stay on line for 100/(1.1)^2 = 80 seconds. If the load went to 160% of rated load it would shut down after 40 seconds.
100MW of batteries could have allowed three things,
a) a lower peak load on the interconnector therefore a longer time before trip
b)some controlled load shedding
c) some useful governor response on the gas plants that were running
b) Ramp time for Quarantine and reciprocating plants so that would then have been stable enough to take up the load.
I don’t know the exact numbers but 100-150MW of batteries (including grid controlled behind the meter systems) may have just halted the slide
I don’t where this has got us.
The situations for September 28 and March 3 still seem very similar, so why didn’t the system go black on March 3.
Fig 7 of the March 3 AEMO report shows a very different impedance trajectory between the two events and that is why Heywood didn’t separate but why the difference in the trajectory?
There was slightly more headroom on the Heywood Interconnector for March 3 but peak flows looked very similar.
It is clear that during the March 3 event there was considerable load shedding that seems to be just a natural function of the connected devices. But why didn’t the same shedding happen during the September event. Was there something in the way the wind farms went off that masked the voltage drop at users and so the same unintentional load shedding did not occur?
The other interesting thing appears to be the reaction of the solar PV systems. The solar systems are happy to drop off but leave the load they were supplying still on line. This seems to have generated an instantaneous hit of 150MW to the load which you would have expected to make the whole state system more susceptible to tripping. This doesn’t seem to be a good idea.
Michael
It’s frustrating that AEMO weren’t able to comment on the reasons why significant load tripped on Mar 3 but not on Sep 28. However the relevant charts in AEMO’s incident reports show that the voltage dips near Adelaide during the March incident were a lot larger than those on the black system day. The nature of the loads would have been different too (more aircon load in March).
On the solar PV question, it’s possible that (some) load associated with those PV systems may have actually tripped – AEMO’s report doesn’t say either way. If you are saying that solar PV systems ought to be set up so that the inverter tripping causes the entire customer supply at that point to disconnect then that seems a pretty blunt approach – nor likely to be practicable.
Allan
I would be very interested to understand what caused the difference in voltage responses – September 28 lost 456MW of wind generation and March 3 lost 600MW of thermal, but the system response was so different. What caused that? The March 3 trips were at transformer level while the September 28 trips were at generator level – does that give us a different response?
On the PV response – if I was a non PV household in Adelaide I would be very pissed off to find that my neighbour (who had installed PV to save money on his power bill) shut down his PV system to protect it and decided instead to pull their whole load from the “high priced” network and that additional load helped pulled the whole network down. I would have thought it would be relatively simple, and reasonable, to trip a householders whole supply in the event that their PV system thought it was unsafe to be connected to the network. I sort of expected that would be the normal response if the PV system was feeding into the network.
It is also a matter of equity. Should users who are trying to distance themselves from centralised generation and network distribution (and save money) be given equal access to that network in the event of an emergency to the extent of compromising that network?
There were two close calls on the Heywood interconnector more than six minutes apart: The initial fault sequence is given as :-
15:00 Heywood 484MW
15:03:46.262 Fault 1: TIPS B4 -130MW tripped
15:03:46.364 PP GT11 -150MW tripped PP ST starts runback
15:03:46.920 Fault 2: 275kV west busbar fault
15:03:47.300 Heywood Max load 962.9MW
15:03:47.812 Fault 3: TIPS B3 -130MW tripped
15:05 Heywood 636MW TIPS B2 starts runback
15:05:15 PP ST tripped -70MW over 89 seconds
15.07:34 TIPS B2 tripped -130MW over 02:34 minutes
15:10 Heywood 861MW, then 2nd peak at 918MW shortly afterwards
It was close to one second from incident start to the maximum load on the Heywood interconnector. There was a second, lower maximum on Heywood just after 15:10.
The first maximum (a 962.9-495=468MW rise) must therefore have been caused by :-
1: the grid transient,
2: the loss of TIPS B4 and PP GT11(a 280MW fall in generation in total) and
3: any loss of solar PV in the first second [1: suggests a delay to about two seconds], together with
4: any loss of demand in the first second.
It is possible/probable that 3: and 4: saved SA from another statewide blackout. Such a result would be pure serendipity. Fault 3 and the trip of TIPS B3 were too late to affect the first Heywood maximum load.
What is missing is an exact time for the 400MW loss of demand and any PV detachment; the report states ‘near simultaneous disconnection’. Was it before or after the Heywood maximum load at 15:03:47.300 or even spread either side of the maximum?
Fig4 has a horizontal scale of one hour and has not captured the first, sharp interconnector load peak of 962.9MW at 15:03:47.300.
It could be that much of the solar PV stayed attached for close to two seconds i.e. after the first Heywood interconnector peak load [1:].
Is this delay what saved the SA grid from another blackout?
How fast does the solar PV reconnect?
‘According to IEEE1547, DERs must wait 5 minutes before restarting automatically after disconnecting due to off-nominal voltage and frequency, provided that frequency and voltage have been restored within tolerance. … no provision for staggered restart’
1: Performance of Distributed Energy Resources During and … – NERC
http://www.nerc.com/pa/RAPA/ra/…/IVGTF17_PC_FinalDraft_December_clean.pdf
distribution system, and, in particular, with distributed wind and photovoltaic (PV) resources. ….. system are likely to trip on under voltage within ten cycles.
Includes references to IEEE Standard 1547 must-trip requirements e.g.
V<50% Max Clearing Time 0.16sec
50%<=V<88% MCT 2.0 sec
Fig 1: refers
Fig 8: shows A/Cs (air-conditioners) tripping on thermal protection and coming back.
Thanks very much PaulD, I hadn’t previously noticed that level of detail.
I can’t access the NERC paper you referenced at the moment but will keep trying.
In your dot point 3 are you suggesting that some of the PV systems could have disconnected within 0.2secs (10 cycles) and made up much of the difference between the rise of 468MW on the interconnector compared with the loss of only 280MW of generation. So 188MW of apparent extra load unaccounted for.
This is not shown on Fig 3 of the AEMO report but that could simply be because the timing of the SCADA could not pick up these rapid changes. This could mean that the 150MW rise we do see on Fig 3 is some of the remaining PV systems falling off. After all it appears there was about 420MW of solar on line at the time and so far we only have 188MW plus 150MW of “phantom” load.
We might get some idea of how fast the PV systems reconnected by also looking at Fig 3. The 150MW rise in load that AEMO has attributed to the PV systems, gradually disappears over the following 2 secs and the load keeps falling, but more slowly for a further 2 secs. So this could be PV systems coming back on rather than additional load simply falling off the system.
But all of this just makes the question of what was different between the March 3 event and the September 28 event all the more perplexing.
May I suggest an addition to the problem list?
Problem #7
Black Start
Solution: A big battery and an OCGT.
===================
If we think back to the September blackout, there was an issue with the black restart arrangements; they failed to work and SA’s grid had to be restarted with a long ‘jump lead’ from Victoria. In the absence of the interconnectors, SA needs a new battery to cold start its first OCGT to re-energise its grid. That allows wind and solar to reconnect. It makes sense to me.
It makes for an interesting operational issue. Do you run down the battery to try to prevent an interconnector trip or do you save the charge for the black restart? If you are leaving the NEM, you definitely need a big starter battery with a full charge at all times.
The battery’s primary job is to cold start the OCGT but it may also be good for supplying inertia/voltage support into the system during grid transients for very short periods of time. I would not expect the battery to be regularly deep cycled, so a long life is in prospect. The OCGT is available to run when prices are high enough for it to make a profit; that might be the end of $14,000MWh prices spikes. Run at minimum load, the OCGT would provide some system inertia, reactive power and spinning reserve, although spinning reserve is useless in SA until after the instant when the Heywood interconnector opens.
If I was rebuilding a grid system for full independence, this is where I would start.
The author’s unambiguous ‘no’ to the question whether wind and solar will make the system unstable is not correct. Sure, power systems are designed to deal with variability, but traditionally that variability has arisen from the demand side, as well as contingency events. Intermittent wind and solar add an additional source of variability. The evidence from overseas market is very clear that intermittency requires additional high frequency reserves to be held.
Thanks for the comment Sabine,
Can you clarify the comment “requires additional high frequency reserves”, either here or via email.
Thanks
A substantial amount of spinning reserve (inertia) will be removed from the national electricity market with the closure of Hazelwood.
It would make for an interesting comparison of say frequency drop rate of change before and after Hazelwood closure upon an event such as a 660MW’s separating from the grid followed in close succession of another 660MW’s
I await the next Bayswater type separation event for comparison.
MrCyberdude
Would be great if a few more frequency drops with greater variance were shown in a graph to get a real feel for the differences between FCAS response from different thermal generators and renewables, especially considering the ever increasing renewable sector.
MrCyberdude
That was one of the harder articles to get through here at watt clarity, for me at least.
While the author concludes that a disproportionate amount of the FCAS costs have been allocated to the wind generators, I would also say that frequency control (FCAS) was not an issue in the past decades before wind generators came alone and made frequency such an issue.
If wind generators brought the problem then they should offset it either as part of their regulatory compliance which would probably mean gas cogeneration at their sites or they should simply pay the lions share for what they have created.
I had never seen so many frequency swings in both directions before wind generators were connected to the grid in quantum.
MrCyberdude
Wind is hardly causing the high requirements and prices for FCAS regulation services. Please see “Fast Frequency Service – Treating the symptom not the cause?” for a technical discussion on how synchronous generation, excessive local requirements and in my article poor AEMO forecasting is much more important than the use of wind as a generation technology. It is simplistic and disingenuous to put the blame on wind generation.
I would add that FCAS became a problem when the synchonous generators relaxed their frequency regulation a few years ago because AEMO penalised one for curtailing when that curtailment was made to assist frequency. So the AEMO decision directly signaled to the market to not prioritise regulating frequency. FCAS goes up. Also aging coal fired generators having problems has increased FCAS. On the other hand Telstra* big battery has reduced FCAS costs.
* Editor’s note – Telsa, not Telstra
Looks like Pelican Point is back in the market at full capacity, as previously modelled by AEMO as a result of Hazelwood closure. Currently outputting approx 420MW. Haven’t seen any official announcement yet.
thanks for the article and also to the site for assisting a non EE person to gain a little understanding.
Dr Mackenzie, does your comment lead to a requirement to revert to the rules on frequency control that applied prior to 1999 as described in the ” symptoms not cause” article as a first step?
Not necessarily. I do think that there needs to be a rethink about causer pays factors generally, and especially for the intermittent generators but there is some justification for the concept. It is the formulation, where the existing dispatched unit causer pays methodology was simply used for wind and solar without really thinking through all of the implications, that is the issue.
OK thankyou for the response. As a person who worked in continuous manufacturing dependent on steady electricity supply for 36 years I do not understand why govt. would allow mechanisms that reduce stability or reliability either from technical or market aspects.
No need to respond to the comment above, just an observation.
Yours and other articles show that there is much that we as the broader public do not know about how our system operates so please keep up the good work.
Thanks for the analysis Paul. This transition really does feel chaotic and would be nice if some level of government could decide what they were doing and plan into the future just to give us some indication of future employment, if nothing else!
Paul
A couple of items for (D) and (E) above:
The AER refers to its ongoing review of regulatory compliance related to multiple aspects of the Black System event in its Q4 2016 quarterly compliance report: http://www.aer.gov.au/wholesale-markets/compliance-reporting/quarterly-compliance-report-october-december-2016
The AEMC has outlined the process for its review – see http://aemc.gov.au/Markets-Reviews-Advice/Review-of-the-System-Black-Event-in-South-Australi – however they apparently won’t start until the AER finishes (whenever that may be), and will then take a further 6 months.
Allan
Thanks Allan – updated above!
We are already seeing some announcements that would partially offset the loss of Hazelwood, such as the additional unit at Pelican Point and large solar in SA (Bungala 200 MW, Riverland 330 MW, Kingfisher 120 MW, Tailem Bend ~100 MW) and northern Victoria (Mildura area 320 MW, Gannawarra 60 MW).
South Australia has tended to be an energy importer from Victoria, but would likely turn to being a net exporter on account of its better solar resource, and Pelican Point being the most efficient gas-fired power station in SA/Vic. It has among Australia’s best solar resource in locations close to areas of grid strength. This would help balance out wind, because days of low wind usually have good sun.
Should also note that while Hazelwood closed, there were 3 large coal units at Loy Yang A and Yallourn. When these are returned to service there would be an additional 1220 MW of coal capacity.
“days of low wind usually have good sun” would appear to be a remarkably liberal use of the word ‘usually’. http://www.sciencedirect.com/science/article/pii/S0960148115303591 https://bravenewclimate.com/2015/11/08/the-capacity-factor-of-wind/
The use of the word transition to me infers an ordered change of state. Instead we have a bunch of fairly random decisions taken on the basis of ideology, politics and virtue signalling.
It seems absent any thoughtful and informed work on a forward plan and the right mix and sequencing of power sources. Right up to this day, I have seem nothing that looks like anyone has any kind of plan beyond what they need to do for the next State or Federal election cycle.
I am in VIC and will be installing what I need to protect myself over the winter period. Probably a reasonably sized generator and transfer switching.
Its great to have such concise analysis; I’m surprised there hasn’t been more discussion since publication. Kate refers to a paper published on 03/02/2017; is it possible to have a link to that? It sounds extremely interesting.
Thanks,
Shane
Hi Shane
Kate’s linked to them below, and I’ve added direct links to them all above.
Paul
The 3 Feb Frequency Control paper is available on the ESCOSA website under the licence review for inverter connected generation and Pacific Hydro has made the Finkel submission available with the paper included via the following link.
http://www.pacifichydro.com.au/english/information-centre/industry-and-policy/pacific-hydro-exposes-a-major-threat-to-power-system-security/
Ever since I’ve become addicted to watching AEMO’s NEM Dispatch overview, I’ve come to a few conclusions rightly or wrongly;
1) SA spot price appears to be consistently higher than a number of States and almost always higher than Qld to a significant extent
2) the contribution towards meeting the power demand in SA by wind turbines is surprisingly poor given all the political hype. It appears highly variable and unpredictable (unsurprisingly) and at times reaching extraordinarily low levels.
3) associated with 2) above, there seems to be a high reliance on the interconnectors with Victoria to supply “coal fueled”) power into SA thus challenging the notion that SA thrives on “green” power.
4) it appears that SA would be in all sorts of trouble if it were not for the ready availability of power from Victoria
5) it appears that due to the relatively low amount of synchronous generation along with various other generation issues, the Heywood interconnector is subject to unhelpful constraints at some crucial times
6) power flows to and from Tasmania together with the constraints applied to Basslink don’t seem to make a lot of sense sometimes when considering the value of premium energy stored in Tasmania especially when Tasmania has been experiencing unusually low rainfalls in the past year or so. It seems that the temptation to make some opportunistic profits because of the price differentials between Tasmania and the mainland might be just too much.
Are my observations reasonable or am I missing some vital clues ?
“Why don’t we ensure (via appropriate rules and technical standards) that all new generation technologies contribute to frequency control and adjust the market to remove penalties that currently exist for generators when they act to provide good frequency control?”
You mean like also ruling that no generator can tender anymore power to the grid than they can reasonably (ie short of mechanical breakdown) guarantee 24/7, 365 days a year for a level playing field market? Now you’re talking but who’s going to point out to our current overlords that it’s a fundamental tenet of engineering that you can’t build a reliable system from unreliable componentry? Unfortunately we just have to wait for the train wreck for them to finally comprehend a national power grid doesn’t run on emotion.
I am hopeful that a well designed system will emerge from the current unsatisfactory situation, and that a reliable system will be built from reliable componentry – but only if people pay more attention to the physics and a bit less attention to the fashionable economic theories of the day.
Well in defence of market economics I’d say it can provide some useful tools for the greatest good for the greatest number but that will depend on the underlying assumptions.
For example it may be the science of Catastrophic Anthropogenic Global Warming alerts the economic practitioner to some social cost that is not being reflected in the much lower private cost of fossil fuel use and there’s a simple remedy. Just call the guru who knows the additional price that needs to be added in order to bring the level of CO2 in the atmosphere back to the level said guru knows it should be. Having imposed that on all players in say the business of producing reliable, 240V, 50Hz electricity as required by consumers is then another set of issues which we would probably agree needs more physics rather than political interference. The economist would say set those basic level playing field parameters and the various players will then come up with the economic solutions for us.
But back to the CO2 assumptions and that guru’s estimate of social cost. With only around 150 yrs of reliable thermometer records our guru has had to rely on such temperature proxies as tree rings, ice cores and varves along with a host of computer models to form an estimate of CAGW and there is some debate about the scientific veracity of all that. However there is one measure of global temperature that clearly rules them all and that’s sea level rise since sea level rises with temperature for only two reasons. Thermal expansion of oceans or runoff from ice melt on land and the polar caps.
Well you wouldn’t want to rely on a tide gauge in Venice or the side of a volcano in Hawaii but here in our own stable ancient weathered continent we have a couple of very reliable sea level gauges. Take the one at Port Arthur(one of the longest man made in the world but certainly in the SH) and it shows us 13.5cm sea level rise from 1841 to 2000 or an average sea level rise of 0.85mm/year. But there is another at Hallett Cove Conservation Park in SA (so proclaimed by none other than our Premier Jay Weatherll as Environment Minister in 2010 because of its world geological significance) and that shows there was an ice age 20000 years ago but around 15000 years ago the ice began to melt and the sea level off SA rose 130 metres until 6-7000 years ago, which could be an average increase of 16.25mm/yr for 8000 years.
So what conclusion did I logically draw from that? There are either some very very clever scientists and their computer models that can extract the anthropogenic CO2 signal from the current Port Arthur temperature proxy, or else aboriginal cooking fires and traditional burnoffs to flush out game have an awful lot to answer for with that Hallett Cove temperature proxy. Now where were we? Oh yes, economics and the underlying assumptions.
I agree with you that market economics can provide useful tools which can be used to benefit society as a whole. I will leave it to others to determine for themselves whether they believe it has actually been applied to this end.
My main point is that some things can easily be traded – e.g. commodities like food, minerals, fuel and (specific to the point of this article) power. For these things designing markets is easy.
However, there are more esoteric, less easily defined benefits that derive from concepts that cannot be easily turned into commodities. In the power system world this consists of all types of ancillary services. These things are not easily expressed in a market paradigm.
Markets that are set up to deliver these types of services easily lead to unforseen consequences which we have seen playing out in the frequency ancillary services, so that now we have worse frequency regulation and a less secure system than we did in 1990. This contributed (albeit only peripherally) to the first state wide black out we have had in approximately 50 years.
With respect to your comments on climate change, I will just treat that as commentary. I am happy to reply to any questions related to Power systems which is my field of expertise and what the article is about. Climate change is not my field of expertise and therefore I suggest you refer such comments to a climate scientist.
Climate change is not my particular field either, but facts are when I go for a stroll through a Conservation Park and my interest is aroused as to just what are the broader facts of the matter. As a result I naturally continue to believe in climate change, just not CAGW anymore after previously reserving judgement on some worrying critiques of computer modelling of lesser proxy variables-
https://www.ncdc.noaa.gov/news/what-are-proxy-data
Science and the scientific method don’t fit in neat little compartments or become the preserve of anyone. The facts speak for themselves until they’re contradicted or otherwise interpreted.
Nevertheless particular fields of endeavour can require certain expert technical expertise to fully comprehend the bigger picture for the policymaker so with that in mind can I ask your thoughts on this interpretation of where we’re headed on the current path with the grid-
https://www.youtube.com/watch?v=gJtv7gkuh1s
There is a legitimate role for good Govt overseeing areas of market failure with externalities (the tragedy of the common and the like) but there is a sense with our national grid that no-one is in charge anymore with a history of incrementalism and the science of muddling through. Whatever level playing field social cost we put on CO2 and bearing equally on all inputs it should (concrete or steel for thermal or wind generators), that still leaves facilitating reliable voltage and frequency at the flick of a switch for consumers and policymakers need to understand and address the obvious shortcomings now. In that respect you’re talking of frequency, I’m talking of reliability of supply while others are beginning to talk about the voltage- https://www.getinvolved.qld.gov.au/gi/consultation/3703/view.html
There must be strong incentive to produce all three at the same time with no beggar thy neighbour outcomes on any score.
References-
Point 6-
http://sa.gsa.org.au/Brochures/HallettCoveBrochure.pdf
and the CSIRO-
http://www.cmar.csiro.au/sealevel/sl_hist_few_hundred.html
Re point 2, the contribution of wind is higher in winter than summer, and higher for the newer wind farms such as Hornsdale that has a higher hub height. Hornsdale has averaged a capacity factor of 41% since the start of January, and most new-build wind will be like this. Once there is more large-scale solar in SA, this will partially balance wind output, being higher in summer and under high pressure cells, and lower as lows pass over with their windy conditions.
Re point 3, now that Hazelwood is closed, Engie’s Pelican Point has been operating at higher capacity factors, leading to much more frequent exports from SA to Vic. This is the most efficient gas power station in SA/Vic. Until about February, most of the load balancing in Vic was from the Snowy. Now it is Mortlake (gas) that seems to be doing some of the load balancing, and some from Newport (gas). The coal generators generally run at close to 100% or not at all (when down for maintenance).
Re point 5, the Heywood inter-connector seems to have much more than its fair share of problems. For example, last Sunday (9 April), export from SA was limited to ~290 MW from 11:35 until 15:00 on Sunday 9 April during a time of high wind output. This caused SA prices to crash to minus $45/MWh over most of this period. At other times of the day, export from SA was never more than 500 MW, and SA prices held up well. Supposedly the inter-connector has an export capacity of 650 MW via Heywood and 220 MW via Murraylink, totalling 870 MW, under a project that finished in March 2017.
How much of this wind correlation penalty arose from SA becoming a net importer to a net exporter of electricity ? The SA market appears to go through 4 stages prices depending on the electricity deficit or excess.
1. When demand exceeds the inter-connector import limit, price is determined by the SA bid stack.
2. When demand is below the inter-connector import limit, and SA is a net importer, the price is ~$10/MWh ABOVE the Vic price. The $10/MWh is probably a margin made by SA Power Networks.
3. When excess is below the inter-connector import limit, and SA is a net exporter, the price is ~$10/MWh BELOW the Vic price.
4. When net supply exceeds the inter-connector export limit, price is determined by the SA bid stack, and is generally negative.
So at most times, the Vic price provides at floor to the SA price, but there is a $20/MWh price difference between net import and net export.
Since the post was prepared, Hazelwood has closed and Vic prices are determined by the marginal cost of gas-fired electricity, first at through Mortlake then at Newport, both of which are more flexible than was Hazelwood. Furthermore, the inter-connector capacity has been increased. Previously during times of high wind production, Hazelwood would keep up full production, leading to a depression in prices from the combined SA and Vic wind production. On Sunday 9 April was the first instance of high wind and low demand in the post-Hazelwood market. Loy Yang A reduced its production and provided stability services, while Loy Yang B and Yallourn W continued unabated, but prices subsided only a little. So these market changes are likely to reduce the problems of the wind correlation, at least until more wind supply is added. Now we need more balancing solar to supply the market under anticyclonic conditions (high pressure cells).
Thanks for writing this up Paul – an interesting situation.
There are a number of things going on here that are worth pointing out.
The transmission line which went out of service, causing the sequence of events, runs between Mortlake Power Station in western Victoria and the main part of the Vic 500kV grid. It is one of two lines that supply the Portland smelter and also the interconnection with South Australia. So with this line out, a trip of the second 500kV line on this corridor (the Moorabool – Tarrone line) would isolate South Australia + generation and load in the south west of Victoria (principally Mortlake PS and the smelter) from the rest of the NEM.
In this situation AEMO invokes a couple of key constraints (amongst others):
1) FCAS constraints requiring 35 MW of FCAS regulation services (“regulation raise” and “regulation lower”) to be sourced from South Australian generation – so that if the second line trips, there will be some generators already providing regulation services in SA to help manage frequency in the “islanded” South Australian system (actually SA + SW Vic in this case). While the DC Murraylink interconnector would still connect the two regions, it cannot currently transfer regulation services due to its design – presumably this might be another technical option for provision of FCAS in SA along with the possibilities you mention of wind or solar generation being modified to provide some of these services. Whether feasible or economic I don’t know.
As you point out the very limited FCAS bidstack for these services in SA plus the requirement to source this locally under the above constraint led to pricing of these raise and lower regulation services shooting upwards. There is clearly insufficient competition for provision of these services within SA to drive prices down to reasonable levels when they are required locally under conditions like those that existed last Tuesday. (Ironically, technically Mortlake PS could provide regulation FCAS within the SA + SW Vic island, and had relatively cheap bids in for these services, but the way the NEM’s regional boundaries are defined, it is obviously not considered a South Australian generator for the purposes of procuring “local” FCAS when this constraint operates! – I wonder if AEMO could make the relevant constraint a little bit smarter so that under the appropriate precontingent conditions Mortlake could be included in the generators from which “local” FCAS is procured – but possibly not, due to the way that regional pricing and cost recovery for FCAS operate.)
2) The risk of islanding also leads to a constraint being imposed on net exports from the SA + SW Vic region, so that if the second line trips there won’t suddenly be such an excess of generation in the island to cause a frequency rise too large to be handled. The limit is 50 MW but because the island would include Mortlake generation and the Portland smelter, the constraint doesn’t simply go onto the SA-Vic (Heywood) interconnector, but includes Mortlake generation and Portland load as well.
At the time of this constraint being invoked, Mortlake was running at 551MW, very close to full capacity and well above Portland smelter load (which appears to have been about 220 MW), so the constraint forced NEMDE to dispatch additional flows westwards on the Heywood interconnector, at the same time ramping down Mortlake generation as fast as possible (nominally 26MW per minute across the two units or 130 MW per 5 minute dispatch interval), in an attempt to keep the net flow to the rest of Victoria from the SW Vic region to under 50 MW. This had the effect of forcing generation levels and dispatch price downwards in SA due to the increased Heywood imports, and briefly upwards in the rest of the NEM (probably due to the rapid ramping down of Mortlake). Hence the “counterprice” inter-regional flow mentioned in your writeup.
Interestingly, within a few dispatch intervals and as Mortlake continued to ramp down to about 300 MW, SA reached the position of being a net exporter to Vic (with Murraylink flows eastwards exceeding westward Heywood flows) so prices in SA being lower than those in Vic were in the expected relationship.
The direction would have been to ensure two large synchronous generators were online. As Pelican Point was offline, and only one TIPS ‘B’ unit bid in to the market, I believe the direction would have been for AGL to start a second Torrens Island “B” unit. The minimum requirements for online synchronous generation have been published by AEMO previously but I couldn’t find the link.
Ian
The initial direction very early on Tuesday morning was for Hallett GTs (8 units) to start. These ran until early in the afternoon when a second TIPS B unit came online, again directed by AEMO.
The paper on minimum synchronous generation requirements in SA is at this link.
Allan
Ian – Allan O’Neil has posted more here today:
https://wattclarity.com.au//2017/05/more-about-wind-output-and-aemos-intervention-in-south-australia-25-26-april/
Was any wind power curtailed or was the surplus all exported to Victoria?
Ben – Allan O’Neil has posted more here today:
https://wattclarity.com.au//2017/05/more-about-wind-output-and-aemos-intervention-in-south-australia-25-26-april/
What impact has resonance on grid stability? Had a look at this guys video where he states that integrating solar and wind into a grid has serious issues. Thought you electrical engineers could explain it to us laymen?
https://www.youtube.com/watch?v=gJtv7gkuh1s
ardy – Allan O’Neil has posted more here today:
https://wattclarity.com.au//2017/05/more-about-wind-output-and-aemos-intervention-in-south-australia-25-26-april/
Wind power in SA tends to peak strongly at night, probably due to “gully winds” in the hills near Adelaide, where most of the wind farms are located, but it will still drop dramatically during heatwave afternoons, no matter how many wind farms are added. Wind power is SA has killed the concept of baseload, without providing much in the way of firm heatwave capacity, maybe there are some benefits that compensate for those terrible characteristics.
An excellent article Allan, makes it a lot clearer to the casual market observer like myself. I’d be interested to understand the impact of the intervention pricing on generators, I assume if AEMO constrains them they would be compensated?
Glad you found the article helpful, Ian.
Intervention pricing attempts to calculate what the market price would have been in the absence of AEMO’s direction(s), and is the basis for the prices actually settled financially with spot market participants.
Participants who are directed by AEMO are entitled to be compensated under the market rules, which also contain provisions for adjustments to settlement amounts for other participants whose dispatch was (indirectly) affected as a result of a direction. The details of how these amounts are calculated are complex, to say the least.
Allan
Why is this not in the news? I know it is a budget week, and I am watching from QLD, but does the market assume that parties will come to the table prior to the 11th, because as a lay outsider, reading the AEMO notices, there could be some pain in store on the 11th onwards.
I can understand that other generators may be hesitant to contract for gas, in order to bid if at the 11th hour AGL could turn around and cancel the outage (I am guessing they can do this via the market rules).
Interesting times indeed.
Two units suddenly shut down at 2:30pm and 3pm today, with what looks like the final one off from 8:30pm. But perhaps coming back for tomorrow evening’s peak (based on predispatch Vic availability). Strange.
If readers are wondering why it’s such an industrial rocky road.
FairWork is broken when it comes to industrial relations and essential service workers, and it is this reason that it is likely to get worse as time goes on. Unless there is a mechanism to deal with scenarios where a company fails to negotiate in good faith then terminates agreements then threatens to lock out workers, therefore effecting 3rd parties at their industrial whim, it will likely escalate on both sides.
MrCyberdude
You should do a follow up for the effect on the NEM of lack of wind in NEM in recent weeks – zero wind generation in the NEM as I write this (with some sites showing negative?). I am not anti-wind but it has seen some radical pricing with no wind in recent weeks post Hazelwood closure.
Simon Benson reports in The Australian, June 3 on the Finkel Report –
“Significantly, rule changes to the National Electricity Market will include the requirement that all new large-scale renewable projects have storage capacity such as batteries to address ongoing reliability issues.”
That’s a start to unwinding this mess with unreliables bludging on thermal and engaging in dumping but what about all the existing ones creating the train wreck at present?
http://anero.id/energy/wind-energy/2017/may
And what politician is game to tell all the mums and dads with rooftop solar that their surplus electrons are not wanted in the communal grid unless they can reasonably guarantee them 24/7 all year round? To do that is to admit they got it badly wrong with half of the Green power equation by not factoring in storage costs to maintain equivalence with dispatchable thermal generation.
There are a couple of models for competition in duopoly / oligopoly markets – two different versions include Bertrand comptetion – where you competing on price, not output, and another is Cournot competion, where you basically compete on quantity, rather than price.
Two very different modes of competition, and behaviours – but both theoretically profit maximising (though in different ways). Maybe it would be too wonking or obscure for the Energy Minister to say “We are going to direct them to behaviour in a manner consistent with Cournot competition”.
…and if they did that, change the mode of competition, CS energy would also be forced to respond (assuming they are also economically rational / act strategically, and try to maximize profit given different behaviour from Stanwell).
Complete speculation – and theoretical. But there is – at least arguably – some logic to it.
Regarding point 2b, it may be more accurate to say “the high prices paid *by retail consumers* for electricity are something that should be celebrated”.
A large proportion of industrial consumers are still insulated from the extra costs associated with supporting the Energiewende, and, apparently, levied with the intent of dissuading electricity use.
I doubt the energy transition can be achieved with a significant presence of wind power. Seems to have been amply demonstrated multiple times , with Germany living through similar doldrums earlier this year, but happily being able to lean heavily on neighbours.
You point 2b is well made. These weather patterns are not new, so plunging into high % wind with 100% backup of whatever type is at the very least irresponsible.
Since solar lulls are never going to go away, and wind power lulls must now be counted as credible contingencies the word “transition” is surely inappropriate. If people insist that the country has to be carpeted in solar panels and wind farms then their output will only ever be the alternative power system, there will have to be another power system that can provide almost full demand when required. Thus, the only “transition” is one from sanity to insanity.
Thanks Paul for a very timely and appropriate review.
As a resident in Victoria with clients in SA and Tasmania, we have all noted that the weather patterns in May and June this year particularly have been dry and regularly still (or low wind).
The BOM has noted this in a number of weather reports and points to a a changed weather pattern where there has been effectively a ridge of high pressure over the southern states where most of the wind farms are located. This high pressure ridge has meant that there has been lots of still and low wind days.
But what has caused this to occur is beyond this simple engineer but I suggest that a question to the BOM might lead to a much more detailed answer
Regards David
BoM reported a couple of weeks ago that the low pressure systems which normally traverse southern Australia at this time of year are (for some reason) further south, closer to the Antarctic, than usual. Hence less wind than usual.
Look up the Southern Annular Mode (SAM) on the BoM website. It has been in a positive phase, which has meant fronts well to the south of the continent, and weak winds within the latitude of most of the wind farms. The SAM shows persistence but can’t yet be predicted. The wind industry isn’t the only one suffering – grain farmers on Eyre Peninsular have had less than 10% of their normal June rainfall. A more strongly positive SAM may be an outcome of climate change. For the power industry, offsetting investments include solar, which would tend to be higher during high pressure periods, and wind farm sites such as Hornsdale that have good exposure to easterlies. Power demand has I expect also been down because of mild weather.
In agricultural research it has been common practice to test new management practices on simulations based on 30 to 100 years of weather data, with the anticipated effects of climate change. Perhaps this should become standard in the power industry. Studies such as that of Andrew Blakers only seem to go back to the start of the NEM, rather than imposing NEM power demand on a longer period of weather data combined with known climate change effects.
Hey Malcolm,
Good point using A longer climate record is always a good idea for planning, however I suspect agricultural research is much more concerned with rainfall which in a farming nation like Australia was obviously a huge focus of our observational stratergy. Perhaps wind wasn’t recorded with such diligence?
I believe SAM is also related to ozone good work out of Monash university on this.
Cheers,
Peter Degorski
Dear Paul
I would certainly be interested in knowing what impact of the recent change in weather patterns has had on the off shore wind data for the same period.
It would also be great to have the MWh total per month of wind which has been rejected due to oversupply in the winter months.
Based on this recent event, numerically, I can see nobody better qualified than wattclarity to estimate the TWh of storage necessary for the East Coast of Australia.
Hi Jonathon,
There are no off-shore wind farms in Australia – and only a couple mooted at this point. We don’t currently have access to off-shore wind speed data either. Hence sorry we can’t help on this.
Thanks for the kind words, by the way. In terms of battery storage, you might find our other site of some value:
http://www.batterystorage.info/
Paul
Dear Paul
It goes without saying that there are no off shore wind farms in Australia.
I am surprised to hear that wattclarity do not have access to off shore wind data.
The storage site referenced is not practical.
Kind Regards
Jonathon Sarah
When we say global warming causes climate change, what that actually mean is climate belts shift. Regions between 30° – 45° will see climate shift by 100-150km per degree of warming. If climate belt has associated wind elements, in the southern hemisphere, these will be moving southwards.
In other words the roaring 40s become the roaring 42s.
Climate belt shifts generally aren’t linear, but rather occur as a step change – a great example of step changes in climate is found in the drying of sw WA and loss of stream flow – https://twitter.com/ProfRayWills/status/877335681038655488
Step changes are of course also points on a bell curve – perhaps this is the first iteration of moving wind resource, and we should expect to see more. In the longer term as climate belts shift, some wind farms may be at risk of being left behind…
Great post. Thanks Paul.
Of all the forecast flops, the biggest howler has to be the Frontier Economics wholesale forecast that made its way into the AEMC’s retail price trends report last year.
It all makes you wonder whether the structure of the NEM is not appropriate. While it seemed to work reasonably well when demand and consumption was increasing with little impact from intermittent generation, the outcomes of Paul’s assessment is that we are seeing a seismic shift that warrants a look at whether the NEM as structured is fit for purpose.
While agreeing with Paul that there is no one cause, what he has left out is whether the degree of competition (or lack of it) that now exists is contributing to the problem. After all,the NEM is predicated on vibrant and vigorous competition.
Another option might be whether the government owned vertically integrated model that existed before 1995 might have resulted in lower costs than we see now
Hi Paul McArdle |
Absolutely excellent report and summary.
Your Someone who understands (or at least may 80% of a complex electrical system). Way more than most .
I’ve been following the AMEO dash Board for some time and predicted that soon as they started shutting down coal stations the price was going move up and much more that the experts said.
The people making these decision seem to have little idea of how this system works and how no or delayed decision are going to effect very body. There seems there was no transition plan mapped out or thought given to spinning reserve requirements. would have though NEO could have just lifted the price for auxiliary service some year or so before . The Governments had chances to leave the major coal station/s (now shut down)on operation stand by until the newables/ storage were bedded in. I note the reason the history SA northern station was build was to give SA security and control of their state Elect Costs from Vic coal supply (strikes), Sounds like they don’t read the history , what’s that old saying , “those who don’t learn will repeat the mistake.” SA even took generation out of the hand of private companies make it state owned,
Maybe run a check how many companies
own all the generators? Can analysis gaming the market?
keep it up
Thanks Simo
well said “truth”. But who is listening ?
Paul
Thanks for getting this – most informative
How is it NOT the RET that’s causing the lack of dispatchable SUPPLY and the higher prices? It’s the RET’s subsidies to the most inefficient power source that are killing off the base load source and causing shortage of SUPPLY…IMO
The gentailers see the huge RET subsidies and favoured status for RE inexorably making coal and gas unprofitable…. and seeing the writing on the wall , they hedge by investing in intermittent sources and putting FF… their coal plants especially…on death row…announcing that they’ll be shut down soon because they can’t compete with the subsidised undispatchable wind and solar..
The privatised gentailers cannot be forced by government to keep unprofitable FF generators operating because they have shareholders and are bound by company laws.
Therefore FF generation shuts down …or is marked for shutdown and SUPPLY dwindles.
Hazelwood closed because Engie saw that writing on the wall…and its closure diminished SUPPLY.
Greens , weak governments and government disregard for the concerns of landowners closed down CSG in NSW and Victoria…diminishing SUPPLY of gas…but had there been no RET ….no VRE with their huge subsidies…and without climate hysteria running Australia’s energy policy there would have been no need to resort to CSG IMO.
Climate change hysteria and the divestment campaigns…and claims that ALL FF will be phased out in favour of RET-susbsidised VRE…. caused a drop in new gas searches and subsequent shortfalls in conventional gas SUPPLY.
The wind drought caused lack of wind electricity SUPPLY.
Gentailers are in the box seat..having covered themselves for all eventualities…and can price the scarce SUPPLY as they wish.
AGL and ORIGIN appear to be running the show now-AGL favouring herding solarPV owners into virtual power stations…to MANAGE their demand…a venture they and their Californian partners SUNVERGE in whose operation AGL is the largest investor at $20million…expect to be very lucrative.
Both AGL’s Vesey and AEMO’s Audrey Zibelman are American RE zealots whose energy plans for Australia are exactly in sync with Labor’s not LNP’s.What sort of government except one seeking to deceive its constituents lets that happen?
The Americans apparently see Australia as an opportunity for their RE experimentation…having not a care in the world that unlike America, Australia has no other country ‘to run an extension lead to’.
Zibelman told her alarmed New York State property-owners in her former job….that they could always run a lead from NYState to Canada if NY became energy-insecure due to her policy there.
They seem to have no knowledge and don’t give a damn that Australia is COMPLETELY ALONE in being the only industrialized country in the world that’s without alternative energy…no massive hydro or nuclear power…and yet is killing off the only energy…coal …that can keep us industrialised and competitive.
There is NO OTHER first world country that’s without hydro and nuclear…. that can do without COAL…NONE!
The shutdown of Australia has already begun as Malcolm Turnbull fiddles.
The LCOE for 100% renewable supply at current component cost and absolute best finance charge is $700/MWh. The prospect of that getting lower is not in sight.
As more renewables are forced into the NEM the cost of wholesale electricity will approach the $700/MWh.
The NEM is a network that takes advantage of economy of scale. It suits coal fired generation. It is inappropriate for a ubiquitous energy source that has no benefit of scale.
The concept of transition to renewable generation supplying the grid is fundamentally flawed. Renewables should be all local supply sources feeding their dedicated loads; never attached to the grid.
Thanks Paul,
Always good to have some insight into what’s driving our weather now that our energy markets are so interrelated to weather.
What about Son of Sam or even Uncle Sam?
But the bigger picture is gas prices have risen due to Green enforced demand policies globally and lack of coal supply locally.
As for correlating price rises with renewables installed the previous point is also relevant in that we have a national grid so a spike in renewables in SA simply lifts all boats with backup expensive gas generation, while SA grabs the localised benefit of the cheap marginal wind production. Basically beggar they neighbour policy which we can’t tell from the simplistic graph. Needs work.
Hi Paul,
Nice work.
2017 pricing to-date are truly scary and this doesn’t bode well for those of us who want an orderly transition to a renewable grid (or even for industry or the grid itself to survive).
I agree that there appears to be a systemic shift in the market, which is not obviously explained with off the cuff simplistic explanations tossed around (without any data). As Simon above mentioned a non-functional market must at least be one of the possible explanations.
I think all the bidding data is public but maybe not easily accessible. Has anyone done a comparison of the bidding stacks for say 2017 vs 2015? OR is there a public forum where I could contribute analysis?
Discussions on topics like – how has the average price tripled in the middle of the night, when there is ample excess coal base load capacity – would be very interesting. Further, it would at least be worth illuminating the bidding behavior of certain market participants.
thanks Adrian
There is an interesting correlation between gas prices and electricity prices in SA available at http://www.theage.com.au/federal-politics/political-news/hugh-saddler-its-gas-not-renewables-that-is-pushing-up-electricity-prices-20170712-gx9vvm.html which provides an answer. While it is expected that gas sets electricity prices in SA (‘cos there is only gas fired generation there), gas is the source of marginal generation in most other regions and so sets the price
Thanks David
Just remember that “correlation is not necessarily causation”.
Peter Martin’s later piece is here:
http://www.theage.com.au/comment/it-not-the-wind-its-the-gas-why-power-prices-are-going-berserk-20170712-gx9lxe.html
In this he comments:
“Wholesale prices and those charged to businesses that buy directly are up an extraordinary 60 to 70 per cent. It’s renewables that are doing it, along with the closure of Victoria’s giant Hazelwood coal-fired power generator, according to Tony Abbott.
The truth is simpler, and speaks volumes about the appalling way we’ve handled energy planning.”
Unfortunately this is a case where there are other factors at work, in addition to just the high gas prices (including both the Hazelwood closure and the intermittency).
It’s not more simple than Tony Abbott’s (incorrect) portrayal – it’s more complex.
Paul
Your chart of utilization has a buried story, because on the far left is the coal, and on the far right is the Wivenhoe PHES which is co-owned, and thus was elected to be less profitable by the coal-power-station owner. Who is in turn owned by the state.
The LCOE of the PHES element is pretty good: its amortized over the requirement for Wivenhoe to exist for flood mitigation. Its only the incremental cost of the PH part, the S part was there anway. So your chart shows how something which a really low PHES, can be turned into a non-earner, if the owner doesn’t want to use it.
Why doesn’t the owner want to use it? Because its only role would be to displace coal power. If it was either split off, or chinese walled, it would have run far more.
Thanks for the interesting comment, George
Some of our more learned readers might want to also add in their thoughts, but you raise the valid comment about portfolio effects in operational decisions.
That’s one reason I am intrigued by the possibilities of shifting Wivenhoe into the mooted “Clean Co” by the QLD Govt. This would potentially deliver both:
1) Market benefits, in terms of increased competition (which, as noted yesterday, has declined in QLD); and
2) A valuable opportunity for collective learning about a real world case study into the use of large-scale storage.
With particular respect to Wivenhoe, it’s important to remember that a true cost of production calculation would need to factor in both:
1) Cost of purchase of energy
Now this might have historically been done at “off-peak” times but, as noted last week, the pattern of pricing has shifted to make off-peak prices more expensive.
2) Round-trip efficiency
On top of this, my understanding is that the round-trip efficiency of the pumping-to-generation process is not as great as it might initially seem.
Hence (1) and (2) add up to a significant (and variable/hard-to-predict) short-run marginal cost of operations for Wivenhoe.
Hope this helps
Paul
As you point out with electricity supply and demand are inextricably linked so LCOE is only a useful measure if all are levellised. That means generators should only be permitted to supply levellised electrons, namely those they can reasonably guarantee (ie short of breakdown) 24/7 all year round.
LCOE as a measure was clearly only relevant to fossil fuel sources in the past but instantly became irrelevant with the enforcement of unreliables on the system. Now while unreliables may be able to outbid fossil fuels for pumped hydro under an enforced levellised supply regime they would still have to contend with the opportunity cost of wasting some water for alternative irrigation,etc with the Wivenhoes and Tumuts or invest in pumped seawater.
Truly levellised electricity with guaranteed despatchability would be agnostic as to their incorporation into any portfolio mix, unlike the situation we have now and LCOE would return to its rightful place in long term investment decision-making.
I’m afraid that’s not correct at all, so perhaps I didn’t make the point clearly enough in the article, or you are too focused on the “dispatchability” issue.
Different generation technologies with different balances of fixed and variable costs can yield changing LCOE relativities depending on assumed capacity utilization. The fact that LCOE is usually quoted for a single, assumed capacity factor for each technology makes it anything but “technology agnostic”.
Have you ever wondered why coal and CCGT are generally quoted as having much lower LCOE values than OCGT (or reciprocating engines), but in real electricity systems there is often plenty of the latter built despite this apparent cost disadvantage? It’s because OCGT economics are much more favourable for the low utilization roles which are inevitable because of the peakiness of demand and the need to cover other plant outages.
There seems to be a common misconception that OCGT plants are built primarily for technical reasons like fast start and ramping capabilities, but in large scale systems these are secondary to the economic driver which is lower capital cost – if something is only going to be used a few percent of the time you want the cheapest capital cost you can get and variable fuel costs are far less important.
LCOE should have a very limited role long term decision making since it unrealistically assumes a fixed or guaranteed capacity utilization.
I understand where you’re coming from here but in judging demand utilisation investment decisions fossil fuel plant investors know they have the capability at their fingertips to respond to various demands anytime. Unreliables don’t but have the imprimatur to dump on the system with extremely low variable costs.
As consumers we don’t want to know that a Tesla battery plus Hornsdale wind might be able to supply 100MW of power for an hour or so but what can Hornsdale supply for a whole week if called upon, let alone 24/7 all year round? Without enforcing levellised supply criteria we are headed for a Green fallacy of composition whereby supply is at the mercy of the elements. My ancestors weren’t so stupid when bright sparks like Tesla and Edison came along to show them the light and the way.
The real issue that our market operators are yet to address is the need for firm capacity to back up renewables and a mechanism for charging for capacity.
A capacity market will do this and will not favour one technology over another.
People often point to the WA capacity market as a failure and yes it was as their capacity market mechanism could not work when most of the generators offering capacity were owned by one entity and their IMO had a completely flawed methodology for determining the cost of capacity.
You don’t have to back up renewables if you simply rule that all tenderers to the grid can only tender up to a maximum they can guarantee 24/7 all year round and leave them all to work it out. Obviously the unreliables would have to invest in storage of some kind to lift their average tender or else partner with thermal generators and pay them their just premium for that.
Yes I know it will need some phase in for existing unreliables to comply with that level playing field and to break it gently to all the mums and dads with their rooftop solar but somebody has to stand up and scream out loud about the elephant in the room. It’s been the greatest moronic Groupthink that got us here in the first place and if you want to get rid of thermal completely then we aint seen nothing yet with the cost of going all renewable with the storage necessary to make it reliable 24/7.
You just look at the graph and ask yourself what sort of battery or pumped hydro storage would they need to be able to tender 10% of installed capacity reliably 24/7 all year round and think about the cost-
http://anero.id/energy/wind-energy/2017/june
One of the price tipping points for Victoria, which flows through to SA and Tasmania, is that there is now insufficient coal-fired generation to support even the night-time load. So even at night, spot prices are set by open cycle gas (Mortlake and Newport). According to the Aneroid Energy website, there are currently only 3,700 MW of coal-fired generation operating in Victoria, with Loy Yang A4 and Yallourn W S2 not operating (with a combined capacity of 900 MW). Night-time load is about 4,200 MW. When Hazelwood was operating, its low bid prices would feed through to low night-time prices in SA and Tasmania. These low-price periods no longer occur.
Another new feature is AEMO’s constraint of wind output in SA to 1200 MW if there are only 3 gas generators operating on the 275 kW network. This has led to curtailment of ~200 MW on recent high wind days (during both night-time and the morning peak), and good prices (~$70/MWh) for non-curtailed wind farms, whereas previously at times of high wind output and low demand, prices would become negative. So again, no tail of low-priced periods.
I would have thought that anyone familiar with electricity costs would already know that only the full system cost is a valid measure. Why does the matter need to be raised? Obviously because there are many cost claims circulating that are simply fake. I hope the next article exposes them. And by the way it is not valid to ‘factor in’ externalities like the damage cost of carbon emissions by adding them to the LCOE. The ‘externality adder’ has different purposes, especially in helping decide what are the best investments to make in reducing damage.
“..it is not valid to ‘factor in’ externalities like the damage cost of carbon emissions by adding them to the LCOE.”
Well it can be that we choose to impose a CO2 tax as a truer social cost on the system provided it bears equally on all tenderer’s CO2 inputs but the level of tax for such a level playing field is one of considerable conjecture.
I’d argue that also bears implicitly on your your other observation- “I would have thought that anyone familiar with electricity costs would already know that only the full system cost is a valid measure”. That’s why I’d argue you can’t possibly have a full system cost allocated fairly amongst all the tenderers without the level playing field of despatchability as well and so far that’s not the case. Finkel recognised that with his recommendation about despatchability for any future unreliables but I’d suggest he was politically silent about all the existing ones for the obvious. Too many red faces admitting they got it wrong.
Interesting, I like it but it is very involved.
Lets assume a very simple system and look at the effect of the RET
In the country of Utopia electric power is produced 50% by wind power and 50% by black coal over a 12 month time period
.
Black Coal produces electricity at $55 per megawatt, but due to the RET they have to pay the wind power producers $85 per megawatt, for a clean energy certificate, so their cost is $55 + $85 + 10% profit, + $154 per Megawatt.
Wind power gets the going rate at that time plus the $85 from the Black Coal Generator.
Thus the power end users Manufacturers , Commercial users and Householdsare paying higher prices than they did previously
Wind powers producers are very happy because they get paid when ever the wind is blowing and then get paid again by the Black Coal Electricity Generators
Not what I would call a ‘level playing field’
Please show me where me I am wrong, it is “doing my head in” and my Electricity Bill is just going up and up and up and up.
Well you might assume the $55 cost for black coal includes a return on capital so make the 10% the GST if that helps.
A timely article, Paul. I am wondering if anyone has done any research comparing price forecasts with actually observed spot prices. As the price pattern changes, forecasts based on historical trends should be getting less relevant
In terms of mitigating the “Crisis” in both price and supply what can be done to rectify the inadequecies of the current process?
Should existing coal fired stations be preserved until the process is improved and the alternative system in place and proven?
I visited the AEMO site today, and screenshotted what i saw. Over $300/MW in NSW! (25/07, 17:45) The other mainland states were not far behind! I certainly hope this does not become a trend! There didn’t seem to be any rhyme or reason for it. Generation figures appeared to be about normal?
The bottom line is, if governments leave critical infrastructure to the market then being free it will move in any direction. Mostly up to protect and reward investors.
The nuclear argument is now essential to the future of our electricity network as it appears ‘new coal’ is almost as unacceptable as nuclear.
Great reporting as always Paul.
Paul
Thanks for the post – as usual topical and to the point
The 6th point made by Sims was about the “loosening”of the network regulation rules about 10 years ago to maintain or boost government revenues. We are still living with the result of this and will do so for many decades to come unless something is done.
But we also need to reflect that the electricity wholesale market is becoming less competitive and this has allowed the wind and coal fired generators to sell their product at prices twice what they used to sell it at profitably, and the benefit of the differential between the cost of wind and coal fired generation and the selling price is not going to consumers
Gold plating would be going on with events like the Victorian bushfires and the SA blackout and mandated compensation in mind. Politicians respond to more quickly to the outraged twitterati.
Had an example of kneejerk compensation when the BIL enquired as to whether I had had a blackout in suburban Adelaide like he had but couldn’t remember having one but the $450 compo cheque said he did. No for me but then I reminded him there’s been a large country blackout including his shack at Mannum and that was it although he regularly enjoys positive cheques from his solar FIT there given its useage.
As for gold plating in general does it also have a lot to do with voltage and frequency management of widely dispersed variable power sources and can they even distill out those costs? Are there any estimates of that around or does it go deliberately unanswered through political censorship of enquiry?
It is unconscionable that the world should be turned inside out like this…but for Australia itself the potential for existential damage is of a magnitude that dwarfs that for all the other first world countries …because we have no alternatives after the Left kills fossil fuel.
Every other first world country has base load alternatives such as massive hydro and or nuclear and connectors to other countries to fall back on in the case of energy insecurity…Australia has none of that!
Australia will be ALONE…out on a dangerous limb as this RE juggernaut demands it commits suicide….the only country in the world being told to commit economic and social suicide !
For the amount of money the ANU’s 100% RE electrical system for Australia plan would cost….all of Australia’s coal and gas plants could be progressively replaced by HELE coal plants and any better versions to come…and there would still be money for the upgrading and/or replacement where necessary of the Australian grid.
There would still be a cut in emissions that’s demanded ….augmented by the elimination of the fugitive methane emissions from the CSG mining that would no longer be necessary.
Any Australian business people who are not deeply secretly invested in the RE juggernaut would have rocks in their heads if they let this insanity go one step further….especially as they should be au fait with the complete unsustainability of it all in the long-term with regard to EROI.
Unless of course they just don’t care about all that…they’re just in it to make their money quickly before it all hits the fan and Australia goes belly-up.
I just simply have no faith in the NEM anymore. I am going to start to consider going off-grid to stop these thieving bastards from my money.
I mean look at the prices, but it is made up on FCAS, absolutely outrageous.
As I understand it, 4 gas generators are required to be online for Wind to be allowed greater than 1200MW but I believe Osborne is not “counted” as it is only connected to te 66kV network. But it does seem surprising PPCCGT is only classed as one unit for this purpose? I cant find the actual exact requirements for Wind to be allowed greater than 1200MW. Is it documented somewhere?
“But it does seem surprising PPCCGT is only classed as one unit for this purpose?”
This post no doubt explains why it’s effectively treated at one unit only-
https://wattclarity.com.au//2017/02/examining-why-pelican-point-didnt-bid-full-capacity-into-the-nem-on-wednesday-8-february/
Given PPCCGT has been running at full load for the last month (2x GT and 1 x ST) your point is not relevant. If you did some research you would know Engie announced it was returning both units to the market a few months ago.
The next question is what AEMO are doing to update this constraint, given that there are another 3 large wind farms under development. They are supposed to follow processes that involve contestibility, but this appears to be an emergency procedure. Can the new big battery gain accreditation to be part of the “3 generators” rule ? Does it have enough synthetic inertia to count as 2 generators ? Can batteries in the new Riverland Solar Farm be included based on their synthetic inertia as measured at the 275 kV network ?
AEMO have also been developing plans for a new 275V AC “Horsham link”, but again there doesn’t appear to have been any contestibility of approaches. The existing AC link via Heywood is limited to well below its 1150 MW thermal rating because of frequency issues, so would a better investment be to upgrade it to DC between Tungkillo and Heywood, which could give a capacity of ~1500 MW with wire return or 3000 MW with earth return ?
There are currently 440 MW of new wind capacity underway – Hornsdale 3 (109 MW), Lincoln Gap (212 MW) and Willogolecha (119 MW). Perhaps one saving grace of these new ones is they are more exposed to easterlies than many of the older wind farms, so while they will produce during the westerlies of low pressure cells, they will also produce during the easterlies of highs, provided the high pressure cell is centred far enough south (as it normally is during summer).
How on earth can thermal possibly plan and predict their costs facing a wind sector varying between 3-85% of installed capacity we typically see here-
http://anero.id/energy/wind-energy/2017/august
Feast or famine and like Pelican Point they’re supposed to be able to enter into cheaper long term contracts for gas. When they couldn’t possibly offer that to the new gas field developers to defray their investment risk, there were plenty of Asian buyers only too happy to fill the void and leave the locals to the vicissitudes of the spot market. Laughing all the way to the bank now with global gas price rises.
As if that’s not hard enough, now we have the SA Govt throwing taxpayer funds at a Tesla battery and diesel generators that will ostensibly be creaming the top off any power price peaks further compounding their planning and contractual input dilemma. Never have I seen so much folly with lack of comprehension of the big picture here and the impending disaster.
There are parallels here with the NBN thought bubble. Utter Groupthink bedazzled by the chimera of ultimate speed and totally oblivious to the elephant in the room with bandwidth/congestion costs and now they’re doing the same again with power despatchability as they’re mesmerised by the lure of ‘free’ power from Gaia. The Devil’s all in the despatchability and he’s about to claim his dues.
You just have join the dots to see the same costly deal with solar-
http://tasrenew.org.au/solar/overvoltage/
http://mcelectrical.com.au/blog/Fronius-Smart-Meter-now-supports-Export-Limitation/
Alligators and swamps.
Metro7 comment on 1/8/17 deserves repeating………
“It is unconscionable that the world should be turned inside out like this…but for Australia itself the potential for existential damage is of a magnitude that dwarfs that for all the other first world countries …because we have no alternatives after the Left kills fossil fuel.
Every other first world country has base load alternatives such as massive hydro and or nuclear and connectors to other countries to fall back on in the case of energy insecurity…Australia has none of that!
Australia will be ALONE…out on a dangerous limb as this RE juggernaut demands it commits suicide….the only country in the world being told to commit economic and social suicide !”
Paul
Thanks for the post – it started me thinking and developing some thoughts I already had. The argument is that the current high wholesale price for electricity is being driven by wind (see above post) and gas.
But where is the benefit of the lower cost coal and wind reflected in the wholesale prices? Using NEMReview v6 I drew up a chart on a quarterly basis of the last 4 years of the total contributions of black and brown coal, hydro, wind and gas and matched these to the total NEM demand. (I tried to paste this into the dialogue box but it won’t paste!)
Interestingly the share of wind and coal to the total consumption have barely changed, ranging from 77 to 84% and the last quarter being 81%. The contribution from wind has also been a fairly constant proportion too ranging form 3% to 5%.
The cost of coal fired electricity has not moved much over the past 4 years but the cost of electricity has doubled. While I agree that the cost of gas fired electricity has increased, the volume of the gas generation hasn’t risen by much but the futures prices for electricity now seem to be based on the current high cost of gas as is the spot price.
So where has the benefit of the low cost coal and wind fired generation gone? The 2017 annual reports from Origin and AGL both show significant increases in contributions from their electricity sectors over the past 12 months.
So maybe he NEM rules need to be reviewed because they may be the cause that has allowed the gentailers to gain significant benefits.
Thanks for this article.
A friend has been helping a medium sized business plan for their energy needs. Despite the fact that they are in Outer Metro Adelaide it appears that their best option is to buy and operate number of diesel generators for n-1 redundancy and disconnect from the grid.
As has been observed previously, “We should start by noting that no-one actually KNOWS, in aggregate, how much small-scale solar PV systems are injecting into the NEM (or indeed in any grid).” (https://wattclarity.com.au//2016/06/the-opacity-of-distributed-small-scale-solar-pv-output/). Given this, how actual are the ‘NEM-wide actuals’? Isn’t this this more just one model converging with another?
Thanks James, very intertesting
There seems to be minimal innovation in retail pricing agreements. For example at present there are substantial price declines in the futures market. An innovative retailer should be able to offer a lower price for say 2 years, but with a switching fee. This way the retailer could purchase on the futures market.
I who’s interest is that ?
energymadeeasy eh? Now I’m pretty good at maths but trying to feed in a service charge, various charging tiers for summer and winter along with off-peak and to cap it off solar FIT scheme credits and the retailer solar bonus credit and on top of that split charging a couple of times a year with price rises part way through the 90 day billing periods and it’s a hulleva task trying to compare offers.
Did it once with a pushy door to door salesman I finally convinced to leave me his you beaut pricing/savings and it was $10 annual difference over my non-contract price with timely settlement discount. Complete waste of time when no domestic contract will guarantee you won’t get a price increase anytime after you sign up.
David
I don’t know if Paul answered your query separately, but let me have a go if he hasn’t.
As you no doubt know, the wholesale market price is set each half hour at a price which draws on enough supply to satisfy the demand over that half hour (“the market clearing price”). AEMO calls the cheapest sources first and then goes up the cost curve until it has enough supply to match demand. The last generator on that finally satisfies demand sets the market clearing price.
As coal plants close (eg Hazelwood), gas fired generators are increasingly becoming the generator that is last on and sets the market clearing price. A number of moth-balled gas fired plants are being restarted in response to the changing supply situation and the higher prices provide sufficient price incentive to do so.
So as gas prices have risen, so has the price that gas fired generators require to turn on and respond to market demand. This is what is driving up the wholesale price – the rising price of gas. (There may also be some pricing uncertainty as well due to the insecure nature of supplies as coal exits the market and there is no resolution on a national emissions policy).
The coal fired generators and renewables (when they supply) are probably doing very well as their costs are not changing but the price they are receiving from the market is much higher than it has been over the last 4-5 years.
Addressing the gas situation (and landing a national emissions policy) would go a long way to addressing the trend in electricity wholesale pricing without having to tamper with the NEM rules, which in itself creates uncertainty.
Why is it so hard to find what the southern annular mode is doing on the BOM website?
To me it is the most important predictor of rain in southern Australia particularly in winter west of the divide.
Wind may be predictable day to day but remains just as unpredictable as ever over the longer term.
The cause of the problem is vertical market integration. Emergence of 3 giant gentailers killed the competition. There should have been some type of “cross media laws” in place which shouldn’t allow generators to buy retailers.
That has been a contributing factor (thanks Alex) but it’s far from the only one.
We all need to use caution when focusing singularly on the [INSERT HOBBY HORSE HERE] that we think has been the reason for the escalation.
After all, mergers were also happening in the first 16 years as well…
Hi WattClarity,
I have spent the past 73 years thinking and trying to understanding how things work on the basis of the facts and the truth. The notion of “truth’ goes back to at least the Greek thinkers but on and off for the next 2500 years a lot of people have thought that the truth was a worthwhile goal in life. After the Scientific Revolution or may be at the same time as the Industrial Revolution these ideal seem to have been part of progress. Then came the Post-Modernist period and now we have moved onto the Revisionist Period. The “TRUTH” has gone – one must be “progressive” all facts if they get in the way of a point of view can be “reviewed” and adjusted. Take careful note and listen carefully about the augments about why one should vote “YES”. This rot started many be 50 years ago with advertising. We let people tell us that they product was the best. They all said that. One brushed that aside on the basis it was only advertising what would one expect someone to to say about their product. But letting such talk go unchallenged was dangerous. Look where we have got to. You seem to think that if you can show someone to be wrong with facts you might win an argument. That was once true but we now live in very dangerous times and such arguments don’t now carry much weight any more. The Unions used to talk about “Direct Action” they knew how to win an argument. May be now words are no longer mightier than the sword!!
At times when the spot price in SA is zero, is there no incentive for VIC to import this very cheap energy? Would this not raise the price in SA materially? Or does the wind supply in SA exceed the export capacity to VIC in these circumstances?
You might find the animation in this linked post of some value in understanding some of the dynamics:
https://wattclarity.com.au//2015/09/yin-and-yang-of-wind-in-the-south-australian-region-of-the-national-electricity-market/
I recollect that in NSW in 1994-95 Pacific Power’s bulk supply tariff for energy delivered at transmission level was $60/MWh.
I watched wind generator output in Southern Aust in May and June and it was pathetic. Oh that’s right we did not have our laughable 100MW energiser battery. Thermal powerstations did all the heavy lifting and ensured reliable supply something which wind and solar will never achieve. I am all for renewable energy but let’s stop pouring taxpayer money into this sector. We are told the technology has improved out of sight so let the proponents put their money where their mouth is without sucking the taxpayer dry. The debate is really ramping up and so it should. The public have been conned long enough and had a gutfull of the climate change religion. Unfortunately many people and the Government have woken up too late.
How does your weighted spot price average track against similar measures for gas prices in Australia?
In Ireland there is a very tight linkage between the wholesale average price and gas prices. Gas prices explain 80-90% of electricity price variations,
http://www.ey.com/ie/en/industries/power—utilities/ey-future-energy-and-consumers
Readers may be interested in an analysis I did on USE recently, the main conclusions being that it is a poor single/zero-figure measure of a complex issue, and that more wind power in SA will have very little effect on load-shedding:
https://climanrecon.wordpress.com/2017/02/05/issues-with-aemo-forecasting-part-3/
Thanks for your link – I’ve posted a comment on it at your blog. In summary I agree that additional information on the results underlying the USE measures would be very helpful and make the ESOO much more useful to readers.
The trick is how to capture and harness this highly variable energy source through effective storage technology, and that’s what our best people are working on
It is regrettable that the energy price debate is inevitably tainted with loose statements. It is even more regrettable that “delivered energy” prices are discovering dizzy new heights. Having said that though, by my calculations, Australia CPI in that period is just shy of 2.5% p.a compared to an implicit 5% compounding increase in wholesale electricity prices. So the generalisation is not entirely untrue…
Hi WattClarity,
When someone asserts that something cause causes something almost always the logic is flawed. WC is correct to challenge Hicks in the SMH, however I would suggest the rebuttal is equally flawed. Correlations or lack of, are a good starting point in discovering how inputs alter outputs but one needs a good deal more than a positive or null correlation to draw a conclusion one way or the other. Hicks suggests the NEM caused the price rise while WC produces good figures to show many years passed after NEM was created before the price rose. But it is equally valid to ask were the NEM to have never been created in the first place would that have prevented the very recent rise? If the answer might be “yes” then it is not wrong to suggest that the NEM has caused the rise. Trying to analysis this matter any further one gets bogged down with the inadequacies of the English language to explain such matters in a few words. The single word “cause” without careful definitions in particular circumstances is always dangerous. Positive or null corrections are an interesting subject. Looking for the causes of tornadoes in the US, one might be impressed with the very high correction between a certain observable single parameter and the frequency of tornadoes. The number of tornadoes per year and this variable tack one another over a 5 to 1 range over the past 100 years. One could well conclude this parameter to be the cause. The variable was the number of successful persecutions per years for having illicit distillation apparatus. The reason for the close collection was that both were tied to US Federal spending. More money meant more Federal Agents and more small weather stations to record tornadoes. Corrections by themselves are dangerous, almost every different sort of food has been shown to be both bad for you and good for you with correlations of one kind or another.
I would suggest that the NEM can be blamed for many negative outcomes; however whether better outcomes might have come from a world without the NEM must be consigned to What If History.
Great analysis Paul. Based on your estimate, LGC spot revenue foregone for those wind farms is roughly $5.2 million! This, plus increased SA spot prices for consumers makes it an expensive constraint for those involved. Be interesting to see what a cost-benefit analysis of the (seemingly arbitrary) constraint would show…
If the wind farm owners are waiting for AEMO to change these arbitrary rules in their favour, they will be waiting a long time because AEMO is under so much other pressure from government. Instead they should contribute and fund a trial, possibly with ARENA support, as Hornsdale wind farm are doing to test the new big battery and the wind farm’s capacity to offer FCAS. It is only through the dedicated short term resources of a trial that there will be enough evidence to convince AEMO that a new set of operating rules will work.
At the time some of the more recent wind farm investments received financial close, there was an AEMO/Electranet project to increase the SA-Vic capacity of the Heywood inter-connector to 650 MW. Yet AEMO continue to impose a 500 MW constraint, citing frequency problems. So they haven’t delivered on their project. Any other contractor would be in legal problems for under-delivery.
The 1200 MW rule is perhaps a boon for wind farms, because although their volume is constrained prices remain strong, whereas without the rule the SA market prices would crash because of the 500 MW Heywood link constraint.
One would hope that with more batteries in SA to maintain frequency AEMO would relax the 1200 MW rule, and allow the Heywood inter-connector to operate at the promised 650 MW, or perhaps even its line thermal capacity of 1150 MW.
You could use non-scheduled wind as a proxy for SS output. The old non-scheduled WFs are less productive though.
Paul very interested to read your thoughts on why it has spiked after 16 flat years?
Do you have time in the next couple of weeks? This is a very important discussion relating to energy prices.
Wind farms are both downward and upward dispatchable. On short time scales, a curtailed wind farm can be confident of its ability to move its output up. Likewise for solar PV.
I couldn’t see it linked in the article but earlier this month AEMO published a report providing some background information and details on the modelling done to establish the system strength requirements that underlie the 1200 MW wind constraints, and the conditions under which it will constrain wind generation.
https://www.aemo.com.au/-/media/Files/Media_Centre/2017/South_Australia_System_Strength_Assessment.pdf
It is fairly technical but worth a read.
It’s a good read. So the SA system needs to have enough synchronous momentum + instant demand response + battery to cover the tripping of the largest single gas generator operating at the time. As more batteries are added, fewer synchronous generators would be needed to back up the others.
There was no modelling of a 100% renewable system on a sunny or windy day, where there would be sufficient ramp up from batteries and curtailed capacity that no unreliable gas generator needs to be “covered” by others.
I’d like to draw attention to two interesting things about the AEMO report. The first is that despite Kate Summer’s numerous analyses of generator dead band issues, there is no commentary on this issue, and no modelling of how the synchronous generators might respond with tighter deadbands, and the role that might play in limiting oscillations. Secondly, they refer to the possible role of synchronous condensers in supporting system strength, and that they will model this in future. It seems to me that these issues are closely linked. Whilst inertia picks up the immediate loss of a generator, additional power has to be added to the system and this can come only from the governor response or across the interconnectors (or maybe even from the wind farms, because under the modelled scenario they are running curtailed) If the governor response remains slow, it seems to me that wide oscillations will follow. The modelling indicates that they used benchmarking for the synchronous generators that was developed for the system black event. Does this mean AEMO continues to overlook this issue?
Link to Kate’s article here https://wattclarity.com.au//2017/03/fast-frequency-service-treating-the-symptom-not-the-cause/
The term “dispatchable” used to (by default) refer to the ability to “dispatch” units to run or not run on command. The term was developed when all generating units had constant access to fuel. The wind energy’s co-opting of the term and bastardizing it’s ,meaning by parsing it from the concept of system adequacy contribution was clever but immoral in intent and in practice. Private enterprise excused from morality is cancerous and ruinous to societies.
This is useful in understanding the constraint. While noting the lost LGC revenue above, I’m curious as to how many of the constrained half hours occur in the wee-early hours, given SA is notorious for having the wind blowing overnight when demand is at minimum. As such how many have actually prevented generation at what would have been low (or even negative) energy price?
Thank you for this very well explained article re FCAS. Who said electricity is dull. Like any other piece of communication, it depends on how it is delivered Great article.
Stefano
“The aim is to put in place a robust scheme that delivers good system frequency control and improves power system security and reliability.”
Overarching number one macro rule- No tenderer of electrons to the communal grid can tender more than they can reasonably guarantee (ie short of unforseen mechanical breakdown) 24/7 all year round.
When you’re putting together a very large goods or freight train that has to get from A to B on time and handle all the inclines along the way there are certain basic performance requisites you require when calculating the number of engines you’re going to hitch up.
It’s a lot like my car when it doesn’t start or I push the accelerator and it doesn’t perform or stalls. I don’t just sit in it hoping for something to turn up, least of all for the weather to change.
Agreed. All engineering systems need backups if they are to be relied upon.
Thanks for the article Bruce, just one quibble in the advanced analogy I would alter the Transient Stability section to say:
“If an engine is decoupled from the train it will accelerate away from it because it does not have any carriages trying to hold it back. If you had a way of rapidly re-coupling it to the train before it got too far ahead, then everything would return to normal, otherwise the engine would escape (and probably crash!). In power systems the coupling is done by magnetic field between the generator rotor and its stator, “escaping” generators is known as pole slipping or transient instability, but if you can rapidly reconnect them before pole slipping occurs; the system is said to be transiently stable.
This should not be confused with circuit breakers suddenly opening causing the generator to unload while the turbine has steam passing through it. The sudden loss of load could cause the turbine and generator to over-speed with catastrophic results due to centrifugal force. Consequently turbines are fitted with over speed trip mechanisms which shut off steam flow to the turbine and release valves which exhaust the steam already in the turbine to atmosphere.”
Unfortunately text colour and cross-outs don’t translate to this comments field so if any of it is unclear I can forward it in a Word doc. P.S. I am a former power station operator so am more familiar with steam than electrical parts of the system.
Regards,
Shane O’Loughlin
Thanks Shane, I generally agree. However I should point out a sustained fault which doesn’t get cleared will also cause the generator to unload with likely catastrophic results.
Fortunately faults are usually cleared within a fraction of a second.
If not dead then dying or perhaps at best every man for himself- http://www.weeklytimesnow.com.au/agribusiness/dairy/dairy-farmers-go-offgrid-as-power-prices-surge/news-story/3f3b42124b2adef064da266255d03af7
Well the SA Govt have set the example rushing out the diesel gennys, albeit with a unicorn Tesla battery to divert attention away from the obvious.
Thanks Shane, I generally agree. However I should point out a sustained fault which doesn’t get cleared will also cause the generator to unload with likely catastrophic results.
Fortunately faults are usually cleared within a fraction of a second.
While some analogies are useful in gaining various insights into the behaviors of a complex system like for example Relativity, this is because what is being describe is beyond our normal experience. Clever people grappling with such concepts can extend their sophisticated understand this way. However, I fail to see why one would want to help those who want to stick their busy beaks into matters which in fact they actuality have nothing to contributed. If someone can’t understand the the system dynamics of a grid using the normal explanations using the appropriate mathematical equations with the correct solutions I think it best that they stay out of the discussion. I am dismayed to read about what I call the democratization of complex technology. I have been lucky enough to spend a working life dealing with various complex technologies that the public has never tried to understand and I would hope it should remain that way. As the clever people of this world do their level best to design, build and run complex systems I fail to understand why they need to taking into account the thoughts and opinions of the what I would like to call the stupid people. There was, once upon a time, nothing more boring than a bunch of power stations supply a grid in the mind of most of the public. All people cared about was that that it should keep working and they did not have to pay too much for the power. While the genie has partly got out of the lamp lets not do any more to let any more out.
Thanks Ken, I think the Duke of Wellington had similar views on democracy
You are kind in ONLY suggesting The Duke of Wellington might share my views – there are others that I would not align myself with for other reason who also thought that the democracy can go to far. On a more technical basis I think your analogy would be fine for a DC grid. But with an AC grid there needs to be two degrees of freedom (voltage and power) and the linear motion of a train only allows for one degree. But, trying to shoe-horn your idea a bit further if you had an engine (loco) to pull a number of carriages (cars in US Nomenclature) and that that train set was coupled to a much bigger train set by a rather long and electric “tow rope” (that could take compression as well as tension) then that might serve to explain the stability problems of South Australia as far as the interconnections are concerned. But your analogy works well for the control of governors and the various types of feeds (1 to 5) of the inverters associated with wind turbines. Some loco just set to a fixed throttle (diesel injector stroke) setting, while some locos with their “cruse controls” are set to a certain speed. But some of the very long ore train had delays of more than 30 seconds on the their brake air lines – not sure how that can be worked into the story. And I am not sure how you can work the quadrature current into the train analogy when dealing with the excitation of a particular alternator.
Hi Ken, I generally agree with your observations. Part of the reason I didn’t dwell too much on the voltage side of things was because it doesn’t fit too neatly into the train analogy, and I also wanted to talk about system frequency issues. However, please stay tuned, I do have a voltage analogy in the works which will be published in the next few weeks.
Add in the solar and wind in generation please to the train . What sort of Engine /carriage will that be? a ship’s wind sail on the a special carriage ? and solar would be solar cells on top of an electric driven carriage?
Simon
In theory Simon. The engine does not care where the power comes from. The power source is determined by politics and economics.
if i remember government forced the Western Power to retire a number of MW to reduce the over supply. Muja was the logical choice for Western Power
Thanks Paul for a good summary and attempt to explain to often inexplicable. But you do raise two important points.
Firstly there is little doubt that private ownership can improve things. The operation of Hazelwood before and after its sale shows that private ownership improved its availability and extended its life after the SECV had condemned it to the scrap heap in the early 1990s.
Secondly, it is the loss of competition between generators that has the biggest impact on prices. What we are seeing now is the loss of dispatchable generation which is the prime provider of forward contracts. There have been attempts (eg by ACCC over the sale of MacGen) to limit the loss of competition in the NEM but state governments have sold the assets to get the maximum sale price rather than deliver a market where competition is strong.
We still have nearly 90% of mainland generation provided by wind and coal at costs of production in the $40/MWh yet the cost in the NEM reflects much higher levels. This is a function of the loss of competition but I suspect that it is also a function of the market structure which is not designed to address the generation mix we now have
That might be a topic for another Wattclarity article
did you mean to say “gas and coal”, David, instead of “wind and coal”?
No I meant wind and coal. These are the low cost sources of generation. Gas is setting prices at the >$90/MWh mark which is where most of the spot prices and futures prices sit.
I used NEMR v6 to get the data for the last FY
I think this is a very even handed approach to a knotty question. I know I bring my bias to the table, I explicitly and frequently repudiate the market in energy, and I believe it was and can be seen to be a demonstrable failure. Alternate realities cannot be proven, so I have to fall back on belief, conviction that a public utility function approach with strong regulation would have delivered a better outcome. But then bring liddell to the table.. would the state have been able to decide to stop sweating the asset and re-capitalize? At what price of capital? at what risk to the lending rate which *is* subject to market process and has follow on consequences.
I know I’m biassed, but I’m holding onto my view right now: we needed innovation and we needed state intervention. What we’ve got is innovation but mostly undercapitalized and the state intervention has been a sorry saga of debt reduction and fantasy roles in the pricing war (Queensland government officials should hang their head in shame, having had substantive control over bidding all the time and simply defaulting to not using it)
So for me? its even odder. I want to agree with the News Ltd Journalist, but to different ends: the price rises were unneccessary but retention of Liddell was not, and will never be a good idea, in either state or private hands.
Seems to me you got your favoured State intervention aka directed innovation and capitalisation of a plethora of wind and solar generators via massive subsidy mining and the dearest power prices in the world. If the State was going to intervene like that and put an end to coal it made sense to flog off aging coal stations to private enterprise to squeeze the last drop of revenue out of them before closure.
The fly in the ointment with the cunning plan was always despatchability plus rising demand and concomitant high prices for alternative gas to replace coal’s traditional role. The Weatherill Govt has the simple solution to such oops! Roll out the diesel gennys and a show battery whatever the taxpayer cost and blame private enterprise for the price rises.
Paul, I look forward to reading this article as it evolves. No doubt you’ll get this completed before the villains even realise that root cause analysis is actually the first step they have to take.
Paul you constantly nail it on the head. However there is one big thing that I wish to add. The bill – 80% of bills that I see can be reduced. Yes, 80% of businesses are paying more than they need to pay for the same power. Why? its all too complicated and its not their core business. Well it is my core business and I am helping businesses save tens of thousands of dollars from the grass roots level. Happy to chat to you any day to discuss this mess. What character am I?
The debate over Catastrophic Anthropogenic Global Warming (CAGW) aside, once policy makers were in agreement about that, they had to ask the hard question- What does it take to replace a megawatt of despatchable thermal power with a megawatt of despatchable wind or solar power? That assumes that tidal and geothermal were way up the cost curve and the obvious rider to the dollar cost answer was how much CO2 would be saved anyway?
Instead we simply had massive wind and solar generator rollout with no thought given to their fickle and unreliable nature, but at the same time bankrupting thermal coal power (not so much bankrupting them immediately but drawing down on their sunk cost with no margin for depreciation and reinvestment)
It wasn’t a matter of ‘no thought’-it was with malice and aforethought…that thermal coal power has been killed off.
That was the pre-requisite…that COAL be killed off as a future electricity generator but be forced to hang around as long as needed so the RE industry can use it and gas to pretend that wind and solar can power a modern industrialised society.
By the time the whole world and the general population finds out that wind and solar …even when supported by batteries and pumped hydro…is unsustainable for the continuation of a modern society…the uber-rich generators will have made their squillions and moved on.
Ask yourself why Germany with infinitely more RE …trillions spent…nuclear power still…connectors to French nuclear power…can’t survive without COAL…and is building new brown coal plants even as they claim to be getting out of coal.
And then ask yourself what sort of madness is it that makes Australia…with no nuclear or huge hydro to fall back on…no connectors to other wealthy nuclear-powered countries… prostrate itself before the carpetbaggers and allow them to make us the ONLY first world country that’s offering itself up for sacrifice at the altar of a LW CAGW hoax.
Villain number one is John Howard who signed and published a fundamentally flawed Energy White Paper in June 2004
Assumption of 100 years of gas
https://pbs.twimg.com/media/DLmukgjVoAAefj3.jpg
Thanks Paul – I posted a bit of a response and a link to a blog post where I wrote about this, here: https://twitter.com/KetanJ0/status/917926993785536512 also included recently updated Australian data on climate science! Hope that helps clarify my comments 🙂
Thanks Ketan!
The public has been voting for the wrong politicians. This is because the global warming debate is circuitous and opinionated. The media indulge in conducting an endless debating club competition. And the reason for this is that we did not have a Prime Minster who himself or herself would have been firm enough in climate science and educate the Parliament and the public about it. A Prime Minister should invite climatologist James Hansen to give us a lesson or two. Lateline’s Tony Jones had an interview with Jim in 2009 but he didn’t mention it in Monday’s Q&A. That’s why we are not proceeding
The real question for the blue line (and conversely the red line perhaps) is- How did you come to believe in catastrophic anthropogenic global warming? The answer lies in the lemming like belief that you can defy a fundamental axiom of engineering, namely that you can build a reliable system from unreliable componentry.
The enquirer can be fooled into thinking that most people made their decisions on a rational scientific examination of the evidence and so we’re now running an electricity grid on e-motion.
Not sure that I agree.
The real problem is that we have had a decade of poor policy decisions by politicians of both sides.
They in turn were poorly advised by Government Bureaucrats who relied on overly optimistic forecasts of gas availabliity.
Just take another look at the Eastern Australian Domestic Gas Market Study undertaken back in 2013 and you can see that there were some heroic assumptions and forecasts about gas availability and price.
Energy Users raised their concerns but they were brushed aside.
Thanks Peter
I appreciate the difference of opinion and that, in all of these sorts of discussions, there is a high degree of subjectivity involved.
I agree that poor policy decisions (over 10 years or more, at state and federal levels) has been one of the main drivers – which is why I nominated our politicians as Villain #1..
In relation to your other points, I’d just note that I haven’t finished my 10 Villains list, yet.
Paul
Paul …I posted a reply laying out why I disagree fundamentally with the explanations for the confusion and train wreck …and with the whole RE takeover of our electricity system and the destruction of Australia for what I believe is demonstrably a CAGW hoax.
Is it because my reply was too long…it WAS long because there’s a great deal more to this issue than is admitted…or was it BECAUSE I fundamentally disagree…not just at the margins.
Could you let me know please for future reference?
“They in turn were poorly advised by Government Bureaucrats who relied on overly optimistic forecasts of gas availabliity.”
But when you believe all fossil fuels need to and can be completely replaced by renewables, why would you care a jot about gas availability, anymore than coal or oil? Caring about fossil fuel availability and the price of evil is akin to heresy for true believers.
“Energy Users raised their concerns but they were brushed aside.”
Ditto. When you’ve seen the light and the way via disciples like James Hansen and know Gaia’s endless bounty of wind and sun will provide the true path to the light on the hill, naturally you have to eschew such impure thought, lest you fall into the sin of skepticism. You’re not continually having these impure thoughts and concerns like we skeptics are you Peter?
You haven’t explained where Howard or that paper were wrong….in the quantities…or just the crime of using a fossil fuel?
If the latter and you are a true believer in CAGW…can you tell me why Australia alone in the world must be the country that commits economic and social suicide because the world warmed a small amount …over a period of 20 years at most…a tiny blip in the scheme of things….starting and finishing with two entirely natural warming events….and followed since then by an IPCC-admitted…recently-reiterated PAUSE??
Can you tell me why that tiny all-natural warming trend set within a steady slight warming over hundreds of years as the earth emerges from the Little Ice Age …warming that wasn’t initiated by a rise in CO2…the CO2 rise having followed the warming …can you tell me why these entirely natural events require Australia to commit hari kari …destroying our children’s futures ?
Can you tell me why the earth WOULDn’t be slightly warming …or did you …IF you’re a warmist…thinks the earth should have retained all that ice from the LIA…and be all iced up now and maybe cooling towards ICE AGE ARMAGEDDON
Exactly right Observa.
The RE industry and other true believers find it very easy to brush aside those little inconvenient facts like fugitive emissions of methane….which is variously described as having 80-100 times more warming capacity than CO2…is ubiquitous and very hard to mitigate…many old wells that have been shut down continuing to release it for decades.
Sierra Club at one point admitted methane was a much worse problem than CO2 but seems to have gone usefully quiet for the sake of the RE industry and the financial and other behemoths with all that investment at stake.
They can always let us know when it’s all too late…our children’s futures destroyed…Australia a 3rd world energy-poor hitherto energy superpower.
The RE industry finds it so easy to put the lid on these pesky truths because they have a completely compliant and protective LW MSM that’s always prepared to destroy a Prime Minister when it’s deemed necessary-to trash the people’s vote…eager to work their guts out to put a stake through the only legitimate PM’s heart in hopes of burying the corpse…anything to keep the heist going…the biggest money shift in history from poor to uber-rich…the millions rolling in and the populace in the dark…literally so before too long.
I’m becoming fed up with the endless prattle that spews out of print and e-media about climate change. I’m becoming more and more convinced that it is all about vested interests and greed. Take away the subsidies and watch the CE industry collapse faster than a house of cards.
Speak of the Devil and impure thoughts-
https://www.spectator.co.uk/2017/08/question-al-gore-on-climate-change-and-hell-call-you-a-denier/
Kogan Creek was originally developed in conjunction with US utility Southern Corporation until CS Energy bought them out. Swanbank E was developed by CS Energy without any JV partner.
It is also worth mentioning the Braemar 1 (450MW in 2006), Braemar 2 (450MW in 2009) and Darling Downs (630MW in 2010) gas fired stations which added significant capacity to QLD.
The overall utilisation of Liddell in recent years may seem similar to the early 2000’s but the operations of the units is quite different. Between about 2000 and 2010, Macgen normally ran two units at up to 515MW (following the turbine upgrade) with two units on standby, with more than two units only used during outages at Bayswater. AGL are running more units (when available) but have derated them to 420MW in order to improve their reliability.
Muja AB was retired by Verve (now Synergy) in order to meet a cap placed on Verve’s generation capacity. It was then brought out of retirement as project between Verve and a private company in order to increase non gas fired generation following the Veranus Island gas explosion. The problems were mainly becaue the boilers had corroded significantly when the plant was shutdown because it had not been laid up properly because no one thought it would be returned to service when it was shutdown.
According to the ACCC the breakdown of cost increases is…”made up of network costs (48 per cent), wholesale costs (22 per cent), environmental costs (7 per cent), retail costs (16 per cent) and retail profits (8 per cent).”
http://www.msn.com/en-au/news/australia/network-costs-the-main-driver-of-electricity-prices-consumer-watchdog-finds/ar-AAtup1Y
But of those ‘gold plating’ network costs and to some extent wholesale costs we have no measure of the added costs to the supply chain in managing voltage and frequency with the addition of a plethora of these disparate unreliables and we need to address that. Admiral Nelson holding a telescope to his blind eye would no doubt approve of the modus operandi here, not least with our Chief Scientist recommending all future unreliables be despatchable, but we’ll overlook the retrospectivity folks.
Nice to read a really high academic quality article – thank you. Only at the end do you mention what I think is the event that will cause the greatest trouble and that is when Adelaide and Melbourne are both extra hot. This is where better demand side management would help. And, by better I mean more carrots and less sticks.
The first paragraph in section 1 “A fundamental concept underlying the NEM’s design was … etc etc” outlines the intended outcome of the NEM and in this paragraph the writer implies that this dream can, in an ideal world, be accomplished. I would challenge this underlying assumption. I believe it is not possible to devise a free market framework where price signals will provide feedback that will deliver a stable and technically desirable out come. I think the free market economy is, for a great many processes, the best of a bad bunch of alternatives, but for making and delivering electrical energy to consumers is one of the silliest ideas I have ever come across. No amount of tuning and fine tuning will ever work – the whole idea stupid. The NEM is a process where a small number of people can make some money. From that point of view it works. But there are no other winners only losers in the NEM landscape.
Well frankly the penny is slowly dropping that you can’t build a reliable system from unreliable componentry. That’s a fundamental axiom of engineering and where in the world did so many bright sparks ever witness a precedent to the breaking of that rule and yet they charged ahead with active promotion of power generation at the mercy of sunlight and weather. If your first year apprentice mechanic hadn’t understood that axiom of engineering by second year you’d have to fire him for the sake of your customers.
Power is an essential service so like police and the fire brigade, if you want to outsource its provision you don’t go round accepting tenders that have fine print saying no one will be on duty at night or when the wind don’t blow or blows too hard, even though the tender price looks good.
Where do we have to go with a communal power grid for a level playing field? Well whatever the supreme guru thinks the price of CO2 should be to get the earth’s temperature back to the temperature only the supreme guru knows it should be is imposed on all the tenderers. Then they are only allowed to tender those electrons to the communal grid they can reasonably guarantee 24/7, all year round.
Yeah I know all those red faces that thought they could ignore that fundamental engineering axiom and have to tell all the mums and dads with rooftop solar that if they can’t guarantee their electrons they can keep them. That’s why even Chief Scientists duck such retrospectivity with the bleeding obvious.
If I read this correctly there is a reasonable chance that Victoria’s peak demand could reach 10-10.5 GW. Total despatchable capacity including VIctoria’s share of the Snowy is 9GW. maximum imports from Tassie, 500 MW which leaves 1 GW to be provided by demand response (150MW likely) local wind and solar (500MW possible) lmports from NSW zero, imports from SA possibly 350MW. reserve margin zero.
“you can’t build a reliable system from unreliable componentry. That’s a fundamental axiom of engineering”
The person who wrote that knows nothing about engineering.
I would agree that the author may know very little about engineering. However what is so unusual about that in the current debate about how the NEM works? There seems to me to be a plethora of economists and bankers and spin doctors with virtually no rational professional engineering input. We would not have the mess we now have if engineers had been left alone to deliver cheap, clean reliable power.
Well try a chain is as strong as it’s weakest link. I’m not saying wind and solar power can’t be made reliable to contribute to a reliable grid system but at what cost and that was completely overlooked or deliberately ignored. Why were so many so stupid or culpable in doing that?
Allan
Thanks for the very useful dissertation. What the article does is to highlight that the sketched outline has a number of wrinkles that have to be addressed, and that addressing them might not be as straight forward as implied..
I am concerned that the NEG overlaid by the incipient 5 minute settlement rule change will have an even greater downward pressure on competition and lead to even higher prices.
Is a GT which can provide energy within 10 minutes a “reliable” source under 5 minute settlement? Is a battery that discharges within 5 minutes a reliable source under 5 minute settlement, noting that it then increases demand as it recharges? Is a coal fired generator with the minimum ramp rate a reliable source under 5 minute settlement?
The more I dig into the NEG and 5 minute settlement, the more I realize that the economists involved really don’t understand the limitations of the technologies that our electricity system needs to ensure that electricity supplies are reliable
“What the article does is to highlight that the sketched outline has a number of wrinkles that have to be addressed, and that addressing them might not be as straight forward as implied..”
Wrinkles! The problem is we have a bunch of beauticians trying to put lipstick on a pig. 5 or 10 minutes despatchability is not the issue but the really big picture- http://anero.id/energy/wind-energy/2017/october
or worse still-
http://anero.id/energy/wind-energy/2017/june
Now while that wind energy sector can produce an annual average output of installed capacity around 30%, just ask yourself what sort of storage investment would they require to even guarantee 5% or 10% of their capacity 24/7 all year round? They haven’t had to but get to dump those peak outputs at almost zero cost onto the market and leave all the thermals to pick up the insurance tab for all the troughs. Nice work if you can get it particularly when you’re further subsidised to do it via cash transfers from Govt mandated taxes on your competition. I guess you could call it a free market and we’ll no doubt all be free of despatchable thermal power sooner or later.
Have they fessed up to cost and pool receipts – are they registered participants.?
Very interesting. The energy balance that Peter Farley makes is very interesting. The whole SA issue depends on probabilities of demand and generation events. The first is that Vic and SA demands are highly correlated. AEMO has summarised the outlook in terms of SA demands, plant failures in Vic (not in these terms) and limitations in solar and wind resources. One issue is that this is fine for one day and all going well but if you get a series of very hot days, then it only requires a couple of these and a plant event to make for a potential loss-of-load event. Any flaws in this logic
Good article.
The NEG proposal has a problem – around 85% of forward contracts are now done through the ASX24 futures, and bilateral OTC contracts are only a small part of the market.
The September 2016 system black was not due to a failure of contracting, reliability or dispatchable generation – it was due to system instability and lack of synchronous inertia.
However, the January 2017 problem was a market contracting failure. There was apparently a lot of customer load that had not contracted with generation, and generators were left undispatched because they had not been contracted.
Why would customers choose spot exposure rather than take out a contract?
I suspect that a large reason is because the $300 Administered Price Cap enabled them to go uncontracted without a significant spot exposure.
The Administered Price Cap has not been reset for many years. It needs to be re-examined, and re-priced to take account of much higher current short-run marginal costs.
The contracts for reliability guarantee may not require financial contracts, i.e. the derivatives. Dispatchable generators and load may sell their dispatchable capacity into the market in 1 MW packets for each 5 minutes dispatch intervals. This can be standard contracts tradable in secondary market. The challenge would be for AEMO and AER to enforce compliance. It would be a serious challenge to verify that a dispatchable generator or load have the dispatchable capacity available for each dispatch intervals and they have not sold more capacity than physically available. This verification would involve checking the technical and operational data records, not merely a reconciliation of contract volumes against physical demand.
It is possible, but I think it would be cost prohibitive.
Ian,
I think you are missing a few things e.g.
1) SA Spot prices compared to Qld for the last 2 Calendar years and 2017 so far yield the following:
SA_SMP QLD_SMP
2017 $108.99 $109.13
2016 $69.67 $67.44
2015 $49.58 $51.97
Which makes me doubt those who spout the economic benefits of coal.
2) . It appears highly variable and unpredictable because it is variable but predictability is improving according to AEMO at least. In any case there are numerous extended periods when the SE Interconnector flow is SA to Vic but it works as intended generally so we get the benefit of lower prices in both regions and the highly flexible gas fired plant in SA is more than capable of operating counter cyclically with the wind generation.
3) Reliance on the Interconnector has always been high but not nearly so much since the wind generators were installed and Victoria benefits considerably from SA exports at times of high wind generation. Far from SA not thriving on wind it does, and so does Victoria.
4) True up to a point but not nearly so much as in the past. The interconnector was initially installed to allow SA to benefit from “opportunity” power purchases from Victoria whenever their price made it economical. Since the NEM start quite a variety of scenarios (draught, plant failure, Basslink OOS etc.) have led to a wide variety of inter-regional power & FCAS provision outcomes.
5) The Heywood interconnector is subject to a lot of constraints for all sorts of reasons but only some are related to wind generation. Apparently Basslink’s availability and its import to export deadband can have significant impacts. Even deep-in-the-system Victorian transmission line, transformer and SVC availability (I believe) can have significant effects. It is even affected by outages in NSW & Qld but so are other interconnectors, though I suspect that SA’s small size and unique location at the far end of a much larger system has an impact as well.
6) I would be surprised if Tasmania hadn’t learned the lesson of the Basslink outage which occurred just after they ran their reserves very low. Remember all those diesels they had to import! I think many of them are now in SA for a couple of years but I doubt they will run now the ride-through settings on the windfarms have been corrected.
Mark,
My attempt below to correct just such a thing in a previous post “Noteworthy analysis of the state-wide blackout across SA 28/09/2016”; the only question I have is what were you looking at prior to the last 10 years? This stuff is standard fare in politics.
Ian,
I think you are missing a few things e.g.
1) SA Spot prices compared to Qld for the last 2 Calendar years and 2017 so far yield the following:
SA_SMP QLD_SMP
2017 $108.99 $109.13
2016 $69.67 $67.44
2015 $49.58 $51.97
Which makes me doubt those who spout the economic benefits of coal.
2) . It appears highly variable and unpredictable because it is variable but predictability is improving according to AEMO at least. In any case there are numerous extended periods when the SE Interconnector flow is SA to Vic but it works as intended generally so we get the benefit of lower prices in both regions and the highly flexible gas fired plant in SA is more than capable of operating counter cyclically with the wind generation.
3) Reliance on the Interconnector has always been high but not nearly so much since the wind generators were installed and Victoria benefits considerably from SA exports at times of high wind generation. Far from SA not thriving on wind it does, and so does Victoria.
4) True up to a point but not nearly so much as in the past. The interconnector was initially installed to allow SA to benefit from “opportunity” power purchases from Victoria whenever their price made it economical. Since the NEM start quite a variety of scenarios (draught, plant failure, Basslink OOS etc.) have led to a wide variety of inter-regional power & FCAS provision outcomes.
5) The Heywood interconnector is subject to a lot of constraints for all sorts of reasons but only some are related to wind generation. Apparently Basslink’s availability and its import to export deadband can have significant impacts. Even deep-in-the-system Victorian transmission line, transformer and SVC availability (I believe) can have significant effects. It is even affected by outages in NSW & Qld but so are other interconnectors, though I suspect that SA’s small size and unique location at the far end of a much larger system has an impact as well.
6) I would be surprised if Tasmania hadn’t learned the lesson of the Basslink outage which occurred just after they ran their reserves very low. Remember all those diesels they had to import! I think many of them are now in SA for a couple of years but I doubt they will run now the ride-through settings on the windfarms have been corrected. I probably just missed it but I don’t remember AEMO fessing up to their lack of diligence in that instance, maybe their lawyers wouldn’t let them.
Does anyone know when the other 7 (or is that 4) units will ne commissioned?
Somewhere the goalposts seem to have shifted from 9 units down to 6 units.
At whose instigation?
From Adelaide Advertiser 2 August 2017
Mr Weatherill has announced plans to lease, and possibly later buy, nine new turbines which can run off either diesel or gas, and provide the state up to 276MW of energy in times of emergency.
[..] Yogi Berra who quipped “you can observe a lot by watching” (he even wrote a book with the title). [..] and I made a little blog post on his passing in Sep 2015 ..
http://eljeiffel.blogspot.com.au/2015/09/on-observation-yogi-berri.html?m=1
So many great ‘observations’ in this post. Will take a while to read, understand and absorb. As always, thank you for your generosity in sharing your experiences and learnings.
Seems like a lot of complexity to tell you what you must have already known really. Solar works best around the middle of the day, less well when its cloudy, sometimes it will supply more than i need , sometimes less. Its not clear to me how i benefit or what i would change as a result of converting basic reality to data and graphs. I must be mising something.
I agree to an extent, but have observed (over many years of user interface design for process plant operators, and frequent reading of other types of data presented in analogue, numeric tabular, or graphical X Vs Y or X vs T forms) that different people have widely different abilities to gain meaning or understanding from that data.
Thus while I might be like you and fairly adept at quickly interpreting data either by comparison with what we expect from our excellent understanding, or looking at a column of numbers and intuitively calculating average and variance, min, max etc, it remains true that there is incremental value in presenting it graphically.
Sometimes you will see learned patterns or changes to patterns that alert you to something significant. As the number of variables and data points increases, the ability of virtually zero cost additional flexibility in display the data in a variety of ways becomes nearly essential.
The trend graphs in Excel can offer even further insights especially if you resist the temptation with the polynomials to go to too many levels of complexity.
Paul,
By way of further analysis; the remark at:
http://australianenergycouncil.cmail19.com/t/r-l-jlbjyll-hyltdykhjy-i/
that “Port Augusta was the ideal location for a hybrid renewables power station because wind strength increased throughout the day and it had a high number of days of sunshine…mainly driven by the temperature differences between land and sea…” got me thinking that with the advent of large solar inputs at the transmission level about to arrive on the scene, including National Grid Metering, perhaps some accurate analysis could be done of the level of cover provided by the intermittency of each e.g. if the wind strength grows at Pt. Augusta as the day progresses it could alleviate the duck curve considerably.
Cheers,
Interested to see your thoughts on this. I regularly see another bias at work – optimism bias – which is where people think that things that have happened in the past to others wont happen in future to them, without any genuine reasons for thinking this.
Last summer was quite eventful in SA, but Oz lived up to its name as the lucky country when a very hot spell, culminating in a total wind power around only 50 MW at peak demand, fell on Christmas day rather than on a full working day:
https://climanrecon.wordpress.com/2016/12/30/christmas-2016-in-south-australia/
at first I thought that reduced residential PV installation data after the end of the RET would be a huge problem. I am a data geek too. but then I thought, PV is really a negative load (at least until it gets to a very large portion of the daytime load).
similarly we don’t have good data every time someone installs an airconditioner, but in aggregate we see their effect with the electricity demand increase on the very hot days. won’t cloudy days impact on PV be just the same as aircon on hot days but in reverse?
I fool around with numbers in the same way to get a handle on an issue. I contend there are two sorts of minds word-smiths and numerate minds. The world is mostly run now by and controlled now by the word-smiths because (1) there many of them who don’t have a worthwhile engagement in life and (2) their thoughts resonate with the pubic. But I can’t thing without numbers being a Professional Engineer and I believe no one else can who wants to make a worthwhile contribution to the current debate about electricity.
But to your topic. Most calculations suggest that cycling electricity in and out of a battery adds between 20 cents and 40 cents per kWh ($200 to $400 / MWh). Those figures assume the battery runs for its expected life at about 1 cycle per day. These costs need to be paid by someone and they are far from insignificant. Whatever the reasons are to have a battery these costs are real and to my mind dominate the other considerations.
Hi Paul,
Olivier here. Thanks for relaying our messages.
We definitely agree that the marginal should be used.
It’s actually already available at pro.electricitymap.org.
Will write another article about that pretty soon.
Olivier
….showing the period of the charge (a couple hours at a rate of about 30MW)…
It’s “a couple OF hours”. This is not America!
Hi Paul,
a couple of observations that I think will make a big difference to your analysis.
1. Most 100% renewable studies accept that if you have a week of really good solar (and/or wind), then you will likely be better off spilling some of the excess rather than trying to store every last kWh. Spilling 10% of your annual resource during the good times will only add 10% to the LCOE of the wind or solar, which is not an outrageous imposition. If you redo your calcs assuming a 10% overbuild, and assume that 10% gets wasted during the good days, then I think you’ll find that storage gets reduced considerably.
2. Our existing hydro is quite flexible, sometimes it puts out less than 500 MW, sometimes it puts out more than 4,000 MW. If you run existing hydro on average 1 GW harder during a week of poor wind and solar weather than you do during a week of good weather, then that reduces your storage requirements by 168 GWh, ie an amount far greater than the 68 GWh of storage that you came up with.
Having said that, your assumption to ignore seasonal differences in generation was clearly very generous. Put that back in and storage requirements increase massively. The AEMO and UNSW 100% renewable studies found that the NEM would need around 100 to 200 GWh of storage, though they both had essentially full back-up from dispatchable gas generation (converted to biofuels) that would run during the periods of poor wind or solar.
The recent ANU (Blakers) and the older BZE 100% renewable studies didn’t rely so heavily on this dispatcable back-up, and as a result they required much more storage, 400 to 700 GWh.
When AEMO says that 2000 MW of additional resources will be available to the market, what are they referring to?
The Summer Operations 2017-2018 document released by the AEMO in November provides details about a number of actions that provides the 2,000 MW of additional resources: increasing capacity and maximising availability of generation and transmission system in the NEM, additional emergency non-market generation, ensuring sufficient fuel supply for generators and encouraging demand-side resources.
So presumably the zero output last sunday through to tuesday selecting HDWF2 only in SA was due to battery hookup/testing?
http://anero.id/energy/wind-energy/2017/november
Please simply explain how the purpose of Big-Battery in ultra-short-term load demand, frequency and voltage-stability.
How even if you were designing a co-gen i.e gas-coal super-critical plant that you would add battery load-equalisation
Well before I wanted to tackle the Everest of electrochemical storage I’d want to prepare for any obstacles like cobalt cliffs-
https://notalotofpeopleknowthat.wordpress.com/2017/11/25/the-cobalt-cliff/
and as one astute commenter summarised it most succinctly-
“I will take the liberty of saying what I think the poster is getting at.
1. That there are solutions available to those Tesla battery design issues mentioned in the article, but the known solutions will kill Tesla’s range claim.
2. That there are three considerations when talking about batteries for E.V. (electric vehicle) designs. Range, charge time and battery life. The most effective solution for one of these considerations negatively impacts at least one of the other two.
3. That there are solutions readily available that solve any one of the three. Any potential method of solving two of them at once is very difficult and definitely not readily available. There is no solution available to solve all three considerations at once that have even reached the stage of a successful test of the theory in a laboratory. Which needless to say leaves a multitude of real world conditions that aren’t even subject to conjecture yet.
4. Unstated in his post is the oncoming calamity which is the underlying goal of some countries to have every single lawn mower, tractor, car, truck, emergency response vehicle, industrial equipment etc replaced by e.v. in the next twenty years. A goal enforced by law in the hope that a solution for some fundamental battery problems will be solved in the next couple of years when the rollout of the conversion will have to start.”
Whether or not Mr Musk will be around to lead this assault on Everest some are beginning to raise the question of his preparedness and sponsorship-
https://www.msn.com/en-au/motoring/news/comment-tesla-roadster-launch-is-an-illusion-nothing-more/ar-BBFZ4FE
Good analysis.
But what amazes me how quickly good engineering management has been forgotten and replaced by market place substitutes. You write how no one in SA had bid for frequency control. This remark is akin to suggesting that on flight XYZ no one bit to put wings on the plane. These are not market place discussions – these are not things that come from “commercial’ decisions – common sense suggest certain things should happen or be done. If this ludicrous idea of leaving critical decisions to the market place is to continue then there has to be consequences of such omissions – like for example what happened in SA should have costed someone something like $50m. There are minimum technical requirements for reliable and cost effective outcomes. If all of this is handed over to the market place then there has to be the service guarantee with heavy consequences for not meeting objectively stated outcomes. What we have now in Australia is not more than 50% of the processes that are required in a completely commercialized market.
Telling one of my engineering friends that migrated to California 50 years ago about the SA event last year and his first reaction was who will pay the $50m that someone in California would have had if that had happened there.
The main problem Australia faces is the rules of engagement were written buy the guys that hate rules. It is as if we had asked the criminals to write the laws. Of course they would have stopped people going to jail and all the nasty consequences.
Our problems will continue for ever no matter how many inquires or investigations until (1) we stop asking people that have vested interest or
(2) we ask someone without vested interest but who know what they are are doing.
Because up to now all they people that know what they are doing are too close to be impartial and impartial people don’t know what is what having never run anything let alone a grid.
“There are minimum technical requirements for reliable and cost effective outcomes.”
Such as no generator can tender any more electrons than they can reasonably guarantee the supply of 24/7, all year round? Furthermore should every generator also be in a position to contribute to frequency control if and as required? Are some specific types of generators being given a free pass with those basic requirements now and if so why so?
OTOH don’t those 2 requirements really matter for overall grid performance and end consumer satisfaction?
Hi observa, the current FCAS rules allow all generators to get a “free ride” as in they don’t have to contribute to frequency control if they so choose. The market is meant to sort that out by price signals encouraging re-entry. However my observation is that the ancillary services market doesn’t work very well, and in my view it was a flawed idea in the first place. It is like setting up a market to determine which side of the road you should drive on, penalise drivers who drive on the wrong side and reward drivers who drive on the right side. Do this on the basis that the most efficient side will be chosen in all possible circumstances. If we organised our road system like this we would have more efficient transport ? or would we have chaos?
Hi Ken, I agree, good engineering philosophies are being overridden by dogmatic attitudes in many cases. In part that is because the engineering profession is not communicating very well but communication is a two way street and it is also the case that regulators and rule makers etc. are not listening. Maybe a few more failures need to occur for the points to sink in.
Hi Bruce,
There are many lessons to be taken from the story of the railways in Britain between about 1830 and 1870. In many ways what we call “capitalism” was born at that time with people buying shares in railways And then in the latter part of that period what was happening in the US. But the more important lessons come from Britain. It was at that time the ICE, the Institution of Civil Engineers moved to near Parliament because the story of the railways is a story of equal measure of Legislation and private investment. The railways wanted land in a continuous strip that could only be purchased by compulsory acquisition by Legislation and many conditions were placed on the railways in exchange for Parliamentary assistance in getting the land for them. And so capitalism in many ways was not born where there were no rules and money bought everything but in a partnership between government and private investment. And the monopolistic conditions of the grid are very similar to the monopoly of the a railway. Our government be it Labor or Liberal have no idea about how to write appropriate Legislation, so they have turned to people that do know what they are doing but they have serious vested interests and endless conflicts of interest. We Engineers need to get off our bikes and advise the various Ministers. I, acting alone, have done a lot of this in the past few years and in my long life I have never met situations where those I have tried to talk to are not even slightly interested in what I have say
Ken.
THANKS CAROL. I LIKE THE NEW FORMAT OF THAT REPORT IT LOOKS REALLY NICE.
A nice increase from the low end constraint for wind from 1200 to 1295MW. I noticed the doc mentions Heywood still only modelled to 600MW VIC>SA. Not sure if we will ever see it opened to 650MW! SA>VIC still also not approved over 500MW. Hornsdale battery mentioned but only the market part of it (30MW). Perhaps with the January update the 70MW component (which bids into FCAS markets) will be able to contribute to system strength?
Not sure you can deduce the duration of the discharge from the 5-minute NEM data. Is it defined what the MW figures are exactly, average over the 5-minute period, or peak, or something else?
Well AEMO’s demand forecasts were more than 2,000 MW out on the high side (over 1,000 in Victoria alone), although capital city maximum temperatures weren’t too far from the BoM forecasts.
Similarly last Wednesday which was hot in both SA and Vic saw demand come in well below levels forecast in predispatch.
Either AEMO is being very conservative this year or demand behaviour has changed – possibly both.
Forecasting difficulties the other way for Victoria on Tuesday – at 2:30pm local time Vic max half hourly demand was forecast to be 7751 MW (although actual 5 minute demand was already something over 7600 MW at the time of the forecast). 90 minutes later the max demand forecast had increased to 8400 MW!
Interesting also that it is only the second December that we have seen values over 13GW (currently 13,010). Given that historically Jan and Feb have been higher, it could be an interesting summer. Off to fire up my BBQ and watch the summer with interest.
So apparent demand was only less than 45% what it was on the afternoon of 19 Dec! 8500 vs 3500 or so. Such variability.
If there were more points of input, would it get harder or easier? If all wind sources from now on are required in (hypothetical) law to have associated storage would that help? I suspect that more points would help but complicate the models and mandated storage would only add hysteresis at the cost of requiring oversupply built into the model at almost all times.
SA may be an outlier in terms of wind power predictability because much of its capacity lies in the hills near Adelaide, and wind there tends to have a very large diurnal variability, which means that it cannot be predicted solely from the spacings between isobars, which generally don’t change much over 24 hours.
Theory Alert (I’m not a meteorologist): The diurnal variability arises from air rising and falling as it warms/cools, a phenomenon known locally as gully winds. I’m guessing that localised vertical air motion is harder to predict than wide area horizontal motion caused by pressure gradients.
The total wind power must be some sort of sum of the local effects and the synoptic effects, sometimes with almost exact cancellation, so even if both parts can be predicted to within (say) 10%, when they are in opposition the error in their sum can be relatively large.
We have done quite a bit of investigation in this area with some promising results. The difficulties lie in the non-linearity of the problem.
The power output of wind turbines is related to the cube of the wind speed, so small errors get amplified. Above a certain wind speed the turbine output becomes constant, and for very high speeds the turbines may trip.
These characteristics vary from turbine model to turbine model.
Local topography also affects wind farm output.
For these reasons, meteorological schemes may be an insufficient approach because they necessarily sample wind speed measurements at discrete locations and extrapolate to areas.
We have been experimenting with approaches which approach the problem from a different angle.
Very interesting Paul, it’s also an illustration of the relatively high correlation of windfarm output in SA, with so much capacity concentrated in the mid-North of the State (the Hallett group 350 MW, Snowtown 370 MW, Hornsdale 315 MW plus Clements Gap 58 MW and Waterloo 130 MW also close by), obviously similar localised meteorological factors can affect a large proportion of the wind capacity in the state near-simultaneously. That clearly puts a premium on gaining a better understanding of what these localised factors are and if possible incorporating them into AEMO’s forecasting models – or perhaps just adapting the very short term forecasting models to better track actual rates of change seen in aggregate wind output, which they currently seem oblivious to.
That said I think that your comparison of the reduction in WF output to a large thermal unit trip – which can happen near instantaneously – could be misinterpreted. The latter has system security implications in terms of the fast-response contingency reserve required on the system, whereas the reduction in WF output happens much much more slowly than that from the perspective of real time operations (the reduction in the half hour up to 13:00 yesterday looks precipitous but was actually only about 20 MW/minute). Of course any extended change like this typically has impacts on spot price, just as do the ramps in demand associated with morning, evening, and (unnecessarily) “off-peak” water heating peaks.
Thanks for the comment, Allan
And thanks for highlighting the confusion I inadvertently might have caused some readers. The comparison with a big coal unit was to provide a comparison of scale – though each event has its own challenges. A trip of that magnitude is unexpected and quick, leading to short-term system security implications as you have noted. The unforecast oscillations (increases and decreases) in wind output is somewhat slower and so represents different challenges in terms of gearing up something(s) dispatchable to wind up and wind back at the same rate.
Well personally I don’t think this macro problem can be solved until all generators are restricted to only tendering power they can reasonably guarantee 24/7 all year round. However we wouldn’t want to start that process from here.
In the interim should wind farms be compelled to install perimeter wind monitoring and advanced warning anemometers to aid short term generation predictability and if so at what sort of distance (radius) would that be optimal for thermal generators to benefit from?
Excellent graphics Paul and an interesting discourse. A couple of comments if I may:
George in the shorter (5 minute) time frame storage can usefully be used to meet enhanced (by technologies such as lidar, skycam and cloud motion vectoring) forecasts. AEMO is exploring this with ARENA.
Climan: yup the best organisations to do short term wind forecasts are the wind farm owners and operators because they know the characteristics of their site best.
Bruce: AEMO produce non-linear generation power curves for every wind farm but due to the effects mentioned by Ciman these are not simple – especially for some farms (refer last comment).
Allan – in this discussion we’ve had wind generation congregated around the Adelaide hills and the mid-north. In reality it’s a mix. We do see pretty fast ramping events (up and then straight back down again) but they’re as fast as asset trips – but often larger. Solar ramping events are hellish fast though. The problem is combining a weather ramping event with another, or a demand ramping event or an asset contingency. These combinations have drastically changed the risk environment for system operators.
Great conversation people – keep it up.
Not as fast as asset trips
Not as fast as asset trips…
A not missing in there…
Yep, this sort of event would certainly give AEMO much bigger headaches and have higher cost implications if SA were not connected to the rest of the NEM.
As things are, I imagine a lot of the ramping capability needed to handle these large wind output swings would be supplied from outside SA over the interconnectors – one of the benefits of interconnection and its diversification of supply/demand.
Even so I would think there is enough ramping capacity and fast-start plant in SA itself to physically handle this sort of event. But as Mike points out, the risk environment for market participants and system operators is heightened because they would need to factor in the chance of this sort of unforecast event in order for enough of that capability to be available.
As I don’t know SA Topo, or places, do the BOM observations concur i.e. http://www.bom.gov.au/sa/observations/saall.shtml
click the place name for complete data, i.e. http://www.bom.gov.au/products/IDS60801/IDS60801.95678.shtml
Can you explain how 7 MW of battery power can stabilize a 2000 mw coal facility 600 miles away?
Hi John,
Are you referring to the discussion around the response of the Hornsdale battery to the unit trip at Loy Yang A?
The battery is contracted to provide Frequency Control Ancillary Services (FCAS) to the NEM. When the Loy YangA unit tripped the battery responded to help maintain system frequency until other FCAS providers could also respond.
Am I on the right track? Can you provide more context to your question?
Carol
Why is the negative price in Tasmania associated with the high prices in the mainland?
I thought the link would cause prices in Tasmania to rise …
What you are doing is essentially applying logical thought and reason to a situation. The way things are done since AEMO stated has never been logical or even according to so called market forces. It is mish-mash undergraduate 101 economics, that is fundamentally flawed. Some things make some sense if you think of the workings of vested interest but not much.
Sorry to spoil a good conspiracy theory, Ken, but there are valid reasons why the prices separate. Too much to go into here, but to do with co-optimisation of 9 commodities (Energy + 8 x FCAS) across 5 regions simultaneously.
Thanks Allen,I,ve only found & been following Watt Clarity for the last 5 weeks or so and commend you on your contribution to enlighten us mer uneducated public on the workings of the national grid system and your unbiased accessments of the vaigrencies with in of its very intricate workings.Please keep your observation coming so some of us can glean some more enlightenment.
Thanks Allen. Great blog.
What conclusions do you draw from the trip at Loy Yang B? Do you think this reflects a broader reliability challenge or just an unfortunate malfunction?
It’s a great question Dean but one for which I haven’t assembled the evidence to judge either way.
What is certain is that large-scale thermal generation units – particularly coal fired ones – are very complicated pieces of machinery with many intricate subsystems. There are plenty of points of failure so it’s no surprise that across the whole coal fleet we do see unit trips and unplanned outages or capacity reductions relatively frequently. And as with (much) simpler machines like cars, it can be harder to keep them running optimally as they age.
But I haven’t seen clear evidence on trends, although I’m sure owners and system operators must be monitoring these closely.
Wouldn’t the reason for the drop in GT capacity be purely the temperature function? As the air gets warmer, you get less mass flow through a unit so the MW go down? For CCGTs, the MW would drop as the circulating water temperature for the steam turbine condenser went up
Hi Paul, great article. I work for Reposit Power and I would like to mention that Reposit takes no cut of the GridCredits earned during a market event like this. We charge retail partners fixed fees so there are no perverse incentives.
The $1/kWh is set by retailers and we are actively working with Retailers to introduce additional benefits. In a trial with one retailer we already offer the customer the full spot price which is how some customers were able to earn up to $37 in a single day just by trading excess solar. We are also looking at other incentives such as reducing the daily access cost to add benefit year-round.
It’s also worth noting in this arrangement that the customer takes no risk – the $1/kWh is paid to the customer regardless how the 30 minute period settles. For example, if the 30 minute period settled at less that $1000/MWh then the customer would still earn their $1/kWh if they were called upon during this time. Due to this risk I imagine we will always see some difference in what the consumer earns in these events vs the spot price.
Reposit is working to empower consumers on the grid and compensate them accordingly. Whilst is not full spot price it’s better than they can currently earn otherwise and it’s better than what they would otherwise buy for. Watch this space – increased incentives will be coming, it just takes some time.
Hey Paul
For the demand revision downwards on Friday, you’ll see that there was a “cool” change in VIC from 2:30pm local. Further AEMO also called ~140MW of RERT which had the impact of reducing scheduled demand (compare the demand_and_nonschedgen and dispatchableload in the intervention/what-if cases – there is no change to scheduled demand).
Just to clarify further for anyone puzzled like I was:
– data published in the “Intervention” case represents how the market was physically dispatched, including the effect of demand reductions in Victoria called by AEMO under its RERT arrangements
– in the “What-if?” case AEMO estimates how the market would have been dispatched without intervention. Crucially, this “what-if?” case sets the market prices.
– you might expect that in the “what-if?” case scheduled demands would be adjusted upwards by the amount of load reduction called under RERT – representing what demand would have been without intervention. In fact the way AEMO deals with this is to leave scheduled demand unchanged and represent the RERT reduction amounts as additional “dispatchable loads” – equivalent to demand from pumped hydro for example. This causes the amount of generation + imports required in the “what-if?” case to increase, just as it would if scheduled demand increased by the same amount.
– this approach also makes the RERT load reduction amounts more explicit in the dispatch data. As HR says, you can see reductions of up to 140-odd MW, averaging 69 MW across the period of intervention.
Hey Allan
Thanks for a fuller explanation.
Not sure if I agree with you about scheduled demand though. There is no what-if scheduled demand – it seems to me that for the duration of intervention AEMO considers the curtailed load of reserve traders is shifted from scheduled demand to dispatchable load – in the what-if case the reserve trader would have continued to consume (and thus there would be a dispatchable load for that period), and the intervention constrains them off.
I never was the greatest at the seemingly infinite definitions of demand… But you’ll see that there is a what-if demand_and_nonsched (which I consider to be the demand required to be served by transmission connected generation).
Btw you owe me for Pony Club 😉
Silly me I think we’re talking about the same thing – I didn’t quite process your comment about AEMO “leav[ing] scheduled demand unchanged” correctly.
Yep we’re in furious agreement about this. Demand_and_nonschedgen includes the demand of dispatchable loads, whereas most AEMO definitions of “demand” exclude the dispatchable load component.
Pony Club?? – You’re confusing me with another Allan O’Neil I think.
Comment moved to
https://wattclarity.com.au//2018/01/if-i-were-going-to-perform-an-objective-review-of-what-happened-last-thursday-18th-and-friday-19th-heres-a-couple-things-i-would-look-into/comment-page-1/#comment-605977
Comment moved to https://wattclarity.com.au//2018/01/if-i-were-going-to-perform-an-objective-review-of-what-happened-last-thursday-18th-and-friday-19th-heres-a-couple-things-i-would-look-into/comment-page-1/#comment-605977
Hi Paul,
Thanks for your articles, I find them all extremely interesting and am amazed how complex the supply and demand issues become.
I am also amazed at the costs involved. Those 2 hot days in SA cost $68M and $60M respectively compared to the usual daily $1.5M to $4M.
This makes the suggested $400M spent on the SA 250MW of diesel generators look cheap.
I can not understand why the generators were not used on 19th.
I estimate if they were used for 7 hours (11-18:00) they would have generated 1.4GWh and saved the state about $4.5M.
Dave
—————
Assumptions:
Cost of power to state from 11-18:00
————————————————–
250MW generators working at 80% (heat impaired) for 11:00-18:00 hours 1400MWh
Average RRP 11:00-18:00: $807000/hour. 7 hours: Total $5.6M.
Obviously the extra 200MWh would reduced the RRP but that is still an actually saving by the state.
Generation Costs:
————————
200MW generator uses 80K litres diesel/hour at $1.75/litre cost for 7 hours: $980K.
Dave,
I think the purpose of the generators is to prevent blackouts, not to reduce prices and they are probably prevented from generating until just before an LOR3 event i.e. load shedding. if they were dispatched before this they would be interfering with the market and would upset the big boys making their killings on the hot days. You could say the heat would cause blackouts which would cause political heat and that is what they are designed to prevent; price is another matter. Sorry for the cynicism.
Hey Paul. Thanks as always. Any chance you could make it shorter – i would love to read this stuff but time is tight? Maybe take as read your frustration with the superficiality of popular commentary, and jump to the gist of your analysis.
Hi Allen,
You are a great communicator yourself – great writings, but I don’t agree with your criticism of the uninformed twitterites. If the information is kept to just the Experts then they can use all the jargon they like if that helps. But if the public is offered the same information in the name of openness then the responsibility for accurate communications rests wholly with the speaker not the listener. If the public don’t appreciate the differences between LOR1 and LOR3 that is the fault of those that use such dreadful jargon. It reminds me of NASA’s jargon with the moon project. They used that jargon for the same reason that scholars once spoke to one another in Latin. It was to impress people not to communicate. If the this industry must speak they way they do then the fault in communications is 100% with industry. Courts are open so that justice is seen to be done. Judges don’t talk to the Jury in long words they speak so they can be understood. This industry if they speak so the public can hear what is being said needs to gets their communications in better shape. I don’t think it adds to the debate to poke fun at how the public does not understand what is being said. The industry needs more folk like you, Allen.
Fair comment Ken, my dig at “instant experts” on twitter could have been better directed at people there – including those who know the subject matter reasonably well – presenting highly partial views and data.
On something like the LOR issue I certainly don’t expect the public to know the jargon or appreciate its nuances but the journalists reporting to the public could generally do a better job of informing themselves IMHO.
AGL says not heat related
“But AGL’s director of operations Doug Jackson said the issues at Loy Yang A were unrelated to the summer heat but were driven by “not uncommon” problems that arose while restarting a unit after a major overhaul to ready it for peak summer demand.”
http://www.afr.com/business/mining/coal/coal-power-producers-defend-record-after-heavy-spending-20180123-h0nbsh
Thanks Rob
… but that was for Loy Yang A. it was Loy Yang B that tripped on Thursday last week – in which case the quote you wanted from the AFR article was:
“Alinta, which earlier this month completed the acquisition of Loy Yang B for about $1.2 billion, said an unexpected trip of one unit on January 18 was rectified and the plant supplied critical energy into the market during the hot spell last weekend.”
Does not explain why the trip, though!
Surely the extra trips of coal plants are due to the fact that the deeper the penetration of intermittents into the NEM…the more coal plants are relegated …by the artificial favoring of intermittents in the merit order for dispatch..coal plants .made to operate outside the baseload mode they’re designed for…as MIT and others have reported.
The extra cycling and ramp-ups damage coal plants and make them less efficient…causing them to need more frequent maintenance…which generators are no doubt reluctant to pay for with their plants on death row anyway.
AEMC has also said that intermittents damage the grid..increasingly as they penetrate further into the market.
The wind and solar enthusiasts are deliberately trying to make Australians believe that all these years COAL has been an unreliable power source when that’s not true.
They MUST KNOW that the system that gives an artificial advantage to intermittents causes coal plants and the grid to be damaged .
Why are they allowed to deliberately deceive the Australian people?
So much deceit on the RE side!
I think its just an unfortunate truth.
The unknown aspect is how it would perform in real life depending on gross variations in how the stored energy is used.
Paul,
Just curious … Is this forecast is using the 50POE curve from AEMO? Just looking at the numerical weather forecasts for reference weather stations, I think we might be heading for a 10POE day on Wednesday. If – and that is a big if – the weather forecasts holds and the forecast thunderstorms stay away, we might start nudging the 10,000 figure. (AMEO operational demand forecast for Wednesday is 9954 @ 17:00hrs on 14/02).
Thoughts???
There are some scary-impressive numbers being generated by AEMO’s POE10 forecast for tomorrow (Wed. 14 Feb). There is a value of 10,115MW @ 17:00 and 10043MW @ 16:30. In the Energex DAPR, they provide a 50 POE reference temperate at Amberley of 30.65C ([max+min]/2).
The reason I keep harping on the 10 POE is that private numerical meteorological forecasts for Wednesday has a forecast average temperate at Amberley of 31C.
A comment out of the Energex 2016 DAPR is interesting. “Importantly, while air-conditioning load growth is beginning to flatten, the sensitivity of the significant >>latent<< (my emphasis) air-conditioning load already on the network remains a challenge to manage.” Out of the 2017 DAPR this comment “Air-conditioner sales have been higher this summer indicating customers are more likely to update or add to the air-conditioning load in their homes in response to the hotter summer.”
And that is the key to what we are seeing. There was a gap in hot summers from 2010/11 to 2016/17 that coincided with the classic peak-demand time of three hot days in a row, in the middle of the working week, in February. The heat has returned with vengeance, and so too has the load.
Perhaps those customers who did run their air conditioners last summer were pleasantly surprised at their electricity bill, or have become re-addicted to the comfort of air conditioning.
Evidence suggests that customers have a “break point” with the heat. When it gets too hot, they will say “stuff the cost” and turn on their units.
Further, the more affluent who also have both solar PV and A/C, also appear to have a new behaviour to run their A/C hard, offsetting late afternoon and evening kWh consumption against export income during the day.
The peak demand is definitely an interesting challenge to manage. If the weather forecast holds, Wednesday will be interesting to watch as well.
Thanks Anthony
We’ll be watching what happens on Wednesday.
Paul
Looks like a new system record, but no 10POE cigar. Sea breeze front pushed in as far as Amberley (33kph ENE at 16:30).
I find it very interesting to read remarks about “managing peak demand”. Remarks like this reminds me of Soviet type think “managing” meaning “controlling”. Supply and consumption of electricity since Edison’s time has been on the basis that the consumer sets the load, rather like a customer enters a shop because THEY decide – not the shop. I think we need to stop talking about managing. Think about the structural steel in a building to resist a storm-wind that might blow at the design speed for 10 minutes every 100 years. At all other times there is a capital investment in steel that provides no tangible return on investment. Those involved in power designs need to get over worrying about peak to average ratios. Many industries have capital investment that is not needed for most of the time. Just spend the money !!!
Ken,
Managing peak demand has been around in the Queensland electricity industry for a very long time. The Brisbane City Council Department of Electricity introduced hot water load control using the Zellwegger system at Holland Park substation in 1957. Yes, over 60 years ago. (I believe the Sydney County Council introduced the same system a few years earlier). At present, there is approximately 1000MW of customer’s load in Queensland connected to Zellwegger demand management schemes. Over the decades, using the scheme has saved system operators from having to do load shedding during generation deficiencies. AEMO’s forbear (NEMMCO) once had the hot water systems in Queensland “instructed off” for an extended period during a significant generation shortfall.
Back in the 80’s, I did the field install for the under-frequency fast hot water shed. (A system that is still in operation today.) I doubt a single person notices when their hot water gets load shed. But everyone certainly does when the lights go out. Ironically, because this demand management system has worked – that is, it has prevented load shedding for standard lights & power – the concept does not attract media attention, and people assume that it does not exist.
I see little difference between centralised demand management and other flexible purchasing arrangements. For example, if I want to purchase a Qantas ticket from Brisbane to Sydney tomorrow. I can purchase a ticket for a 5AM flight out of Brisbane for $177, or I can purchase a 6:30AM flight for $490. If I can management my demand, I can save 64% of the price. Ditto cheap movie tickets on Tuesdays, and so forth. This is free market economics in action, straight out of the textbook for first year Uni students. There are options available for both buyers and sells, and both adjust their behaviours (ticketing prices and willingness to purchase) based on supply and demand. It does not bear even a passing resemblance to centralised control “Soviet” economies.
The thing that I find damn irritating is the collective social amnesia – most annoyingly from senior leaders inside the electricity industry – that parts of Australian electricity grid has a world class demand management system that works. Instead, the industry is accused of being behind the times and that we should do demand management by TCP/IP, or something like that …
Looks like QLD schedule demand peaked at a new record of 9928 MW today.
But returning to Feb 12, it’s worth noting that while scheduled demand peaked at 18:50 (9550MW), once you include the effect of rooftop PV, you can see demand (scheduled + rooftop PV) actually peaked at approx 9835 MW a couple of hours earlier around 17:00. So rooftop PV reduced the peak by about 300 MW, and moved it back a couple of hours.
Correction, looking at your data it appears that peak demand on Feb 12 (scheduled + rooftop PV) was around 9868 MW at 16:55 on Feb 12.
Well if you keep adding quarter of a million new consumers into Australia each year and you don’t build anymore base load coal fired power stations there’s a certain inevitability about all this. Well given the low hanging fruit of power efficiency and economising has largely been plucked due to hefty price rises.
… just wait till everyone wants to recharge their electric cars in the evening.
If all Australian cars were converted to EVs they would need an additional appr. 3,000 MW for recharging
Fear Not!!!!. South Australia is showing what a Battery can do to keep things in check.
Yes a 100MW one, that does not produce power obviously but stores it. It actually takes like 10% more power to fill it and then it also has 10% less than finally got into it come out. Thats if its not hot, if its hot, well then it depends on how hot, as to how much can actually come out!!!!!.
Anyway, if you start to think about a battery being used in a peak period of high demand for a long time, its, well actually useless.
In fact the whole setup will actually need power, so its kind of good most the time but in peak times it adds to the demand. Bit of a over sight. However in the real world thats reality.
there is plenty of capacity between 8pm and 6am for EV charging. I have been charging my EV by timer each evening sometime after 8pm for 5.5 years. It’s trivial load shifting and works for everyone as long as their car is topped up by when they need it the next day.
Trivial, until all of brisbane wants to charge their EV at the same time.
Then I doubt the wires that have been run and the transformers that are in place will be enough.
But Thats for the future.
Got to get there first.
Maybe users of EVs can assign control over the timing of their charging to their retailer, the retailer can stagger the charging of EVs through the night so that they don’t get every single EV in Brisbane being switched on at 8pm.
I daresay most people won’t need a full charge every single night.
Fair comment, but there is a LOOOOONG way to go, as we also need to look at heavy EV’s too.
Meaning trucks and buses.
They will need to also be charged.
To replace fossil fuels is going to be a tough gig.
Mind you, Tesla was a leader in wireless energy transfer, so maybe we will not need to charge anything ever, just get charged as we drive.
thank you Russell
I agree with your assessment
I prefer knowledgeable and experienced engineers to run the electric grid
Curious that Russell is able to come to a conclusion as to what is the most “cost effective” option for emissions reduction and dispatchable power without mention of the costs involved in the solutions he proposes and no comparison of costs of the alternatives.
Thank you Russell for an interesting and thought provoking article. Your exploration of the possibilities for improving the performance of the existing coal fleet is great. Don’t we need to estimate the cost of these improvements and compare to other options before concluding that it would be the most cost-effective, dispatchable, low-carbon solution? Internationally new gas fired plants are the common dispatchable companions to high penetrations of intermittant renewables. Gas emissions are around half that of even a finely tuned coal plant but still not low enough to meet CO2 targets in time. Going even further, a genIII or genIV nuclear plant would cost more but produce zero emissions and greater output. New gas or nuclear plant would have service life far in excess of reworked coal plant. An analysis of total lifetime cost, emissions reduction, dispatchability and other key parameters would determine the most cost-effective way forward. Investing large sums in coal plant mods Australia-wide could be akin to rearranging the deck chairs on the Titanic both economically and environmentally.
Or we could burn wood in coal plants, like germany is currently doing.
Or as South Australia has just done, buy big diesel gensets!!!!!!!!(Yes I know they also got the biggest battery, but really, does it actually do anything?, except use power)
I seriously have to laugh, better weather forecasting and wind and solar only have small units go off line.
No they have both, small units going off line and the whole plant due to weather and problems.
Controlling wind and solar, compared to base load power is like adding another two numbers to power ball!!!!!!!! (Meaning its a lot more complex)
Interesting article, but it seems to assume that all power generation is equal. As long as we can balance generation with demand, then all will be well.
Hugh doesn’t take into account frequency stability. A key factor in the SA system black event was that SA wind generation couldn’t stay in sync with VIC, once SA lost a large amount of wind generation.
The Heywood interconnector tripped because the frequency of SA generators fell below that of VIC, not because VIC couldn’t supply the load.
What followed was a rapid sequence of under frequency load shedding events, as transmission systems tripped off the remaining renewable generation that was unable to remain synchronised.
His conclusion that “the value to the system of large, inflexible coal-fired generators will shrink” seems to miss this point.
There have been some interesting articles on WattClarity about how AEMO now manages the situation when the total SA generation becomes mostly wind. Following the system black event, AEMO can now direct gas generators to come on line, so there is a minimum of synchronous generation to ensure reliability (for example: 2 x Pelican Point generators, or 1 Torrens Island and 1 x Pelican Point generator), even though there is enough wind generation to supply the load.
The value of large thermal generators is that they stay in sync, even when subject to a large changes in load, unlike wind generators.
A great article all the same! I enjoyed reading it.
Thanks Hugh for a nicely written and factual article. I wonder why you omitted the inconvenient truth behind the system black event in SA. In this case AEMO expected a strong contribution from wind turbines, but they shut down unexpectedly (tripped) because of multiple short circuits to ground (aka lightning).
The operators of traditional generators and distribution systems have a social obligation to increase output, often under duress, to supply their customers loads. On the other hand operators of variable renewable energy generators are only obliged to maximise the revenue available to their owners and place protection of their assets ahead of supplying the load. Consequently SA is now a circus ring for spruikers like Elon Musk.
I have great difficulty getting my head around much of the commentary on the electricity system in Australia. I could follow much of the sentiment if it was 1925 and the grid and all the vagaries associated with the “new” electricity were not yet understood. But the system is now a very mature discipline in 2018. Mention was made about the high temperatures for example, with the implications that there were no materials capable of working at 50 degrees C or above. We (humans) might find 50 degrees hot and we do turn on our air conditioners more. But electrical systems which over heat under these conditions are themselves wrongly designed or maintained. Class H insulation work well above class A insulation of cotton-covered-wire. It is not 1925 !
The public can reasonably expect to have electricity for a good deal of the year. They can also reasonable expect to be the sole determinant of the load – just as a customers chooses to go in to a shop to buy something – solely when the customer decides – not the shop.
As the country transitions into a period when more and more electricity is provide from renewable sources they also a right to expect that the transitions will be done by professional engineers who typically get it right first time and hopefully on budget. Sadly this is not what is happening. We have policy written by people who are self styled experts who often have no idea of what they are doing. We have commentary offering derision of the public that are asking for nothing more than it works. Of course when it falls short of reasonable expectations the public offer their advice which often is wrong. But the public are not the cause of the problem because they offer the wrong advice – it is the silly people that have got into running the show who are behaving like what happens when the asylum inmates take over the asylum that are the problem not the public.
Paul C writes that when SA was blacked out that the SA frequency fell below that of Vic. This is not correct. The Haywood inter-connector is not a DC link – it is synchronous connection albeit a weak connection. The two grids must at all times share exactly the same frequency. It is true that in times of instability that there will develop an unacceptable phase difference between SA and the rest of Australia but the frequency remain the same. If frequency is defined the reciprocal of a single period then may be the problem could be seen as a frequency difference but that is not the normal meaning of frequency. The same thoughts apply when thinking about frequency modulation and phase modulation. Over a very short period of sampling then it is hard to tell the difference between the two but they are not the same. The wind turbines took themselves off line because someone had written software that asked for this to happen. But the owners of the turbines and the AEMO had no idea this was the case. After multiple lightening strikes like what happens on most days in Far Nporth Queensland in summer the wind turbines threw the towel in. The system was never deigned to loose so much capacity so quickly and that triggered a text-book cascade fault condition, which involved the Haywood inter connector also tripping. After the inter-connector tripped with the systems now separated the death-throws of the SA grid obviously involved frequency collapse but not until after the separation.
It makes perfect sense that whatever power generation is currently in place at a certain location should continue, unless there is a VERY good reason to change. The fuel supply chain will already be there, as will the skilled workforce, and the power distribution infrastructure.
Why is common sense no longer in fashion?
Paul, re blown LEDs, could I suggest that you check your mains voltage. I my case the mains voltage was sitting at 253 volts on winter days – but it showed a weak correlation with PV output. I wrote to my electricity distributor with a spread sheet of 5 minute average voltage readings during a week, and to my surprise they responded immediately. I was also surprised when they said they had to setup their own monitoring (I expected that in suburban Melbourne it might already have been monitored). Within a few days of seeing with their own eyes they changed a distribution transformer tap. The voltage is now right on specification and with very little variation over 24 hours (except for the notable air conditioners causing ‘fuses’ to ‘blow’.
Thanks for your contributions.
A suggestion: try looking at cumulated data. A first cut would be simply to subtract the overall average from the daily production, and cumulate the daily differences. The difference between the highest and lowest points on the chart show how much storage is required if you assume that demand is constant and you are depending only on solar+storage. The next step in sophistication is to subtract instead daily demand, scaling solar to the same annual total, and once again cumulate the differences. You can then look at various percentages of overbuild. You should also account for the round trip loss in any energy stored – probably 25% is a sensible starting point. Of course, you need to add in the storage required to meet overnight demand on the biggest demand night of the year – that has to be in store at dusk.
I think you may find that the storage equation starts looking more formidable – particularly if you look to cover worst cases. The real killer is that seasonal storage only gets used once a year – not 365 times as in storage to cover overnight demand – which makes it 365 times as expensive. Worse still, if you want to cover a 1 in 20 bad year, the added storage you need to do that only gets employed 1 year in 20 on average – which makes its economics another 20 times worse. You can reduce the storage requirement by also including either other renewables such as wind, or varying amounts of dispatchable generation. You may be surprised how much you have to add back to deal with those pesky worst cases.
An interesting article. I think our national electricity market and renewable energy certificates go close to getting the mix right. I’d like to see the target higher but we should also consider efficiency improvements at existing generators, both fossil fueled and renewable. Continual improvement can only be a good thing.
Right exactly. That”s why it”s *NOT* Internet of Things. It”s just Connected Things On A Private Network. We used to do that with leased lines and AX.25. Big wup. Thanks for re-inventing SCADA. You are still tied to a single wireless technology, to a single MVNO, to using SIM cards, and your devices only talk to the cloud, not to each other.
Is this poor frequency control a reason why we get such poor performance from our inter-connectors ? For example, between NSW and Vic only about a quarter of the line rating is made available to the market, whereas between SA and Vic about half the line rating is made available. The market constraints are contained in equations that are derived by regression from records of frequency. I expect very few market participants understand what these constraints are. The business case for the SA/Vic upgrade was for it to release 650 MW of capacity in both directions, but this has still not been delivered to the market. How many other transmission upgrades might we make on flawed business cases, when the least-cost path to increased inter-connector capacity is simply to set the governors properly ?
To a small extent only-
I think you are referring to stability limits that would apply even if governors were well tuned.
The lack of frequency control does impact a bit on inter-regional constraints but these may not be fully taken into account in the existing constraint equations, which puts the system at risk.
part of the justification for interconnectors is the ability to share reserves between states thereby making the system more robust and cheaper to operate. The interconnector constraints are more related to contingency reserve sharing where a contingency in a state can be managed by sending additional generation from interstate hence interconnector capacity for use in the energy market is set at a level below the maximum interconnector capacity. This would not be the case if the interconnector was set at maximum capacity for the energy market. The interconnectors are basically working as per the business case.
Ron R
I agree, however if we have sloppy frequency control that also manifests itself as sloppy control of interconnector power flows. This has a significant negative economic impact on the delivery of power.
Great article Bruce. As a long term proponent on renewable energy and the benefits it can and will bring to the Grid I find it amazing that here we are in 2018, with all the technology advances that allow us blast a tank to nothingness using remote controlled drones, we still have major issues with power and the grid. At “watt” point will all these competing disinterests come together and devise a sustainable solution…that provides security at competitive prices.
This is one of the main issues holding back the introduction of renewable generation. If we had a more rational approach to frequency control I am confident we could successfully negotiate the technical obstacles currently holding back the industry.
Great article Paul unfortunately some of these companies have been disrespecting customers for some time.
It is amazing to see some of the assumptions made in their marketing. This is also the case for large commercial solar so companies are given incorrect information to make the business case look more attractive.
Try http://www.smappee.com for a real power monitor. Hasn’t missed a beat for me in 4 years. Class 1 energy meter that actually measures the voltage (unlike most).
Unfortunately energy pricing is complex and only understood by a relatively small proportion of the population. If an advertisement promises $10,000 per year return it is easy to see the attraction, and harder to understand that it may not be realistic.
Hey Paul, you are turning into a grumpy old man. Of course its complex, in electricity more so than in others. But the incentive to mislead is huge. Here is an idea, why don’t you keep a register of the most egregious claims. I am sure regulators and others would be willing to pay for that. Cheers, Bruce.
Absolutely brilliant analysis.
In terms of savings of greenhouse gasses emissions it is correct to discuss the total quantity of energy generated by the different methods. It is also enlightening to look at the coincidence of generation with load. But discussion like this always chooses to ignores what happens when there is a long period when the renewable energy production falls short of demand. The larger the amount of energy produced from renewable sources the less often alternative generation will be called into service, but what seems to always get ignored is to note it will be called on every so often and it is essential to retain this backup capacity or even to deploy new backup capacity when the old generators become US. What needs much more discussion are the economics of this infrequently called upon backup standby generation plant. Plant with a very low duty cycle is always expensive to operate. It is wrong to assign these high costs of non-renewable production to the bad old guys. Backup capacity is an essential component of renewable generation and as such the high costs of the low duty cycle production needs to be added to the renewable cost and not sidelined as the cost of old legacy plant – out of sight out of mind.
I guess that Solar thermal with storage will replace most of the coal generators over the next 10 years – solving the economic problem of uncompetitive coal and the need for legacy plant backup capacity.
Justin, if coal is noncompetitive into what market are you referring? In a market without RET then coal is the best solution. I refer to the collapse of the solar industry in Germany and the falling % of renewables being used there.
So lets delete the subsidies immediately and see how this market goes.
BTW Solar and Wind technology was developed in the late 1800’s along with the first coal fired generators. They were dismissed due to intermittency.
They are all old technology, I am amazed that we cannot do better than this after more than 100 years!
thanks Alan. I am not sure I understand your point here. So thermal generation on AGC responds. Yes, we know that. Surely the interesting here, as pointed out in Renew is the relatively superior responsiveness of batteries. The fact that the volume of their response is small misses the point.
Bruce, in line with WattClarity’s theme of “making Australia’s energy market understandable”, the purpose of the post is primarily to present a clearer overall picture of how the system responds to a trip, especially for readers not as familiar as you might be with how the electricity grid operates and the characteristics and behaviour of different types of generator. I don’t think anyone could say that the original Reneweconomy article had that aim in mind!
In this particular incident, as Andy points out below, the response of the battery is neither remarkable (a battery can quickly switch on from zero output – who knew?!) nor critical, and of course nor is the response of any other individual generator. There are definitely all sorts of valuable characteristics and demonstration effects that HPR and other new technologies bring to the energy transition – but this event doesn’t really illustrate them. In fact it’s my understanding that the battery’s capabilities have had to be “hobbled” somewhat to fit in with operating in an old-style predominantly thermal system.
Allan
(PS Update 7/4 – see the just released report by AEMO on initial operating experience of HPR at http://www.aemo.com.au/Media-Centre/AEMO-Hornsdale-report – gives a much fuller overview of the battery’s capabilities and services)
Thanks Allan, perfectly reasonable points. Thank you.
Why dont we just have a heap of batterys as the back up. Money is not of concern, because we have much cheaper power with renewables and its also saving on co2.
Yes we might need a lot more capacity for batterys because so much is wasted getting it in and out of them. Lets also not forget they add to demand because they do not make any power so need power for fans and lights and the compound they are in and the control room.
So just to have a battery you need more power.
But they are fantastic because they save co2.
Germany and South Australia are proof it all works.
The fact that several generators started to increase load before NEM frequency decayed thru 50Hz implies their loading instructions were being received via AGC as participants in FCAS (AGC usually also receives digital signals from stations such as trips to pre-empt frequency decay). The rate of these load increase will most likely be driven by the control algorithms in the AGC rather than the governor actions of the active units. Little natural governor action appears to be present. (And many generators in the NEM not participating in FCAS market will actually have their governors switched off) Hornsdale response seems to have commenced after frequency fell below 50 Hz and looks to be proportional to the (0.2Hz-deadband)/2 x Capacity – about 9MW based on 4% governor droop. All looks about right. Not sure there is a story here for the battery guys (or anyone else really). Very good, practical analysis however Allan.
Great writeup Alan, your stacked bar chart of unit responses is excellent. Have to point out something this explainer overlooked though: there’s two ways to rebalance supply and demand, and recover to nominal system frequency following a generation contingency: you can increase generation, or reduce demand (or both). Indeed, in this trip, both types of responses would have occurred. In the dispatch interval where this trip occurred NEMDE had enabled 19/50/58 MW of R6/R60/R5 FCAS from demand side loads (across three dispatchable units)… so the generation responses in your chart are only part of the story. (Of course, you’re forgiven for being unable to represent the demand side contribution in your chart, as the data isn’t immediately published in AEMO’s 4s SCADA logs)
Really good point Matt – I should have mentioned this, along with the intrinsic “load relief” that accompanies a frequency dip. I did actually check APD (Portland Smelter) which does show up in the 4s data and was enabled for 17/11/11 contingency raise, but its response was minimal. As you say, the other demand side DUIDs enabled in that interval don’t appear in the 4s data set.
Really interesting and useful analysis Tristan. A couple of thoughts:
1) Until there is a lot more interconnector capacity, regional demand peaks are also important reliability drivers, so perhaps looking at some regional peak days would be instructive.
2) In the medium term reliability may become less a function of demand peaks and more a function of “supply lulls” – eg extended periods of low wind / sunshine. In other words reliability is driven both by the energy and capacity characteristics of the “residual” demand to be served by storage and other forms of generation (I don’t think demand shifting addresses the extended lull issue). I know the Blakers / ANU study dealt with this, but it made some pretty heroic assumptions about diversity of RE and supporting grid investment, as well as relying on demand reduction for the last piece of reliability.
Allan
This graph gives a different perspective by including black as well as brown coal.
https://www.aer.gov.au/wholesale-markets/wholesale-statistics/generation-capacity-and-output-by-fuel-source
Wow that graph puts the whole article’s message in a different perspective.
It’s unfortunate that this article was put together without black coal on the graphs, to the innocent viewer it would appear that renewables generation were higher than coal, when coal generation by far exceeded renewable generation. Even though the author mentions that black coal is left out, it still appears to me that the article is still intended to create a positive bias on renewables.
I knew this was going to be the case, thanks for the graph. There is so much BS in this market when it is a critical part of Australia’s economic survival. We need a brand new approach to cheap and reliable electricity.
Don’t have an answer but the problem is there for some smart person to come up with something. Nuclear will be buried by red tape and coal needs a rethink at best. Wind and Solar offer little for the future growing demands without subsidies.
Another point worthwhile doing a post on is that some of the new solar farms will be producing at near-maximal rates until nearly the end of the summer evening peak in summer. Data from the Moree solar farm, which has tracking, shows that it in summer it produces at 80% or more of its rating until 18:00 NEM time (except when affected by cloud). Because of being further west, a new solar farm with tracking at Balranald will produce to 80% of its rating until 18:24 NEM time, while at Port Augusta until 18:48 NEM time. These times are equivalent to 19:24 and 19:48 Eastern Daylight Saving Time, and are about the end of the evening peak. Spot prices during these periods would tend to be set by open cycle gas, unless undercut by hydro. There would also be pass-through of Eastern States peak price timings into South Australia, except on the rare occasions when export from SA to Vic during the evening peak exceeds inter-connector limits.
This has two implications. Firstly, the solar farms will be getting peak prices for about a quarter of their output (~3pm to 6pm NEM time), which would improve the economic viability of solar farms with tracking in the western part of the NEM. Secondly, the amount of dispatchable backup for these solar farms would be much less, because of reliable production during the evening peak.
The world’s best engineers and technologists have an inborn aptitude for the disciplines in which they have studied and worked.
Until such people are given their due respect and tasked with planning the future for Australia’s electricity supply, we are going to slip further into the fiasco now unfolding.
It’s also clear from the many comments posted to its website, that ignorance and naivete are prerequisites for believing Reneweconomy’s propagandist messaging.
With the design and development of the nation’s electricity supply being taken over by politicians and activists, is it any wonder that the horrendously expensive and fragile grid that has resulted comes about?
It appears that collapse and chaos is the only way to bring the nation to its collective senses.
That would be great Rob if some of the ultra conservative energy professionals of yesterdays-year hadn’t been so resistant to the need to replace fossil fuel generation with renewables and storage. Like it or not means big changes in the operation of the grid.
Fortunately the tide seems to be turning and even the older engineers resistant for a great time to renewables are finally seeing that back to the future is not gonna fly, New solutions to old problems and new problems will need to be found. That’s surely what engineering is about, sticking with what they already know (i.e. fossil generation only) is not engineering so much as a basic fear of change, and it’s no longer an option for engineers, politicians or finance people.
It’s a great shame it took a rag-tag crew of activists to smoke out the blockers within the gentaliers, network owners and operators and regulators. I work with a brilliant group of engineers, coders and finance people who model what energy grids of the future may look like and assess where the risks and opportunities lie. It’s great when engineers can solve problems in this way and show creative initiative in contradistinction to modelling ostriches with heads in the Nullarbor sand.
Yet even with these great solutions, engineers, modellers and financial people are still needing to play activist to get the recognition and reform at the level of government policy and mechanisms put in place to take advantage of the now cheaper renewable generation and storage possibilities for better economics. If RenewEconomy becomes cheer squad at times, you need to recognise the mammoth task in converting old thinking about energy in not just the energy industry but also finance, heavy industry, resources and politics.
Thanks Allan for the great analysis. Very clear. I echo with you on the problem with the time tags, which makes difficult to see how much increased power of generators is from the governor control and how much is from 6sec and 60sec services. The 6 sec and 60 sec services should kick in only when the frequency deteriorates to a level below 49.85Hz. Is that right? Although many generators choose to quit their governor control because of wear-and-tear cost and not being properly awarded in the market, I believe there are still some royal generators fulfill their responsibilities.
Thanks for the feedback Jin. You’re correct that in most cases the 6s and 60s raise services would not kick in until frequency drops to below 49.85 Hz, although I think in the case of providers with Switching Controllers (smelter potlines might be an example), there might be lower frequency setpoints before raise services are triggered.
While I didn’t go into this in the article, looking at which generators were enabled for fast and slow raise services vs the responses actually seen was interesting – there were some significant responses from units not enabled for 6s/60s raise – HPR fell into this category – while other units enabled for 6s/60s raise appeared to provide smaller amounts than expected, based on their enablement level. However this last is difficult to be certain of from the 4s data, and accurate verification of contingency responses requires higher resolution data as set out in the MASS.
Thanks Allan! Great comments!
Allan,
Great article.
http://www.abc.net.au/news/2018-04-06/tesla-battery-outperforms-coal-and-gas/9625726
Our ABC giving their un-bias view as usual – I don’t know why they get away with it unchallenged.
I’m not sure of your point there Dave.
The ABC article was based on a report just released by AEMO (http://www.aemo.com.au/Media-Centre/AEMO-Hornsdale-report). On a first skim I think the ABC article is a pretty fair summary of that report.
As the AEMO paper makes clear, the battery does provide very fast and high quality response for both regulation and contingency FCAS, and in very different circumstances from the LYA3 trip referred to in my post, that fast response (harnessed via the newly developed SIPS scheme, see p4 of AEMO’s report) could be very important in maintaining South Australia’s connection to the rest of the NEM and to keeping the lights on in that state.
The point of my post was not to knock the battery at all, simply to explain in full context what happens when there’s a large generation trip in the NEM, and to correct the possibly misleading impressions people might have gained from the original Reneweconcomy article.
A huge problem in the energy ‘debate’ at the moment is people’s tendencies to take sides and hold to their positions regardless of facts – something applying across the spectrum of what Paul has termed the Emotion-O-Meter (https://wattclarity.com.au//2016/07/the-emotion-o-meter/ and covered in many subsequent posts).
Allan
Thanks Allan. I fully agree with you that people should not take sides when looking at the FCAS. Although battery storages could kick in very quickly, but it is unfair to say the conventional generators are too slow to respond. Although governor control is slower than the battery, but think about the big inertia that kicks in immediately after the big disturbance. The kinetic energy stored in the conventional generators go nowhere, but convert to the electric energy trying to fill in the gap. Compared to the battery contribution to capture the frequency drop, the contribution of conventional generators should also be applauded. Saying they are slow is a little bit unfair to them.
Fantastic analysis Allan.
That second graph with the same scale really shows how much the battery has been overrated in the media. In the first graph, it appears that the battery ‘prevented a blackout’ by meeting the shortfall when in fact it was only a small part of many FCAS responders. The capability of the battery to respond so fast to stabilise the frequency should not be ignored however.
This shows the bias towards renewables on the RenewEconomy website, they disregard selected facts to make renewables look good.
Interesting analysis, but I disagree with the implication that maximum demand coincides with maximum rooftop solar generation. This is factually incorrect, with demand increasing around 6pm when rooftop solar is well down from maximum output.
Rooftop solar actually exacerbates the evening peak i.e. the baseload generator ramp up curve is steeper because it starts from a lower value.
Large scale solar with tracking is a different story (as mentioned by Malcolm M), and I predict that solar tracking can also be utilised in reverse to reduce solar output if required for grid stability, to allow room for synchronous generation.
INteresting post Ben and will have a look at this.
Very interesting. I live in Adelaide. I fitted many years ago an over voltage and under voltage relay to my a/c. It has never tripped with over voltage set to 265V in 25 years. A few hours after that event at night with no neighbourhood PV feed the relay tripped. I measured the voltage with my multi meter and it was 268 V and stable. I went back to be bed and can add more more info. More than just Hz went feral that night.
Thanks Paul. This is all excellent but even exceeds my nerd quotient. I do generally find the water analogy helpful. But one nit picky thing is that Image #2 refers to “electrons delivered down the pipe”. It is an interesting fact that in a 50 Hz AC circuit electrons don’t actually travel very far. Energy is transferred at the speed of light despite that fact that individual electrons only jiggle back and forth a short distance along the conductor.
In G) the first sentence isn’t finished.
Three paras up there is another unfinished sentence “We just need to keep in mind that Scheduled Demand is”
PS this article says “The truth is way more spectacular: the energy doesn’t travel through the wires at all — it shoots through the space around them, at the speed of light.”
http://www.abc.net.au/science/articles/2014/02/05/3937083.htm
Not sure I can get my head around that!
Thanks for your time in putting together this insightful article.
I would love to know more about how the different States are using electricity, for example, why does QLD sometimes use more electricity than NSW, when they have a much lower level of population? Is it due to Aircon usage, or industry?
Thanks.
The NEM has high business electricity use, as noted here:
https://wattclarity.com.au//2017/08/in-the-nem-at-least-business-consumption-of-electricity-dwarfs-residential-consumption-of-electricity/
The ratio is higher in QLD than in other parts.
Thanks Paul, very useful article in explaining some of the more abstract concepts.
Just picked up on one thing; in section B, the units of the last two lines below should be watt and megawatt?
3,600MJ/h
= 3,600,000,000 J/h
= 1,000,000w
= 1MW
Thanks Pei
1J/s = 1W (that’s the definition of a Watt) . I have added 2 rows in the table to clarify.
Paul
Let the graph speak for me-
http://anero.id/energy/wind-energy/2018/april
Indeed, as stated above, “We have many dashboards … “.
One bit of transparency we no longer have is the comprehensive Hydro Tasmania weekly report on dam levels. This report showed the installed generation capacity and whether dams were rising or falling.
With Basslink still out of action and Tasmania totally reliant on its own resources and unavailable to assist with load balancing on the mainland, this report would be making very interesting reading.
Why was this regular report “disappeared”?
Thanks Rob (and for explaining more when you called yesterday).
For our other readers:
1) it’s Hydro Tasmania’s spreadsheet (which used to be on their website) which you’re missing
2) it seems to have been removed when the website was upgraded
3) to understand why, we’d need to hear from Hydro Tasmania.
Paul
ARENA has done a wonderful job with its seed capital in establishing the RE sector.
Proponents of new projects in that sector now need much better meteorological tools to assist in planning the location of their wind and solar farms and managing the output.
They also need storage so that they don’t have to spill energy because of constrictions in the network. Even better, they need much cheaper connections to the main grid from the locations where the sun shines and the wind blows (all pretty obvious).
Perhaps a step by ARENA might be to stimulate research into much cheaper methods of construction of spurs from the main grid to these locations.
great post paul, thanks
1. Storage will need to become an increasing part of the energy scene. Windy periods in southern Australia are separated by lulls of about 3 days, so there needs to be enough storage to cover these periods.
2. Solar would be partially counter-cyclical to wind. Solar definitely needs storage, and this storage will be helpful more wind into the energy mix.
3. While the wind profile in various parts of Australia may be decoupled, showing no correlation, there are probably few negative correlations, which would indicate complementary wind patterns.
4. Bureau of Meteorology wind data would be of limited usefulness in conducting studies of wind complementarity, because the measurements are at 10m, whereas the newer wind farms are harvesting some energy at 200m. The relationship between wind speed at different heights is a “bent stick”, with the lower height becoming becalmed when there is still significant wind higher up.
5. Since most new wind farms will have a height of 200m (hub + blade), wind output data from older wind farms will be of limited value in conducting simulations of a grid with a higher renewable component. Wind farms with the greatest relevance are Kiata and Sapphire, which are both nearly 200m high, and I believe are attaining capacity factors of close to 50%, which should reduce the periods when the wind farm is becalmed.
6. New wind farms under development with the higher hub height include Silverton (NSW), Coopers Gap (Qld), Lincoln Gap (SA) and Stockyard Hill (Vic).
7. Professor Blakers of ANU recommends storage close to major load centres. Yet the largest storage development underway is Snowy 2, a long way from load and other generation. The Murray power station on the Vic system is close to the feed-in point for Snowy 2, and has a marginal loss factor of 0.92. So round trip transmission losses for Snowy 2 could be 8% + 8% = 16%, in addition to the round trip losses of about 30% within the pumped storage system when operated at its full 2000MW (but lower losses at partial capacity). There would need to be some substantial transmission investments to provide for both the capacity of Snowy 2, and reduce the round trip transmission losses.
Thanks Malcolm for highlighting (clearly) a number of points worthy of further consideration.
“Storage will need to become an increasing part of the energy scene.”
Blind Freddy can see that but at what cost? SA now joins the elite having the highest power prices in the world and hasn’t paid for the storage side of the equation yet.
Who is going to break the bad news consumers aint seen nothing yet with their power bills? The obvious reaction will be- Whatever happened to all the cheap power we were implicitly promised from Mother Nature’s limitless free bounty?
Storage for 3days Wowa!!!!!!! I am not sure many people understand these two facts about the world saving storage.
1. They add to the overall demand.
2. They produce nothing but to put energy it into them and then get it out of them the process wastes a variable amount which is depented on a number of factors. But allow 10% minimum, that’s bare minimum like all the stars have aligned. The worst thing is on hot days when the demand is its highest the storage is worst. Not only will nothing be going into them or coming out, the storage project needs power!!!!! So adds to demand.
I am sure however all the pro RE people will have a different view, but unfortunately that’s the reality.
Good article, I would have liked you to diversify the SA sample by using Hornsdale and one of the EP WF’s instead of the three Lake Bonney’s.
In this particular case, Josh, it would not have made any difference as everything on the mainland was down at very low output.
History is history but the additional interesting fact concerning the remaining 32 year old Muja C/D (854MW) and 17 year old Collie (340MW) is although these units are nowhere close in age to Liddell and Muja A/B (52 years as indicated in the article), they are now in the crosshairs of the looming “duck curve”. As consumers desperate to avoid increasing power cost are installing solar PV at unprecedented rates, midday load has fallen off to the extent that on many weekends, the minimum load of the day is now 12-12.30pm and not as historically, 3-3.30am. The point is rooftop PV is robbing these baseload clunkers of their needed demand and within 3 years, the ramping load on the remaining coal units in WA will see them in terminal decline. The daily load swings simply cannot be accommodated by coal fired generation. The government must now act to actively plan the transition to a dispatchable power system philosophy and bid farewell to the baseload one that has served us well in the coal era of the past.
The decisions are not going to be easy but they are mandatory for the security and cost of future power on the SWIS.
It is good to read about some of the cost/price implications of significant PV and wind feeds. I have contended for a long time that generation cost comparisons of renewable and non-renewable energy have been very misleading. As the sales opportunities for traditional generation become impacted by renewable feeds the consequences of these changes should applied to the renewable side of the ledger. The cost of generation of what ever flavor is still largely the repayment of the capital costs. These capital cost need to be met even if there is no product sold. A cut of 50% of the product almost exactly double the unit cost. Even if a renewable feed was zero, for example, because there still remains the need for standby generation then about the same number of dollars need to flow to run the standby plant for it to remain viable. Cost comparisons seems to imply that when renewable feeds are less than traditional feeds the total cost of generation will drop. It will not. Even if renewable feeds were zero the cost of product will be almost the same because of the need to pay back capital on the standby plant. It will never be that cheap. Good to see some discussion about this very badly presented subject.
Nice analysis. When we (Gamma Energy Technology and Red Vector) looked at this issue using historic wind data matched to existing and planned wind farms we found a very significant correlation between wind droughts in each state. In fact there was a period in May 2010 when the output from a diverse portfolio of wind farms in each state would’ve averaged just 5% across the entire NEM for 8 days.
That is not something that storage can help with (the volumes are far too large), other forms of generation will be needed during these periods if wind becomes significant, and if the purpose of wind is to decarbonise the system then these will need to be low carbon too.
Wivenhoe pumped storage (500 MW) could be used much more for balancing, and questions need to be asked as to why it is used so sparingly. So far this month one of its generators has been used 6 times at between 60 and 100% of capacity, and the other has not been used at all. For comparison the Shoalhaven pumped storage has generated 23 times. Wivenhoe is operated by State-owned CS Energy, which also operates several coal-fired power stations. It has a conflict of interest in despatching Wivenhoe because this would reduce its income from the coal-fired stations within its portfolio. But as an asset owned by a State government with a target of increasing renewable energy, its management by CS Energy seems to be detrimental to implementing government policy.
The graphs show a typical power price of $70/MWh, apart from price spikes and daytime troughs when there was an excess of solar. With current steaming coal prices an operator competing for potentially seaborne coal needs to recoup at least $47/MWh just to cover the cost of coal. Furthermore the power stations near Rockhampton and Gladstone get paid a 5% discount on the State price because of line losses. So there’s little margin to cover the other costs, show both a return on capital, as well as profit. Hence their reliance on price spikes.
Paul, interesting, thanks. Not sure I like the title though – its pejorative and ideological. There is nothing terrible special here: production increases when the sun shines. We have seen the same thing repeated whereever solar becomes significant – Italy’s main trasmission has been recording peak demands at night, not day, since 2011. This provides incentives to switch load into the day time period.
I would love to know where the $70 per mwh is needed to break even for power stations comes from.
Somehow over the last 15 years they have operated for up to 4 consetive years where the average spot was around $30!!!!!!!!!!.
To anyone who wants to debate the cost of coal fired energy, I would first like to level the field, and allow to discussion to be around, coal fired plants being able to be built as easily as wind and solar and the bidding for power being filled in order of lowest bid, as it used to, not renewables first.
I am quite sure in Queensland if that were the case the $70 per Mwh, would quickly become $30 again.
Remembering that coal generation has not sat still and has advanced a huge amount over the last 15 years.
The constraint to coal is simply the belief that co2 is a problem, there is no cost problem where renewables are cheaper.
Hi Paul,
I did use NEMReview to get a bit more granular and looked at hourly average data from 1st April. It gives another great perspective. It does show that the Rule has effectively put a floor under the spot prices during periods of high wind generation and low demand. It would be interesting to model how much this has increased average wholesale prices in SA, that ultimately also flow into retail prices, and the shift in wealth from end users to generators.
With the cost of solar components continuing their global descent coupled with the move towards storage and the increasing awareness/consciousness of the “average” Australian, these trends will continue and magnify over time. Will they in fact help with reduction in grid electricity prices? Time will tell. Interesting article, thank you!
As prices rose to pay for decentralised renewables and their delivery the low hanging fruit of consumer demand was plucked with energy saving appliances, lighting, judicious RC aircon use, etc and now the hard yards of increased population and concomitant inelastic demand begins in earnest. With our thermal generators wearing out due to freeriding renewables we not only have to pay for more wind farms and solar panels but the elephant in the room- storage. We live in interesting times now.
When doing this sort of thing it is valuable to normalise at least for customer number. Having done so we see steep declines throughout the NEM except in Qld, where CSG compression has swapped demand contraction.
Thanks Bruce
Normalising (e.g. by customer number) would provide useful additional details. That would be food for another post at some time.
The numbers in aggregate (as in this article) provide a useful measure in their own right – such as, in my case, helping to identify that in aggregate average demand per region had stopped declining in NSW.
This will have implications for the future in a number of respects.
Paul
Paul
When you do the sums on a financial year basis, the growth in demand in NSW has increased by only 100 MW over the past four years supporting the view that Bruce M makes
The only real growth that has occurred is the addition of the CSG compression in Qld
Thanks David
I did try to make it clear in the article that the growth rate in NSW was very modest. Sorry if that was not as clear as I meant it to be.
What my main point of surprise was, and is the thrust of the article, was that the decline in demand had stopped in NSW.
As you know, I have a particular focus on QLD and have noted many times on Wattclarity about the impact of CSG compression. My assumption had been that the decline was continuing in VIC and SA (true still), along with NSW (not true for the past 4 years, which is embarrassing for me).
Paul
Might be interesting to put these numbers against economic activity in the respective states and also compare total energy consumption by state. Given the recent national statistics, there may well be a simple explanation – that we should all be happy about. People forget that at a high level, domestic energy consumption is only about 35% of total energy but causes over 50% of maximum demand. The NEM was well protected during the last economic cycle by a large over supply of generation. Retirements in the last 5 years have removed that. This next cycle might be more interesting on days when the wind isn’t blowing. First sign will be forward pricing on industrial bilaterals. Might be interesting to watch from a distance in WA !
the only measure from an electrical system point of view is the “native demand” which is the underlying demand and energy consumption or the total of the demand and energy consumed by customers. In this case, for the NEM connected customers. It must by definition be a sum of the NEM generation plus embedded generation plus behind the meter generation. On this basis, the energy consumption of the NEM is increasing. To measure it in any other manner for system security and planning purposes is to ignore a large growing segment and an action that will lead quickly to reduced reliability and higher pricing.
A key thing that I believe is worth mentioning is that Tuesday night LOR was down to the Bannaby-Gullen Range transmission outage (they were cutting in a new substation), which reduced flows into Sydney by ~1000MW. This outage was supposed to finish at 17:00 but kept being extended, finally finishing at 19:15 or so.
I don’t believe that RERT was dispatched – there would have been constraints #RT_NSW1_OE/PE reflecting this. The decrease in demand following the MPC spikes was most likely Tomago pulling potlines, and the 2nd spikes due to increased demand which was probably them putting them back in.
Gas availability coincident with peak was Colongra bidding unavailable to avoid uneconomic start. Not sure I agree with how Snowy managed this in light of LOR2 – someone probably had a word with them given they ran on Friday night.
The SA LOR was due to some planned maintenance by ElectraNet to take a key 275kV line out which would reduce Heywood capacity to ~100MW.
Thanks HR. Hopefully not too many other things I missed in this summary.
My project is on FCAS and FFR markets and I’ve been reading plenty of material online but never seen such clear information. Thank you.
mmmm…lets not have reliable dispatchable power, lets have an innovative set of derivatives. Typical of rent seekers circling this mess.
“That’s not getting us anywhere closer to a properly managed energy transition.” While talking about your emotion meter and others stongly held beliefs you expose your own.
I would say the Author will be nailed to a cross for saying that wind and solar and not reliable energy.
Quite a lot of people believe it’s the other way around. Coal and Gas are intermittent and wind and solar are predictable. Due to the weather being able to be forecast so accurately. Yes I am a fond boater and the forecast for boating weather is so broad and covers all most every possible out come, yet my experience is that they still have a very poor accuracy.
I find this article really fascinating given all the talk and evidence of rooftop solar booming nationally. How is that we get so much generation from solar yet NEM demand increases. Where would we be without solar and other renewables…probably all living with constant blackouts, shortages etc. So much for the current government line…
Paul, is there experience in other jurisdictions with prices going negative, or is it happening in Oz for the first time round the world?
My crystal ball has a blurry image for about 2020 where the price bounces around at a much lower level than now, especially when the interconnectors are in heavy use and the renewable generation is high. I do hope that the crystal ball the investors pouring into the Oz market have, shows a different and sharper image.
Let’s always entertain that there is a bust around about with every boom and to plan for that too, whether it be Dutch tulips, mining construction or renewables.
Your prognosis is valuable to the industry as well as to your own business. Thanks for sharing.
I suppose all the price pain will be worth it in the long run, we are trying to change the way we supply power, to reduce Co2.
With gas prices at $9.34/GJ the fuel cost for gas generators alone will be $9.34/GJ * 3.6MWh/GJ @ 30% efficiency = $112/MWh. Which makes it even more interesting as to why TIPS would be running 7/8 units at less than 35% CF as shown in the figure at that price. But the real issue is the interconnector outage – it would be much harder to get these outcomes without it.
The interconnector is no issue. There was always available capacity all day, physically at least.
(Only one of the two circuits was out for maintenance).
Available capacity is not indicative of market outcomes. And the interconnector was limited to less than 100 MW throughout the day (currently limited to 37 MW).
TIPS would be running units at part load so they can quickly ramp up and meet AGL’s own peak demand in the evening. They are probably not running at higher loads during the day because they are long on capacity during the day and want to retain line pack in the pipeline to meet the peaks. Just because the pool price and gas price is high doesn’t mean that this reflects a gentainers real position.
But if they could sell more power at close to max wholesale price now, and potentially fall short of their retail demand later (assuming interconnector/wind issues ease by then and prices are more typical), wouldn’t that still present a good arbitrage opportunity?
Maybe they didn’t want to assume the interconnector/wind issues would be overcome by the evening peak? AGL are probably a bit cautious following the issues in NSW a few weeks ago when they were very short on capacity due to multiple unit outages.
Great read Allan, very informative.
At least the battery was there to keep the gird stable.
Great analysis Allan.
Allan
Thanks for the very good analysis and I agree that the primary issue is lack of competition. There is a view that 5 minute settlement will also depress competition so perhaps that is not the answer.
The work you did also highlights
that gate closure (say 1 or 0.5 hours before dispatch) would also address the problem
Another solution could be having the lower SE a separate region,or even part of Vic region
Regards
David
Hi David
Thanks for your comments.
1) Re Gate Closure
You’re not the only one talking about some kind of gate closure mechanism (e.g. the recent Grattan review) – and these calls tend to do the rounds periodically (e.g. when the SA Govt sponsored review at the AEMC into bidding a few years ago). However each time the call is made, no-one seems to be able to help me understand how no change would avoid the fact that there’d always be a “last bid” (hence the motivation to try to maximise profitability by changing bid). Doesn’t moving gate closure further out from “now” only end up making AEMO procure more Regulation FCAS services, given increased uncertainty about what demand will actually be?
2) Re Zonal and Nodal
My sense is that you’re onto a solution that would have a chance at being more effective, in terms of creating a separate Region for South-East of South Australia (and will need same solution for North Queensland in 18-24 months as well). However the politics of this would prove very tricky – which is why (despite the fact that it was allowed for in the rules) States have never been disaggregated into Regions.
Paul
To the layman, this post is incomprehensible, which is not in the least the fault of the author, but was almost certainly intended by the crafty folk who created this thicket of legislation, pseudo-market prestidigitation and regulatory gobbledygook. A mere two or three decades ago, the average voter could understand how the electricity system operated – a handful of large, often state-owned generators produced reliable electricity at low cost, which was distributed to everyone via an equally simple distribution network and sold directly to consumers as a regulated, essential monopoly product. State governments each operated their networks separately, and were thus responsible for the outcome. After two decades of massive subsidy, interference, distortion through hundreds of pieces of legislation and the creation of a dozen or so unaccountable regulatory bodies, all this simplicity has been swept away. By whom, and for what purpose? Can sanity in this essential industry ever by regained?
Thanks for the article Paul.
I will add a couple of points increasing the complexity further:
1. Climate change and potential for an increased number (and duration) of extreme weather events e.g. heat waves.
2. Increased temperature responsive load, driven by renewed housing stock which, by being generally well insulated and installed with airconditioning, is space heating/cooling load efficient in mild temperatures, but significantly increase demand when temperatures are outside the mild range.
I’m constantly amazed at the blurring of Energy (MWh) and Peak Demand (MW) in the discussion.
50% renewables (ie MWh) does not mean 50% installed capacity (MW).
There is very little understanding about diversity, and capacity factors, especially for wind generation.
Queensland’s 50% renewables target is a 50% energy (MWh) target. In essence it’s also a 50% non-renewables target, and the policy relies on the contribution of coal generation to stabilise the network and provide reliability, but no-one talks about that
Your widgets are amazing. keep up the good work.
You’re not wrong when you have a play around with wind outputs for Australia and compare say July13 with July17 overall. That’s bad enough to cope with but then knock off all the State’s outputs except South Australia for July13-
http://anero.id/energy/wind-energy/2018/july/13
How on earth is a national grid (ie a rational system) supposed to cope with that because MLFs and gobsmacked doesn’t begin to do the problems justice and where is the omniscient benevolent dictator to solve it all or wind back the clock?
Is villan no. 5 the sheer rapidity of change, to the extent that there are very few people that really understand some key aspects of the power system? The best example of this is in “grid security”, which is often used in South Australia to curtail wind output, but also limits most inter connectors to well below their thermal capacity. How many wind farm developers in South Australia included this as part of their financial assessment? Even the NEM rules were until recently were for an energy-only market. As an example of this, on 13 November 2016, just a few weeks after the system black in South Australia, the SA market operated with only a single gas generator, and all other energy from wind. This would now be considered “not in a secure state”. Even AEMO don’t seem to have many people skilled in this area.
WattClarity does a great job of helping to reduce the evilness of Villlan #4 (including me). I would be wary of the strength of the emotion-o-meter villain #3. I have witnessed how it has ruined the gas extraction and GM foods industries and no calm reasoning of fact based website has found a way to bring them back from the margins to the sensible centre. It’s a little like a story about a horse, a gate, and a bolt left unlocked.
Paul,
Under Q2 you say “In the chart above I have assembled monthly aggregate production from generators for which AEMO has data”; this doesn’t seem to hang together for the only chart in the article. Am I missing something or is there meant to be another chart
Sorry for inadvertent confusion, Shane
Only one chart intended!
Paul
I like the premise of this article.
Do you have any thoughts on the QLD government proposed “CleanCo” – a new government owned clean energy generation company?
great post. I didn’t realise how the 5/30 split was leading to that incredible volatility. Looks like AGL is doing an ‘Enron’, and they’re being allowed to get away with it.
A few more novice questions, which I hope are addressed in future issues.
1. Why are interconnectors between regions constrained to half or less of their thermal capacities, whereas within regions there don’t seem to be problems using transmission lines up to their thermal capacities? In other words, does physics work differently between States than within States?
2. Why are interconnectors often constrained to well below half of their thermal capacity?
3. Why are there now system strength requirements requiring a minimum number of gas generators to be on-line at all times in South Australia, whereas in the recent past it seemed to operate without problems without these system strength requirements? For example, for several hours on 13 November 2016 there was only a single gas generator operating in South Australia, with all electricity generated from wind, and there didn’t seem to be any problems reported.
4. In determining the system strength requirements in South Australia, why have none of the simulations to date have included the 70MW of battery capacity from Hornsdale Power Reserve to include it within the minimum system requirements.
5. In the Victorian market, why would the brown coal bids be so low that night-time spot prices drop as low as $11/MWh, and at this price none of the generators are at minimum operating load? Surely it would be in their interests to put in higher bids down to the minimum operating load.
Thanks Malcolm. Will get to these in the future “Intermediate Level” posts.
Paul
It may be less a case of the windfarms _actively responding_ to negative prices, than simply having capacity offered (bid) into the market at prices well above the -$1,000/MWh price floor. In cases like today where there is low demand, a strong constraint on exports – due to yet another 275kV transmission line outage in SA – plenty of wind, and the system strength constraint requiring fixed levels of thermal generation to stay online, economic market dispatch will necessarily reduce generation from the “higher priced” windfarms to keep supply and demand balanced.
Effectively the windfarms in question would have set up their offers so that their output is automatically reduced by AEMO if the spot price is lower than some threshold, which may be related to the value of LGCs, or the bundled price of energy & LGCs. Not that different to how other generators bid except that for contractual reasons these price thresholds tend to be negative for windfarms.
Allan
Will have a look tomorrow, Allan. Thanks for the reminder that there might be simpler actions!
It is a pity the SA/Vic boundary wasn’t moved from Heywood to Tailem Bend for the duration of the works. What is the point of curtailing the SE wind farms when the transmission restriction is between Tailem Bend and Tungkillo, and there would be more than enough capacity on the Heywood interconnector to export all their output to Victoria.
Good question Malcolm. I haven’t looked but if the SE windfarms are being curtailed by the Tungkillo-Tailem Bend outage (rather than just dispatched down on price), then that will be due to a transient stability constraint not a thermal limitation on exports to Vic. The issue is that if the remaining 275kV line in SA trips, the SE generation remains connected to SA on the much weaker 132kV network, and if that generation is running too hard at the time the sudden change in impedance leads to transient issues impacting those generators. The constraint’s generation limitation effectively controls what is flowing back into SA from the South East rather than what is exported to Vic.
Won’t there be even greater problems of managing the generation/load balance within the SE SA region if it is part of the SA market than moved to the Vic market for the duration of the works? For example, if SA prices turn negative and the SE wind farms turn off, AEMO needs to quickly find perhaps 200MW to support the SE load, while the capacity from north of Tailem Bend is limited to 100MW. So it would have to find the balance from the Heywood interconnector. According to an AEMO diagram, there are 3 lines from Tailem Bend to Adelaide
– a 275 kV line to Tungkillo (which is being worked on)
– a 275 kV line to Mt Barker
– a 132 kV line to Mt Barker, which has a likely thermal rating of 100 MW
The latter line is no doubt the limitation, so if the Mt Barker 275 kV line trips, there is still 100 MW of capacity via the lower voltage link.
Malcolm
As Ben says below, location of the regional boundary is really just a settlement issue. The dispatch algorithm loads generators according to least-cost principles consistent with all constraints, regardless of which region they are in.
In the scenario you paint, the “missing” 200 MW would simply be provided by increasing loading on other generators (including possibly in Vic / NSW / Qld) and network flows (including across Heywood interconnector) would adjust accordingly.
Also note that it’s not only the thermal capacity of the 132kV network that is the limiting factor on flows to / from SE SA, it’s the much shorter term transient impacts of a trip on the remaining 275kV line that have to be protected against.
Allan
The dispatch system should still dispatch the generators downstream of the constraint correctly assuming they are bidding below the marginal cost of supply in Victoria.
The constraint orientation should recognise the generators’ locations OK.
The regional boundary affects settlement, but should not directly affect the efficiency of the dispatch process. Sometimes generators will rebid to avoid being run and settled below running cost, but probably wouldn’t occur in this case.
Why were they bid in at $-1000 in the first place if they are not covered in some other way like a PPA? Is there some technical reason that wind turbines cannot increase or decrease their output on demand, such that the owner would be prepared to take such a hit?
Wind farms will bid in at -$1000 for two reasons:
1. PPA covered all risk, as you mentioned.
2. When there are too much wind in the region, AEMO issues a constrain to all wind farms, and WF with the most expensive bids would be curtailed out of the market. WF would typically bid low to stay in the market because prices are still likely to be profitable for them – that is until the -$1000 hits.
With 130km/hr winds and severe chill forecast across the southern States we may be about to live in interesting times again, especially with the NEG up for debate.
So does this mean that there will be a inverse effect on the value of dispatchable storage capacity as the opportunity value of zero/sub zero PV energy represents an even greater earning potential for storage projects such as Kidston?
If I were a wind generator, I would bid in at -$1000 to ensure you get dispatched for most of the time and take it on the chin when prices a negative for a number of negative dispatch intervals. Almost all of the time wind gets dispatch at positive prices and unless they have a ppa, I would prefer to take pool risk.
Speaking of power risk it’s claimed the scalp of another PM who’s tried to marry reliable affordable power with Green energy and on that note another interesting resignation-
https://reneweconomy.com.au/agl-boss-andrew-vesey-steps-down-immediately-96485/
All in the too hard basket gentlemen?
Did the two interconnectors actually trip at exactly the same time or did one trip fractionally earlier? Rapidly falling load on the Heywood interconnector could cause it to open to protect the SA network with consequent drop of frequency in VIC & NSW causing the QNI to trip.
So we didn’t lose a single generator on the system but had to trip more than 1GW of load and on a low load Saturday afternoon.
Mick, as noted to Liam (comment below) I don’t know the exact sequence.
All we know is:
1) Two trips occurred within a 5-minute window. We don’t know how close (but suspect within seconds).
2) Which one came first, we don’t know.
3) What caused the first to trip, we don’t know. There were storms in northern NSW so perhaps lightning was a factor, but that is purely speculation (and was not implied in Market Notices).
4) It seems likely that the 2nd trip was a result of the 1st (perhaps due to rate of change of frequency, as Liam suggests in his comment).
Per AEMO’s market notices, simultaneous tripping of 330kV transmission lines in QLD 8L and 8M Dumaresq to Bulli are now considered credible contingency, synchronous export from QLD to NSW have been constrained. The cause is unknown and AEMO not satisfied this event will not reoccur.
Perhaps the chain of causality is opposite to your suggestion, ROCOF in VIC was too fast, overloading Heywood? I will certainly await the report into this one.
Sorry, Liam. Yesterday when I posted, did not intend to suggest either sequence of interconnector trips – i.e.:
Scenario 1 = whether QNI was first and then Heywood, or
Scenario 2 = whether Heywood was first and then QNI.
The Market Notice you refer to was published at 15:44 (more than an hour after this post).
My reading of it still does not confirm which trip came first (i.e. either Scenario could hold). Your hypothesis is one possibility in Scenario 1.
Like you, we’ll be awaiting the report. A sensible thing for any reader to do, to reduce the risk of forming wrong conclusions….
Cheers Paul, my comment was intended to be a reply to Mick above, adding the later information from the market notice.
A small update. One part of the mystery is solved.
“64012 26 Aug 2018 17:52 RECLASSIFY CONTINGENCY
Cancellation of a Non-Credible Contingency Event: Simultaneous trip of 8L and 8M Dumaresq – Bulli 330 kV Transmission lines as a credible contingency lines in NSW due to protection mal-operation.”
At the time I just happened to check my solar inverter on a sunny day in southern Victoria, expecting to see high output. But instead its screen was going through what seemed like a boot-up procedure, then after finishing this there was the “clunk” of the relay and it began showing its power output. This is worrying from a system point of view, that a new inverter (Fronius) tripped out with a rapid change of frequency. At that time Opennem shows 802 MW of roof-top solar in Victoria. If much of this were to trip because of bad fault ride-through settings, a lot more load-shedding would be required.
Perhaps AEMO needs to commission some physical tests of inverters under simulated rapid changes in frequency, rather than trusting manufacturer specifications. Inverters that fail are then banned from the Australian market.
If it is “quite probable” that mistakes are made, is this article suitable for publication?
WRT the question in observation #4, it doesn’t look like there is a significant time stamping issue because the small frequency deviations seem to match quite well between the measuring devices.
The drop then rise you are talking about in observation 10 (and others) looks like inertial response to me. That will look like a response proportional to the rate of change of frequency. It is governed by the physics of the machines. So not really a controlled response.
So I would say for Loy Yang, Bayswater, and Gladstone there was no governor response or there was but it was too slow to see in that time domain.
Thanks Kate, I have been following your conversation on this topic since I saw a presentation of yours in 2017.
Obviously the deadbands need to be set and prioritised correctly and I hope that your advice is acted on in record time.
I have two very minor critiques, pleased accept them in good spirit!
1. You referred to renewable intermittency a couple of times when referring to frequency, and seemed to imply that intermittency applies to poor frequency control. In my experience the term intermittency applies more often to weather dependent generation capacity, not so much frequency control capability.
2. You mentioned FCAS contingency market and spinning reserve, and obviously you are correct that a generator has to have some spare capacity in order to increase active power output to manage frequency, but my pedantic nature forces me to point out that there is actually no market for spinning reserve for generation capacity.
Ok, thanks for listening, looking forward to some proper governor control soon!
Thank you Paul for this excellent article.
Regarding Observation #15, I believe the decrease of the output of the 6 renewable generators is consistent with the frequency control they must implement to meet the requirements of clause S5.2.5.11 of the National Electricity Rules. Given the capability of power electronics based generation, these generators are probably registered with an Automatic access standard to clause S5.2.5.11, i.e. they must reduce their active power output by at least 20% of their maximum output for each 1% frequency increase above the upper limit of the normal operating band (being 50.15 Hz on the mainland).
This frequency control is a security measure to prevent the power system to collapse. All generators implement such a control regardless their participation to the FCAS market.
Solar farms won’t be able to propose an Automatic Access Standard under clause S5.2.5.11 since they generally cannot increase active power output in respond to a reduction in system frequency (S5.2.5.11 (b) (3)). It is more likely to be a Negotiated Access Standard which can meet S5.2.5.11 (b) (1) and (2). And if they are not going to participate in the low FCAS market, they do not need to activate this function. They simply just need to demonstrate that they are capable of doing so. But I know that some plants enable the function without fully understanding the consequence.
As you wrote, solar farms only need to demonstrate they are “capable of”. If they are not dispatched at their MPP, they can increase their output in response to a frequency fall. So they can propose an Automatic Access.
AEMO June 2018 report which discusses non credible event of a QNI trip from high load has some interesting facts.
I’ve carried out lots of plant failure analyses even back to the days of sprecher & schuh relays and you’re on a hiding to nothing trying to use 4 second data especially when you don’t have access to the source of the scans.
There will always be timing issues when different systems are involved. And even though a time is assigned to a scanned value you have no idea of the exact time the data was actually collected.
Modern failure analysis relies on high speed monitoring systems runnng independently of the process control system. While all the above comments are interesting, I suspect none of the data AEMO is currently providing has the required level of integrity to allow for the sort of analysis required.
Re Observation #4
Yes, I think you have a timestamp issue there. I log line frequency in SEQ with 10 second intervals, and here’s what I saw:
2018-08-25 13:11:20, 49.980
2018-08-25 13:11:30, 50.000
2018-08-25 13:11:40, 50.542
2018-08-25 13:11:50, 50.642
2018-08-25 13:12:00, 50.592
2018-08-25 13:12:10, 50.582
2018-08-25 13:12:20, 50.622
2018-08-25 13:12:30, 50.612
2018-08-25 13:12:40, 50.642
2018-08-25 13:12:50, 50.642
2018-08-25 13:13:00, 50.673
2018-08-25 13:13:10, 50.693
2018-08-25 13:13:20, 50.733
2018-08-25 13:13:30, 50.753
2018-08-25 13:13:40, 50.753
2018-08-25 13:13:50, 50.733
And here’s the output of my SolarEdge inverter in watts, you can see it bailed early….
2018-08-25 13:11:20, 1750.0
2018-08-25 13:11:30, 1750.1
2018-08-25 13:11:40, 1658.8
2018-08-25 13:11:50, -3.9
2018-08-25 13:12:00, -4.3
Thanks Watt Clarity for your excellent blog. This may interest evertyone as I stumbled onto this website last night. Looks like this idea has been a long time coming….. they talk about a ‘reverse ebay style’ auction to help drive down energy prices. Looks interesting.
http://www.givemeyourbestprice.com.au
Do we have relay flags from Heywood and Qld as to the causes and PRECISE timing of the tripS?
Perhaps SA should consider high-speed inter-tripping to shed SA LOAD when Heywood is overloaded in the import direction and to trip WIND GENERATION when OVERLOADED in the export direction.
Stated power supply from QLD to NSW prior to the incident: 866MW
The QNI capacity of both[1:] circuits is 1078W[2:]. Both circuits must have been in operation to carry 866MW.
Q1: If you want a single failure proof QNI interconnector, then what is the maximum permissible load on both circuits of the interconnector?
A1: 539MW, preferably shared half and half.
Q2: If you do not operate in a single failure proof manner, then what are the consequences of a single circuit failure for any reason?
A2: A cascade trip of the second circuit, that leads to a systemwide transient that is only halted by a AUFLS trip of both Tomago and Portland smelters. The QNI trip shed 866MW of supply and the smelter load shed 850MW of demand. A neat fit that made the transient survivable without a total system black.
Tomago drops 600MW
Portland drops 250MW
Q3: Why would AEMO expect any different result?
A3: Because they have not learnt the lessons of the SA statewide blackout. Is AEMO going to make a habit of doing this[3:] ?
[1:]
https://en.wikipedia.org/wiki/Queensland_E2%80%93_New_South_Wales_Interconnector
The Queensland – New South Wales Interconnector (QNI) is a 330 kV AC interconnection between New South Wales and Queensland, Australia. The link was commissioned in 2001. It consisted of double-circuit 330 kV lines between Armidale, Dumaresq, Bulli Creek and Braemar, and a double-circuit 275 kV line between Braemar and Tarong.[1
[2:] Stated power supply from QLD to NSW prior to the incident: 866MW
The QNI capacity of both[1:] circuits is 1078W[2:]. Both circuits must have been in operation to carry 866MW.
Q1: If you want a single failure proof QNI interconnector, then what is the maximum permissible load on both circuits of the interconnector?
A1: 539MW, preferably shared half and half.
Q2: If you do not operate in a single failure proof manner, then what are the consequences of a single circuit failure for any reason?
A2: A cascade trip of the second circuit, that leads to a systemwide transient that is only halted by a AUFLS trip of both Tomago and Portland smelters. The QNI trip shed 866MW of supply and the smelter load shed 850MW of demand. A neat fit that made the transient survivable without a total system black.
Tomago drops 600MW
Portland drops 250MW
Q3: Why would AEMO expect any different result?
A3: Because they have not learnt the lessons of the SA statewide blackout. Is AEMO going to make a habit of doing this[3:] ?
[1:]
https://en.wikipedia.org/wiki/Queensland_E2%80%93_New_South_Wales_Interconnector
The Queensland – New South Wales Interconnector (QNI) is a 330 kV AC interconnection between New South Wales and Queensland, Australia. The link was commissioned in 2001. It consisted of double-circuit 330 kV lines between Armidale, Dumaresq, Bulli Creek and Braemar, and a double-circuit 275 kV line between Braemar and Tarong.[1
[2:]
https://www.aemo.com.au/-/media/Files/Electricity/NEM/Security_and_Reliability/Congestion-Information/2017/Interconnector-Capabilities.pdf
QNI nominal capacity From To Nominal Capacity
Queensland NSW 1078 MW
[3:]
http://wimnet.ee.columbia.edu/wp-content/uploads/2015/10/Analysis_of_Failures_in_Power_Grids.pdf
Analysis of Failures in Power Grids
https://www.aemo.com.au/-/media/Files/Electricity/NEM/Security_and_Reliability/Congestion-Information/2017/Interconnector-Capabilities.pdf
QNI nominal capacity From To Nominal Capacity
Queensland NSW 1078 MW
[3:]
http://wimnet.ee.columbia.edu/wp-content/uploads/2015/10/Analysis_of_Failures_in_Power_Grids.pdf
Analysis of Failures in Power Grids
Thanks for the analysis!
AEMO’s 4s data has been referenced a few times, which contains two measurements of frequency – one each for NEM NORTH and NEM SOUTH. Do you know where these are measured from?
Just after the event I happened to look at the display of my inverter (a new Fronius), and was surprised to see it rebooting. Then after the reboot there was a “clunk” of the relay, and it gradually began to show the amount of power from DC to the home and grid. It was a sunny day here in southern Victoria, so there would have been a reasonable contribution of rooftop PV to demand. The inverter specifications show it tolerates frequencies down to 49 Hz, and there is no setting for the rate of change of frequency.The 4-second data show it didn’t quite do down to 49.0. Their data are presumably the average frequency of the 4-second interval. However what AEMO should be logging is the max and min frequency in each 4-second period, because it is the max or min frequency that will turn off the small-scale inverters. As inverters become an increasingly larger proportion of daytime power supply, should we be requiring a wider range of fault ride-through settings?
Can we get some analysis on what impacts the Hornsdale Power Reserve had on the spot price in SA during hte islanding event.
It was meant to be the saviour during events like this but it seems they were low on charge even though Hornsdale Wind Farm had been operating above 75% capacity factor for most of the 2 days before the event.
Hi Paul,
although I can’t contribute to clearing up these obscurities, I’d like to thank you for yet another great article that helps anybody trying to get a better understanding of the NEM.
I especially liked the “quick notes” in the squares as a kind of glossary.
Many thanks, keep up the great work!
great article Paul
As an aside tThis might interest the number crunchers at Wattclarity as they’ve released the cost of the Tesla big battery and Hornsdale wind farm facility-
https://www.msn.com/en-au/news/australia/tesla-battery-cost-revealed-two-years-after-sa-blackout/ar-BBNBBaD
That’s an investment of $900/KW but if wind can only produce 30% of that on average that jumps to $3000 and then the question of what sort of investment cost per guaranteed kilowatt 24/7 all year round? Marginal costs are low but what’s the bottom line for a level playing field kilowatt of battery backed wind power here?
Thanks – Can you give an example of what you mean by “pushing risk anywhere but me” under Theme 9?
Really nice document Gents.
Actually, I think the most glaring example of villain #5 is not focusing on mitigating the impacts of rising temperatures. As stated above, villains 1 thru 4 ensure that, if you believe CO2 is the primary cause, that prevention is not possible. Mitigation is therefore the only viable option and it also addresses the issue that, if CO2 is not the problem, rising temperatures are going to occur despite all of our efforts.
Paul, you mentioned carbon pricing, therefore I take that as an opportunity to mention the logical inconsistency in the IPCC position. I explain it thus:
– CAUSE: IPCC position is that carbon emissions are currently the greatest threat to humanity, a threat created by developed economies and to be solved by developed economies
– ACTION: IPCC advocates reducing carbon emissions, mostly by the following mechanisms regardless of the effect on developed economies: increase renewables, reduce meat consumption, increase electric vehicles, reduce poverty and increase equity
– LOGICAL INCONSISTENCY: if reducing emissions is the most important action for humanity, and the energy system of the western world is the largest source of emissions, the logical solution is rapid transition to nuclear power generation, globally.
Since the IPCC does not recommend rapid transition to nuclear energy (which incidentally would provide electricity to more people than renewables thus reducing poverty and increasing equity), I conclude that carbon emissions are NOT the greatest threat to humanity, and therefore carbon pricing is a purely political strategy designed to restrict developed economies.
But I agree that if our chosen politicians insist on carbon pricing then it should be like the GST but with less exceptions.
Wind output was quite low at the same time, so SA exports may not have tested the limits there, but AEMO stability requirements will still require some synchronous generation at the SA end (to mitigate risk during unplanned interconnector trip).
The AEMO Q2 dynamics report showed some significant curtailment of semi-scheduled (wind) generation, so it would be interesting what the system would look like during a scenario of the same low demand but high wind.
Thanks for putting this together, Paul–quite interesting. I will preface by saying I’m working from knowledge of other systems and have no specific familiarity with AEMO, so I hope you’ll correct misinterpretations.
If I understand correctly, it looks like they are abusing the Target, which seems to penalize injections above the desired dispatch point but not below. Since this underproduction has to be balanced by providers of FCAS, I would expect there to be a penalty of some kind–does that exist in AEMO? If not, given how quickly they seem to be able to ramp down, the -$1,000 bid is strategically correct (although maybe not good faith).
I am a bit surprised by the speed with which the rebid is incorporated. Based on the timeline, a rebid at 14:33:12 gets incorporated into the SCED in time for the 14:35-14:40 interval. It’s also confusing that dispatch instructions for that interval only arrive at 14:35:20. Am I misunderstanding this timeline?
A semi-related question: does AEMO use a multi-period dispatch model?
Hi Tristan, regarding your comments about the rebate. I quote you in two lines below.
Firstly the word rebate. The rebate is actually a subsidy, which is actually a levy. The levy is not applied exclusively to those who install solar panels, the levy is applied to all electricity users, but not referenced on their electricity bills (no transparency), increasing electricity bills. The levy then has GST applied to it (no transparency). Those with solar installed then have some of that levy returned to them, which reduces their electricity bills. Typically the people and businesses with solar panels are not in the lowest economic group. The net result is your “rebate” increases cost of living for the lowest income groups and reduces cost of living for the higher income groups – that is effectively transfer of wealth from poor to rich (in simple terms).
Secondly your implicit assumption that subsidised solar panels on roofs in Australia can have any effect on global warming is far removed from reality and deserves to be called out as fanciful ideology. In other words, “you’re dreaming mate”. I encourage you to provide some evidence to support your position on this topic.
“So are we faced with a serious problem of there being too much solar and does that mean we should scrap the rebate?”
“Then there’s also an issue about do we care about the dangers to our children and others around the world from global warming, and is this solar rebate a worthwhile expenditure to address these dangers?”
Hi Ben,
I wonder if you could clarify your second argument a little, as I’m not sure what the angle is.
It seems intuitive to me that energy produced from rooftop solar would otherwise have come from fossil-fuel power sources and so has reduced power-related carbon emissions from a business-as-usual scenario. The subsidies would seem to have accelerated the uptake of rooftop solar (this makes economic sense) and helped get the cost of solar down.
If your argument is more around this phenomenon only representing a fraction of carbon-intensity of the Australian power industry, which is a fraction of Australian emissions overall, which is a fraction of the world’s emissions, then I sort of see what you’re saying, but I would offer two rebuttals to that:
1. There’s an important difference between something having a small effect and having no effect. A lot of small effects added together make a big effect. Arguing that this is too small to make any impact is like arguing that there’s no point voting in elections because one vote won’t sway an election.
2. Global warming is a global problem and demands that human society works together across borders, continents, political divides, etc. to solve it. Australia is a developed country with excellent renewable resources, and so is well-placed to lead the global economy through a transition to sustainable, fossil-free energy systems. We may be a small player in terms of things like GDP and population, but small players can punch above their weight particularly in kick-starting things (indeed as Australia did with solar PV R&D). It’s a depressing thought to believe that we might as well not bother trying to do anything meaningful or important because we’re too small to make a difference.
Feel free to correct me if I’ve misinterpreted what you were trying to say, or shoot down any of the points I’ve tried to make!
Tristan, loved the article, there’s some graphs and data analysis here that I haven’t seen anywhere else in this debate and help cut through the spin and misinformation.
Hi Jonathan, the renewable incentive schemes have certainly absorbed some of the market share of coal and gas generation, and hence emissions, there’s no doubt about that.
Your point 1 (the voting analogy) only works if each solar panel has an equal value in addressing emissions (incorrect because of global and regional solar resource variability), if each individual has equal access to a solar panel (incorrect because of differences in subsidies / cost / income / infrastructure), and other implicit differences e.g. are human emissions the root cause of global warming in the first place… plenty of debate there! This analogy also only works if each individual (or even country) has an equal say in emissions – plainly incorrect when you consider the EU, India and China who have no emissions reduction agenda.
To your point 2: if emissions are directly “the greatest threat to humanity” then the IPCC should recommend the logical action that will have the quickest and greatest emissions reduction – rapid deployment of nuclear power generation. The fact that the IPCC does not recommend nuclear, and instead recommends variable renewables is telling. If emissions are the greatest threat, surely the most logical response is to reduce emissions rapidly on a large scale, therefore surely nuclear energy would be the recommendation?
Alan Finkel stated that Australia can make no measurable difference to global temperature. This is an incontrovertible fact.
There is no “global community” working together to reduce emissions.
The IPCC does not recommend the logical obvious step to reduce emissions.
Nothing I’ve said above is controversial, hopefully that helps you to understand my objection to subsidies(or levies, these are not rebates).
They are bidding in good faith because they intended to honour the bid before market conditions changed.
Their rebid describes the market change that required a change of bid. Maybe Torrens should have rebid a little earlier eh?
Insofar as the constraint is a submarket within the NEM they have to sell at -1000 to remain generating. When the submarket begins to impact the actual dispatch price they have to change bid.
Paul, I trust you will be highlighting all of the gas generator rebids for “avoid uneconomic start” where they fail to ramp up 100 MW when prices spike?
I find it laughable that you expect Infigen to cop massive losses due to extremely complicated market conditions. Keep in mind that during these periods of Intervention pricing that predispatch price is totally misleading. Traders are flying blind and NEMDE is trying to minimise their revenue. Good luck out there!
Hi Tristan, regarding a couple of statements in your blog post, I feel obliged to highlight what I believe could be some factual errors.
“Solar – shown in yellow – is simply knocking off the top of the peaks in demand that would otherwise be serviced by what has become quite expensive, and hard to secure gas.”
The afternoon load peaks in the NEM occur after combined solar output has begun to reduce. I agree there is some overlap between combined solar output and afternoon load peaks, but the amount of overlap differs across the regions and the time of year. I believe it is misleading to state in simplified fashion that, “solar reduces the load peaks”. The duck curve created by the day time solar output forces dispatchable sources to increase their output at a faster rate than without the solar.
I agree with your analysis that in terms of bulk quantity of electricity generated, solar is nowhere near the amount of electricity demanded. No argument there!
But what you might have given is a summary of the various recommendations and concerns from AEMO about inertia, fault level and other grid stability matters. For example in SA there is a minimum combination of synchronous machines required to remain on the grid, with AEMO regularly and increasingly being forced to intervene to ‘direct’ renewable asynchronous generators to reduce output and synchronous generators to start or remain online. There are plenty of reports on this in the public domain.
I suggest that the intent of the conversation being directed by the various stakeholders in the NEM is more around the quality and stability of the grid, rather than the absolute quantity of solar power being generated.
Regarding the voltage, the allowable limits for voltage are 230 volts +10/-6%, which is the range set out in Australian Standard (AS) 60038 (Standard voltages).
230 x 1.1 = 253 V
Another note on this is the transformers closest to the loads are only manually adjustable off-load, so they have to be set to cater for voltage swings between day (solar inverters pushing up voltage) and night (no solar inverters). Transformers at higher voltages and substations can be adjusted online from the control room.
This could be why there are limits on both the capacity of individual household solar panel installations and the number of solar panel installations on a particular distribution transformer.
Hope that helps?
Ben,
If you look at the first chart illustrated patterns on December 2017 and in particular the days with the highest peaks it shows quite clearly that the absolute peak occurs where there is a whole lot of yellow. The peak which occurs with the red colour is lower than the peak that has yellow. You see similar results in January and February which tend to be the other months where peaks in NEM-wide demand occur.
Secondly, if you look thoroughly at the voltage distribution charts you’ll see they provide two lines – one for night-time voltage and one for daytime voltage. In both cases the voltage distribution is very similar and reflects the fact the voltage is clearly not remotely aimed at achieving nominal voltage of 230V. The point of this is that in most cases the networks could adjust the transformer taps (yes I know this is done manually) to deliver lower voltage and not be at much risk of falling to voltages that were too low, while allowing significant room for solar to push voltages up without much risk of reaching excessive voltage levels.
On your point about maintaining frequency/fault levels etc. yes these are legitimate issues in the case of SA’s very high levels of renewables penetration but not a major issue in other parts of the NEM and were not the subject of media reports which I am addressing in this article which mentioned issues such as overloading transformers and causing blackouts and producing far more solar generation than could be absorbed by demand. I would note that many of SA’s issues (exception of fault level) will be largely resolved with SA-NSW Riverlink interconnector that has been recommended in AEMO’s ISP. I find it quite irritating that the media uncritically report complaints by some sections of industry to support a nation-wide change in policy which when you scratch the surface are based on solar reaching very high penetration on boxing day in a single state representing a small portion of NEM demand and which will be resolved by undertaking actions that AEMO has already recommended and are in the process of being implemented.
I won’t bother to address your argument about emissions because you make a series of assertions that reflect someone who I suspect has made up their mind a long time ago that they find the conclusions of Academies of Science across all major countries so unpalatable that you have become impervious to logic, evidence or just simple ethical regard for your fellow human being, particularly the poorest of the world who are most vulnerable to the impacts of global warming.
Tristan,
With regard to the street level transformers averaging above nom 230v; would you agree that firstly, current is lower, therefore transmission more efficient. Secondly, domestic PV presents as a sudden load when a cloud covers the sun momentarily, at the same time the line voltage drops, so the line loses are momentarily compounded. Voltage drops because loss of PV and sudden increased load on the grid causing any inductive load to draw even more current. The upshot is the swinging loads caused by domestic PV at a local level are problematic. Power factor can also swing due to local conditions. Batteries will help but they are an extra cost.
Thanks Tristan, I’m happy to be corrected on the peak load, you are right that the ‘behind the meter’ solar output from rooftop panels is greatest during the day. I had been looking at the AEMO dashboard which of course only shows the solar output as a reduction in demand during the day. The unintended consequences I have mentioned (dispatchable generation ramp rates) still apply.
I’m pretty happy though with the logic I have presented on subsidies, nuclear generation and emissions. This logic is, after all, not open to attack because it is based on facts.
Switching the argument in an attempt to prescribe guilt at the theoretical unproven sometime in the future (maybe) plight of the world’s poorest is difficult to align with the argument that for zero effect on global emissions, Australia has some of the highest power prices in the world, which disproportionately affect our own poorest citizens. Choosing this style of argument does nothing to help the world’s poorest, even if they are prone to some consequences of ‘climate change’ – refer my argument above on the IPCC recommendations – since actions in Australia to subsidise renewable energy can have no effect on those people.
If you think it will have an impact, and that I am “impervious to logic”, then please take the opportunity to educate me and other readers.
Regards
Ben
Ben
Thanks for methodically taking this claptrap to task. Tristan traditionally takes a one dimensional view when developing his pro renewable propaganda. That the ESB, AEMO and other responsible authorities need a lesson on the virtue of solar panels is a preposterous assertion.
Tristan, I think you need to understand how distribution low voltage feeders are necessarily designed. Bottom line is domestic demand peaks always occur after rooftop PV is essentially gone for the day. So the design criteria for the feeder is around the peak load. Now a transformer is actually a voltage source – built to a price with fixed tap changing. The lengths of low voltage feeder runs are determined by the allowable voltage drop along its length. It is a compromise. A customer connected near the transformer will see high voltage (at the top of the allowable range) whereas a customer near the end of the feeder run will see voltage at peak demand near the bottom of the allowable range. The transformer off load tap changers (normally 2.5 % steps) allows a small degree of tuning to this profile over time. This is simply good engineering design – function at least cost. The overnight voltage difference will be small simply because domestic overnight load is small – volt drop goes away as an impact. The problem with distributed generation is suddenly the volt drop issue runs in reverse with power flows back towards the transformer in mid morning to early afternoon – very little domestic load (empty houses) and peak solar irradiance. The design criteria is basically now completely wrong. The solution is the installation of on load tap changers to distribution transformers (usually a transformer change out) – not normally very attractive as distribution transformers are generally set and forget – tap changers are not – or you install more transformers and shorter LV cable runs. Both solutions are going to cost money that will end up in the capital base of network service providers (what happened to “causer pays”?). This stuff is not free.
3. 230/460 volts (single phase) with an allowable margin of +10/-6 per cent (between 216 and 253).
http://www.parliament.qld.gov.au/Documents/TableOffice/TabledPapers/2018/5618T56.pdf
Paul, can I suggest that in the case of blown LEDs, you attempt to get the temperature in the suspect downlight fittings, and compare that to lights that work fine? Voltage will affect all lights equally (and any other electrical appliances), while temperature and moisture can affect individual fittings separately.
I have had combinations of flouros and LEDs for years with no issues, and my power sockets read 245V pretty regularly (I have 3kW solar panels and am about 50m from a pad mount substation).
I recently changed 8 flouro donwlights to LED (Edison thread) and they are all fine, except for one that lasted 1hr and went kaput. I put this down to a manufacturing fault, since any other bulb works fine in that socket.
I have found that temperature, moisture, vermin, dirt and build quality have played a greater role in bulb life at my place than voltage.
Your general quest seems to be to understand more about power quality in general, in which case there are some terms that you might need to understand in more detail:
– flicker, unbalance, harmonics, overvoltage, undervoltage – there is more than just voltage, and any number of IEEE articles discussing the power quality effects of solar inverters and similar non-linear power electronic devices.
Q1. Does solar output drive voltage higher – yes! I visualise this by thinking about the situation with no solar – all the power is coming from a fleet of generators far away, with associated losses and voltage drop. With solar being exported locally you are effectively reducing the load and the losses, so the voltage increases.
Q2. Voltage is supposed to be 230V +10% -6%, but its a big deal to change all the off-load tap changers in distributed transformers in all the suburbs… takes time and coordination and priorities
Q3. Implications for equipment – see above discussion on power quality and local conditions separate to voltage
Q4. You can buy separate power metering devices that plug into a power socket, gathering information might be the first step.
There is a lot to unpack in these questions.
A1) Voltage rise during load is a characteristic of Line Drop Compensation for customers at the front end of the 11kV feeder. Wikipedia has a good article on LDC and associated problems due to distribution generation. https://en.wikipedia.org/wiki/Voltage_regulation I strongly suggest reading this.
A2) While some claim that the 230V standard was introduced to “allow more solar” this is a furphy. As a matter of fact, the 230V standard was introduced back in the year 2000 (well before widespread adoption of PV). It was actually introduced to facilitate more voltage drop along the feeder, to squeeze more out of existing electrical assets. The old 240V standard (only recently superseded) was 240V +/-6%. This makes the voltage range from 254.4V down to 226.4V. The 230V standard has a range of +10% to -6%. This makes the voltage range from 253V down to 217V. As you can see, the upper limit hardly changes (1.4V), yet the lower limit drops considerably (9.4V). To be more specific, the system voltage has to be below the upper limit 99% of the time, and above the lower limit 99% of the time. If it is outside these limits, then complain to Energex. They are generally very responsive.
A3) Maybe. But there are a multitude of factors. As Ben has pointed out, temperature, moisture, vermin, dirt, and build quality all play a part.
A4) I would start with ensure adequate ventilation around the LED bulb and buy good quality components.
Just some other points;
Energex has been installing power quality metering equipment on the 11kV/415V transformers for quite a while now. I would suggest reading their public domain document titled, “Power Quality Augex Forecast 2015/20”. It is very detailed and also addresses some of your specific concerns in the article.
The old BC/ES fittings have significant safety issues. A child or adult can readily put their finger onto energised terminals. The newer GU10 and/or 12V screw fittings are much safer.
Hi Paul, I’m suspecting that your down lights aren’t on when the sun is out, so a higher voltage due to solar output is unlikely to be the issue. Heat build up during operation on the other hand would be, as Ben suggests. I’ve had some flush fitting down lights now for a few years and not a single one has failed to date. I suspect it’s because the switch mode power supplies are separate to the lights themselves and live on top of the ceiling insulation. They were about $100 for a box of ten from ebay, great value! The previous LED’s they replaced were bulb type lamps installed in can type down lights and they got as hot as Hades and failed regularly. LED bulbs fitted to open holders on the other hand have been fine.
This from DMRE explains in detail all about the voltage settings, history, reasons and future.
https://www.dnrme.qld.gov.au/__data/assets/pdf_file/0005/1279571/decision-ris-qld-statutory-voltage-limits.pdf
LOR2 back in force for 4:30pm to 7:00pm
References 95-100 in our paper related to the issue of voltage control at distribution level with rising PV penetration. They are a little dated now, however it is indeed an issue, and one that has been on some radars for a while. https://www.researchgate.net/publication/315745952_Burden_of_proof_A_comprehensive_review_of_the_feasibility_of_100_renewable-electricity_systems#pfb
Great use of Solar Analytics data, Paul. I couldn’t help but dig a bit further. I’ll send you some more info.
For the moment, on question 1:
Yes, PV generation will correlate with grid voltage to some extent. Your inverter needs to raise the local voltage in order to push current back upstream towards the transformer. How much depends on what else is happening on your local feeder. So if you and your neighbours are all generating lots of energy, voltage will go up but if you are all consuming lots, it will go down.
You’re seeing some pretty massive swings throughout the day, so I’m guessing you are towards the end of a feeder with plenty of solar on it.
Possible solutions:
1. Balance the generation. You’re on three phase, so a three-phase inverter might have helped reduce the voltage rise on any particular phase. I also see that the voltage on one of the other phases is much lower, so putting the inverter on that phase would even things out a bit. In these cases, some installers put Solar Analytics on before they install the system so they can find the phase with the lowest voltage.
2. Reactive Power Control. Qld now requires power factor of 0.9 lagging to “consume” reactive power, which helps to limit voltage rise. This doesn’t affect your real power generation unless your inverter is maxed out. Perhaps check if your installer can enable a reactive power setting on your inverter.
3. Load shifting. Balancing generation and demand at local level helps to balance voltage and increase the amount of solar that the network can handle. We’re working on a virtual peer-to-peer product to help non-solar customers participate in that load-shifting. Most networks also have demand response trials but they need to get a move on.
4. Tap change. As noted by others, the standard is now 230V +10%/-6%. It looks like they could easily go down 5-10V. It is a tough gig as a DNSP though as they have so little visibility and these changes cost money. We’re trying to help as our database grows with loads of useful data like you’ve demonstrated.
I would like to know from someone who writes here, can solar from a house go into the grid, and back through a transformer? to say the next area and then to another house. Or does it have to stay on the transformer supply side?
In short, yes. The transformer will change the voltage (and therefore current) but the power will flow throughout the grid to wherever it’s required. Generation and Consumption need to be balanced across the grid as a whole (e.g. the NEM connecting Qld-NSW-Vic-SA-Tas) otherwise the frequency will rise or fall accordingly.
So solar power from a roof can go all the way back to say a main generation point, being stepped up?. Lets say the Local Wind farm that is for this example say 100K away?
Yes. Electricity will always flow from a point of higher voltage to lower voltage. Solar inverters push power into the network by injecting it at a voltage slightly higher than what it’s connected to. Hence the topic of the article.
Transformers don’t care which way power flows. Either direction is fine, although because they were designed for one direction, they may be a little more efficient in that direction.
There are a number of feeders (11kV lines from sub-stations) that already go to net positive areas. Currimundi for example, as a suburb has periods of good solar that generates more power than it consumes and that power goes back to the sub-station and out to another feeder.
Technically, there is nothing wrong with all of a sub-stations feeders being net positive and the whole sub-station would push power further up the transmission network. So long as voltage levels throughout are within acceptable range, all is fine.
Thanks for sharing, Paul.
Your plot shows a very large difference between day-time voltages and night-time voltages. It made me wonder how many sites have a similar range so I ran some numbers and found that you are indeed at the upper end. I’ll send you some more data.
Banjo,
The bias against nuclear technology in the IPCC is changing, albeit too slowly.
Using it to legitimise the idea that mitigating greenhouse gas emissions is not a very important thing to do seems just as fallacious as the thinking you seek to identify.
Bad thinking doesn’t cure bad thinking.
For only the latest group who argue forcefully for greenhouse gas mitigation AND nuclear power, see this open letter. http://environmentalprogress.org/big-news/2018/10/25/open-letter-to-heads-of-state-of-the-g-20-from-scientists-and-scholars-on-nuclear-for-climate-change
I certainly share your frustration. I think the inclusion of those technologies, rather than causing conflict, could prove a remarkable circuit breaker in Australian energy policy.
I would like to present an idea which makes a convincing argument for extending Australia’s existing National Grid as soon as possible.
Extending Australia’s National Grid should be our highest priority.
Technology today includes UHVDC, which allows for transmission across great distances (3,000km), with very little transmission loss.
All renewable energy projects by private enterprise will automatically follow.
“Build it and they shall come”
Please check our website for detailed information on our proposal.
As far as you emotion-O-meter goes, I’m aware of at least 5 organisations that have made some kind of detailed study of 100% RE and one of these organisations keeps the lights on 99.998% of the time for the NEM, the AEMO. Are you suggesting BZE, UNSW academics, AEMO, SEN, UTS for GetUp! are all “loopy left” organisations and not competent at engineering analysis and modelling?
Trying to claim credibility by asserting one is part of a “sensible centre” is nothing more than rhetorical spin. I could deconstruct the “sensible centre” concept if you wish but my point is more about you positioning 100% RE (or say 95% RE for that matter) as some kind of unrealisable ideal only subscribed to by hippies who don’t know what they are talking about.
Alastair Leith
Outreach Coordinator
Sustainable Energy Now
In amongst the questions that form the premise of the article could be another question:
if we are providing subsidies to large scale generation of RE including utility solar plants (i.e. to corporations) in the form of LGCs (soon to be quite modest if RET target is not extended past 2020) then should small businesses and householders be able to participate in the same kind of benefits?
So far it’s only 1239 MW of wind, which would be negatively correlated with solar, so there should be enough local demand and transfer capacity for it not to be constrained. However, there are some other proposals not on the list such as Gennex and Kennedy.
Sun Metals have plans to increase the size of their zinc refinery, and securing cheaper electricity is part of their business case.
The North Queensland region is similar to the South Australian region in terms of demand and transfer capacity to the adjacent region, but there are no system strength requirements for local synchronous plant to be spinning. Is this a historical quirk of it being a single Queensland market, or does physics work differently in Queensland to South Australia?
Interesting point Malcolm M. North Queensland as a region would be a similar size to South Australia. It is only because SA is a separate region with the possibility of islanding from the rest of the NEM does it have an arbitrary requirement to have x amount of synchronous generation for system strength.
North Queensland has no such requirement at the moment (despite operating just fine with little synchronous generation), yet one could argue that it is just as prone to islanding as SA.
So if it became a new region, would such a requirement be imposed on it? (to be able to continue islanded in case of interconnection failure).
Regardless of becoming a new region or not, it would be wise to establish the capability to be able to operate islanded. Traditionally this capability would only be done with sufficient synchronous generation, but that is very unlikely to happen at all. So other methods must be explored.
There are alternative solutions. But it isn’t likely to happen with slow incremental evolution of the network. Planners have to step back and consider much more revolutionary changes to the way the whole network operates, and then start planning the changes required to achieve that. Thinking only one or two steps ahead as is the case now will not work.
At some point, the train has to jump rails. The current track is on course to a disastrous dead end. Time to design and build a new track that keeps going far into the future.
Strange comments.
In reply to yarpos:
The purpose of derivatives is to make solar power more reliable and dispatchable. Who are the rent seekers? Perhaps you are also against lending money by banks.
In reply to Shilo:
Solar and wind power is predictably intermittent, but not controllable. Coal and gas are controllable. The proposed derivatives will bring more private funding into solar and wind forecasting. This should also improve your boating experience.
Factor 5 = Early celebration of the 20th birthday of NEM commencement (13/12/1998)?
Ah, Dean, is it sad that we remember the days of NEM start?
Any commentary on the Tesla battery (Hornsdale) and it’s inability to reduce prices in SA during this event?
Follow-up: I came across this report which essentially confirms Kate’s work and also acknowledges further risks.
https://www.aemo.com.au/-/media/Files/Stakeholder_Consultation/Working_Groups/Other_Meetings/ASTAG/371100-ETR1-Version-30-20170919-AEMO-Review-of-Frequency-Control.pdf
It will be interesting to see if this forecast holds. It is the Friday before the Christmas break, and historically, a lot of industry (both heavy and light) closes up shop at midday and opens the beer fridge. I am not sure if AEMO models beer fridge loadings correctly. Further, the numerical weather forecasts from BoM ACCESS-C and ECMWF have the sea breeze front pushing in somewhat deeper across metro areas of Brisbane, Sunshine, and Gold Coasts. By 1600hr (time of forecast peak demand) the sea breeze is starting to push past Brisbane’s western suburbs. Finally, there is a vigorous south east change moving through, and by 1700 it is forecast to be sitting in a line from Southport through to Boonah. Shrug. If AEMO forecasts do hold, then we could see some amazing system loads if there is a hot spell in February. (Isn’t the electricity industry supposed to be in a death spiral?)
At least somebody is discussing this risk. Not likely to be discussed much over at the toxic bubble that is RenewEconomy…
I’d also draw attention to the AEMO SA Electricity Report 2018, which amongst several other worrying statements, has this to say on page 5 within the executive summary:
“AEMO forecasts risks to supply in the next five years in South Australia, due to the region’s interconnectedness with Victoria. To avoid unfairly penalising one region for a supply deficit spread through several interconnected regions, an equitable load shedding principle applies in the NEM. This principle states that load shedding should be spread pro rata throughout interconnected regions when this would not increase total load shedding. Therefore, while the reliability standard is not expected to be exceeded in South Australia, high risks of load shedding projected in Victoria are likely to result in any supply deficit being spread across both Victoria and South Australia.”
Equitable load shedding principle?
Load shedding should be spread pro rata throughout interconnected regions?
Deep state is upon us.
Wow, look at that, 9602MW @ 16:25. And on a Friday afternoon that is part of the Christmas shutdown. I wonder how much the cloud band coming across Brisbane contribute to this peak. For example, the western facing bank of my PV system rolled back from 2462W @ 16:13 to 445W @ 16:18, all in the space of five minutes. (No way I was turning down my air conditioners.)
AEMO final report is out 10 Jan 2019. Eight recommendations, the first and most obvious / important being to reinstate primary frequency control.
https://www.aemo.com.au/-/media/Files/Electricity/NEM/Market_Notices_and_Events/Power_System_Incident_Reports/2018/Qld—SA-Separation-25-August-2018-Incident-Report.pdf
AEMO final report is out 10 Jan 2019. Eight recommendations, the first and most obvious / important being to reinstate primary frequency control.
https://www.aemo.com.au/-/media/Files/Electricity/NEM/Market_Notices_and_Events/Power_System_Incident_Reports/2018/Qld—SA-Separation-25-August-2018-Incident-Report.pdf
AEMO final report is out 10 Jan 2019. Eight recommendations, the first and most obvious / important being to reinstate primary frequency control.
https://www.aemo.com.au/-/media/Files/Electricity/NEM/Market_Notices_and_Events/Power_System_Incident_Reports/2018/Qld—SA-Separation-25-August-2018-Incident-Report.pdf
From the last panel it would seem that an extra 2 GW of low marginal cost wind and solar generation would push the bid stack far enough to the right for prices to be determined by black coal rather than gas. There would be at least 2 GW of wind under construction at present, and with designs that produce well under moderate wind conditions. So it would only take moderate wind conditions in the evening peak to squeeze out gas and peak prices would be determined by NSW coal generators.
I expect there would then be some repricing. The coal units that cycle their output regularly (such as Eraring) would put in low bids for perhaps the first 60% of their capacity, and bids just below the cost of gas for their remaining capacity (used in peak). This would be necessary to protect revenues. Meanwhile the Snowy hydro stations would also need to reprice their output in order to maintain downstream flows, again to just below the price of gas.
“In fact 2018 has seen a generally rising trend in underlying spot prices even as growth in renewable output has accelerated.”
It’s almost as if closing reliable coal fired power stations and increasing penetration of renewables drives up prices.
The claim by the renewable lobby of increasing renewables while driving down electricity prices is nothing but lies.
Why would anyone be surprised that hydro and gas largely set the bid stack pricing when you look at wind output- https://anero.id/energy/wind-energy/2018/june
It’s going to take a mountain of those Tesla 2170 lithium lego bricks to smooth that lot out and keep the lights on but if you can get a sizeable heap of them up and running early you can clearly make a motzah creaming FCAS returns off the top.
Any commentary on the Tesla Battery and it’s impact (if any) on keeping prices down like it was supposed to do?
For the battery, you’d be looking at FCAS prices, which have been reasonable so far. That said, forecast price spikes to $10000 per MWh did not eventuate yesterday.
What I’d be interested to know is what “negative settlement residues” are and why they can lead to constraints being set on inter-regional interconnections? (for example Market Notice 66430 issued by AEMO this afternoon).
Thanks for the article Jonathan, found it very informative. I’m wondering if you might have come across a company called Faraday Grid, and what you might think about their products / claims if you have read anything about them? They claim to have developed power flow control devices called “Faraday Exchangers” (I believe they replace existing transformers) that autonomously maintain voltage, frequency and power factor efficiently in distribution and transmission networks. The Exchanger is also said to generate its own inertia providing greater security in a grid that is increasingly reliant on VRE. I wonder if you think the “Faraday Exchanger” could revolutionise NEM infrastructure, as they seem to claim, and could displace ancillary service providers? Keen to hear your thoughts!
A mountain of Lego Bricks (Tesla) will only increase the overall demand, they are like a water tank with a hole in the side of them. 20% minimum, leaks out to onto the ground. (And no one that i know of has worked out, how far the power coming out of them can actually go. Meaning what really are the longer distance losses from operating them. (Yes I know they can be put near the load, however you must look at the average and when longer distances require power they must deliver it)
Another fun one today, at dispatch interval 16:40 to 16:45 on 2019/01/22 the price SA, Vic and Tas hit the ceiling of $14500, it then plunged to -$1000, and oscillated between positive and negative price for a few intervals. There appears to be no market notice associated with this event, except for “prices subject to review”.
That Tesla battery sure helped…….
As the ‘gap’ between power demand on the one hand, and the some 20GW of power still available from coal, plus solar plus wind power on the other, increases above 5,000 MW, on the east coast, large peaks of gas and hydro output are called on for the difference. Given gas/ hydro have little competition in responding to these peak demand situations, these sources will continue to charge very high prices at such times. Whether the market actually goes crazy or not seems more a function of market characteristics.
New solar power capacity, presently being rolled out, should reduce such high price occasions, during daylight, if the network is up to it. But the naturally high incidence of 5% to 25% wind load factor occasions, will ensure high price periods will continue to occur often, when power demand is high. These occasions will increase too, as coal power capacity is driven out of the market and the ‘gap’ to be filled by gas/hydro, is thereby increased at low wind times.
Storage may become a solution if the increasing cost of grid power is seen as of little consequence.
The situation was extreme, amd also imports from NSW seem to have been constrained, possibly due to a bushfire.
LOR3 declared in Victoria, per market notice 66670号
“The maximum load to be interrupted is 266MW at 1810 hrs 24/01/2019”
“Unfortunately we also see a significant loss of supply from aggregate wind generation that comes before both the sustained price spikes (800MW lost before spikes) and the peak in demand (>1,000MW lost prior to peak demand)”
That is the nature of renewables and why we are in this mess to begin with.
I rarely if ever see the mention of capacity factors for renewables with wind averaging 20-25% of nameplate capacity. so regardless of the fact we have over 5600MW of wind installed, we on average get 1000ish MW and sometimes much less.
Thank you for including the Hornsdale battery information.
It is important for people to see that even large scale battery storage cannot fix these problems.
Is that correct? During peak demand we had 2000+ MW of wind and solar and AEMO can always constrain wind or solar if there is going to be a forecast shortfall…
http://anero.id/energy/wind-energy/2019/january/24
https://anero.id/energy/solar-energy/2019/january/24
Australia wide wind power oputput had been declining all day, hitting the low at 3:35pm (16% output or 939MW) just when solar was coming off its peak.
SA wind power began its decline from 7am. It dropped from almost 70% (1300MW)in the early morinng to 50%+ (100ishMW) output at 9am and then plumeted to around 15% (300MW) from 2pm, falling as low as 4% (80MW)later in the evening.
Victorian wind power output was relatively stable at between 30% (500MW) and 40% (700MW) for the morning but then suddenly declined after 3pm and stablised at the low point of 20% (360MW) between 4:30pm and 9pm which was when demand was rising and solar output was declining.
The fact is that you cannot rely on solar and wind to provide reliable supply on hot days with warm nights as these usually occur when there is low wind.
Wind capacity factor in SA is typically 28-32%, long term average. Go to the AEMO website and download the SA Electricity Reports, that tells the story very well.
Yesterday there was more diesel generation (400 MW) in SA than wind generation (270 MW) around 7pm after the solar generation went to 0 MW.
https://anero.id/energy/wind-energy/2019/january/24
https://anero.id/energy/solar-energy/2019/january/24
2000MW of output compared to a nameplate capacity of 5661MW for wind and 2241MW for solar is 25%. It only got worse as the afternoon progressed.
South Australia was woeful with just 5% of wind output between 7:30 and 9pm, victoria was at sub 20% for most of the peak demand period.
Wind power generation plumeted during the afternoon to a
Does anyone know why the Vic – NSW interconnector has been at zero at various times over the last couple of days?
Bushfires in New South Wales were allegedly the cause.
Constraints during peak demand today were for net flows toward NSW, even while load was being shed in Victoria. That one might take some explanation from AEMO.
Rereading the post, it is likely the same constraint was in operation as yesterday N^^V_NIL_1, preventing voltage collapse on the NSW end should basslink or a Victorian generating plant trip off. Perhaps investment can be made to improve this situation (for example STATCOM equipment at Darlington Point).
Being retired and from the “old school” where things were centrally controlled I am endlessly amazed at the discussion about the complexities that go with this market system. If the control were under one roof and basically up to one person supplied with all they need to be told about the weather how simple thing could be. A list of serviceable plant. Forward planning not much different to a steam engine driver about to go up a hill and getting as much pressure ready as needed. Not rocket science. Now with forward trading – derivatives – no useful integration – AEMO only now seeming to realize that they need to take some form of control there are complexities that are simply created by the design which is not fit for purpose. Thank goodness I don’t have to work with something that I could never believe made sense. You guys have my total sympathy keeping a straight face as you seem to do, or may be being too young to know that it need not be like this. Ever done some work on a washing machine or the car at the weekend and wanting to meet the idiot that made something hard to do. Who was the …. that made it like this?
The Battery is being used like a water tank, with a hole in it. Why on earth could it be cycling on a day like yesterday. WHERE was the power coming from to charge it. WHY was there power allowed to change it??????.
That has to be looked at in detail, because which ever way you look at it, on a day like yesterday, the battery is only making the situation worse. It takes more power to charge it than it can put into the grid, ITS SHEER MADNESS, to allow it to run yesterday!!!!!!!!!
Agreed, just like Snowy Hydro 2.0 which will require huge amounts of energy to pump water uphill and then only be able to provide 70-80% of that energy back when it is needed.
These storage solutions aren’t fixing the problem of renewables and prices
It’s true that Snowy 2.0 is a net load, with no additional generation. What they are hoping is that wind power can be used to replenish the pumped storage, instead of curtailing the wind at nights. Not much being said about the additional transmission line CAPEX and OPEX necessary to ensure wind farms can send their power to the grid without being curtailed, and what happens when the wind disappears for days…
I did not see it “cycling” during the high prices. It discharged at 30MW until it was spent. No surprise there. There is simply a shortage of dispatchable generation. Wind and solar is somewhat predictable although not necessarily dispatchable when required. Fossil fuel generators are generally dispatchable but their reliability is unpredictable. We see coal generation having issues and other generators severely limited in output due to heat. If there was a more storage available in pumped hydro (for example) the system would have been much more stable. High output – but only needed for a couple of hours. Blaming wind,solar, coal or any other form of generation is a bit like blaming who lost the football game. Was it in the last 2 minutes when the team missed that goal and hit the post, or was it the 1st quarter when the losing side didn’t kick any goals??
@ Ian: regarding your post your words are largely correct but the context matters, in this case the context is blackout occurred because of a lack of power generation. Power shortages are a logical result of weather dependent power generation.
Lack of maintenance and zero investment in dispatchable generation are a logical result of subsidised renewables.
The whole scenario is entirely predictable.
But we have commentators like Nicky Ison from The Guardian, with zero technical knowledge in the subject, touting an old AEMO report discussing what it would take to achieve 100% renewables. The report says the idea is rubbish, but the article goes on to say it’s a great idea and perfectly achievable.
What is wrong with this lot?
https://www.theguardian.com/commentisfree/2019/jan/28/what-would-australia-look-like-powered-by-100-renewable-energy
Hi Ben, yes I agree with you pretty much. Paul regularly talks about people at each end of the “Emotion-o-meter”. Some people blame wind and solar (whose output was low which unfortunately does happen at these peak times). Others blame coal/FF generation, eg X units were out of action due to failure, or were limited to X amount of output due to heat (all true and correct). What will it take to change the situation, while the small amount of storage introduced recently in terms of batteries I think has worked well for it’s intended purpose, the amount of stored energy is very small when this type of event occurs. Do you think we will see a change of market rules in the future introducing capacity payments perhaps? We certainly need something to incentivise storage, although if we have many more days like this I think the monetary incentive might be already here!
The 600 transmission constraint equations seem like a black box that are probably only understood by a handful of people within the industry. When a constraint equation is imposed how much analysis is there of risk vs consequence? For example on Friday 25 January Victoria had to shed 200 MW of load, but there was spare generation in South Australia that could have been used if there were a temporary lifting of the 500 MW constraint on South Australian exports to Victoria via Heywood. The physical works on this link were completed over 2 years ago to allow 650 MW capacity in both directions, and an export of 600 MW from Vic to SA is now allowed. But are the risks that the 500 MW constraint is trying to avoid sufficiently great to start load shedding in Victoria? Or would it have been better to try increasing the constraint to 600 MW and monitor its effects. If there is problem, the worst that can happen is it is reimposed to 500 MW and load is shed.
THe low output of wind during Victoria’s price peak needs to be contrasted with the high output from solar farms, which until 5:30pm were producing 200-250 MW. By this time next year this should increase by 500 MW through new solar farms under construction, and completion of the current solar farms. If there is an equivalent amount of roof-top solar, we would have quite a different situation of perhaps another 1000 MW of solar into the Vic market, and a much shorter period of price peaks.
Inefficient infrastructure attracts capital
There is astonishment in the industry about the timing of EnergyAustralia’s maintenance program for one of its Yallourn units, and concern about the ability of both Yallourn and Loy Yang A to provide “firm” power when needed, particularly given their owners intend to keep them operating for another 15 and 30 years respectively.
But there is also anger at the bidding practices of some of the big generators in the lead-up to the heat-wave and over much of summer, something that has been building up for the last six months following AEMO’s warnings, the spike in futures prices and the scarcity of contracts available on the market.
https://reneweconomy.com.au/fury-over-generator-bidding-this-time-they-have-gone-too-far-39712/
The coal generators had been running at 100% output for the last 2 months due to the closure of Hazelwood.
Blaming coal plants for not being able to run at 100% when wind and solar are only able to produce on average 20-25% output and aren’t dispachable is disingenuous.
reneweconomy is a garbage fire of a website and should be ignored.
According to an article in the Herald Sun, AGL had just spent $136m on preventative maintenance on the generator unit at Loy Yang A that had just failed. There must surely be some questions within the company about the returns on such life extension investments in coal-fired stations, when similar investments in solar or pumped-hydro would be much more reliable. The Loy Yang maintenance is $0.25m per MW, compared to about $1m per MW for a pumped hydro that would probably last 30 years.
Hi I read yesterday on http://joannenova.com.au/ that the cost for electricity in Vic and SA was nearly a billion dollars FOR ONE DAY! Amazing and one poster there claimed ““Three days and you could buy a HELE plant with the money wasted.” That’s a power plant that could last 70 years, and provide electricity at under $50/MW”.
Renewable Wind and Solar was developed around the time of the first coal-fired electricity generator circa 1895. It was very limited engineering then and it’s the same now!
Paul, is there any room in your work to look forwards and consider this scenario occurring in 5yrs time? There is another 2100 MW of wind committed for VIC in the AEMO generator register, which will be in addition to the existing 1050 MW of wind. Back of the envelope calcs tell me that if existing wind capacity was delivering 400 MW (Thu) and 650 MW (Fri) during peak demand, then a total of 3000 MW wind would have delivered around 1200 MW and 1950 MW for the same wind conditions.
This additional wind capacity is already being planned, and if > 42 degC is about a one-in-three year event we can expect similar conditions to happen again.
Ben.
You have to consider that the wind farms in place now are at the best locations.
The more wind farms we install the worse the actual output will be because they are going to be located in even more subprime locations.
You also have to consider that while the average output (20-25%) is useful indicator, the system is only as strong as its weakest link which means the numerous periods where we have 24 hours or more at less than 10% output need to be the focus point.
Just look at Victoria wind output for 21, 22, 26, 28 and 29 January. It is way lower than output on the 24th.
Bear in mind that Victoria has 1740MW of wind farm nameplate capacity:
The period from 8:15am-12:15pm on 21st the maximum wind output in Victoria was just 40MW
The period from 6am-10:40am on 22nd the maximum wind output in Victoria was just 23.8MW
The period from 9am-12pm on 26th the maximum wind output in victoria was just 20MW.
https://anero.id/energy/wind-energy/2019/january/21
https://anero.id/energy/wind-energy/2019/january/22
https://anero.id/energy/wind-energy/2019/january/26
https://anero.id/energy/wind-energy/2019/january/28
https://anero.id/energy/wind-energy/2019/january/29
Batterys, and heaps of solar and heaps more wind will fix it. How any one makes money from it, does not matter, because the bigger picture is that the world will be saved from catching fire, even though the sea will be so high everything will be covered with water.
An observation. The price for the NEM electricity grid is very volatile, for a number of well documented reasons.
It may be comparing apples to oranges in some ways, but the Western Australian stand alone grid is much much more price stable.
Paul do you have information and data about the WA electricity grid ?.
I understand you are very busy, but the differences may be of interest.
I noticed that on those hot days where Victoria was struggling, there were a few occurrences of NEGRES CONSTRAINT on the interconnector between NSW and VIC. I don’t quite understand what that means, but from the little bit of reading I did at the AEMO web site I get the impression that flow from one region to the other is intentionally reduced because of some sort of accounting consideration even if a region really really needs the electricity to be flowing on the interconnector. Did this cause VIC to be under-supplied on the 25th of January?
I guess another reason for the low flow from NSW to VIC on that day might have been the fact that MURRAY power station was going hell for leather, and that meant that there wasn’t enough capacity on the NSW to VIC interconnector(s) for NSW to export to VIC?
Perhaps you can write a brief article that helps us to understand why NSW wasn’t exporting to VIC at VIC’s time of need?
The big change over the next year will be an increase in solar into the Victorian market. Over the last few days large-scale solar has produced 200-250 MW until 5:30 pm, and there is another 500 MW under construction in addition to rooftop solar. Add to this another ~200 MW nearly complete in South Australia (Bungala 2 and Tailem Bend), which would be able to help out Victoria if there is capacity on the interconnector (which there usually is around 5:30 pm).
Utility solar cracked 250 MW once in the last 7 days, typical is about 170 MW peak. You are correct though, that more solar will help the grid for 6hrs between 9am and 3pm. And of course rooftop solar will continue to increase, with around 900 MW peak now. And solar is relatively predictable, much more than wind.
And wind is the big problem for the grid, in my opinion. Another 2000 MW committed for Victoria, eroding competition when its windy and rarely available for peak demand.
I wish people would stop talking about batteries when discussing bulk power delivery. Batteries are good for small power islands, short duration low capacity backup and frequency stability services.
And even 10,000 homes with a 5 kW battery is only 50 MW total, and only short duration, so even time shifting loads or peak shaving is a drop in the ocean compared to Vic peak demand of 8,000 MW.
Too bad demand during the 24 and 25 January peaked until well into the evening so solar won’t do squat to help that.
The loss of coal generator units is going to happen more and more as we run them at 100% to make up for the closure of other coal plants (hazelwood) And the lack of reliable supply from renewables.
Of course we will get people like Di Natale blaming coal but the reality is that when we had hazelwood in operation there was spare capacity so you could take a unit offline for repairs and maintenance and they only had to be run at 80-90% output so they wasn’t breaking down as often.
Look, Wind/Solar and Batterys, do not have any issues, they work, when the sun shines, or the wind blows, it depends on the amount of wind, and direction, also the amount of sunlight and temprature, as well as how clean the pannels are, the battery’s work when ever they are needed, not as good when its hot, and as they age, but they still go. Now all of this stuff has to go through inverters and they work better when its cool, not so good when its hot, and the inverters can maybe at times not work, but there will be so many, you will not miss a few here and there not working and they will not be down very long because they will just be fixed quickly. So all in all, RE, just keep working all the time, they never stop. Vs Coal plants that stop. Cannot anyone understand that.
John, no rational person despises wind farms and solar panels, but a power supply network has to be able to supply electricity to match demand. When demand increases, either the power generation increases to match, or a storage option is called on, or the power is not delivered (blackouts, load shedding, demand management).
There is a place for demand management (as long as it is voluntary and doesn’t cost more than generation) and there is a place for storage (as long as it doesn’t cost more than other forms of generation and does not reduce the overall capability of the power supply network e.g. harmonics, fault level, inertia, reactive power).
Wind and solar cannot increase output on demand.
Power electronics (inverters) cannot deliver fault current, and only the latest technology can support frequency.
A business cannot operate when its power source depends on favourable weather.
Trying to compare a 40yr old coal plant suffering tube leaks with vagaries of the wind is not a suitable comparison.
If we are going to crunch the numbers…
Clean Energy Council – $10b wind and solar investment in 2017
Environment.gov.au – 2.8% emissions reduction in the electricity sector across 2017
That’s $1b per 0.28% emissions reduction
Wind and solar generation increased by around 3000 GWh in 2018
That’s $1b for every 300 GWh increase of wind and solar
Australia’s total electricity consumption is around 200,000 GWh per year
The ALP 50% renewables target is therefore 100,000 GWh
Current wind and solar total generation is around 16,000 GWh per year
100,000 – 16,000 = 84,000 GWh remaining to achieve 50%
84,000 GWh / 300 GWh/$1b = $278b
If they do that all with wind and solar over ten years, this will cost $28b/yr
Not including any costs for transmission lines or substations or SVC or reactors or condensors or batteries…
And not touching emissions from agriculture or transport or industry…
Ben those numbers are quite large. But if the will is there sometimes a way is found. Germany is also still going forward with a lot of will.
next 10 years its going to get interesting.
But I will guess this, if the will falls short, it will only be, becuase of this or that, and it can still all work, it just need this and that!!!!!!.
John, investment in generation technology and assets has to be performed at the lowest cost that meets the system requirements. Government subsidies have effectively removed the function of the market.
Despite the massive subsidies and investment in renewables:
Electricity cost to consumers has increased
Electricity reliability has reduced
Electricity sector emissions have reduced
I put it to you John, what in your opinion is the priority for a country’s electricity system? I suggest that “reducing emissions” is not the core business of an electricity system.
Grid
Appreciate the analysis Paul, and I’m looking forward to the generator report card. Many thanks – Jeremy
We are keen to deliver it to you, Jeremy. A load of data crunching, analysis and explanation to be done before we can do that, though!
Hi Ben, I think number one is, that we need power, I think number two we need as close to the lowest cost we can get in the world. We with coal had number one sorted, but failed with number 2, even though we could have had the cheapest in the world.
The path we are going down now, I just hope gets number 1 covered, because useing history as my guide for number 2. We are not going to have cheap power.
Hi John, I agree completely. Priority 1 = delivery of power when it’s needed. Priority 2 = lowest possible cost.
These priorities are closely linked through cost-benefit, with the largest deciding factor being the cost of not delivering power when needed. If the cost of blackout is less than cost of power, then power cost is too high. The reverse is where we normally sit – cost of power is less than cost of blackout, hence the NEM 99.998% reliability target, normally most at risk by network issues (much more wires than there are generators).
The “renewables at any cost” advocates and lobbyists want greater reliance on interconnectors and try to dismiss reliability concerns with “distributed generation”, except then we are at the mercy of a greater percentage of power in transmission lines PLUS the vagaries of the weather.
This is not a rational objective.
Hi Ben,
Yes plus just how much needs to be maintained its simply mind blowing.
I have been trying to work out Rooftop Solar, and am wondering if you might know about how far it can travel within the grid.
So by that i mean, if say all of one area at midday is awash with solar, and the solar has to travel backwards through a transformer to the go where ever it has to go, to then go through a transformer again, and to another person house.
What are the losses going backwards through a transformer, and also what distance can it go, once it has gone that way.
So lets say the transformer it goes through is 400 meters away, to start with, what would that loss be?, then being stepped up, what is that loss, then say another 400 meters to another area, in HV, then back down through another transformer, what is that loss?, then to someones house, say another 300Meters.
And for fun lets say that houses demand is 80% to charge a Tesla Powerwall battery?
What amount of solar power is going to make it into the battery?
Then what amount will make it out of the battery.
So all up the solar power only has to travel 1100meters?.
(I am going to guess the % is going to be lower than going just one way from a power station that is designed to deliver a long distance to a lot of people)
Good info. Could any updates please include a quick look at:
– interconnector activity, including any generation change in NSW that would have affected QLD loads
– coal and gas performance during elevated temperatures i.e. any derating occurring
– which gas peakers stepped in and any abnormal activity there i.e. fail to start
The system seemed to work well and coped with both high ambient temperature and high demand, with technical and economic components operating as required.
To answer your question at C1, the loss could easily be due to temperature derating. If the effective cell temperature of the solar panel is 60ºC the derating factor could be around 0.82 so your 5kW output is derated to 4.1kW On really hot days the cell temperature could even be up to 70ºC which would derate even further.
So long story short, renewables weren’t able to supply the demand and exports from Vic were lower than expected and couldn’t make up the difference.
That meant the domestic dispatchable power with a bid at $13,000 was required, sending the prices skyrocketing.
The answer is no more intermittent power connected to the grid.
Hmmm… seems like you overlooked Contributing Factor #4?
If you’re holding a hammer then every problem looks like a nail …
Snowy loaded up Upper Tumut (gatekeeper), then swapped volume to Lower Tumut (contributor) – they made V>>N-NIL_HG bind and imports reduce.
Looks like how someone would explain economics or something that is really is too complex to work properly.
Just my opinion.
It would seem that the system has lost its way on price. Things used to be far more stable and the price much lower.
Day by day its getting more and more all over the shop and the price just keep creeping up, yet i read how well everything is being run, and AEMO is doing such a wonderful job.
Agree on the complexity, not sure why there are separate prices in each when all states are connected.
I think criticism of AEMO is a tough call, since they just have to deal with the policy, without being able to influence it or influence any investment decisions. Although I believe the CEO panders to certain interest groups unnecessarily and against the intent of the AEMO reports that are escalating the language around reliability and cost.
There is no indication that cost of generation will come down at any stage.
Sorry Ben
Did you read into something I’d written in this article that was a criticism of AEMO?
Paul
Hello Paul,
I think Ben was commenting on what i said
Hello Ben,
Who then has the control over the price. Someone must have some kind of input into it, I am saying that is AEMO, they are the Pilot.
John,
On what basis are you saying AEMO is controlling the price? Have you read this article? AEMO dispatch generators according to their bids and (generally) in line with the market rules. What you are saying is like blaming the ASX for the stock market for being too high/low/unstable.
Just because something is complex does not mean it cannot work properly.
Interesting that the LOR was cancelled yesterday but is back up again today with a 5 hour peak price period forecast in Vic and SA, similar to the 24 and 25 January 2019 event.
SA and VIC are set to experience another 5 hour price cap event this afternoon.
If this occurs it means that in just over a month there have been 2 events that will wipe out 6 months of predicted savings.
That assumes these events do not continue which is highly unlikely.
Residential customers are extremely unlikely to benefit from spot pricing.
SA and especially VIC are sitting on All time high for a year of energy. With SA, the question has to be why?. What has gone wrong with the planning for renewables? They have known about them and must have planned for them, and have had a long time to plan for them, but the price is SA just keeps going up.
John, the price is high because the penetration of renewables reached a tipping point forcing the closure of cheap coal generation.
With Hazelwood no longer operating, SA and Vic can’t reliably meet the demand in the afternoon/evening peak if we have a low wind day.
Just imagine what will happen when Liddel closes in 2022.
But, renewables are cheaper?. In every single study done from CSIRO, to banks, to industry.
It is more than just coal being shut down. It is a lack of control.
John.
Renewables are only cheaper when you exclude the capacity factor which is about 3-4 times lower than fossil fuel generation.
If you multiply that through the levelised cost you get a true cost comparison which shows renewables are way more expensive than coal and gas
Yet another crazy price day. Its just nutty. The whole pice system is CRAZY.
I hear you about the market complexity vs physics argument (twitter chat) but those extreme prices don’t seem to be triggering investment decisions. I suppose the RET has somewhat superseded price signals…
‘I flagged a propensity (by us all) to reach for some Magic Wand (or Silver Bullet) as the answer to the tricky problems inherent in this energy transition’….
‘I’d like to just reiterate that the prime objective of the Energy User in the electricity grid is not to “keep the lights on”, but rather to focus on the result of whatever useful work they want the electricity to do in the first place.’
Ah yes the big picture considerations and the raison d’etre for that energy transition in the first place-
https://www.dailycaller.com/2019/03/01/tucker-carlson-wrecks-renewable-energy/
No magic wand indeed.
Paying a business not to use electricity seems completely counter-intuitive. I understand that if generation cost goes high, that a business might have a threshold beyond which it is cheaper to stop producing. But surely this is cave dweller stuff? How can we justify developing an electricity supply system that deliberately restricts business output?
“Hi, sorry, we can’t make that coffee for you right now, we are practising demand management, since the cost of power is greater than what we can sell the coffee for…” That’s not a helpful conversation.
(And I know it’s a facetious analogy, but I’m practising humour demand management… 🙂 )
I enjoyed that write up, good job thinking outside the box. It is correct to point out that recent planning reports have escalated their tone and language around reliability.
So, a separate market for demand response sounds ok to me. What about a separate market for semi-scheduled generators?
Hi Ben, would you be able to help me understand what you mean by “separate market” please?
There is only one dispatch process that runs every 5 minutes.
Hi Paul, we already understand the AEMO bid stack where the power supply tenders are accumulated using the lowest price first, then the next lowest option until supply meets demand.
I was visualising two bid stacks, one each for dispatchable sources and another for non-dispatchable sources. The dispatchable demand would be total demand minus the non-dispatchable supply.
I believe this would calibrate each market correctly based on their true value. It would also remove what I believe is the masking effect of the RET on dispatchable investment signals.
I haven’t bottomed this idea out in my thinking yet, so I thought I would drop it here where somebody might give me some deeper insights.
Thanks for reaching out.
I think the cost of an underinvested network should include the cost of blackouts due to capacity shortage. This is logical because an overinvested network shouldn’t have a capacity shortage.
How you’d measure or estimate the cost of a blackout I’m not sure.
I’d like to hear perspectives on the cost of network not related to demand, but attributed directly to reducing congestion from intermittent generation or facilitating new intermittent source connections. The demand is not increasing (rooftop PV) but the networks are…
Absolutely. The cost of blackouts is not just limited to the cost of fixing the infrastructure and getting it back up to code, but also the serious flow-on effects of power outages to the broader economy which will not occur with over-invested networks.
On the second point, the impact of increased intermittent generation is a major concern for regulators. As well as the impact on dispatch forecasting which I mentioned above, the AEMO is concerned about:
-the effect of DER generation in excess of regional demand during periods of minimum demand. There is no current mechanism to constrain down generation from DER;
-the progressive displacement of synchronous plant by inverter-connected DER plant removing inertia from the system. This challenge for system security and reliability was a major concern for the Energy Security Board in developing the National Energy Guarantee as well;
-the challenge of maintaining voltage within technical limits without visibility of DER.
Dealing with DER is going to be costly. Whether that cost is best passed on to retailers through a National Energy Guarantee is another question
For more discussion see the AEMO report (https://www.aemo.com.au/-/media/Files/Electricity/NEM/Security_and_Reliability/Reports/2016/AEMO-FPSS-program—-Visibility-of-DER.pdf) and the AEMC report (https://www.aemc.gov.au/sites/default/files/2018-09/ERC0227%20-%20DER%20register%20-%20Final%20Determination%20%28final%29.pdf).
Thanks Drew (for the original article and the response)
The AEMO report has some particularly interesting language around conservative safety margins and larger reactive reserve thresholds – all reducing efficiency of the system.
It seems the current market and regulatory environment are not set up to pursue the most efficient system of generation, transmission and distribution.
I find the “gold plated” argument frankly annoying. For starters, those who perpetuate the myth clearly have no knowledge of the external boom, bust, boom cycle that caused these gyrations in network investment.
Unlike Drew, I was there. I doubt Drew has ever stood beside a 50 year old creaking 110kV switchyard running above full load, hoping like heck the nearby circuit breaker wasn’t about to explode into a fireball, and wondering if he would be alive to see his wife and kids at the end of the shift. I did. More often than I cared to. (So, please excuse the passion.)
It is not a problem these days, but it sure was back in the summer of 2004. Every night, the local TV news would broadcast pictures of power transformers in Brisbane suburbs being sprayed with fire hoses in a desperate attempt to keep cool. How badly overloaded? Well Albany Creek had 30MVA load on two by 12.5MVA (25MVA firm), and these transformers were originally 10MVA uprate to 12.5MV.A with fans (no oil pumps).
Or how about 260MVA daily summer peaks at Victoria Park substation on two by 120MVA (240MVA firm). Note this is well above N-1. Even better, one of these 120MVA transformers had an electrical age of 50 years and wet windings. But under the severe capital rationing that Energex was under at the time, there was no funding to replace it. At the time Victoria Park supplied critical load like the Royal Brisbane Hospital (arguably Queensland premier hospital), St Andrews Hospital, and Brisbane Central Hospital, and a quarter of the Brisbane CBD. I could go on, but you get the idea of what the network was like in 2004.
Which is a nice segue into the AEMO report that Drew mentions. Why does the AEMO report select 2006 as the start year? My view is because it slants the story in the direction its author’s want it to. Why not start the comparison in network expenditure from, say, 1996? Because it tells a completely different story, and one that runs counter to their argument, and deflates the “gold plated” investment story.
But it goes back even further than that. The electricity industry in Queensland had a significant investment boom in the late 1970’s and early 1980’s for an anticipated mineral boom that never really eventuated. (It did, 25 years later.) This lead to severe capital rationing in the mid-1980’s and coupled with a strong push on cost savings via use of contractors, lead to the SEQEB dispute of 1985. After the government won the SEQEB dispute, there was even more rationing of capital in the late 1980’s across the entire Queensland electricity industry. Buoyed by their success, the economists kept up their severe capital and operational expenditure rationing during the 1990’s. And why shouldn’t they? After all the engineers were just gold plating the network.
In Energex’s instance, along came Greg Maddock in 2000. Greg was an economist. The short story is that he too believed in the “engineers are gold plating argument” and decided to cut network investment even further.
But something else was happening at the same time. Yes, electricity tariffs were low (around 14c/kwh), but the Queensland economy was moving along at a great clip. Price of electrical appliances, in particular air conditioners, was decreasing. Net migration into Queensland was running at 1000 people per week. Mining investment was starting to increase as well. But investment in the electricity grid did not follow.
Something had to break. And it sure did in the summer of 2004. Yes, the EDSE report provided some details into how bad things were. But it is what happened behind the scenes that explains what happened next. Economists and regulators all over Australia saw clearly for the first time how their “regulation” had resulted in significant under investment and this had taken the electricity network to the edge of the abyss. They peered over the edge into the black(out) below and knew what they had been doing was clearly wrong, and if it continued they were about to be publicly flogged.
So, the economists threw the problem back into the lap of the engineers and said; fix it, money is no object. So, not only did the expenditure in 2005 to 2015 have to match the phenomenal grow of the day, it also had to fix up the under investment of the last twenty years as well. Hence, the investment megacycle.
This is the true reason why the investment in the grid – yes “investment” not excessive gold plating expenditure – in the (very suspect) AEMO chart looks the way it does.
Now, if you think this is a fairy tale told by a grumpy old electrician, I point you to two excellent papers.
Simshauser, P. and Catt, A. (2012), “Dividend policy, energy utilities and the investment megacycle”, The Electricity Journal, 25(3):4, 63-87.
And the classic “death spiral paper”
Simshauser, Paul & Nelson, Tim. (2012). The Energy Market Death Spiral – Rethinking Customer Hardship.
The death spiral paper is more accessible outside of paywalls. Check out figure 3 in this paper (which has CAPEX investment from 1955 to 2011). Note the investment peak in the early 1980’s. Note the investment blackout in the 1990’s. And note the investment mega cycle (Simshuaser & Catt’s term) from 2005.
But, there is more.
This investment mega cycle in the Queensland electricity grid (hopefully the gold plating argument is now looking exceptionally weak) occurred in the midst of a once in a lifetime mineral boom. Not only was coal seam gas (and LNG trains) coming on stream, there was rampant mining activity in Queensland. Mining companies too were chasing electricians, electrical contractors, and things that contain copper and spin. The prior under investment had also result in under investment in apprenticeships. Guess what happened next? Wages boom for sparkies.
Take the price of copper (the stuff transformers and power lines are made of). In late 2003 the price of copper was around $1500 per metric tonne. By 2007 it was near $8000 per metric tonne, peaking just short of $10,000 per metric tonne in 2012. The raw material – copper – that is integral to the electricity network has risen by over 600% !!!!!!
The icing on the cake was the regulator’s decision on how to calculate WACC. Despite being warned by electricity utilities in their regulatory submissions, the AER decided to change the WACC calculations. Then along came the Global Financial Crisis of 2008, and the cost of capital went through the roof. This too added to customer’s electricity bill.
So I see a completely different story. Where some see “engineers gold plating” the network, I see inept and clueless economists causing bust and boom cycles in investment. Where some see over investment in the decade from 2005, I see chronic under investment and missed opportunities (when costs were at record lows during the late 1990’s) to make sensible investment decisions for the future.
As for the future, well, that is another story.
Dr Anthony Spierings is a registered electrician. He holds an Associated Diploma in Electrical Engineering. A degree in Economics, Banking and Finance. Master of Project Management. PhD in Information Systems. He previously worked at Energex for 35 years; including extensive stints in substations, project management, program management, and network operations.
When will a MLF, get applied to rooftop solar??? Bit of a elephant in the room that one.
What is often overlooked is that end users see massive swings in their MLFs and have done so since the NEM began. What is even harder for consumers is that the MLFs are driven not only by generator locations but by their bidding pattern as well.
What we have also seen is that MLFs consistently over recover the actual losses incurred but the reallocation of the residual back to consumers does not reflect the amounts each consumer paid.
Generators have never been interested in minimizing consumer MLFs so I find it a bit hard to be all that sympathetic to the current out-pourings of being hard done by
Anthony, I was ‘there’ too. And I can tell you, at that time in history, gold plating is an apt description. The other difference in our experience was my employer took a little more care with WH&S. Why an earth did they put you in such situations?
Anthony, I appreciate your perspective on the QLD system. Here is one source for the “gold plating” label.
https://www.abc.net.au/radionational/programs/backgroundbriefing/2014-06-29/5549868
Thanks very much for this, Anthony. There is a lot there that I didn’t know about pre-2006 investment and thanks for the links to other articles. It is a good reminder for me (and maybe all of us) that regulators and policy-makers are often partial to a certain narrative and we always need to look at their reports and reviews with a critical eye.
More on “gold plating”…
https://www.smh.com.au/national/nsw/electricity-industry-threatens-farmer-who-dared-to-fight-20121109-293wf.html
https://www.smh.com.au/business/power-sector-primed-for-overload-20120530-1zitp.html
A bit more from one side of the debate
https://reneweconomy.com.au/new-solar-wind-projects-may-stall-in-face-of-network-bloodbath-49402/
Interesting discussion Paul. For my part the way i conceptualise a 2-way market is for demand side (retailers, large users) to bid in their demand in tranches of decreasing prices (an old-style downward sloping supply curve) with non-curtailable load bidding in at the MPC. Retailers’ bid stack would reflect the sum of their agreements with customers. One further advantage is that there is less onus on AEMO to forecast demand accurately, that responsibility is devolved to market customers. This sort of approach would require a lot of retailers, of course, but at least initially, if they wanted to be pure price takers they could.
I’m not sure your idea of mandatory bill info would in itself do much to jumpstart greater customer interest in DSP. I think the ACCC-led process for 3rd party data access is more likely to help – with customers’ permission other parties can take their consumption data and use whatever analytical/presentational tools they like to present the DSP opportunity to that customer.
Paul-Can you or anyone here tell me what the role of the existing and augmented grid will be once almost all or perhaps ALL present grid customers…households and some business…are largely independent of the grid …ie with rooftop solar and batteries formed into VPPS orchestrated by aggregators…as AEMO urges?
Who will be paying for the even more expensive grid then…with no regular customers?
Solar and wind farms will be in the AEMO-designated Renewable Energy Zones presumably with their own microgrids.
Who but big industry will take and therefore PAY for power from the grid,,,and will they be enough to sustain it.
The whole ongoing costs of the grid would surely be a huge impost on industry and ultimately on all of us….and especially on Australia’s competitiveness.
Can you or someone here give us an overview of who will take power from the grid and who will be the regular customers of it in the DER Australia that AEMO hopes for please?
That’s an interesting question, because governments are actively promoting policies that reduce grid demand – rooftop solar, batteries, demand management and efficiency measures – while at the same time incentivising large scale intermittent sources that require massive investment in networks.
It is all starting to get exciting.
Quite interesting article!
I’d agree with HR about the fundamental reason for this spike being Snowy moving generation from Upper Tumut to Lower Tumut. Essentially when V>>N-NIL_HG is binding, shifting 220 MW of generation from UT (coeff -0.3) to LT (coeff +0.2) forces a reduction in Vic exports to NSW of 110 MW. Snowy rebid this way effective from 16:00.
Part of the reason for the export forecasts & price forecasts being out is that the representation of the constraint for predispatch is only approximate, and was overstating the export limit from Vic by about 100-120 MW. My guess is that this systematic gap was as much a factor in the “surprise” element of the spike (and in plenty of others like this one) as the other possible causes identified by Paul.
Just to clarify, the fundamental reason for price volatility on the day was the level of demand and overall set of supply offers for NSW; Snowy’s impact on exports from Victoria was then the proximate cause for the spike occurring at 16:15.
Dear me, all of this simply for the price of electricity. I really think we have lost touch with just how simple things can actually be.
I’ve been struggling for a while with the concept that all generators get paid the marginal price. Why don’t all generators receive the price they bid?
An alternate model would be “pay as bid” as noted here, Ben:
https://wattclarity.com.au//articles/2018/08/different-approaches-for-dispatch/
… but it’s important to realise that generators would bid differently in that case (i.e. to recover all their long run marginal cost, and make a return).
Thanks Paul, I agree that “pay as bid” may tend to increase some bids (and therefore wholesale prices), but if that’s the case, then would you say currently we do not have a market that prices cost of generation accurately?
A nodal model such as in New Zealand could lead to a better dispatch outcome for power stations near regional boundaries. For example the Murray Power station is in the Vic region but is penalised if sending power into NSW by a Victorian marginal loss factor of 0.9 (based on sending power to Melbourne), multiplied by the dynamic loss factor for export from Vic to NSW (which would be calculated from Melbourne to Sydney). Likewise the solar farms near Mildura are penalised by a Victorian marginal loss factor of 0.8, based on sending the power to Melbourne, multiplied by the dynamic loss factor of sending the power from Vic to SA (because Adelaide is the nearest major load centre). What are the downsides of a nodal model, and why wasn’t it used in the NEM?
The diagrams you show of the marginal loss factors are close to the physical layout of most States in the NEM as it was 20 years ago, before the near-border power stations were built, and Snowy had its own region. But has it come time to update it to a system that more closely approximates the physics?
Malcolm, market operators and designers have been grappling with the challenge of accounting for losses and constraints for as long as there have been electricity markets. It was Strategic Issue #2 internationally back in 1997 when I was involved with the Association of Power Exchanges, for instance:
https://wattclarity.com.au//articles/2011/05/6-enduring-challenges-for-competitive-electricity-markets/
There is a trade-off between engineering/physical specificity in locational pricing, and the ability of participants to effectively hedge risks that arise in the market (the current concerns about Marginal Loss Factors being just one manifestation of that trade-off).
With COAG Energy Council having requested the ESB to design “NEM 2.0” for a start date of 2025, I’m sure these 6 Strategic Issues from my APEx days in 1997 will be focus areas for consideration:
http://www.coagenergycouncil.gov.au/sites/prod.energycouncil/files/publications/documents/ESB%20-%20Post%202025%20Market%20Design%20-%20Scope%20and%20Forward%20Work%20Plan%20-%2020190322.docx.pdf
I will say again, the elephant in the room is roof top solar, and paying 100% for the export of that. As every day goes on and more and more is installed and more is exported and all still getting 100% of the export at the silly prices. The higher the cost to the system.
Someone needs to work out some kind of average loss of this stuff coming out of 240 (230volt) inverters and going back into the grid to be used, where???, and if its being used how much of it is able to be used. Keeping in mind that as more and more of it is going into the grid, the average distance its having to travel is further, on average.
Well unless its every second house in every street thats getting solar and on average its only haveing to go next door.
Looks like the teeth are getting bigger and sharper
Meanwhile on 20 March (just 5 days before this topic was posted) wind output was:
Below 800Mw for the period 00:00 until 15:20.
Below 400MW for the period 06:30 until 12:40.
Below 200MW for the period 9:00 until 10:15.
https://anero.id/energy/wind-energy/2019/march/20
Celebrating a peak output is an odd choice given the system is only as reliable as it’s lowest output.
Paul, thanks for your balanced view. Facts are necessary! The complexity of our power system proves that time after time.
I can only see forces imposing upwards pressure on consumer electricity prices. Removing a big chunk of the bottom of the ‘bid stack’ can only force more generation in at the top of the stack. Seems obvious.
The only way to reduce generation cost will be:
– massive over-capacity in low cost intermittent sources (but with massive increased cost of the grid)
– gas market intervention (or huge over-supply of gas)
– large coal units adjacent to their own mine, not exposed to international coal prices, using existing transmission eg Liddell and Hazelwood replacements
The grid is expanding (to accommodate intermittent sources) while the customer base is shrinking (solar PV and batteries). The logical conclusion is that customers who can afford it will get off the grid, the rest will see spiralling costs and companies/governments will face grid asset write-downs and customer subsidies.
Government policies are effectively paying people to leave the grid while paying to build more grid.
It’s going to be ugly in 10yrs.
https://www.esdnews.com.au/new-centre-drive-energy-policy-victoria/
Isn’t the NEM a “free” market?
Private business exists to make as much profit as it can and the Vic governments decision to privatise the network made sure of it.
How much cheaper would the average Vic bill be if it were still government owned and run!
Victoria had the cheapest prices in the NEM for years until the closure of hazelwood.
Privatisation doesn’t drive up costs.
Paul
Thanks for the assessment – I found it very useful.
It is important to note that in its June 2018 report, the ACCC considers the NEM to be highly concentrated and this leads to higher prices than a competitive market might deliver. The Major Energy Users attempted to get the ability to exercise market power that comes from a highly concentrated market addressed but all it got was market monitoring. The issue of gaming the market in Qld was addressed by the owner directing its generators to cease the practice and sudddel]nly it stopped.
We all forget that as demand increases, so does the number of generators making offers reduce and we have seen in other markets that there is a recognition that competition reduces as demand increases and that actions are needed to limit the potential to exercise this market power – something we don’t see in the NEM.
While we haven’t yet seen what the outcome of the 5 minute settlement process might be, there is a real fear that it will result in further market concentration in each dispatch period.
The ACCC in its report came up with a number of sound recommendations to address the issue of market structure and market concentration but there doesn’t appear to be a strong commitment at a Federal level to these recommendations other than “the big stick”, brow beating of some gentailers and getting more coal fired generation in north Qld.
Probably most of all, we need to have a clear policy on carbon emissions and how this will be transitioned to as a nation, rather than the various state governments doing there own thing. Providing policy in this was a recommendation of the ACCC too. See page 101
“The ACCC considers it critical for there to be a stable energy policy in Australia which incorporates the need for Australia to meet its climate policy obligations but at the same time does not distort the way in which new generation capacity enters the market”
Hi Paul, can I add my name to the list of people thanking you for writings. I can’t agree more as a Professional Engineer with your remark about using non emotive wording. The facts are the facts and facts have no feelings. If the rules are to be reviewed then I would drill down to the very bottom. Bids are noted, they are stacked in rank order of price and that part of the stack that matches the load is activated. I am using my words deliberately. Then all the bidders are paid the same price. That is the most illogical process I can image. If I were to buy a string of contiguous properties for a property development I would negotiate each purchase in a confidential manner and try to stop the different property owners from talking to one another. I would expect to pay a range of different prices for what might be identical properties because of the different positions different owners would have taken. If I can and am willing to supply power for $30 per MWh then I think that is what the transaction should take place at. If someone else wants $100 per MWh because they have different cost structures and might even be making a smaller profit but that is their bottom line then that is what they should be paid. I know the model of stack pricing is based on the share trading model where some shares are traded as a result of an offer and acceptance ( the big guys ) while others are traded at the current spot price ( small traders ) and that in this model there is only ONE price at each moment of time but that is shares or could be Brussels sprouts in a vegetable market but it is not electricity. To follow the vegetable model a retailer might have purchased two bundles of Brussels sprouts from two different market to fill his purchase order. The two prices he paid need not be the same. Paul, if I might be permitted to use one emotive word – I think the present model is STUPID. Change that and the price of electricity will be harder to “Game”.
Dr Ken, I have also come to a similar conclusion. I accept that if a generator were to receive income relative to their own cost, then the bids may look different to how they look now. I suggest the overall generation cost would be lower than it is now.
However, is it generation cost or network cost driving consumer prices? Bit of both maybe?
So network cost has to be dealt with too.
Current policy is rewarding consumers for moving away from the grid while at the same time forcing a larger grid to accommodate intermittent sources.
This is divergent policy – paying customers to reduce grid demand while expanding the grid. This has to reduce efficiency, ruin market signals and increase consumer cost.
Complimentary policy would be rewarding those customers far from generators for leaving the grid, while also rewarding customers close to generators (or existing permanent large load centres eg cities) to remain on the grid.
At a macro level complimentary policy is logical. Divergent policy is not.
Paying all the clearing price (as the NEM does) is a standard auction mechanism. Pay-as-bid, which is the alternative you are suggesting, changes bidding strategies. It therefore does not necessarily result in lower costs to consumers. Additionally any evaluation of the wholesale cost of electricity has to include the contract market. The spot and contract markets influence each other – you can’t look at one in isolation to fully understand the dynamics. Given you started by agreeing with Paul about “using non emotive wording”, it’s unfortunate that you ended by calling a model you clearly don’t understand “STUPID” (sic). I’m not saying there is no case for pay-as-bid, I’m saying you haven’t made one.
Paul, I’m not sure what you missed in your statement, “I did not the broader recommendations in the ACCC report with respect to rooftop solar”. As I understand it the only two recommendations about solar were for states to take on board the cost of the excessively generous feed-in-tariffs and to terminate the Solar Credit scheme in 2021 rather than wind it down progressively until 2029. ACT electricity electricity consumers still have 10 plus years of paying $15 million annually towards eating the subsidy on these schemes on top of the cost of solar credits.
I think this is a very reasonable way to write about a complex situation. But I think the take-away would tend more to the John Quiggin view of utility function pricing and economics than any other: if we want to constrain market pricing of electricity supply then we should be asking some pretty big questions about why we had a market in the first place. If the market pricing is not within bounds which the consumer is willing to accept (and I think “everybody wants it cheaper” is a bit flippant, yes, true, but we’re also open to realistic pricing:cost relationships and what is not being exposed is the cost:price differential, or the curtailment costs, and we’re certainly not being offered price discounting for supply variances except in the case of the very large industrial consumers)
The Market in electricity wasn’t created to make market value: it was made to try and avoid state capital investment costs by inducing commercial investment for a long tail profit. The price controls were meant to satisfy consumer needs. They patently aren’t working, but the legalisms around varying the model are escaping the ACCC. Regulators that don’t want (to be seen to, let alone actuall) act, is very distressing.
So I too wouldn’t entirely blame the generators: Acting within the rules to maximise profit is not “rorting” but it is “distorting” and I think this kind of price distortion should be controlled, and constrained.
Am I banned from this site?
Ben…intermittents as it stands now are paid the same averaged price as other bidders even if they don’t deliver are they not?
So when intermittents fail to deliver they leave a shortfall and the extra bids to meet demand would be higher than those of unfirmed RE so the averaged price would be higher?
Intermittents can’t do without coal backup ….but the government-mandated priority dispatch for intermittents [ that are favored by law in the merit order process despite the fact that they’re not dispatchable]…. relegates coal to standby mode for ramping up to rescue the NEM when RE fails…wherein coal plant is economically and physically damaged …metal fatigue…tube leaks …boiler and chemical problems etc…as well-documented in MIT studies and many others.
Because COAL is indispensable right now…without it Australia would be blacked out…it must get a high enough price so shareholders will agree to its sticking around to be used and abused in these ramping up and down ways it’s not designed for.
Many RE enthusiasts try to make Australians think the increasing maintenance problems coal plants have are due to their being old obsolescent unreliable technology…in order to ensure Australians are gung ho for getting rid of coal.
Nothing could be further from the truth.
But who will pay the operational costs of the grid when household customers are thin on the ground?
No one will answer that question.
Truth: I’ll attempt to paraphrase your first two paragraphs, so we can see if I understand your point.
– the design of the market ensures that all the contributors to the 5min period, including low cost intermittent sources, get paid the same as the highest contributor – I think we both understand that equally
– your point about the effect of an intermittent source “not meeting its bid” is interesting and i’d like to know how often this situation occurs – obviously there are forecasts for weather dependent generation, but your question seems to be aimed at how often these forecasts are wrong
– I think its a fair question, but I also think to be completely fair, then you would have to assume that there are occasions when the weather is both less favourable and more favourable for generation i.e. sometimes there will be less generation than forecast, sometimes there will be more generation than forecast
– AEMO will have the data, I’m not sure if it’s publicly available
– the effect on wholesale price I think would be marginal, since the forecasting error would usually be less than the volume of marginal generation eventually setting the pool price
Your final sentence rings true – that we currently have contradictory energy policies resulting in less customers (rooftop PV and batteries) AND more gird to accommodate widely distributed intermittent sources.
I think the only way to stop network price increases is to freeze network investment, then write down the asset base.
Generation cost reduction (under the current market) needs lower cost marginal generation across all the 5min periods or rule changes (market change) to have generators paid what they bid.
No option seems to be on the table with any political team.
BM: Qld coal plants are government-owned and they WERE gaming the market under government ownership until the QLD government was shamed into putting a stop to it
The NEM is anything BUT a free market…there’s too much government intervention for that…not the least in its mandate that intermittents must be favored in the merit order for dispatch even when they’re unfirmed and therefore not dispatchable.
That changes everything…as everything IS changed…rules…enforcement of safety standard requirements for Li-ion batteries…everything…to prop up RE and damage COAL.
You say….’despite the enormous volume of discussion and debate about the climate change, and the causes of it all over the airwaves and internet (and even a full length movie or two) over the intervening 10 year period. I find that staggering, and quite explanatory.’
I find it staggering that you could think that has happened…..especially the reference to the movies.
Questioning and dissent on this issue was shut down from the moment the dodgy consensus was declared decades ago…when we were told the science was ‘closed’ …before most of the science was even done…before clouds and oscillations were researched….when temperature of 70% of the earth system…oceans…were not even able to be reliably measured…long before the advent of the Argo float system in 2005…at which time they found there was no significant rise in ocean warming….and the earth cannot be warming significantly if the ocean is not.
Other explanations for the slight warming that there is….are infinitely more logical and likely than the CO2 hypothesis…but any scientist …no matter how highly-credentialled who questions the political hypothesis that’s designed by warmists as a means to a political END…any scientist who offers alternative theories ….must risk his job and future to do so.
If the debate has been so open here , why did BoM so fiercely resist an audit when Australia’s PM wanted that done…shortly before that ‘problem’ was solved by COUP.
The universities’ massively politically-charged treatment of dissenting scientists and their views has been a disgrace to Australia
Why are the Australian universities so militantly closed to alternative views on this absolutely seminal matter of science? Nothing could be more vital to get right than this….and yet science is being unprecedentedly corrupted in order to prevent anything but the politically correct dogma …..that just happens to require the suicide of Australia as a 1st world nation ….to drive our public policy .
If a foreign enemy had somehow gotten hold of the energy policy levers in Australia in order to deliberately destroy our country for the foreseeable future, they could not IMO have devised a more existential threat to Australia than this ‘transition’ from simplicity and efficiency and affordability to unfathomable ad hoc…cut and paste complexity.
The problem of growing wind is that the absolute power gap left when the wind drops off on a national scale, also grows. When this occurs at times of high demand and with coal/gas being driven from the market also by increasing wind/solar, what is there to fill such ever larger gap events? Such shortfalls at times of low wind/high demand seem to be in part behind a number of the soaring spot prices in the last six months. Even $50 billion of grid changes and costly storage could only do so much. Disappointingly, it seems more wind = higher electricity prices.
A physicist weighs in with some calculations and analogies with the really big picture-
https://www.manhattan-institute.org/green-energy-revolution-near-impossible
It pays to keep reminding ourselves that electricity is a means of delivering energy and it’s EROEI that rules it all and replacing every ICE vehicle with battery stored energy is only adding to competition for battery production and output. It would appear from Mr Mills’ calculations that as fast as all the Gigafactories we could muster for the DUAL task to get rid of hydrocarbons altogether they’d be also be trying to keep up replacing spent ones. That’s before we even deal with all these spent incendiary items that at the moment are predominantly going into landfill unlike the recycling of lead acid batteries.
John: I suggest you read this article from Stanford University that talks directly to yours and the original posters statements. https://quillette.com/2019/02/27/why-renewables-cant-save-the-planet/
or a video if you prefer https://www.youtube.com/watch?v=N-yALPEpV4w
I am amazed that wind and solar generation which was invented in the 1890’s, along with the first coal-fired generator is thought to be the solution. I want to go forwards not backwards to intermittent energy. Nuclear energy and hopefully Thorium is the way to go.
I also think electricity should be much cheaper if we are going to power our planet.
Hi Michael, I just came across this article (April 2019…!), thanks for sharing it. The benefits to the end user and the generator from contracting power seem obvious, I guess also the retailer when the difference between spot and contract works in their favour.
Do you have any insights into what effect the contract market has (if any) on the spot market? Does contracting power tend to reduce spot prices or increase them? How much of the market is contracted at any time?
Also I am trying to understand how the gentailer arrangement with contracting power is affected by / affects the sport market.
Any insights appreciated.
Hi Ben,
Thank you for your feedback on the article.
The contract market would have an impact on individual generator bidding strategies. Generators would typically bid more competitively up to their contracted volume and then more aggressively (higher prices) for the uncontracted proportion of their generation capacity.
I cover this in more detail in my book Power Profits that you can find here https://www.amazon.com.au/Power-Profits-Comprehensive-Framework-Electricity-ebook/dp/B07C4HB5XL
I’m not sure how much of the market is contracted at any one time but I would expect that it would be a very high percentage. The number of users fully exposed to spot prices is quite small but has been growing over the last few years. Some end users have spot exposure plus some contracted positions (hedges).
In regards to gentailers, my observation has been that the retail arm of a business is not necessarily given more favourable contract prices to any other party that wishes to bilaterally contract. “The price is the price”. That is why retail prices offered by competing retailers are often very close.
Michael Williams
Altus Energy
Thanks Michael, I have your book on my list now 🙂 I read your bio and it’s safe to assume you are aware of the ongoing debate about renewables, cost and reliability.
In terms of the spot market and the bid stack, I can see that more low cost generation at the bottom of the stack should push some higher cost generation off the top of the stack, so the spot price is set by a lower cost generator. This will result in a higher margin for the retailer in the periods when SRMC is low, so retailers like this to happen. On the flipside, when intermittent VRE is not available, the SRMC is expected to be higher.
Since there are market caps for high and low prices, but the high cap is far higher than the low cap, it seems logical to me that retailers would hedge more against the periods of high wholesale prices than against low prices i.e. there is risk to retailers on the high wholesale price side, but rewards on the low side.
If this logic holds true, then intermittent sources result in higher retail prices, and the market cap on the high side can act like a target (not a cap) for SRMC generators.
So two questions (finally):
1. is it appropriate for intermittent sources to bid apples-for-apples against dispatchable sources in the same market?
2. if the market high cap (target) was removed, would extreme pricing be prone to self limiting?
Cheers
Ben
Does all this mean that Queensland,particularly South East Queensland, should switch to daylight saving in the summer,thereby shifting the demand an hour earlier and make more use of the solar generation?
Mr Mills, is but presenting the glass half empty. Lets see just where we can get resources that the group of people that were happy to believe Oil/Coal/water/food and just about everything else was going to run out, yet, are now happy to believe that there are enough resources to build all the stuff they think will give us zero Co2. I find it amazing that all of a sudden all of the minerals that are needed for RE are not even talked about.
Has any of them added up the tonnes?????. or guessed at them. Wind turbines use a huge amount of steel, and there will need to be a bloody lot of them.
(They were so wrong about Oil and coal running out it was a joke, yet steel for wind turbines is a snap. The coke and iron ore)
Hi, wondering about a couple of questions on the combined solar tracking installations vs rooftop solar:
– average capacity factor (all periods all stations)
– capacity factor per hour of the day for all seasons
– contribution to peak demand e.g. the 2017 AEMO SA renewable energy report discusses the 85th percentile of peak demand periods only receives 1.6% of solar nameplate capacity
– stations with highest and lowest capacity factors
Cheers
Ben
Ok so its not wind, which when it is hot normally goes to close to zero output, because there is no wind.
Solar as it goes over 25c gets a little less, so when hot a bit more.
Coal plants do
Gas plants do
Battery’s once hot after discharge, struggle to recharge if hot, so their 24 hour cycle will go down when its hot
Inverters suffer when its hot, both ways, the cooling needed in them goes up a lot (Fans), so to charge a battery must get a double whammy, the battery struggles in the heat, and the inverter to charge it and then to re-send its power back out at the set voltage.
So which one do I pick, hmmmmmm
Steam turbines – constraints due to ambient temperature can manifest in the condenser hotwell (impacting STG condenser vacuum), the cooling tower (also vacuum), the lube oil and jacking oil coolers (heat exchangers). Ambient temperature of 45degC doesn’t have much effect on a boiler operating at hundreds of degrees.
Gas turbines can be directly constrained by high intake air temperature – like a turbo vehicle GTs are sensitive to density and exhaust temperatures.
For a steam turbine plant there are some external factors not directly related to the equipment that can force reduced output. The most obvious being cooling water discharge temperature – if that is tightly monitored, as it is on discharges into rivers and lakes, then reducing load might be the only option to keep those temperatures within their regulated environmental thresholds.
My guess: Loy Yang A or Kogan Ck
Ben it is more likely Millmerrian than KC. Just my opinion. If its one of them and not a battery/ inverter change discharge cycle in hot weather
The behaviour of the Mystery DUID looks very much like what happens at the wind farms around Goulburn in NSW. My own analysis of weather data and AEMO data shows that Taralga, Woodlawn, Gunning, Gullen Range, and Cullerin Range all have maximum power versus temperature curves that look a lot like the Mystery DUID.
Solar farms show almost constant maximum a.c. power output as a function of temperature. Probably because they tend to have oversubscribed d.c. power and so the a.c. output can still reach registered capacity even on very hot days.
The Mystery DUID sure ain’t Kogan Creek. It managed to provide about 94% capacity factor at 40C in 2018. Both units at Millmerran have been at about 82% capacity factor when the temperature was 40C in 2018. Keep in mind that Paul’s graph shows MAXIMUM power, and so while some coal DUIDs have been known to suffer from heat stroke, they seem to be able to keep going some of the time when it gets hot 🙂
You will have to give is a clue so we can narrow the field Paul!
Correction to my previous post: For solar farms the “registered capacity” in AEMO’s Registration & Exemptions List appears to be the d.c. photovoltaic module total capacity. “maximum capacity” appears to be the solar farm’s maximum a.c. (inverter) output power. When I said registered capacity in my earlier post I meant the thing AEMO calls “maximum capacity”.
Hi Mike, How about the Big Batt, the cells that have discharged in 40C, how much actual power comes out, vs what was put in when it was cool, plus, how hard is it then for them to cool and take a charge. Meaning lets say the 30MW they use (If they in fact only use the same 30mw of the battery) to cycle in say 20C, what is the charge losses, then discharge losses, and speed in which they can recharge and then discharge again over a 24 hours period. Vs a period of time where its 40C for 12 hours and over 30 for the rest.
The Hornsdale Power Reserve battery has only been called upon to deliver its full output power of 100 MW on a few occasions. Most of the time its output is limited to 30 MW so that it has 70 MW in reserve for emergencies.
You’d expect it to struggle most when it’s at full power because the heat dissipation will max out when the power is at its maximum. It’s possible that they use some of the batteries at 100% when they are delivering 30 MW. Or they might only use 30% of the power capability of all the batteries when they are delivering 30 MW. I don’t know how they manage it.
These are some of the times when it has jumped to 100 MW and the weather has been warm to hot at the time:
* 13 Dec 2017 at about 11:30 and about 37C
* 24 Jan 2018 at about 16:30 and about 32C
It was delivering 100 MW on both occasions.
It hasn’t gone to 100 MW in 2019 so far.
On a typical day its output is mostly zero. It seems to sit there looking for opportunities to make money on the spot market (energy and FCAS). You can often see it jump to 30 MW when wholesale prices are high and then recharge at 30 MW when wholesale prices are low. Sometimes you see it at 30 MW in one 5 minute AEMO SCADA interval and then recharge at 30 MW in the next 5 minute AEMO SCADA interval. It has done a few thousand of these +30/-30 MW cycles since it was commissioned.
Tesla designs electric cars. Cars have to work at quite high temperatures (air temperature plus insolation). Tesla’s batteries have a liquid cooling system. My Tesla battery at home has the same sort of cooling system. I can get the full 5 kW for my air conditioner on days that are 40C.
So I reckon the Tesla battery systems and other properly design grid scale batteries will have no problem on hot days. They will be designed not to overheat when they are charging or discharging. You just need a well designed cooling system.
I haven’t looked at how much energy is lost in the Hornsdale Power Reserve battery but it kinda doesn’t matter if you can buy $50/MWh and sell at $300/MWh.
PS: The best match I’ve been able to get for Paul’s mystery DUID is Waterloo Wind Farm. That’s my final offer! 🙂
Hello Mike, Thank you for your messages. Ok firstly. What i am talking about is simply the maximum rated amount of power the battery can deliver in 24 hours. It does not matter if its the whole 100, or just 30MW, as long as its just the same sections of battery’s that are being used. Just like a gas plant or coal plant have a rated output. Kogan Creek is like 750MW. And as you know that is 750 by 24, to equal what it can do in 24 hours. CF (Capacity factors are done over a year) So its 750 by 24 by 365. would be 100%. On a hot day as we know Kogan creek has to wind down, it cannot run at 750 during hot weather it can only roughly get out 680 to 700. (Roughly)
So the battery must have a set maximum it can run at in 24 hours. Like it can be charged and discharged lets say 10 times so in 24 hours it can deliver 1000MW / 24 equals an average of 41MW per hour. (I am guessing at that, what is its rated day capacity? 100mw is how much it can hold, which is diffrent than what it can deliver) However on a hot day, its not going to deliver 41mw per hour, its going to suffer from the heat, its not the batterys will be hot, the inverters will be hot, the power lines will be hot. The battery must also have a reduction in its rated capacity. What is it? That is all I am asking, or alluding to.
Now as for the battery buying low and selling high, that is 100% fine, to make money. Its not the best, if you are short on power because the battery takes in less than 100% to start with, and then lets out less than 100% of whats in it. So there is a loss both ways. The battery produces zero power, it uses power to deliver at a later date, or as you put it to buy low and sell high. But not matter which way you cut it, unless its being charged with excess power that would otherwise be lost, its adding to demand. Which that battery has been doing most the time as it mostly chargers from the grid.
Your tesla will also have a maximum amount of times in one day it can be charged and discharged I would call it the cars 12 hour driving range or 24 hour driving range.
Think it’s Loy Yang A or Yallourn Power Station.
Hi John, I think you might be trying to compare apples with oranges? Capacity factor does not appear to be a relevant figure of merit for a battery.
At the risk of sounding like a Tesla fan boy (and with apologies to everyone for taking up a lot of space) here are a few thoughts regarding your doubts about modern batteries (…ok, I admit it; I am an Elon Musk fan. Anyone who can nail the synchronised landing of two booster rockets deserves some respect if you ask me).
In an electric car the battery discharges whenever you are on the accelerator and charges whenever you are on the brakes thanks to regenerative braking. That means it might switch between discharging and charging hundreds of times a day if you’re driving in city traffic like Tesla hire car drivers do, or like Toyota Camry Hydrid taxi drivers do.
As I have already mentioned, the Hornsdale Power Reserve battery switches between charge and discharge many times each day (and night).
My home battery switches between discharging and charging pretty much instantaneously. On hot afternoons when my solar panels are generating not quite enough to meet the peak power demand of my air conditioner, as the air conditioner cycles between almost no power and peak power, the battery has to switch between charging (using excess solar power that the house is not currently using) and discharging (when the air conditioner needs more power than the solar system is currently producing). This can happen lots of times per day and Tesla guarantees the battery for 10 years. By the way, my battery has carried me through a dozen power interruptions in the nearly two years I’ve had it. The most recent event was 3 hours in January this year in suburban Sydney. Uninterrupted cricket viewing in a nice cool house. That battery is worth its weight in icy cold XXXX Gold. 🙂
In summary, as far as I can tell there is no _practical_ limit to the number of times per day a Tesla battery can switch between charging and discharging.
On the subject of the Hornsdale battery acting as a load, that is not necessarily a bad thing. It can save inflexible generators (e.g., thermal ones) from having to rapidly follow fluctions in demand. That sort of help should save the owners of thermal plants some money and _should_ drive down the cost of electricity (sigh). A big battery could have saved Queensland from some pain when it recently disconnected from the rest of the NEM and the sudden loss of a large lump of load caused the frequency in Queensland to go troppo for a while. Also note that when power is in short supply prices tend to be high and so it is unlikely that a battery would be charging and adding to demand.
I know it’s politically incorrect to say so in Australia but all the data seems to be saying that batteries might just save the world (or at least what remains of the Hunter Valley).
Hi again John, apologies for not addressing one of your important observations. In 2018 the AEMO dispatch SCADA data says that for the Hornsdale battery about 49 GWh went in and about 40 GWh came back out again, so it returned about 82% of the energy that went into it. 82% is comparable to pumped hydro plants (Wivenhoe, Tumut 3, Shoalhaven, Snowy 2.0, etc). See https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricity If you’re worried about the energy losses of a battery they pale into insignificance compared to the energy inefficiency of any thermal generator be it coal, gas, diesel, or whatever. Coal is typically only 33% to 40% efficient even for HELE (ref: https://www.worldcoal.org/reducing-co2-emissions/high-efficiency-low-emission-coal). My home battery returns about 94% if I deep cycle it and about 85% for the shallow cycling that happens most days. You might think I’d be unhappy with 85% but it’s a lot better than the 25% return I’d get if I sold my excess solar energy to the grid at the NSW IPART rate of 8 cents per kWh during the day and had to buy it back again at 32 cents per kWh during the night.
I think that’s enough from me.
Hi Mike, Thank you for your message. You do like Mr Musk.
As for a Battery not having a maximum capacity, and not having losses, you seem to some how have that part missing. The way you write is as if no matter what time of day, temp, the battery works the same, and so does the inverter thats between it. Not matter what.
Really there will be a difference as there is with everything else.
I am not saying a Battery is not excellent, I am just saying its also suffers in the heat. As do inverters. As do thermal FF plants, as do solar pannels, as do aircons and lastly as do humans.
As stated on a 40c day, a batterys ability to meet its maximum supplied 24 hour total amount of power, will be less than on a 20c day. Due to the increased losses charging it and increased losses discharging it, both from the battery itself but also from the inverter / as well as the speed to recharge it will be reduced by the battery and also by the inverter.
Another candidate for the magic wand category is power electronics. While it can support fault levels (part of AEMO’s termed “system strength”) and provide reactive power control (including distribution and transmission voltage control). What it can’t do is let grid operators escape power engineering design.
On the one hand, the capabilities and limitations have not been considered in VRE location or connection requirements, or protection design. On the other it is just expected that *this or that feature* (synthetic inertia, fast frequency response) will allow wholesale replacement of synchronous plant. The result is many “surprises” in plant response appear at inconvenient times.
Hello Mike, I am sorry I did not see your second reply.
Yes those numbers are correct. The battery lost around 18% of the power that went into it. Thats a average. So as I have said above. On a hot day, its going to be higher than 18% and on a cool day its going to be lower.
Hi Alan, do you have any insights on negative pricing in SA in the last 24hrs (30/4/2019 – 1/5/3019)?
Hi Ben; I haven’t looked in any great detail but the quick answer is high wind generation in SA, modest grid demand, plus significant restrictions on exports to Victoria due to transmission works on the Heywood interconnection. SA prices were probably being set by negative offers at one or more windfarms. Cheers, Allan.
Thanks Allan. Apologies if these are dumb questions, but how does a negative bid from a wind farm still sit on top of the bid stack to set the price, and why would a wind farm bid negative anyway?
Are the ‘forced on’ gas turbines exposed to the negative wholesale price?
Appreciate your insights, thanks in advance!
Ben, good questions, not dumb at all. Answer to the first is that a negative offer can still sit on top of the *dispatched* subset of offers and thus set the spot price if there is enough volume to nearly satisfy demand (including exports) bid at even lower negative prices.
Answer to the second (in short) is that when a generator is directed to run by AEMO, that generator subsequently receives compensation for the difference between its costs and the spot revenue received (or paid out in the case of a negative spot price).
Allan
Good job, something that’s needed and will add to the conversation. I think this will be good for the country. Let’s see what the various partisan commentators pull out of it!
https://ab.co/2EKS54M
Don’t worry Paul, in a former life, my organisation stated that there were 5.3m too many which got rounded up to “nearly 6m”
Dear Me, some people are going to be upset with your information. Maybe you should have done what the Polling groups did or what humans almost always throughout history have done when information does not fit what is believed to be true, fact or scientifically proven. Changed it, so it fitted in.
The news did lead me to wonder where are all the stop Adani voices, et al calling for said generator unit to be left in mothballs like they all have to be by [fill in last chance to save planet date here].
You forgot to mention the “planned” outage in late 2017 which resulted in the “super load” that Victorians seems to get so excited about when a generator was sent to Germany for repairs in early 2018 and returned earlier this year – this was a generator outage and the planning was done following the unplanned failure.
When prices are suppressed in pursuit of political ends (and compensation denied), necessary investment lags, with resulting, adverse reliability risk/consequences. It ultimately catches up and customers suffer.
Yes – I have seen it happen in various countries/jurisdictions where I worked – and we never seem to learn. In the book I am (not) writing, the antidote to this kind of ‘populist’ government is a properly educated/informed electorate – but governments around the world measure their education programs’ performance by how much they are spending; not the results.
The 20th of May was swaption expiry for FY20. I believe the volume you’re seeing on the exchange is the delivery of underlying for in the money options. (If you look at the ASX trade log, you’ll also see a crazy range of transaction prices in that daily volume, which relates to the different strikes exercising. And the volumes for option exercises come in pairs (of 2, or really strips of 4) of the same volume.)
Does your analysis distinguish between a correlation of wind intensity with temperature and an effect of temperature upon wind generator performance?
Ha, good one! I hate being wrong but I like knowing the real answer too…
On reading this I immediately reviewed some old files I have for one of the grid connected wind farms in NSW, turns out there are several temperature sensors in the nacelle for the mechanical systems.
Below are the two high temperature alarms described in the technical documentation from the WTG supplier (in this case RePower). It is possible (and likely) there are others:
gearbox oil sump temperature
generator stator temperature
Page 3 is interesting.
As all of the best sites are used for solar and wind, you will get diminishing rates of capacity factors for future installations being in lower quality areas.
I’d love to know what the failure was. I’m betting it’s in the electrical side rather than the gas side. And a rewind means a long outage.
Re those bids dispatched where offer price is higher than spot – it looks like they’re mostly coal units limited by their offered ramp-down rates (typically ~3 MW per minute; note that this is a bid parameter not a fixed technical limit). The dispatch engine would prefer not to dispatch them but can’t move their targets by more than 5 * ramp rate in a 5 minute interval.
And for anyone wondering why those bids don’t set the spot price at their higher offer prices – it’s because those bids aren’t marginal – change demands by 1 MW and their dispatch levels wouldn’t change, constrained as they are by ramp rates.
In QLD there is still over 4GW of coal online during minimum demand and very low spot prices. For the coal generators to bid their capacity so low, is it fair to assume this capacity is already sold as PPA, so they don’t care what the spot market is?
And if that’s correct, or close to correct, then is there a ‘shelf capacity’ of PPA where the spot price changes from ‘care’ to ‘don’t care’?
And what happens when the unscheduled rooftop PV generation eats into the PPA capacity and the coal plants start to get dispatched out of delivering even their PPA volumes?
The same will eventually happen to utility solar and wind – they will be constrained like any other generator – when the rooftop PV eats into their market share. At least a coal plant can make money overnight and on cloudy days and can be held online for FCAS.
None of this gives me any confidence the market is working to price generation correctly.
Whenever generation is running at spot prices below its marginal cost/value, someone is losing money ie someone cares. In a true PPA situation where the generator is paid a fixed price per unit of production, the offtaker is receiving spot price (+ possibly LGCs if it’s renewable generation) for sale of that electricity into the pool, and that offtaker would prefer the generator not to produce once spot price (+ LGC value) is less than the PPA price. The generator is earning PPA price less its marginal cost, so in principle doesn’t care about the spot price. But work through the cashflows and everyone can be better off with no generation once spot price + LGC value is less than marginal cost (the offtaker could compensate the generator for its value of lost output, [PPA price – marginal cost], and still be better off than wearing a loss of [spot price + LGC value – PPA price]).
More typically, traditional coal fired generation is not contracted through PPAs but via swap contracts where they receive fixed price and pay spot price to a swap buyer. Again the incentive for these generators is to not generate once spot price < marginal cost. (In effect they can "buy" electricity from the pool, for financial delivery to the swap buyer, more cheaply than by generating themselves. These contracts are not for physical delivery and carry no obligation to actually generate – it's a financial decision).
So why do these generators bid $0 or -$1,000/MWh and keep running through zero or negative price periods? Typically if their costs of shutting down then restarting units once prices recover are likely to exceed the losses incurred while remaining online. For coal fired units these startup/shutdown costs can be quite significant and their marginal costs are relatively low so they choose to keep running through short periods of low price. But if you look at gas-fired generators with higher marginal running costs and lower startup costs they regularly decommit units rather than stay online at a loss.
In your great piece on price setting, you explain how MLF doesn’t affect the actual output but the price paid to the generator.
My question is: what is the output measure used in a PPA? Do you know if most PPAs take into account the output at the RRN (if so double counting the MLF on both the output and the spot price) or do they use the generator connection point?
You’ve also asked this here:
https://wattclarity.com.au//articles/2018/07/helping-to-tame-villain-4-by-understanding-more-about-mlfs/#comment-733520
“We hope that the results presented in the Generator Report Card can raise the level of awareness of what’s actually happening..”
Well all I know is I just got my last power bill in suburban Adelaide and they’re charging me 51.2c/kwhr peak in an all electric household (19.05c for off peak HWS). It was 43.2c peak last quarter so the poor folk must be wondering when the cheap renewables are coming down the wires.
On Friday night there was a grand total of 150MW wind output across the entire NEM. Low correlation won’t help with that.
Correlation is a popular argument used to advocate for more transmission lines. But interconnectors expose regions to price volatility and system strength issues as much as they expose generators to more load.
Has SA found their interconnection with VIC reduced their c/kWh? Or gas QLD being linked to NSW helped QLDs c/kWh?
I think a region with low volatility and generation over-supply would be better off without an interconnector.
The inherent variability of wind turbines and its effect on price volatility and system strength would be better off not spread throughout the NEM.
Spot the correlation with AGL having a change of heart over closure of Liddell-
https://www.msn.com/en-au/money/company-news/agl-delays-closure-of-liddell-power-plant-after-battle-with-government/ar-AAFcr5X
The math is changing, larger wind turbines, can drive water turbines, to pump water uphill, for power storage and do it economically. If we look at the challenges, resource shortages, we can develop ways, of using more plentiful inputs, that’s how economies of scale work.
The cost per kWh, km, gram of protein, compared to a competing method is lower. An electric engine uses 1/5th as much energy, as an internal combustion engine. A solar panel can capture 10X as much energy as a natural plant.
Improve the cost factors, with research and mass production and the inherent physics advantages shine powerfully. Near worthless desert land, producing 10X the energy, of the best agricultural land. Vehicles using 1/5th as much energy, requiring 1/10th as much maintenance.
Moore’s law gave us the tech boom, 1997-2000, until market saturation, gave us the tech crash. Strange as it sounds, in a decade boom, most cars will be self driving electric, most power will be renewable.
But these vehicles will be low maintenance, as will be the solar transistorized power. At market saturation, we’ll need the basic minimum income. Not difficult, with cheap transportation and energy, as well as a doubling of assets, due to Clean Disruption.
A second roaring twenties, taking 3rd world countries, to second world standard of living. Second world countries to first world standard of living.
Solving pollution, noise, transportation, infrastructure problems, in the first world countries. It won’t be paradise, but people, will live to 110 years old.
Perhaps the longer testing periods required to before new generation can feed into the grid at full capacity have affected their decisions. AGL must be smarting from delays in allowing the Silverton wind farm to run at full capacity. Better to keep an old unit in reserve until the new capacity is fully commissioned.
The extension of 3 units of Liddell gives “cover” for decommisioning one unit early in 2022, which would presumably be the least reliable unit. This decomissioned unit can then be cannibalised for spares to keep the other units operating.
AGL’s results announcement on Thursday said re LYA2 “Work under way to return to service, expected mid-December 2019”. As this is material information in an ASX release you would have to assume it remains AGL’s best estimate for return date and would have been carefully vetted, probably ruling out worst case scenarios like manufacture of a new rotor or stator. Of course it doesn’t rule out risks of the return date slipping by weeks or even months. It will be fascinating to see how AEMO deals with delay risk in its ESOO update.
In Example 1, Very Simple, the ROP is $65.44. But then the Trading Price looks like $80.32. What give? I’m already lost. I’ll go back to the simple guide now. But this may be worth explaining here, no?
1. The decision to reduce flows on long distance transmission in the face of storms is common in many jurisdictions and had the practice been followed before the system black it would not have happened. Between 100 and 150 MW of gas generation around Adelaide would have solved the problem.
2. The anti-renewables rhetoric may delay the transition at the margin but economics will win so before the end of this government non hydro renewables will provide more energy than gas and brown coal combined and black coal will fall below 45% of total generation
oh, and as a quick PS, readers might also find the comments on this LinkedIn thread useful:
https://www.linkedin.com/feed/update/urn:li:activity:6566537197816766464/
Paul
Nice analysis. We shall have to see what the E(3)C, ESO and Ofgem investigation(s) produce; and how much is published.
Stepping back from the specifics, my own take is that (1) there is inadequate competent strategic oversight of the transition in an industry that tends to predict the future from past experience, and (2) the “energy wars” you describe make this worse. The German experience with frequency-related PV disconnection is another example. I anticipate that the investigation will find that no one body was to blame but they collectively failed to act on a fairly obvious risk.
I am currently looking at the systems implications of the possible transition to EVs in the UK by 2030 (to meet the Fifth Carbon Budget). Lots of opportunity for things to go wrong.
Andrew
Andrew – you absolutely nailed the problem both here and there. As to why there is insufficient competent strategic oversight, well here at least
a) our participants and institutions are almost exclusively mired in detail,
b) our politicians are almost exclusively preoccupied with picking fights and picking winners.
Please explain what is wrong with my understanding of inertia
In practice inertia is a store of energy. If a generator goes offline then the energy stored in the other rotary machines on the grid, (motors as well as generators)as well as the energy in the windings of motors generators,transformers and the transmission links (inductance) forms the inertia of the grid. Referring only to the mechanical or rotary inertia it is Jw^2/2 which J is the polar moment of the machine w is the angular velocity.
Now because the frequency is only allowed to fall to about 48 Hz before disconnection, the useful inertia of a turbo generator is J x 2 xpi x (1-(48/50))^2. or roughly 4% of the system inertia. So if there is 4,200 MWs in the system then only 168 MWs can be used before cascading failure begins. Assuming a rate of change frequency ROCF of 0.5 Hz/s that is 168 MWs delivered over 4 seconds (2 Hz/0.5).i.e. 42 MW.
If a battery can deliver 42 MW for 4 seconds that effectively provides the same energy as all the available system inertia so one can argue that the Hornsdale battery actually provides 129 x 3600 = 464,000 MWs far more than the available rotary inertia in the system. Even with a ROCF of 1 Hz/s inertia only supplies energy at less than 100 MW and only for 2-3 seconds. After that if governor response has not kicked in, goodnight folks.
Similarly, synthetic inertia can be provided by wind and solar farms via slightly different mechanisms.
a) A typical modern wind turbine has an Inertia constant of 4-5. i.e. a 4 MW wind turbine with an inertia constant of 4 has 16-20 MWs of inertia, but because most wind turbines are asynchronous machines, their speed can be allowed to drop by 10-15% while the power converter still puts out a constant frequency. So if you allow the rotary speed of the turbine to fall 15% then the available inertia per MW is 1 x(1-(42.5/50)^2) = 0.38 MWs/MW. so 1,000 MW of wind turbines can contribute about 380 MWs whereas 1,000 MW of gas turbines with a typical inertia constant of 3 can only contribute about 3 x 1,000 x 0.04 = 120 MWs
b) a solar farm can similarly provide synthetic inertia by running slightly below peak instantaneous capacity. If 1,000 MWs of solar farms are running at a nominal 800 MW but deliver only 760 MW, it can ramp up output almost instantly to 800 MW thus providing 160 MWs over 4 seconds just as the gas plants would.
Thus a properly engineered wind/solar/storage system can provide much faster disturbance response with lower inertia. In fact because the inverters don’t have inertia they can respond to deviations much more quickly.
There is a whole other discussion about what happens next but that is for another time
The UK company National Grid is the AEMO of the UK, and is very heavily involved in “The Transition”, i.e. makes a lot of money from all the poles, wires, controls, analysis, etc that is required.
If you regard these events as match-meets-pile-of-wood, you can be sure that any report from National Grid will blame the match for the fire, and will hide the involvement of the pile of wood.
Wind power earlier in the day was 67% of demand, uncharted territory, so claims of great system resilience are specious.
Hi Paul,
Great article. In relation to your question about the UK equivalent of FCAS, National Grid ran a tender process for Enhanced Frequency Response services in ~2016 (1s response time, 4 year contracts offered, taken up by various technologies, many BESS projects). There is more info here:
https://www.nationalgrideso.com/balancing-services/frequency-response-services/enhanced-frequency-response-efr
In addition it looks like they are also trialing the procurement of Firm Frequency Response services (slower response times).
https://www.nationalgrideso.com/balancing-services/frequency-response-services/firm-frequency-response-ffr?overview
It would certainly be interesting to know how the EFR plant performed. Perhaps the capacity just wasn’t sufficient to mitigate the frequency drop.
Thanks to WC for the insightful analysis.
I cannot possibly see how a train wreck can be avoided without mandating that every tenderer of electrons to a communal grid has to be able to reasonably guarantee them 24/7/365 or they can keep them and that includes frequency and voltage. Anything else is delusional State sponsored dumping in my book and normally the ACCC would be all over it for the obvious but alas.
It seems that the initial report on the UK outage has been leaked and it notes the wind turbine failure occured first, which tripped the gas power plant.
https://www.thegwpf.com/leaked-report-points-to-wind-farm-failure-causing-uk-blackout/
“Experts have said giant batteries providing instant back-up energy could help.”…
“The costs are sizeable but if there’s a risk of this happening more frequently, I don’t think businesses or consumers will accept a lower standard of continuity of supply.”
https://www.thisismoney.co.uk/money/news/article-7367413/Renewable-energy-blackout-risk-warns-National-Grid-outage.html
I would think so when it takes 6.8 million pounds for experts to work that out.
Lots of focus on “unreliability” of fossil fuel plants while ignoring cause which is the market distortion due to government subsidies for renewables.
AEMO’s plan is to spend tens of billions reconfiguring our energy network so it can do what it did fine before renwewables were introduced. And we all know that just means the cost will flow onto consumers, driving our electricity prices even higher.
Ms Zibelman’s AFR contribution states that transmission is needed to replace generation.
Unfortunately it doesn’t work like that. Yes lots of wind power across SA sends power back to VIC. But the power comes from Traralgon when the wind drops.
It is disconcerting that the AEMO CEO did not mention intermittency once in her article. She gives more weight given to factors that will increase costs than factors that will reduce costs.
“Pay more accurately for firming capability” – well you don’t need to pay for it when it’s inherent in the technology.
The move to inverter supplies asynchronous intermittent generation might provide cheap power when the weather is favourable, but it’s also forcing the firming capabilities into separate pieces of equipment – fault level, inertia, voltage and part-time bulk power delivery.
So now we need all that plus bulk power capacity available all the time, plus lots of transmission, plus batteries plus hydro.
Sadly the internet has given a voice to people who have no idea what they are talking about. The energy market is extremely complicated and the information in the reports aren’t readily accessible to the layman so they read headlines from media who either don’t understand the topic or willfully spread misinformation.
The only way to fix this is for AEMO to release a simple/layman version where they explain things in plain english as this post managed to do quite well.
Yes, the 0.002% reliability threshold for the NEM’s 200 TWh annual demand is 4 GWh, and that can happen all at once or in pieces over time.
Large-scale solar was modelled as having a 40% capacity factor at the time of maximum summer demand, and wind a 16% capacity factor. The summer peak for unserved energy in Victoria is ~4:30pm for 2020. Yet Angus Taylor wants like-for-like replacement of old dispatchable with new dispatchable based on nameplate capacity, not taking into account that both new solar and wind contribute something to capacity at the time of summer peak. This will eventually lead to an expensive oversupply of dispatchable plant.
Anyone notice that Basslink went out about 11 am this morning. Hope it’s fixed quickly, but if not it’s another factor affecting reliability In Victoria not considered in AEMO’s report.
Hi Malcolm, it’s well buried in the detail (p71 of the ESOO report) but AEMO does consider forced outage risks for key transmission links including Basslink in its reliability modelling. Obviously a special factor with Basslink is that some forms of outage can take a long time to repair and it’s not clear if and how fully AEMO’s modelling captures this low probability / high consequence risk.
Re your other comment on solar and wind contributions to meeting demand peaks, the modelling does not assume fixed percentage factors but instead runs simulations based on historical half hourly traces of demand and renewable output (scaled for changes in installed capacity) in an attempt to better capture their variability and correlation. The figures of 30% for large scale solar and 16% for wind (quoted on p76) are ex-post average contributions (as percentage of generator capacity) extracted from those simulations during periods of unserved energy. Across those simulations there would be a wide range of modelled contribution levels from solar and wind.
Your general point that variable renewables (VRE) effectively contribute some proportion of their capacity is correct. However it’s a smaller contribution percentage than from “firm” dispatchable generation, and the contribution percentages of VRE at times of system stress will tend to reduce as the VRE share of generation increases, because instances of tight supply-demand balance will increasingly be driven by low VRE output in addition to high demand and outages.
Best description of the NEM I have ever seen : “There’s no doubt it’s a sophisticated piece of machinery with a lot of complex science and technology behind it. We then wrap it in a highly regulated market founded in arcane mathematics and towing a boatload of jargon and acronyms, operated and overseen by a plethora of organisations with obscure names and functions, trying to navigate and administer a morass of laws rules regulations procedures codes and other legalese.”
Keep up the good work
I wonder why Ms Zibelman puts the boot into thermal plant without also mentioning weather dependent intermittency?
Also no mention of Basslink which has been offline twice in 3 months.
I would have thought there’s more to reliability than simply how many blackouts and for how long. The other side of the coin is how much to pay productive users to shut down and knock off for the day in order to cope with unreliable supply. Greenouts but whose wallet are they coming out of?
Take that to extremes and pay us all a pile to do it and at some stage those of us with the capital can afford our own diesel or town gas generator all year round and dump what we don’t use back on the grid to also amortise the investment. Now where have I heard that before? The same place I’ve heard about a fallacy of composition.
I note the recent UK experience where 1000MW loss of generation capacity resulted in 500MW of behind-the-meter generation, thus necessitating greater response which ultimately included significant blackouts.
AEMO, IMHO, should specifically allow for loss of embedded generation in their ESOO models, yet I have seen no mention of that.
John
Thanks for the comment. I tend to agree with your analysis of the recent UK event, but note that was a “system security” problem – real-time robustness of the grid against a sudden transmission or generation failure.
The ESOO does not deal primarily with this kind of risk but with “capacity adequacy” – having sufficient dispatchable capacity (generation and potentially demand response) to deal with *forecast* future levels of demand. Unavailability of generation is allowed for in the analysis, based on known outages and failure statistics, but the ESOO modelling itself does not pretend to assess the second-to-second robustness of the system against outages. So things like “cascading” failures don’t need to be explicitly modelled for the ESOO. However the models do incorporate many constraints (for example on interconnector flow levels) that ultimately reflect system security considerations.
Allan
This is what happens when industry shuts down because of the worlds highest energy prices,and renewables make so much money in subsidies they just bid in the negatives to ensure they are in the bid stack.
It’s interesting that you note renewables contract structure is such that it’s someone else’s* cost (and risk) when prices are negative – so are happy to sail through with their own revenue covered by some form of PPA independent of spot pricing.
The below article from earlier in the year outlined some of the issues with the distortion of the market due to government subsidies for renewables
https://www.dailymail.co.uk/news/article-7021119/Solar-farms-millions-taxpayer-handouts-make-selling-electricity.html
Great Article. It’s really confusing to understand why wind and solar bid so far into negative. It really highlights the need for bulk storage. in a fully developed market energy will be extremely cheap in QLD. competition between renewable and coal = cheap power.
Looking forward to seeing what new attacks our polies have on wind and solar. Hard to dispute the benefit here.
>It’s really confusing to understand why wind and solar bid so far into negative
When they receive so much money in subsidies and power purchase agrements, they can bid negative and still make money.
You can’t have competition between coal and renewables unless you remove those subisides to renewables that distort the market.
Some thoughts:
– what was the proportion of gov owned generators losing money to stay online compared to private companies, because government backed generators with years of huge profits behind them can afford negative prices for longer than most
– a coal fired generator can turn down to the point the boiler air flow can no longer sustain stable combustion, unless it has a turbine bypass, in which case the boiler can keep its firing above the critical threshold and the excess steam can be bypassed straight to the condenser hotwell – a few tonnes of extra coal on top of the negative prices isn’t much
– I’m pretty sure at least one of the QLD coal fired generators is fitted with turbine bypass systems
– to get below the critical coal firing rate requires support from oil or gas burners which are more costly than coal
– what effect does wholesale price volatility have on hedging contracts – I imagine volatility increases risk for both retailers and generators, therefore both of them cover this risk with increased prices to consumers
– what percentage of the QLD demand is supplied through contract
Seems much more that the CANNIBALISM that’s an inescapable feature of intermittents has begun in QLD.
Solar bids the price all the way down to -$1000/Mwh to compete for each solar plant’s hoped-for share of the favored dispatch they enjoy in the market that’s corrupted by government interference….to prop up intermittents.
It’s all artificial.
When the price gets so low and they either have agreements to switch off at that point as they’re propped up by government anyway…or they just decide it’s in their interests to do so…the solar farms just pull the plug…knowing the QLD GOVERNMENT will order coal plants to bid and generate to fill the gap even though it’s unprofitable despite the hedge..because if they didn’t….the lights would go out and everything would stop in QLD and the SYSTEM WOULD THEN COLLAPSE across the whole NEM…as forecast by the former AEMO boss…the late Matt Zema just before his death.
It would collapse because QLD coal plants…the newest in the country…. are all that’s keeping Australia’s East coast from being in SYSTEM BLACK now.
As Bloomberg[ BNEF] said…
‘Challenge for policymakers is to figure out how to pay for the bulk cheap RE units that run at close to zero marginal cost & ALL GENERATE AT ONCE cannibalizing their own energy prices-& at the same time keep conventional thermal plants online & available’.
How long will coal plants make themselves available in the OTHER states where they’re NOT government-owned and are subject to abuse-economic…material…real and verbal.
Murphy’s Law can pop up too – say the spot market is low and a generator has planned testing that requires it to generate through periods like this.
Braemar 2 was pigging its gas inlet last week, meaning it had to guarantee a constant gas flow for a time period – some of that was through negative prices when they wanted to be offline.
I wonder if some of the actions by the coal plants are aimed at highlighting what is bound to come more often in the future? Not in the realms of conspiracy, but more like, “let’s do this and see what happens, so we may educate ourselves and better prepare for the future”.
Lol if I was into conspiracy theories I would be more interested in the August 2018 islanding of SA and QLD, at almost the exact nadir of SA wind output, which is the only time an event like that could occur with the lowest risk of unintended consequences 🙈.
Stanwell at it again today (Monday 9th September), during the daytime producing well above their night-time generation levels despite negative prices.
There is a problem when the WeekEnd Australian which is known as the weak lend of you Australia publishes the opposite type of article.
Perhaps with edited quotes to support the view point.
The None Australia paper says the reason for -$1000 price is because of Solar.
Paul has pointed out Solar turned off.
I think it is time for truth in articles frankly.
Interesting discussion about storage.
Meanwhile Basslink is out of action again and won’t be back online until end of October.
Missing from the worked example is the cost of fuel. A generator that increases output to crush the spot price will incur higher fuel costs. The rising cost of fuel will offset any gains from the swaps. So the question is the degree of offset.
There could be a system security issue due to system strength, requiring the big generators to be on line even at low spot price. I am not sure whether there is a market mechanism that force the generators to stay on line regardless of spot prices and then get compensated for the loss. The market operator got caught by the huge influx of asynchronous generators and trying to catch up in their plans.
There is a non-market mechanism for that Nadia – direction by AEMO – and we’ve seen negative price outcomes in South Australia when it is invoked there on days of high wind generation. However AEMO has to publicly notify the market when it directs generation to remain online and that certainly wasn’t the case in these Queensland events. As most of the non-synchronous generation in Qld is embedded small-scale solar I don’t think it would be leading to general system strength issues – although these may emerge in sections of the transmission network with concentrations of utility-scale solar and not much synchronous plant. Allan.
Before my time, but… Wasn’t 1999 a time when SA max payment was say $1,000/MWh and the market max was say $10,000/MWh and anytime the price spiked above $1000 SA would have rolling blackouts to cut demand? Maybe this is just an old NEM tall tale…
It’s interesting (or ironic?) that reducing demand is being discussed in the same era as some others discuss the benefits of 200% and even 700% installed capacity of intermittent generation.
It’s a tipsy topsy world these days:
– pay some users to reduce load, to reduce costs for others
– generators pay to generate, while consumers costs go up
I’m not convinced electricity should be this difficult.
If we had our time again, with what we know now, would we allow individual rooftop PV and unconstrained access to the grid by intermittent generators?
I’ve been following this saga with interest.
Of course it only takes a matter of minutes for deadband settings on a governor to be changed.
Primary frequency control is only one of the inherent advantages of synchronous generators.
It’s a shame they are all so under appreciated.
Its seems totally inevitable to me that inertia on the grid is going to keep falling and that as with so many other things we need to start moving forwards.
A workshop from the University of Washington earlier this year contained some excellent presentations and youtube lecture videos that I’ve been working through. I also got a working paper from the workshop convenor that was readable to a determined layman. My own view is that Australia could be moving more in the direction that Europe and the USA are already going. I don’t yet think there is a role for markets in this except maybe for fast frequency control, but I might be wrong. To me its about getting the design principles right for the next 20-30 years at a time when technology in the area is moving quickly. Seminar reference attached and if anyone want’s Brians roadmap document let me know. I’d appreciate input from people who unlike me can speak with authority on an important topic https://lowinertiagrids.ece.uw.edu
I remain perplexed that this problem you have highlighted, not only remains unresolved, but seemingly the AEMC still doesn’t get it. The response to the double islanding event surely underlines the desperate need for this to be rectified. AEMO in their report noted that the last Qld separation event that happening in 2008 involved the loss of power transfer from Qld to NSW of 1091 MW vs 879 MW in the recent event. In that event even though the loss of power was greater the frequency deviation was less in both Qld and NSW and no load was shed to rebalance the system.
Surely AEMC should be able to join the dots.
Kate you make a good case for re-empowering the power engineers — the market design has failed us and its about time that good engineering was given a dominant role again.
Paul,
I have just come across this article and noted the remarks about demand in some places being characterised by higher proportions of industrial demand relative to domestic. Many years ago I used to predict daily demand for SA and it was pointed out to me that our demand was relatively lightly affected by industrial load, despite having the car industry and its ancillaries, because gas was such a large contributor to energy usage by industry. Adelaide was the first to take gas from the Moomba gas field and its development was enabled by Torrens Island PS taking 2/3 of the gas from Moomba in the 1960 – 70s; the rest being taken by industry, commercial and domestic.
When predicting demand the major impact was due to air-conditioning so demand was highly correlated with temperature since there was a small industrial base load for electricity and we became quite adept at ‘adjusting’ for number of days of high temperatures preceding, overnight temperatures etc.
It was also necessary to take account of school holidays, Christmas and Easter as domestic, school and commercial air-conditioning all had varying effects.
It was apparently quite a different process with unique considerations compared to the Eastern states.
Fascinating article and great explanation of a complicated process. Apologies. I am little confused about the undershoot references (i.e. “so far under its target?”) above and possible “large-error trigger” non-conformance. In a previous article .. https://wattclarity.com.au//articles/2018/09/a-case-study-in-active-management-at-a-semi-scheduled-plant-tuesday-24th-july-2018-at-lake-bonney-2-wind-farm/ it was indicated that ‘For the Semi-Scheduled plant, the Target effectively acts as a Cap – i.e. AEMO is telling them to “don’t produce any more than Target MW” at the end of that dispatch interval.’ I suspect I am missing something about the semi-schedule dispatch process, particularly when the semi-dispatch cap is set. Any chance of a clarification would be appreciated. Thanks .. Geoff
Nathan
Thanks for this, a really good summary of the traps. One that seems not mentioned but has caused problems to others is the cost for addressing earth fault levels at the nearby substation
Thanks David. Good point on fault levels. Agree this can be an issue and perhaps something that early engagement with the incumbent NSP may help to identify.
Well done and well presented Nathan; you probably regard it as an integral part of the network connection process but one part of that process which seems to escape notice or be regarded as trivial is National Grid Metering (NGM), especially the role of Metering Coordinator. The whole NGM process is complex and riddled with acronyms and contradictions (some more apparent than real) so again early engagement with the NSP who usually act as the MC is highly advisable. More than one proponent has had a shock on learning how complex and regulated the metering process is and the desire to have total control has over-ruled good sense in making use of those who have done it before and have the contracts with other NER defined roles in place.
Alan,
Re “why call it something as impenetrable as the “Electricity Statement of Opportunities”? I would add why do such documents always get moved to locations in the bowels of AEMO’s web site often requiring hours of searching to locate since such moves always seem to lead to any previous links being broken. it is often quicker to google the document than search on AEMO’s website.
Thanks Nathan – lots of challenges for those unfamiliar with the NEM.
Worth noting that one of our other guest authors (and also collaborator on the Generator Report Card 2018 released May 2019), Jonathon Dyson, penned these 10 tips for new generation developers as well:
https://wattclarity.com.au//articles/2018/09/lessons-from-the-trenches-10-tips-for-new-entrant-generation-developers/
This Guardian report would suggest National Grid has a frequency control problem, with three significant frequency excursions in the three months leading up to the event. https://www.theguardian.com/business/2019/aug/12/three-blackout-near-misses-in-three-months-says-national-grid The watt logic article above points out that UK nuclear does not provide a frequency control response, so the amount of MW and MW.s available for Raise regulation may be quite small.
great note
Thanks Warwick – given the (relatively small) sizing of the battery, how is the shape risk of the flat swap managed by the solar farm? i.e. in periods where generation is insufficient to deliver the floating leg of the swap, how does the solar farm manage the risk that the cost to procure that power at spot is higher than the fixed swap price (particularly if the price is spiking above $300/MWh, and the battery is depleted)?
Thanks David,
Hi Nick, your question relates to the valuing of risk as there are often periods where there is insufficient energy in the battery to deliver to match the contract level. What needs to be appreciated is that the battery is providing the best financial solution not providing an engineering solution for an islanded system. So yes, you might not always have enough energy in the battery to “defend” your contracted position, but in measuring the balance between the capital cost of the battery vs a desire to match every possible half-hourly price spike, it seems the best compromise is to make the battery as small as possible. Also, the battery might charge multiple times or make arbitrage over night or on weekends, making additional revenue for the project.
This is similar to generators in the wholesale market (including baseload coal) that sell contracts against their output and face the risk than an outage will mean that they could lose money. They take a calculated risk that in the long term, contracting is best and accept the non-zero risk of an unplanned outage.
regards
Warwick
Thanks Warwick, it is an interesting concept.
I am referring to the ability of the solar farm to defend its contracted position, noting it is not dispatchable. Because the flat peak swaps traded on the market (used in your example) do not follow the shape of standard solar production, there is a mismatch in the shoulder periods (irrespective of outages). Is your solution to effectively self insure, I.e to reserve the surplus revenue above the contracted volume to settle the swap in the shoulder period?
Hi Nick,
It’s true that the solar farm is not dispatchable but the battery is. You are correct that the there will be times when we expect no solar production (i.e. after sunset and until 10pm) or limited production (early in the morning/late afternoon) yet there is a contract position to defend. You could say we “self-insure” in that we accept the risk that there may be a time when the battery is out of charge, solar is below the contracted level and the spot price is above the contract price but this is a calculated risk, mainly weighed up against the expectation based upon observed history that we would expect to earn close to the contract rate of $87/MWh (before battery cost) vs a PPA price of $40-50/MWh if one is available.
The alternative is to negotiate a swap contract that better matches the expected solar production but this would be difficult as the profile does not usually match a customer load, so we’d expect a discount to the standard contract, which is exactly the case with a PPA which is a perfect match to the profile.
In the end, if you’re willing to prepare a more complex integrated business model, you may be able to achieve a better return for your investment with less uncertainty. Alternatively, you can take spot revenue with absolutely no certainty or you can negotiate a fixed low price PPA with no price uncertainty.
It’s also worth noting that usually wind or solar projects “self-insure” even under a PPA as they do predictions of the expected annual energy yield and do a probabilistic estimate of the potential outcomes. Financiers may also require a P90 or P95 (i.e. an annual number that is satisfied 90 or 95% of years). This means that whilst they have no price risk, they do have volume risk, which can actually be insured against through bespoke weather derivatives or other insurance products.
Hi Paul. We should have a chat at some stage. I like the work youy are doing.
John, our number is +61 7 3368 4064. Will email you directly.
PPA offtakers effectively pay unders and overs. For example, if the strike price is $70, and the pool price is $80, then the PPA offtaker gets a net revenue of $10 per MWh. If the pool price is $60, the PPA offtaker gets a net cost of $10 per MWh.
If the pool price goes to minus/negative $100, then the PPA offtaker incurs a net cost of minus $170, while the generator continues to get paid $70 per MWh. So PPAs have a few ways to deal with this: one method can be to limit the pool price used in the calculations to $0 per MWh. In this case, if the pool price goes to minus $100, and the generator continues to generate, the PPA offtaker pays $70 per MWh to generator, but the generator is also responsible for AEMO payments below $0, so would see a net result of (70-100) minus $30.
(You may already know all this)
Of course, AEMO payments are settled weekly, and PPAs monthly, so 30 minute pricing is less important rather than overall Dispatch Weight Average Pool Price is what counts.
But I do think this incentivise’s better outcomes. Supply should be geared to meet demand, rather than just run of plant generation. If the price is negative, it is a sign of excess supply, and supply should throttle and reduce output. This was less possible for brown coal generation historically, but theoretically much easier for solar farms.
I am mystified by the AEMO requirement for synchronous condensers at new PV and Wind sites. All of these new sites use inverters to couple to the grid and inverters can generate VARs or absorb them just as effectively as synchronous condensers. In any case the overall power factor of the grid used to be set by the network not the generator
Another weird period in the SA market was on Saturday 16 November. In the 18:20 to 18:25 interval the price was -$387/MWh, then 18:45 to 18:50 it was $5000/MWh. The Port Stanvac generator turned on, and there was no flow on either of the interconnectors for several hours. The Heywood link came back just after midnight.
Thanks Malcolm
There was much more to it than that – here’s a look at (essentially) just the first 4 dispatch intervals:
https://wattclarity.com.au//articles/2019/11/sa-islanded-on-sat16nov2019-firstlook/
Paul
“if we start with an incorrect understanding of what the actual problem is, then it’s a very low chance we’ll ever come up with a solution that actually works?”
Without level playing field power generation I’d suggest the big picture can never work. That simply means every tenderer of electrons to the communal grid must reasonably guarantee them (ie short of unforeseen mechanical breakdown) along with a minm FCAS standard 24/7/365 or keep them. The fact we don’t demand that is the root cause of all the problems we currently face.
It’s a political problem and not an engineering/ economic one even if you want to levy a level playing field CO2 tax on them all which is further compounded politically with the international cooperation required. Nothing like protracted brownouts to call forth the political will nationally to face the truth at some stage I suppose while the international politics is a second order priority.
Looks like VIC gas picked up the missing 300 MW that was coming from SA at the time.
Plenty of gas on in SA as the solar disappeared and wind dropped off too.
First class note. Still not sure it needs to be that complex.
Hi Tom, a few thoughts.
– on your point about the broken hill solar farm I would argue that despite the initial MLF being above 1.0 that was not a good investment. The risk of that MLF being downgraded significantly was always high, when it was built there was no plans for upgraded transmission or greater load near the source so any addition near them would undercut them. I don’t think it is unreasonable that they should have assessed this risk. I would liken the Broken Hill solar farm to investing in a mining town in the boom.
– on your point on building where resources are best, I agree that would be ideal. The establishment of REZs and adequate transmission will assist this. But development of projects that are better located to supply that electricity, I would argue, is just as important as locating them to generate the most. No point having a perfect spot if you just waste a heap in losses and get curtailed all the time.
This is what happens when you let accountants out of the box and give them control of technology. They forget that the initial intention was to drain the swamp. The financially most efficient place to build any generator RE or fossil fuel is where the recourse is optimal. This applied when we were building thermal power station, near coal deposits, and the grid was reinforced to cope no one applied loss factors. The obvious place to build RE generators is where there is either a lot of sun or wind. These areas are readily identified and measures should be taken to reinforce the network to handle the increased generating capacity. RE generators in these places have the following economic benefits, more local employment, lower marginal cost of generation and reduced carbon foot print all good outcomes for the economy and the environment. Unfortunately we have a mob of twits in charge who cannot see this
I would like someone to explain how the Big Battery has such a high MLF, can it really deliver its power to a end user at a high MLF. What loads are around the battery that can take instant 30MW off and on day and night?. Or can it and does its load go long distances?.
By looking at its output, its pretty obvious its being run to make money, Mainly, not to deliver power, although at times it does, but thats mainly with the reserve.
How far can its power actually go within the grid is I suppose my question, say compared to a coal power station.
Hi John,
The lines in SA were built to take coal power from the north down to Adelaide with limited losses.
Multiple high voltage, parallel lines makes for a great place to connect generators. The wind resource just happens to be excellent too! Nearly all generators between Adelaide and Port Augusta have good MLFs.
Hi Tom,
I thought there were also other factors such as voltage that played a role?
I wonder if the battery is say being charged at the same time as suppling?
I wonder if maybe some of that is in essence going around in a loop?
Thankyou for a very good analysis but if we take Jan 25th as a the base case where we were short 200 MW and coal was only supplying 3,400 MW instead of a rated 4,700 MW, we have already experienced a situation with 4 of 10 generators being off on a hot day.
One can say that if we had had 300 MW more supply/ less demand we would have been OK. We will have almost 400 MW of behind the meter solar, 600 MW of utility solar and about 900 MW of wind additional capacity. Usually hot days are more like today where minimum wind generation in Victoria between 10 am and 7 pm was about 850 MW whereas 25 Jan was very unusual in that wind varied between 380 and 810 MW
Due to difficulties connecting wind and solar farms the installed capacity growth has been much faster than energy delivered, some wind plants are also being constrained down as well as the solar.
I think it is hard to see that demand will not be reduced by at least 200 MW from 9 till 3 due to rooftop supply and new utility wind and solar will provide a minimum of 1,000 MW and an average of 1,500 MW, vs 450 MW min and 800 MW av on January 25th. Today for example wind in Victoria only averaged a bit over 45% CF so it wasn’t a particularly good wind day and yet wind generation was more than double that of January 25. Further at 1 PM on January 25 imports from NSW were zero. NSW will be generating at least 400 MW of new wind and solar in the west and the Riverina, surely some of that will find its way south to Victoria even if it is only 100 MW.
But it is also a reasonable bet that if there is really a small reserve margin, that ways will be found for at least a temporary increase in allowable power from these constrained northern and western Victorian assets, so I suspect if coal drops by another 500 MW at least 250 MW of wind and solar would be found, adding to the 850 MW swing in the demand supply balance mentioned above.
Farley, VIC wind contributes about 10% of nameplate capacity to peak demand. Solar contributes zero.
You don’t have to exaggerate here like you do on the other blog, it’s not necessary, just the facts are fine. Everybody know that no matter how much nameplate capacity of wind is installed, it may not be there when it’s needed.
Having spent years commissioning power stations and other large rotating equipment, I suggest the risk of the two replacement generators having teething issues is understated.
Additionally there will be performance testing and bedding in periods required by the OEMs. “Online” is a good milestone, but doesn’t necessarily mean “fully available”.
It seems to me that last summer was the harbinger of things to come. Tight supply is the new normal, regardless of how much wind and solar go in.
Thanks for the comments Peter & Ben – they both identify important issues and illustrate the complexities of the reliability puzzle, particularly as your conclusions tend in opposite directions. I’ve tried to put some thought into a response.
First some context on Jan 24 and Jan 25 last summer. On both days there was forced load shedding (smelter on day 1 and Victorian consumers on day 2), as well as AEMO calling on load reductions from its portfolio of contracted RERT customers.
Demand and supply conditions were different on the two days. Jan 24 saw a combined SA/Vic shortfall with extreme SA temperatures (~46 degrees) and very hot weather in Victoria (~41 degrees in Melbourne). Two large Victorian coal units were out. The supply shortfall emerged in the late afternoon / early evening as rooftop PV production fell away, increasing grid demand, and wind production across both states reduced to about 15% of installed capacity. Smelter load cut and RERT calls in Vic reached about 400 MW of demand reduction.
Jan 25 saw a Victoria-only event, with temperatures of 43-46 in the Melbourne area following the previous day’s high temperatures – consecutive days of heat also elicit significantly higher demands than the first day of a heatwave. Cooler weather in SA enabled exports to Victoria up to interconnector secure limits, but offsetting this a third coal unit had failed. RERT load reductions of up to 450 MW were called by AEMO but as demand continued to climb and wind production reduced to around 400 MW (roughly 20-25% of Vic wind capacity), rolling area blackouts commenced from midday local time – these reduced demand by around 200-300 MW. Significantly longer and deeper load reductions would have been necessary had rising underlying demand not been arrested by an early afternoon cool change accompanied by increased winds and higher windfarm output. Victorian rooftop PV and large scale solar output was over 1000 MW during most of the load shedding period.
Both days were good illustrations of low probability but high consequence events that can contribute disproportionately to reliability risks. Very hot weather is not always coincident across SA and Victoria. Based on historical outage data for the ten Victorian coal units, the chances of three units being out on the hottest day of summer, Jan 25, was of the order of 1.6%. (Even with doubled forced outage rates that risk would still be only around 8.5%). Combined with that day being one of extreme temperatures – well above those expected on the hottest day of a “median” summer, this may all look like a very improbable outcome.
But there were also a multitude of favourable factors that don’t get as much notice as the negatives – there were no significant gas, hydro or transmission outages, renewables output was broadly speaking in line with or even above median levels for the time of year, the cool change arrived before the afternoon / evening demand peak on Jan 25 and so on. And even if only two coal units had been out on Jan 25 there still almost certainly would have been shedding, perhaps even the case with only one unit out.
Because of all these moving and uncertain parts it’s basically impossible to draw clear conclusions about reliability by using “single point” estimates – even quite conservative ones – of what levels demand, thermal output, renewables etc etc might reach, or how many units might be off, to conclude that there will or will not be supply shortfalls this summer. I don’t see much alternative to the type of probabilistic modelling that AEMO runs for the ESOO which can test literally hundreds or thousands of combinations of the myriad factors with associated probability distributions.
The result of AEMO’s modelling for this summer is that – assuming LYA2 and Mortlake 2 are back in service – they estimate that the “risk x consequence” of supply shortfalls (USE measure) is a little lower this summer at 0.0013% than it was leading into last summer at 0.002%. Behind that number are many scenarios – probably a majority – where there are no shortfalls (Peter’s point) but still a significant proportion where there are (Ben’s point). In a sense you could both be right.
A lot to digest here. I guess paying people for their solar output at a mlf of 100% (or 1) in the long term will more than likely have to be reduced down, as more and more people get solar and on average the solar from houses has to go further to be used.
This is very useful, just picked up on it. Could we perhaps have a think piece on renewable PPA’s from the point of view of the buyer, particularly a gentailer buyer? Renewables are increasingly achieving low NEM prices as their weight in the bid stack increases. To make this worse, I suspect that correlation is rising as geographical dispersion of assets falls. The PPA buyer is paying a fixed price for variable volume, and will frequently have been better off just taking spot – perhaps one reason for large hedge expenses booked by AGL and Origin. So what exactly is the point of a PPA for the buyer? If PPA buyers start to get cold feet, could it make it harder for new projects to get financing?
Posted this on Allan’s ‘negative prices’ piece, repeating it here. Could we perhaps have a think piece on renewable PPA’s from the point of view of the buyer, particularly a gentailer buyer? Renewables are increasingly achieving low NEM prices as their weight in the bid stack increases. To make this worse, I suspect that correlation is rising as geographical dispersion of assets falls. The PPA buyer is paying a fixed price for variable volume, and will frequently have been better off just taking spot – perhaps one reason for large hedge expenses booked by AGL and Origin. So what exactly is the point of a PPA for the buyer? If PPA buyers start to get cold feet, could it make it harder for new projects to get financing?
I think that low point this year tells a story, my guess is that rooftop solar have started to impact, and have created a lower low than the year prior. So there is a chance, the high will also not meet last years high.
However, weather will play the bigger role this year, but going forward in time, both the lows and the highs should track lower due to rooftop solar/ batterys?.
John, batteries are generally fully charged by 10am and fully discharged by the afternoon. The effect of batteries is therefore to accentuate the duck curve.
This is due to the default charging/discharge curve. An optimised curve would spread the charge over longer to help with the trough and delay the discharge to help with the peak.
Long way to go with batteries.
Hi Ben,
Is it not just a matter of putting more Batteries in and we have all the power we will ever need?
I thought there was no need for a MLF because inverter settings are meant to reduce output to zero when the voltage at the inverter reaches 258 Volts. This occurs occasionally in my unit between about 11 am and 3 pm on sunny days when there minimal load in our house. However there is no process to formally check inverter settings. How is this overvoltage curtailment captured in the AEMO and APVi data? Our curtailment data are captured in the Fronius database, which may be made available to these data sources.
Earlier this week there was an interview on the 7:30 Report with a spokesperson for SA Power Networks, where he said there was a time recently when there was so much rooftop solar output that the entire southern half of Adelaide fed back into the transmission network.
Aunty has been running a series on power problems and challenges with the grid and the obvious for summer is airconditioning demand management-
https://www.msn.com/en-au/news/other/power-companies-will-soon-be-paying-you-to-cut-your-energy-use/ar-BBXLRSN
Either that or peak pricing which makes you wonder how we’ll handle the EV revolution supposedly replacing our current mode of transport. Most will want to charge them at home at night but the solar duck curve suggests they should be charged during the middle of the day-
https://www.abc.net.au/news/2019-12-01/rise-of-rooftop-solar-power-jeopardising-wa-energy-grid/11731452
Interesting that the penny is beginning to drop with Aunty that something does not compute with aspirations vs reality when you want to change the climate or perhaps just hold the fort on that.
Hello Observa, I thought all the cars were going to be charged during the day, so the batteries in them could be used at night to power the grid. Or at least when there was not wind or sun for a few days?
It is interesting that the two interpretations of the same event both contain elements of truth. Part of the deal with Aluminium smelters, paper mills etc is that if you get power at 1.4-2.5 c/kWh, you are expected to drop demand in emergencies. It is common practice throughout the world including in Germany where customer downtime is an average of 12 minutes vs our 2.5 hours per year.
As you say wind output on the day was a bit higher than the previous 10 days but only half that of December the 9th, the first hot day of this summer. The cool change was definitely a help but another coal generator Loy Yang A 2 went offline at about 5 pm, more or less offsetting the reduction of demand.
I am not sure how you can claim that there would still have been load shedding if there were only two coal units out, because Yallourn 3 and 4 which were both offline by 1 am were 375 MW each and Loy Yang A1 of 500 MW was offline. Loy Yang A2 never got above 75% capacity and finally gave up at about 5 pm so during the critical period from 10:30 till 6 PM, coal generation was down by 1550 to 1850 MW. One Loy Yang unit and one Yallourn unit is a total of 880 MW. Even if we had one of each out and Loy Yang A2 struggled along at 75% capacity we still should have had 3,500 MW of coal which would have been enough to avoid blackouts
However on the overall point it is ridiculous to demand that we eliminate all risk of USE in a spindly grid like Australia’s where 95% of outages are T&D related. Back in 2009 we had three load shedding events of about 1,000 MW. Did the world stop.
Increasing use of local resources can in fact reduce the likelihood of cascading widespread blackouts. For example in Adelaide during their storm blackout, when the storm warning was issued if they had increased gas generation by a mere 150-200 MW, thereby increasing the headroom on the interconnector, no more than rotational shedding for half an hour to an hour would have been required. If South Australia’s current battery capacity had been available at the time probably no urban or Southern blackouts would have occurred at all
The most charitable thing to say about Ben’s comments is that he hasn’t looked at renewables output graphs for the last few years
Peter my comment that load shedding would still have been likely on 25 January with only two units out is supported by AEMO’s 2019-20 Summer Readiness report which quantifies general customer load shed that day as up to 538 MW (on top of RERT reductions of 625 MW). An extra Yallourn unit wouldn’t have covered this gap, while an additional Loy Yang unit would have left the balance very marginal.
And I don’t disagree that in the majority of circumstances the additional wind and solar commissioned in Victoria should leave supply-demand balance better placed than last year.
Allan
If the collapse of day-time prices as seen in the last 4 months was the cause of the lack of completed projects, then it was obviously well signalled and forseen.
But perhaps there are other factors preventing the 2019 completion of QLD projects, including:
– difficulties in getting network connection arrangements agreed;
– lack of purchasers interested in middle of day power volumes;
– State government electrical works licensing changes that dramatically increased construction costs;
– failure of building contractors;
– delay in QLD government renewable auction scheme;
– reluctance to compete in a market where the QLD government holds 85% of generation assets.
Thanks for noting some other contributing factors
I’m on my phone so I can’t check easily right now however what is the average MLF weighted by solar farm capacity between NSW and QLD?
It should be no surprise that wholesale prices will be affected when new generation enters an already oversupplied market. The day-only characteristic of solar determines when it impacts the market.
I’m not convinced a renewed subsidy scheme for wind and solar to reduce global temperatures is a positive step.
I’ll look forward to your assessment of price spikes in Vic and SA on the 20th. They coincided with what appears to be curtailment or poor production from wind farms during the heat. I live near the Macarthur Wind farm and it was blowing a gale at 40 degrees, prices were up to $2500/MWh, but output was only about 20%. Other wind farms were similarly affected, and some also had periods of constrained or nil production during the day including Mt Mercer, Morton’s Lane, Mt Gellibrand, Murra Warra, Oakland, Portland, Waubra, and the SE SA wind farms. If wind and solar are the new baseload we need to make sure their output can reach users during the heat when it’s most needed.
Here are Bureau of Meteorology data on wind speed at Hamilton Airport, which is their nearest observation station for the Macarthur Wind Farm, Morton’s Lane, and Oakland.
http://www.bom.gov.au/products/IDV60801/IDV60801.94829.shtml
Solar, wind, batteries, no doubt great for moving money from customer to industry bank accounts, but the real hero gets mentioned last, the Torrens Island gas power station, rapidly adjusting its massive output as required.
The real problem in SA, and elsewhere in the climate change obsessed world, is that nobody is going to build another Torrens, so the clock is ticking, not to any climate tipping point, but to a breakdown of the electricity system.
Hi Climan,
There is going to be so much cheep power 24/7, we in Australia are going to be making hydrogen, plus exporting heaps of power direct to asia.
Well thats what I am getting from all that i have been reading.
Per market notice 71933, the Heywood M1 500/275 kV transformer (taken offline at 13:55 on Friday 20th) did not return to service at 17:00 when scheduled in notice 71906. This continued to constrain inter-regional power flow from Western Victoria during the period of peak demand.
I live near the Macarthur Wind Farm, and was surprised that it was blowing a hot gale but the wind farm was only producing at <20% of capacity. For example at 17:20 NEM time, when the 5-minute Vic price spiked to $14,000/MWh, BoM's Hamilton Airport weather station recorded a wind speed of of 30 km/hr, gusting to 48 km/hr. Temperature was 40.6 degrees, having declined from a maximum of 44.5 degrees at 16:00 NEM time. The wind farm was only producing 60 MW (14.4% of capacity). It is rarely subject to grid constraints because it feeds into the 500 kV line to Heywood. Several Vic wind farms were off at the time including Mt Mercer, Mt Gellibrand, Murra Warra, and Waubra, while the operating Vic wind farms had low capacity factors despite being exposed to a similar wind resource. The SE SA wind farms, exposed to similar high winds and high temperatures had capacity factors of 60-80% much of the day, but went off at 17:20 NEM time with the loss of 200 MW.
Two hypotheses
1. When developers ordered generators for the Vic wind farms they didn't believe it was worthwhile paying any extra for high-temperature performance, because their contracts are only for the energy output without any premium for production when prices are high, whereas SA developers are more acquainted with such temperatures and have paid extra for better cooling systems in their generators.
2. TNSP's underestimated the degradation of their equipment performance over 40 degrees.
Until the pricing of power is changed, there will be times when, yes even solar and wind producers will, try to help the spot price up, its human nature.
The top price is simply just way too high, its too tempting.
But can the price ever be right without guaranteed dispatchability from every player at the regulator’s beck and call? Also presumably TNSPs would no doubt like to know what’s regularly required of them for that too.
I know Aldi have some cheap hardware from time to time but if it’s a necessity I go to Bunnings. It’s not that hard to work out the difference.
Some thing is wrong, no matter which way its looked at. The price of power has gone up 3 fold, and the fluctuations are simply unreal. The whole system, is a dogs breakfast.
We are now hitting the CRAZY high prices, and pretty Crazy low prices, more and more often, and everywhere in between.
Power is a needed for our system and country, and it needs to be stable in Both supply and price.
It is simply crazy, as it is 2019, almost 2020.
Obviously wind farms use monitoring software like this one I came across- http://www.scadaminer.com/
when searching for answers to temperature response of wind turbines and came upon this-
http://www.scadaminer.com/wind-turbine-temperature-performance/
“When purchasing a wind turbine, the power curve is guaranteed up to a certain ambient temperature (often 40°C). Beyond this temperature, the operating temperature of some components monitored by the wind turbine’s control system may exceed safe levels and initiate a temperature-induced constraint on the wind turbine.”
If it is indeed – as it appears at this stage – that a lot of existing wind farms suffer pretty severe temperature related output constraints above roughly 40C ambient that is… not going to help the public image of renewable energy. It will certainly knock dead any supposed advantage that deployed VRE has over thermal generation in terms of temperature de-rating. Can we put this down to equipment designed for e.g. European climates? Deploying gear in Australia that goes into limp mode once temperatures pass 40C seems foolish/near sighted. We all wait with bated breath…
Also worth mentioning that Loy Yang B unit 1 went offline yesterday with a boiler tube leak. It will also be interesting to see if there’s any repeat of the heat derating of the windfarms that we saw on 20 December. A couple (Mt Gellibrand and Yaloak South) are offline as I type – possibly not heat-related – while output at Waubra and Macarthur is declining.
I happened to be holidaying with family at Coobowie/Edithburgh in SA today in view of Wattle Point wind farm with its 55 turbines and for the last couple of days they appeared to be working well. Except for around 11.30-12.00 today when we happened to be returning from Troubridge Point in a dust storm with the car temp outside reading 40-41 degrees and the turbines had all shut down which you can see clearly here leaving only WPWF graphing-https://anero.id/energy/wind-energy
Not the only severe dip with the wind blowing strongly all day as you can tell by the wind farm output over 90% of installed capacity alongside the shut down. What is going on with these wind farms?
Re: (E2) Is it true that “Lines from Snowy to NSW have been destroyed by fire”?
Minister Matthew Kean was on ABC Sydney radio just before 6pm saying that at this stage he was being told that the ability to control the output of snowy was lost but that no expensive equipment had been destroyed so there should not be a prolonged outage.
The interesting point is that contrary to predictions it was NSW which went to $14,700/ MWh and Victorian prices went negative, indicating a strong oversupply in Victoria. At 5:25 pm after transmission was was restored, NSW was importing on all three links at limit values, power was still at $14,700/MWh and Victoria only rose to $34.
My summation is as I predicted a couple of years ago that NSW may actually more vulnerable than Victoria
Transgrid (via Twitter) reports upper and lower Tumut substations were subject to fire damage. Lower tumut restored, but upper awaiting road or air access to site.
When a transmission line is reclassified as a credible contingency due to lighting, is the transmission line removed from service or does AEMO have a plan/constraint equation ready in case it is hit by lightning?
Good question; no the line(s) would not be removed from service but a constraint would be applied to ensure the system was secure against loss of both circuits. This might for example involve limiting total loading on the line to levels below normal capacity. Allan.
“Teams of people from regulators and industry participants beavered away under the aegis of the ESB………… I haven’t looked at the relevant sections but I guess they run to dozens and dozens of pages, and associated guidelines and procedures produced by the AER (Australian Energy Regulator) and AEMO run to many hundred more.”
Oh dear when what we consumers really want is tenderers of electrons to the communal grid to be able to reasonably guarantee them 24/7/365 both with their frequency and voltage or else keep them. A level playing field methinks but perhaps I’m being all too simple minded here in the face of great complexity?
For the evaporative cooled plant, it might be worth plotting the capacity factor against the Wet Bulb temperature, as that limits heat rejection of the cooling tower and minimum condensor temperature/pressure.
Thank you Allan for an excellent summary of the situation on January 4. I feel compelled to acknowledge AEMO for its highly professional management of very difficult conditions over an extended period. And I should add to that the people at TransGrid who worked tirelessly on the ground in the the control room to keep the system up an running as much as possible.
Paul, thanks for the feedback. Those of us watching events could well imagine the extreme pressure TransGrid and AEMO personnel were under on that day, as well as many others over this summer. They certainly did a great job. Allan.
Paul
A couple of quick responses to your questions on pricing under RERT dispatch and observations on the southern NSW constraint.
When RERT is dispatched, intervention pricing (also known as “what-if?” pricing) is used to determine spot prices. This attempts to back out the effect of the RERT contracts (so add them back to demand if they were demand reduction agreements) and work out what the price would have been. If enough supply was offered below market price cap to meet demand with RERT added back in this way, the resulting price will be below MPC.
This differs from the situation where AEMO orders *involuntary* load shedding in which case the price *is* automatically set to MPC.
On the southern NSW constraint, its effect as I understand it is to limit aggregate flows from southern NSW into the rest of NSW, due to reductions in transmission capacity caused by the fires on January 4th (specifically damage to the Upper Tumut – Yass 330kV O2 line). Whether those flows come from Snowy generation, Victorian imports, or other southern NSW generators doesn’t really matter. Unless I’m missing something, Snowy’s generators bidding in more volumes at low prices doesn’t *reduce* the overall flow available into the rest of NSW, it just displaces other generation or imports within the same overall flow limit. When we see “forced exports” from NSW to Vic in this situation it isn’t because less power is getting to the rest of NSW, it just output from generators on the south side of the constraint which *exceeds* the available northward capacity finding its way into Victoria instead.
Allan
I should add though that Snowy’s generators being paid the NSW price even for the excess generation they force into Victoria isn’t really an economically efficient outcome, it just reflects one of the “compromises” in NEM market design where all generators in a region are paid the regional spot price, regardless of where their energy actually goes.
Thanks for the reminder about the mechanics of pricing under RERT, Allan. I was thinking more philosophically …
Viewpoint #1) I’d been thinking of RERT as an “out-of-market” intervention (i.e. in essence a sign that the market had “failed” to deliver sufficient capacity – and that, as a next-to-last resort, better to pay those who volunteer to turn off rather than proceed to (much less surgical) Under-frequency Load Shedding.
Viewpoint #2) However perhaps today is an example of a more risk-averse & precautionary approach to dispatch (i.e. the prices delivered might might be read to suggest that whatever was triggered in Reserve Trader was not actually have been required, in the wash-up (of course, with hindsight being a wonderful thing that just does not exist ahead of time – especially with it needing to be triggered (at 13:26) 120 minutes (or 90 minutes ?) ahead of dispatch starting 15:30 (or trading period ending 15:30?))).
Interested to know if others perceive things the same way, or differently?
The economically rational approach for Snowy would be to run the Jindabyne pump, which pays the Vic pool price (negative $200/MWh at times today so Snowy gets paid to use power), while also generating at capacity in Tumut 3 and getting paid $14,000/MWh in the NSW pool. These economic absurdities will strengthen the case for Humelink and Keranglink as transmission investments to increase the Vic-NSW capacity.
Is there a correlation between wind strength (not power output) and high temperature?
Because the charts show wind dropping off in the afternoon.
Several wind farms turned off during the afternoon – perhaps through transmission limitations. Boco Rock, Crookwell, Gullerin Range, in all about 350 MW missing, or perhaps displaced by more strongly negative bids from Tumut 3.
Several wind farms turned off during the afternoon – perhaps through transmission limitations. Boco Rock, Crookwell, Gullerin Range, in all about 350 MW missing, or perhaps displaced in their access to the transmission network by more strongly negative bids from Tumut 3.
Ben & Malcolm
There were a lot of different things going on. Extreme wind conditions were affecting some of the wind farms throughout the day, causing them to reduce output to quite low levels to protect equipment. Paul’s article mentions the constraint limiting flows north out of southern NSW – there was also a constraint limiting southward flows into Vic to 500 MW, and when this one bound as well it forced offloading of generators in southern NSW, including the windfarms. On a quick scan most of those generators had offered at -$1,000/MWh so should have shared the offloading proportionately, but Snowy had craftily rebid its downwards ramp rate at Tumut 3 so as to limit how much of this offloading it actually wore every five minutes, likely meaning that windfarms took more than their share of the output reduction. This sort of ramp rate rebidding has occurred in the past and needless to say has caused some controversy.
Allan
The abrupt shutting down of Wattle Point wind farm in SA I noticed on Dec 30th travelling to Troubridge Point was clearly a congruence of the high temperature (41-43degC on my car external temp gauge) and an obvious large dust storm that suddenly blew up above and beyond steady wind already turning the turbines. That seems to suggest that it’s not just temperature that automatically constrains these turbines but a range combination of wind speed and temperature which would vary with turbine type. That information for each wind farm should be available to the AEMO for forecasting purposes.
Operational AEMO staff called RERT providers on Wednesday afternoon, anticipating the possibility of breaching the LOR2 threshold.
On Wednesday the PASA LOR forecasting was playing up – putting the LOR2 trigger in the 1100 MW range (due to issues accurately forecasting intermittent generation in the PASA process), however operational staff anticipated that the actual LOR2 trigger was 750MW (an Eraring unit) and that supply/demand balance would be heading into this region.
Ambient temperature has zero effect on a boiler or a steam turbine. Heat exchangers in auxiliary services such as lube oil coolers and the condenser hotwell is where the effects of high ambient temperature come in.
High oil temps can trip a unit quickly, especially if a cooling fan fails, but gradual temperature increase can usually be halted by reducing load.
Gradually losing vacuum in the condenser is easily managed by reducing load.
(B2A) WE’VE SEEN NO EVIDENCE ABOVE THAT AMBIENT TEMPERATURE IS A SIGNIFICANT FACTOR IN FULL UNAVAILABILITY!
I was once an Operator at Vales Point Power Station
There are two things which can limit the load on the Generator
1 Outlet temperature of the cooling water from the condenser and
2 The back pressure in the condenser.
The cooling water outlet temperature on the condenser is limited to about 37 degree C. I retired nine years ago and that is what is in my memory. There is a lower outlet temperature (about 35 degrees C) which was the normal limit. If the grid was in danger ie…load was not matching demand then permission from a higher authority was required to raise the limit (37 degrees C) and I do not think that authority was from an executive of the Power Station. This is to reduce any damage which may occur of any marine life within the water source.
The cooling water outlet temperature of the condenser was directly proportional with load and the cooling water’s inlet temperature at the condenser’s inlet. During continuous high daily temperatures (as in summer) the source of the cooling water entering the condenser increases and as a consequence, the cooling water outlet temperature will also rise.
To maintain the condenser’s cooling water outlet temperature within limits will require the load on the Generator to be reduced. So, the volume of the steam entering the condenser, lowers the heat energy contained in the steam, matching the capacity of the cooling water. Energy in equals energy out.
If, as at Liddell, the water source is confined to a fixed source, then that source will increase in temperature as the cooling water is recycled around the condenser, also, as during the previous drought, over time that water source will diminish as a result evaporation. Another problem, the condenser tubes must remain highly conductive to allow the heat to be transferred from the steam/condensate to the cooling water.
At Vales Point the cooling water is drawn from Lake Macquarie. I always suspected a current was developed, in a stream, so that a proportion of the water flowing into condenser’s inlet had a component of water from its outlet.
Another suspicion I had we were drawing cooling water from Earing Power Station’s cooling water outlet. Eraring Power Station is located on the western side of Lake Macquarie, relative speaking, not too far north from Vales Point.
THE BACK PRESSURE IN THE CONDENSER
The most desirable back pressure in the condenser is 5kPa absolute (on the steam side) at all loads. As the load increases so the back pressure will also increase. The back pressure is maintained by the cooling water passing through the condenser. When the cooling water temperature is comprised and is unable to maintain the back pressure within the limits, then back pressure will increase as the load increases. There is a limit on the backpressure (which I cannot remember but I am pretty sure it is not higher than 10kPa absolute). When the back pressure reaches 15kPA, both the Boiler and the Turbine will trip and the Generator will come off the grid. The 10 and 15kPA pressures are estimates of what I remember those pressures to be.
This back pressure must be maintained to ensure the steam leaving the last blades of the LP Cylinder is in a superheated condition. At low loads, the vacuum in in the condenser reaches up way back into the LP Cylinder and as the load on the Turbine increases, the pressure in the LP Cylinder also increases. If the steam leaving the LP Cylinder has no superheat then the water partials in the steam will damage the blades in LP Cylinder (which are spinning at 3,000 rpm) that are exposed to the partials.
Now, as the cooling water entering the Condenser’s inlet becomes warmer, the capacity of the cooling water to maintain back pressure within a satisfactory limit becomes more difficult. When the back pressure is exceeding these limits then the load on the Generator must be reduced to ensure reliability (some load or none).
A factor in the transfer of heat of the steam/condensate in the condenser to the water cooling the steam is relative of the condenser’s cooling tubes through which the cooling water flows. If the tubes become dirty the efficiency is lower.
At Liddell, this cooling water is sourced from a relatively isolated lake and the cooling of their condensers rotates around the different parts of this lake and doesn’t have the benefit of a free flowing river or the high and low tides of a lake with a connection to the ocean. When considering the server drought in the recent past, these two conditions could have contributed a rise in the lake’s water temperature. Because Liddell Power Station does not seem to reach 400MWs implies to me there is a problem with their condensers
(This is my opinion) At Vales Point the cooling water rotates in a stream from the outlet side to the inlet with some dilution of cooler water in the lake depending whether the tide is flowing in or out also Eraring Power Station is on the western side of Lake Macquarie, not too far north of Vales Point. I always thought we were pulling water in from Earing Power Station’s outlet and this increased difficulties in the summer months for Vales Point.
SOMETHING ELSE
With renewable energy creeping into service and consuming energy demand, the coal fired Power Stations must be satisfied with what is left over. As this renewable energy takes more and more of the share this continues to reduce the Power Stations’ share. During summer months, the days being longer and the renewables performing better will leave less capacity needed to be made up.
Boilers, at coal fired Power Stations, have a low load limit. This is to ensure the safety of the fire in the Boiler.
At Vales Point this was once 300MWs with three mills in service with their feeder running not less than 50%. It was later reduced to 250MWs Oil torches where placed into service and the load reduced from 300MWs to 250MWs and when the fire in the Boiler was considered safe the oil torches were removed from service. This has since been lowered to 200MWs indicating the space load share was becoming difficult.
Because of this low load limit and the share over time diminishing there will come a point when a Unit from a Power Station will need to be removed from service. If it is Liddell Power Station, then 400MWs will be removed from the grid. If it is from another Power Station, it will be at least 660MWs. This can be serious because if this Unit “which has been mothballed or it is too inefficient or is too old” is required for service, it can take two to three days to synchronise and close on to the grid.
(Another opinion) In earlier years of commissioning Power Stations the Operators were sourced from steam ships. These Marine Engineers developed a lot of experience operating steam engines and the older “gingerbeers” passed their knowledge to the newer and younger recruits. All these Marine Engineers had a Marine Engineering certificate (part A and/or Part B). These certificates taught on the properties of steam which is very helpful when considering the condenser and other parts of the Unit to analyse what is going on. There are now no steam ships and no Technical Colleges courses supplying that help. Most of the Operators are now Electricians.
I hope this is helpful.
Thanks Allan for this excellent summary of the issues that – to some extent – all generators face in managing their revenue. From a policy perspective the key question seems to be: should generators be facing these headwinds – i.e. are they a fair reflection of the costs and benefits they offer to the grid? If so, then they would seem to be just the cost/risk of intermittency/weather dependency and it’s up to the generator to look for cost effective ways to manage this risk – removing these risks would stimulate more renewable development, but at a significant cost to consumers who would end up bearing those cost/risks instead. That would probably take several further articles to discuss, but would be interesting to read your thoughts on this question. I recognise that there are multiple parts to the question – in the case of MLFs, say, it seems reasonable that generators be given some locational signal so that losses are accounted for, but does the specific method of calculation do a good enough job of doing this? And then are there risk management tools available to the generator – and if not, why not?
What do you think caused the 700MW drop at 13:30 NEMTIME?
will get to that…
It looks like Snowy was holding back a bit to me. Curious stuff!
Hold the phone “The Heywood interconnector linking Victoria and South Australia was cut at 1.24pm”
What do you think caused the 700MW drop (VIC) at 13:30 NEMTIME?
After the trip of the 500kV transmission line at 1:45 pm Friday due to 8 towers being blown over just west of Geelong, it is interesting that the Mortlake gas power station now fully supplying the Portland aluminium smelter as a local circuit independent of both the Vic and SA networks. While the transmission structures are intact to connect this circuit to SA, by operating as a local circuit it doesn’t need to meet the NEM standards of requiring alternative energy paths in case of circuit trips, as well as standards of inertia, frequency control, etc.. These protocols of a local circuit from Mortlake were presumably developed in response to a power outage several years ago that was long enough that caused potlines to solidify.
Malcolm – some market view dashboards (eg AEMO’s own website) may give the impression Mortlake and Portland smelter are isolated from both regions, but they remain connected to the SA network through Heywood. The zero target flows shown on the Heywood interconnector are an artefact of the way dispatch in the SA + SW Victoria “island” is being managed. I’m sure Paul will post an update giving more details later this weekend. Allan
I can indeed imagine some fairly rude comments about the Federal Govt owned Snowy rebidding entirely for profit. There is a lot more than could be said. I also think that Friday proved once again the point Wattclarity has made on numberous occasions and is now “internalised” that wind output tends to be low in hot weather (ie summer peaks). However it is worth noting that system wide electricity consumption in mid winter is about the same as January and that wind tends to be much more effective then (but of course solar is down). Every generation fuel has its place in today’s market even if not in a future market.
Thanks for diving into that Paul.
I was going to provide some comment on how fast start plant is dispatched in the NEM, to partly explain what might have motivated Snowy to rebid this way (not necessarily to justify or approve of it!), but realised that’s probably a subject for a whole WattClarity explainer of its own.
Suffice to say here that when that $14,700/MWh dispatch price arrived (published just after 16:50) for DI 16:55 , Colongra would have received a signal to start up, but its “T1” time to synchronise is 10 whole minutes so it would not have delivered actual output to the grid until some time beyond 17:00, after the end of the 16:30-17:00 Trading Interval. The settlement price in *that* Trading Interval would have included the impact of the DI 16:55 $14,700/MWh dispatch price (six 5 minute DI prices are averaged to produce the 30 minute TI settlement price; the price for the half hour ending 17:00 ended up being $2,559/MWh), but Colongra would have produced no output until the *next* TI (17:00 – 17:30), when much lower prices might have applied.
Apparently Snowy was not willing to take this economic risk and thought the better thing to do was rebid the plant as unavailable in both the 16:30-17:00 and 17:00-17:30 Trading Intervals, allowing it to ignore the start signal received – and just as a by-product of this “optimisation decision”, triggering an LOR2 event as explained in your article.
Allan
Thanks Allan – so many layers of complexity to explain!
My focus was more on why the LOR2 arrived, not so much the motivation for the rebid decisions (though I understand people will go on their own tangents).
Thanks for the analysis Paul. I was aware that some generators game the system by withholding capacity to boost prices but letting the spare capacity fall to the rating of one of the larger thermal sets is really playing chicken with Mac trucks. If say Mt. Piper tripped at that time it would have really put the fox in the hen house
1. Will the 5 minute rule make such bidding practices more or less likely
2. SA prices have become much more stable with the installation of their batteries would 500-600 MW of batteries across NSW under different owners have defeated the gaming
The factors that drove Snowy’s rebidding are peculiar to the current 30 minute pricing / 5 minute dispatch setup, along with Colongra’s 10 minute startup time. The first will go away with 5 minute settlement, but how plants like Colongra will adapt to that regime are not clear. Can their startup time be shortened? How would they have to change their bidding and operating strategy to remain viable if it can’t? I’m not sure we have answers yet. And no doubt other issues will arise under 5 minute settlement – for example incentive to ramp output immediately, rather than linearly over 5 minutes, when prices go high or low.
Here is the Portland Smelter (APD) Load from 13:00 hrs on Fri 31 Jan 2020 through to 00:00 hrs on Sun 03 Feb 2020 .. http://nemlog.com.au/nlog/portland-smelter-load-on-fri-31-jan-2020/
The Smelter was offline from 13:24:56 to 16:09:32, a total of 02h 44m 36s with load gradually restored over the next 6-7 hrs. We can also see the output of two 294 MW Gas Units at Mortlake and the Heywood to South East Line Flows where it is apparent that AEMO moved to the reduce this flow to zero by approximately 16:47 hrs on Sat 01 Feb 2020.
Thanks Geoff!
That’s really interesting Geoff and Paul.
From the logs you’ve provided it looks like the V-SA interconnector remained connected through the fault and started supply back into the SW region. AEMO have then re-synchronised Mortlake to SA which is supplying the bulk of SA the FCAS (which explains the request for large amounts of FCAS in SA Friday night; can you check that please Paul?)
Seems that SA is still connected to SW Victoria & Mortlake. AEMO’s dashboard is not showing flows though (they’ve disabled it when the website went down?).
I’m not sure on Mortlake’s black start/grid forming abilities but I suspect they’ve used SA to do it. I’ve also noted the SA batteries got a hell of a workout Friday too.
It does appear that the design of the towers is nowadays inadequate to withstand extreme weather events. I would have thought that a engineering assessment of the transmission towers would have been done after the 2016 blackout. Maybe it was but the results not released? Regardless, the towers’ ability to withstand extreme weather needs to be upgraded.
Thanks for the insights Allan. I presume that there is some requirements as to how generators react to this type of situation, and that these requirements have become more detailed over time as grid management matures?
Chris – there has certainly been a lot of work done on protecting the South Australian system in the case of “islanding” events where AC connection to the rest of the NEM is lost, for example an over-frequency generation shedding scheme (OFGS) that starts to disconnect generation when frequency exceeds 51 Hz (and corresponding schemes dealing with under-frequency). However the circumstances of the 31 Jan event fell well beyond most scenarios explicitly considered – because the loss of both 500kV circuits in Victoria would have normally been viewed as extremely improbable. It will be very interesting to see AEMO’s detailed report on exactly how the SA system survived the event as well as it did. Allan
Hi,
When you remark “no guesses on that one” regarding the 15:30 NSW demand reduction, do you mean that the demand reduction is obvious or that you’re unsure about what caused the demand reduction?
The main reason the SA system formed a stable island is because SA was exporting at the time. If SA was importing from VIC at the time, things would have been more challenging. The portland load connected to SA following loss of the 500 kV lines would have actually helped limit the over frequency event in SA.
Agree that exporting pre-event likely made things easier Winodh – however the 4s data on the first chart above indicates that Portland tripped off very close to the time of the 500kV fault (as indicated by both the Portland load trace and the 1000 MW reversal in Heywood flow). We’ll need to see AEMO’s detailed report with high resolution data to know the exact sequence and timing.
Hi Paul thanks for the insights. I’m interested in the winners and losers of this islanding noting that the Hornsdale Power Reserve has made over $14 million on FCAS markets since the separation. Looking at the GSD2019 this is more than half of what it made on the FCAS markets in 2019!
Paul
There are lots of folks more knowledgeable than me on the fine detail here, but a few thoughts anyway:
1. As you show, the behaviour of PV in SA broadly assisted the situation by reducing aggregate output after a rapid jump in frequency when loss of the Vic 500KV lines disconnected SA and SW Vic from the rest of the NEM, and then during a sustained period of over-frequency (your earlier article on SA frequency showed it exceeding 50.25 Hz for about 20 minutes after the fault). Some of the reduction is due to systems tripping off, and some possibly due to systems reducing then holding down their output while the period of over-frequency persisted.
2. Expected behaviour of PV inverters in the fractions of a second after the fault depends on a lot of factors that we can’t yet see in the data – how rapidly the frequency jumped (RoCoF), the peak frequency reached in SA (almost certainly above the 51 Hz in your 5 second data), voltage disturbances caused by the line fault etc. It wouldn’t be surprising if many of the trips related to these currently non-visible factors.
3. Expected behaviour also depends on age of system – there were important changes in PV inverter standards in late 2015.
4. For inverters in SA complying with the new standard that remained connected through the fault, these ought to have reduced output (relevant to pre-event levels) while frequency remained above 50.25 Hz. At 50.5 Hz, where frequency sat in SA for most of that 20 minute period, this reduction in output should have been about 15% (if I’m interpreting the standard correctly – a big if). It would be interesting to know if you can see anything like this behaviour in the sample data you have.
Allan
The missing solar analytics data also probably tells a story. Although the monitoring units have a 3G/4G network connection, they are not battery powered and need grid power to collect and upload stats. if there are holes in the data (no samples) it most likely means the local grid was off. I am not sure how resilient 3G/4G towers are to power outages but they probably have some battery backup in at least some of the towers but towers might get congested if some towers run out of power in an extended power outage which was not the case in this instance.
Apart from lack of cooling for wind farms the density of air declines and as the available power is proportional to mv2 and m declines from 1.2 kg/m3 at 20C to 1.1 @ 45C there is almost 10% reduction in available energy, the high temperature also reduces the efficacy of cooling of the generator as well as the gearbox so a 15% reduction in output at a given wind speed is not unexpected in high temperatures. The change in air density combined with Carnot efficiency losses is why gas turbines can be even worse than wind farms.
Coal plants often have to heat and partially dry the fuel so although the ISO ratings are usually performed at 15 C some coal plants particularly those with high moisture content coal often reach peak efficiency at up to 25 C because the higher boiler efficiency offsets the lower Carnot efficiency of the steam cycle. The brown coal plants probably have a better thermal co-efficient because the boilers are oversized to cope with poor quality coal and in Victoria unlike SE Queensland and even the Hunter, high temperatures are less likely to be accompanied by high humidity so the wet bulb temperature will usually be a bit lower.
If nuclear plants were built they would usually be de-rated more because they operate at 260-320 C vs 550-650 C for modern coal plants so the % temperature difference between Tmax and Tmin in the Carnot Cycle is higher.
Finally IC engines with properly sized cooling systems probably have the least temperature degradation of all because Tmax in a diesel is around 2000 C so a 20-30C change in Tmin is not a big amount although there will also be a reduction in charge density in the cylinders due to reduced air density
Excellent, thanks Peter. Thanks for investing the time to be so detailed!
Should you not also include Inverters? As they also suffer in hot weather as far as i know?
Thanks John, essentially thought of them as a bundled package with the upstream energy conversion tech.
So part of a battery system is the inverter.
So where you say the output of the battery in Power is not affected, from your data, a battery and inverter (the inverter is the one making the power ac, from the battery which is only dc.) system does not suffer peak delivery power reduction at all? Just length of time at peak supply.
Thanks John – have added the figure from the GSD2019 showing full-year data for GannBESS discharging above in the article. Thought it was the easiest way to illustrate the answer to your question. In terms of instances of output (and availability) there is not a derating seen in the same way as it is seen for other tech. As noted on the image, the way this analysis has been done it does not highlight worsening round-trip efficiency.
Very interesting article Paul. I just want to point out that it is important to beyond direct correlation between output and weather. In Texas, we’ve seen fossil plants show up at higher than expected rates during periods of extreme weather stress. But, especially in July 2018, these events were fairly well separated. When I talked to ERCOT (AEMO’s equivalent) folks, they were clearly worried about similar summer conditions happening with more clustered extreme heat conditions, as they had some concerns that some of the older fossil generators might not be able to hold up under much more continuous use. So it seems that cumulative use can be a predictor of generator outages. Have you guys looked at that kind of analysis?
What is interesting is that Mortlake Power Station has stayed online and this seems to be the main contributor to the Westerly flows across Heywood post the trip (the 500 kV system was severed east of Mortlake). The Portland load should have been able to be maintained under this scenario so suspect there is something else that has resulted in Portland load tripping. The answer to this might sit in the first plot where the Portland load hasn’t reduced as soon as the Heywood flows reversed, but need the smaller timestep data and sequence of events data to confirm. Looking forward to the AEMO operating incident report
PS: This information and plots are great, no doubt a bit of effort goes into putting them together, but they tell us a good story.
Great article Alan.
The graph for Raise 6 Second FCAS shows the Hornsdale battery providing what looks like around 60 MW continuously for days. I thought it was a roughly 100 MWh battery. So how does it provide a continuous 60 MW FCAS supply? Any ideas? Is it being kept charged by the Hornsdale wind farms?
Fair question Paul. The so-called “contingency” FCAS services, including Raise 6 second, only require a supplier to rapidly deliver (or consume) extra energy when there is a significant frequency disturbance. So 99.x% of the time, a battery can provide this service without actually having to discharge. This is one of the major reasons, along with quick response and excellent control, that batteries have been able to occupy a large share of the market for this service. Allan.
PS So just to clarify, for these services, providers are being paid to provide a #response capability# measured in MW over a period of time measured in hours – hence paid in dollars per megawatt-hour of capability not energy, if that makes sense.
What drives the difference in FCAS between different solar farms?
Given renewable projects typically have contracted their output via PPA and thus don’t have spot exposure, wouldn’t it be more informative to show the FCAS cost they have incurred divided by their output?
Thanks for the feedback HR … could do. As time permits, we might issue an update
I see the ESB is worried-
https://www.abc.net.au/news/2020-02-24/solar-power-means-cheaper-energy-but-grid-instability/11993776
There certainly are a lot of acronyms in the power field nowadays but they seem to spell out the same message.
Near the beginning of this great explanation you stae “… subset of Ancillary Services supporting frequency control – keeping supply and demand in close real-time balance, reflected in system frequency staying within a narrow band around 50 Hz and returning quickly to that band if a disturbance shifts it.
Isn’t more of a phase shift correction that’s required? A frequency shift would have generators being at opposite phases for some number of cycles, thus a current short-circuit that increases and decreases as the generator phases slip through the 180 degree point — or the 120 degree point assuming this is a three-phase grid. Can transformers, generators etc. survive such frequency slippages?
Ike, I think you’re talking about “pole slip” or “loss of synchronism” events which have to protected against – very quickly – for individual generators or across interconnections. When supply and demand get out of balance for some reason the natural response of a synchronous system is to slow down or speed up through inertial effects, manifested in system frequency falling or rising (my original wording about “frequency shifts” may have been slightly rough, it’s really frequency *deviations* that FCAS respond to). Frequency control ancillary services provide controlled injections or withdrawals of power over various timeframes to arrest and correct these frequency changes. But if rates of change of frequency (RoCoF) are too high then yes that can lead to those phase angle problems you refer to which may even lead to system collapse.
Overall I think this is an excellent blog and provides some much needed balance on several issues.
However I was a little disappointed in the analysis of which energy source “saved the day”. In my view the coverage there is a little unhelpful as it merely seems to be a criticism of the various claims from “competing “teams”. But surely in analysis of an event like this it is important to determine which energy source you would like to have more of at your disposal. This for instance might allow energy planners to determine which source of energy might require more incentives than can be achieved by the current operation of the NEM.
As implied by this blog, you probably can’t determine what this energy source is by looking at the actual contributions. But it would be a useful “what if” analysis. For example at any given time, what energy source would have been called on if the load had been 10%/20% higher (or what generators might have been cut if the load had fallen by 10%/20%). It is probably that this answer would vary across the analysis, giving each of the various “camps” a win, but it is also likely that one energy source might be identified as the “preferred” across the majority of the period.
Thanks, Paul, for your excellent review of the SA islanding in February. As I understand, only a single 500 kV circuit has been restored west of Moorabool and as a consequence there are stricter export/import limits on the Heywood interconnector as pointed out by Allan O’Neill in his post yesterday. During the islanding there was a period when the normal dispatch outcomes for gas peakers appeared to be disrupted. Osborne was absent, possibly for maintenance. Barker Inlet and Ladbroke Grove also seemed be reduced in output, whereas diesel generators were contributing to greater extent than usual. Daytime energy spot prices of $300 were not uncommon in this period. This may have been a consequence of the greater FCAS requirements that were imposed in SA.
Whatever saved the day as The Advertiser reported-
“Electricity consumers will be charged a combined $93 million to cover the cost of stabilising SA’s power system while we were cut off from the national grid last month”
At last count SA had a population of 1.743 million but how that converts to the number of electricity meters and bill payers is not easy to find for some reason.
The SA Power Networks Distribution Annual Planning Report 2019/20 to 2023/24 on page 23 states the number of customers is around 900,000.
Will the forthcoming synchronous condensers soften AEMO’s islanding risk reduction requirements for SA, or reduce the cost of FCAS when interconnectors are out of service?
If the network owns the sync-cons, how do they charge or bid the services, and do they compete on equal terms with gas/batteries ?
The Electranet network appears to be only 10% Australian owned, with the rest shared between China and Malaysia. 65% of Transgrid is owned by Canada/Kuwait/Abu Dhabi. All the investment in this equipment will be recouped through the Regulated Asset Base (Electranet’s 2018-23 RAB was $2.7b without the new gear).
Past assessments have criticised network ‘gold-plating’ where RAB growth has outstripped demand growth. Demand is still not growing, but transmission is, with investment planned / progressing in several NEM locations, with the blessing of renewable lobbyists.
There is not much incentive for networks to maximise the use of existing infrastructure. The AEMC proposal to have generators reserve network capacity is a positive, ad is their decision to maintain the MLF specific to each generator (rather than averaging the losses). Building transmission purely to connect intermittent generation can only reduce network utilisation. That’s less efficient, that’s more cost than necessary.
It seems to me that we have a government guaranteed return, on privately owned monopolies, paid by consumers without a choice.
The foreign ownership of monopoly networks is worth noting, because of (a) the lack of competition and (b) the guaranteed returns.
The National Electricity Objective seems to have been forgotten.
“Will the forthcoming synchronous condensers soften AEMO’s islanding risk reduction requirements for SA, or reduce the cost of FCAS when interconnectors are out of service?
“If the network owns the sync-cons, how do they charge or bid the services, and do they compete on equal terms with gas/batteries ?”
They’re good questions and I don’t know all the answers. The syncons won’t be supplying any FCAS and therefore not needing to bid into current markets, but will indirectly influence energy and FCAS costs by reducing the incidence and impact of constraints currently used to maintain system strength and/or inertia in SA – both under normal operation and when islanded. Allan.
Thanks Alan. I’m also wondering if there has been any assessment of rooftop solar’s impact on FCAS?
Is there any correlation between FCAS and time of day?
Today is a public holiday in SA, VIC, ACT and TAS.
For once I disagree Paul. If there was an economy wide carbon price I might have more sympathy. I think you have misdiagnosed the problem. Cheap wind and solar is the appropriate source of energy but a system built around that, and that is where we are going, needs other elements, a bit more transmission and as inertia falls, a change in the control technology.
I’m not sure we’re in disagreement here, David.
1) An economy-wide carbon price would clearly assist in a number of different ways – however that’s stuck in political never-never
2) Wind and Solar is an appropriate source of an increasingly large share of energy
3) My comments are more related to the form of the incentives which have, in my view, led to a new wave of entrants who have come into the market without really gaining a real view of the risks (and opportunities) involved – and their obligations for participation (such as in this instance, as just one example).
This, unfortunately, has led to unpleasant surprises – and it is also likely to have led to a deployment pattern of wind and solar that is different than ‘optimal’, and might have actually accelerated these unpleasant surprises.
In simple terms (on the outside) it seems so much easier to understand an LGC price than the Spot Prices for Energy and 8 x FCAS (and the opportunities and risks involved), so that is where a disproportionate amount of attention has been focused.
The various regulatory bodies are also responsible for part of the problem. AEMO insists on calling emissions reduction a core function, and all their planning includes an emissions reduction focus. But AEMO’s role is not to plan emissions reduction.
It’s role is to maintain focus on the national electricity objective, which does not mention emissions. The ISP does not include a ‘do nothing’ scenario for coal plant life extension, not even as a foil for the radical ‘step change’ scenario. Investors will obviously take that as a signal that renewables have a favoured position in the NEM. But it’s clear as day that lowest cost electricity is to be found by extending the life of some coal plants and limiting the capacity of intermittent asynchronous generators, and maximising the utilisation of existing transmission.
However uncontrolled rooftop solar will be the worst market force, impacting all forms of grid generation.
Another part of the puzzle is the never ending funnel of “100% renewables” reports and academics spruiking 100% renewables with near zero engineering experience or input.
The CSIRO GenCost is hopelessly biased against the cost of coal as a fuel source, you have to wonder how that happened.
Nobody seems to mention Lazard LCOE since they made it clear they don’t account for ‘balancing cost’ or extra transmission.
I’d prefer states to take back the decision making process in the power system, I’m not convinced centralised unaccountable decision making bureaucracies are giving us the best outcomes.
Paul
You are clearly right. In my own very small way I have been writing about this on comments on Renew Economy etc. for only two to three years, clearly not as prescient as you.
Worse it won’t be fixed by new transmission. Most of the proposed new transmission projects will add to system costs rather than reduce them. For example Kerang Link has two functions,
a) to maintain a robust connection between Victoria and NSW in the face of fire and storm. On recent performance this would have had some value for about 24 hours over summer and would have saved a few million dollars of high price power
b) to improve connections to the Rhombus of Regret that would allow about 600-800 MW of power or about $1.1 bn of generating assets to connect to the load. The link itself will cost about the same again so in effect doubling the cost of energy
Paul
I suspect you are in Sydney. It was a public holiday in SA and Victoria today. So the low demand, while unusually so, is not unlike a Sunday.
Sun was up, so domestic solar was producing away, at least in Adelaide.
Sorry John, missed that one – my mistake.
Peter and Ben, who paid for coal fired power plants to connect to major demand loads from the location of sizable coal deposits? Who actually built the majority of these old power plants? The state did. So it’s reasonable, and makes most commercial sense or the system, that the state centrally plan and fund REZ and new transmission for REs. Contrary to assertions it doesn’t “double the cost” of solar power. In WA Sustainable Energy Now has modeled various penetrations of RE using hourly calculations for weather dependent generation and projected demand based on hourly historical demand and the cost of transmissions to go to 85% RE or 100% RE would be around $2B amortized over 50 years or more.
At some point in time, hopefully sooner not later, societies will factor in the high cost of GHG emissions, and the longer we wait the higher that will be. US$200/tCO2 would be reasonable today, if not much higher. With BAU to 2030 there we wont even be able to put a reasonable price on GHG emissions, the Great Barrier Reef and most other coral reefs will be destroyed beyond recognition and irretrievable? People point to the billions of dollars annual income from GBR tourism and (over) fishing, but is that all it’s worth to us? Again it comes down to the difference between measuring value versus costs. When the world is at 1.5 ºC and beyond $200 a dollar for CO2 will seem like a rounding error on the global carnage and disruption to economies.
(we’re already at 1.2-1.3 ºC above pre-industrial temperatures and 1.0-1.1 above the 1850-1900 average temp IPCC most commonly makes reference to — BAU has us heading for 3-5 ºC by 2100).
You think CORVIR-19 is a big deal economically? Think again. The US$5.3 Trillion with a ‘T’ dollars gifted to the Fossil Fuel industries by governments globally (source IMF) would be much better spent on planning and building not only 100% RE grids but fuel switching our entire economy off FFs.
Better still read this report “What Lies Beneath” by Ian Dunlop and David Spratt
link to Dunlop and Spratt report: https://www.breakthroughonline.org.au/whatliesbeneath
Unfortunately, we see some of the same misguided thinking in the comments to this article. I reiterate Ben’s point that none of the numerous 100% RE studies are of any value. None of them that I have seen paid any retail attention to the adequate supply of security services and so none of them contemplated the issues that have arisen at much lower penetration levels. This includes the ill advised AEMO exercise that the greens got them to do a while back (there were a few caveats buried in a footnote on p79 or similar…). Accordingly, these desktop exercises are of no value in assessing the actual physical outcomes.
Further, the comments about how older plant got built are fairly irrelevant – we all know there was a fundamentally different ownership and governance approach then so unless your point is that we should return to that fully centralised publicly owned model (in which case there would be no private developers to worry about) so what? In any case over 10 GW of dispatchable capacity was built post market start ( new coal, coal expansions, new gas etc.) so the market and the rules can’t have been that much of a problem for parties who knew what they were getting into.
I have some sympathy with the idea that renewables developers responding to government targets may have thought that those governments would follow through with a comprehensive plan to make their target effective, but at the end of the day if you are building a multi-million dollar asset, you do your due diligence and base your decision on the physical system and the rules as they are not as you would like them to be. Of course you can seek to change them since anyone can lob in a rule change. But don’t be surprised that if you predicate your rule change request on a plea to just have all your risk taken away and given to someone else that the AEMC will not go along with it.
As for the delay in building out transmission to potential REZs (important to remember that is what they are until all the generation turns up) that is not down to a lack of central planning and funding. The RIT-T essentially is that. But the vacuum in national carbon policy for several years has made it hard for any party to plan with confidence when and where to build billion dollar transmission links. Rightly or wrongly, the ISP has got around that by ‘deciding” when and where they should be built and the energy market bodies have since then largely got on with the job of accelerating the investment decision process while keeping some sort of oversight in place.
Finally, thanks Paul for pointing out the irrelevance of the ACT model to the NEM-scale transition. One so rarely sees this acknowledged (and certainly not by the ACT government)
25C max and sunny and 16C minm at night in Adelaide was picture perfect for solar rooftop with no aircon demand on a public holiday so you’d expect that’s as good as it gets.
“my understanding is that the round-trip efficiency of the pumping-to-generation process is not as great as it might initially seem”
I’d like to know more about that. Has it been the subject of a subsequent article?
Not a specific article, Mark – however there was discussion about several aspects of storage following the release of the Generator Report Card 2018 during the middle of last year:
https://wattclarity.com.au//articles/2019/09/some-observations-about-storage/
With the statistics being updated and extended with the Generator Statistical Digest 2019 we have the opportunity to delve further, and AEMO’s Quarterly Insights reports are another good reference.
There are many interesting developments yet to unfold.
1. It is likely that with Queensland gas prices falling below $3.50 with possibly further to fall that Queensland’s CC gas plants will undercut older coal plants and Pelican Point will export to NSW via Victoria. Combined with overall falling demand and increasing renewables, particularly rooftop solar, coal powered demand will fall faster than anticipated which may result in medium term withdrawal of one or more coal units in NSW.
2. Work continues on new wind and solar plants and methods will be devised to gradually reduce constraints on existing farms while demand stays depressed. Q3 may well see 30% renewables and coal share falling below 60% particularly if a wet winter tops up hydro resources
3. The economics of Snowy II, always dodgy at the best of times look like a disaster now.
4. There are conflicting reports of demand for rooftop solar but a sure way to put a lot of people to work quickly after the lock-down will be to broaden support for new rooftop solar and local storage.
5. Although the stink of the Pink Batts fiasco still lingers, a wide scale home efficiency upgrade program is another quick high employment scheme that might see the light of day, further suppressing overall demand.
Welcome aboard Marcelle. Fantastic article. Easy to read, supported with many relevant links and recommendations. Learnt a lot. Many thanks.
Terrific article Marcelle, contains a wealth of information about a set of processes that are pretty opaque from the outside – and it turns out not all that well engineered on the inside. The recommendation that bid max availability for S/S plant be treated as a cap and not totally ignored seems like a very sensible and easily implemented first step in addressing some of the problems we saw over summer with the current processes.
I wondered if you had any insights into the disconnect we saw in AWEFS forecasts feeding 5 minute and 30 minute predispatch / PDPASA processes on Friday 20 December – see https://wattclarity.com.au//articles/2019/12/out-of-the-blue-an-lor2/
the dispatch-level (5 minute) forecasts generally tracked actual output levels (but with some strange exceptions), but the 30 minute forecasts were off in la-la land for much of the critical period. Any thoughts on how this happened?
Allan
Thanks for your comments Allan. On your question about Friday 20 December and the 5 minute vs 30 minute forecasts, have a look at section 4.2 of the first document (“Australian Wind Energy Forecasting System” from 2016) at this link on AEMO’s website.
https://www.aemo.com.au/energy-systems/electricity/national-electricity-market-nem/system-operations/dispatch-information/policy-and-process-documentation
While the document is a bit out of date (doesn’t include the SCADA Local Limit for example), it explains how the 5 minute and 30 minute forecasts are calculated based on different inputs (real-time SCADA for 5 min, weather forecast + availability portal entries for 30 min). If there isn’t a semi-dispatch cap on a generator the 5-minute forecast inputs include the current MW generation, which would explain why these tracked well.
If I read this correctly could it be said that the battery is detecting the phase error between its internal clock and the grid and that phase error initially is much easier to detect than frequency change. There is still a measurement delay but it is much shorter, it can be a few degrees of rotation rather than a few revolutions. In that case it could be analogous to a servo control position system which has an outer position loop (frequency) and an inner velocity loop (phase). Then there is the question of voltage which must also have risen but I suspect that detection of voltage changes is much slower than phase differences.
Whichever way you activate it it is quite important in the first 20-200 msec and would reduce the chance of overshoot in response quite significantly.
Can we assume the same applies to synthetic inertia on solar farms and wind turbines
Paul, the media need stories and “nothing much has changed really” is not a story. Reputex know this and know that if they can give the media a story they will get some free advertising of their wares. Like you I’m not going to pay for the analysis, so I’ll take it with a huge pinch of salt because I can’t sense check it myself they way I do with publicly available analysis. Before you get too frustrated with this state of affairs, please bear in mind that today’s news is tomorrow’s chip wrapper (if only metaphorically these days with most of us reading online).
Your note in passing that 31 January was probably an underlying demand record for the NEM is worthy of more analysis if you ever have the time and inclination. In any case thanks for another interesting piece.
This article seems to claim that the grid forming inverter deployed at the Dalrymple BESS are actually able to provide an instantaneous “synthetic inertia response” performing like any synchronous generators. Let me just quote two peer-reviewed results on this topic :
1)”Effects of decreasing synchronous inertia on power system dynamics” – 12 June 2019
https://onlinelibrary.wiley.com/doi/pdf/10.1002/2050-7038.12128
“One of the main questions is how efficiently synthetic inertia can substitute direct physical response of the system (synchronous inertia). Published results show that response times are in the range of 50 to 200 ms, which are very fast, but in relation to time constants of dynamic power system events, they are far from being instantaneous.”
2)”Electric power system inertia: requirements, challenges and solutions”
https://www.researchgate.net/publication/327105525_Electric_power_system_inertia_requirements_challenges_and_solutions
“For power electronic connected technologies, delivering synthetic inertia and/or fast frequency control, there is a delay between the event the device’s response. The response time can be represented by a combination of the following four different parameters:
– Measurement time, which is the time needed to detect and measure the RoCoF (or frequency in case of FFC)
– Signal time, which is the time required to get the activation signal from the measurement device to the FAR device.
– Activation time, which is the time required from the FAR device to deliver the initial power response once it received the activation signal.
– Ramping time, which is the time required from the FAR to ramp up to the required active power setpoint.
However, an adequate response time depends on the system dynamics and requirements such as system inertia and RoCoF limits.”
Please can you provide the response time, to any Ro.Co.F event, of the grid forming inverter you are talking about?
Many thanks
Hi Francesco.
Thanks for the comprehensive comment. Your question strikes at my motivation to write the article – to clearly distinguish between virtual inertia and FFR.
Due to a lack of clear definitions – many people use the terms virtual/synthetic inertia and FFR interchangeably or as one in the same. However, I describe two very different mechanisms in the article, each with their own benefits as discussed. Grouping synthetic inertia and FFR as the same service doesn’t allow people to distinguish between them and undermines their use for certain applications. This is echoed in many papers eg:
“Many studies on synthetic inertia do not focus on distinction between inertia and fast frequency response. From a system perspective, important measures of frequency stability are the minimum instantaneous frequency, and RoCoF. A relaxed view on synthetic inertia might include services which contribute to improving the response of the system, such as lifting the minimum instantaneous frequency, and reducing RoCoF. However, a more strict, unified view of this term would lead to a clearer picture of services being offered to improve system stability.”
(https://www.researchgate.net/publication/321104126_Synthetic_inertia_versus_fast_frequency_response_a_definition)
Your Paper/Point 1)
This paper seems to refer to FFR as synthetic inertia. It is clear in the article’s plot from Dalrymple that there is no 50-200msec delay they mentioned for the commencement of the response – the frequency measurement resolution in the plot is 20msec and the frequency rise and the real power response occur within one of these periods. I would expect such a delay with FFR for the reasons I discuss in the article and believe this is what they are capturing in their literature review.
That paper was submitted in 2018 and FFR has been a much more prevalent service in the market to date. I’m not aware of another grid forming inverter with Virtual Synchronous Machine software running in parallel to a large interconnected power system such as the NEM. This technology has typically been used for off-grid or smaller power system applications. Dalrymple is among the first and as such, there is still limited awareness and information on what can be achieved with this approach – hence my article.
Your Paper/Point 2)
Again, they are talking about FFR here. They say “delivering synthetic inertia and/or fast frequency control” which groups them as the same mechanism and then they discuss the same measurement and activation delays only relevant to FFR that I call out in the article. With the virtual inertia delivered at Dalrymple, it doesn’t rely on measurement to respond to RoCoF. There are two voltage sources (Dalrymple and the Grid) and whenever there is a difference in frequency between them, there is a real power transfer. Real power transfer is equal to the product of the two voltage magnitudes of the two sources divided by the coupling impedance multiplied by the sine of the phase angle difference – just like a spinning machine’s inertial contribution. (We have a paper that describes this in detail that I can potentially share with you once made public – appreciate that this comment forum may not clearly resolve all your queries.)
So the papers you reference are correct but discuss FFR – they aren’t applicable to the virtual inertia mechanism I discuss. There needs to be clearer definitions established to help understanding here and hopefully this happens soon.
To your final question, the response time in terms of the commencement of real power response to a RoCoF event is instantaneous as seen in in the article. The amount of real power, rise time and settling time are all tuneable in the Virtual Synchronous Machine software layer and are optimised for the application and point of connection through connection studies – but the start of the response is immediate.
Thanks for taking the time to comment – I hope this is useful.
Cheers
Stephen
Grid Forming / Grid Following are not types of inverters, they are modes of control. They can be changed/configured as requested. There are also other characteristics of being in grid forming mode here which are ignored. A hybrid model/approach is not discussed here at all. Some sudo-science wothout factual comparisons.
Presumably the report prompted the same flurry to get some newsworthy comment from any sundry experts about islanded WA here-
https://www.msn.com/en-au/news/australia/falling-power-use-due-to-coronavirus-risks-system-overload-and-blackouts-experts-warn/ar-BB12NV6b
Dooming sells it seems but I note the negative correlation between MSM articles on global warming and coastal/esplanade RE prices so there does seem to be some considerable immunity among our movers and shakers. But with Covid19 and relevance deprivation syndrome for all other messages I guess you have to try a lot harder to rise above the milieu of bad news just at present.
Thanks for you answer Stephen, the papers I reported missed some details on clarifying differences among virtual inertia / grid forming / virtual synchronous machine, I agree on that.
I’ll try now very quickly. Virtual Inertia emulates the inertia response of a traditional synchronous machine (SM) mathematically via pulse width modulation (PWM).
Virtual Inertia based inverters and their controls strategy can be implemented in various applications such as BESS, statcom, solar plant etc.
The objective of the VSM (Virtual Synchronous Machine) is to provide grid forming employing frequency droop and Virtual Inertia by using swing equation.
The emulation of the synchronizing torque in the transient domain is still subject to a delay because of the inherently ‘time constants’ of the inverter.
Therefore it’s impossible to have the same instantaneous synchronous inertia response of a synchronous generator.
This paper should definitely clarify every doubt on this topic:
https://www.mdpi.com/2076-3417/9/24/5300
About the plot and the measurement, using 20ms measurement resolution would never be enough to catch delays of 5ms or less.
Best
Thanks Paul
these previous AEMO reports come to mind
2013 Wind integration
https://www.aemo.com.au/-/media/files/pdf/integrating-renewable-energy–wind-integration-studies-report-2013pdf.pdf
2015 emerging technologies
https://www.aemo.com.au/-/media/Files/PDF/Emerging-Technologies-Information-Paper.pdf
2017 Advice to commonwealth government on dispatchable capability
https://www.aemo.com.au/-/media/Files/Media_Centre/2017/Advice-To-Commonwealth-Government-On-Dispatchable-Capability.PDF
Plus several ‘system study’ and ‘dispatchability’ type reports.
Well switching off their rooftop solar dumping payments will ruffle a few feathers as they’re already grumbling about falling feed in prices being paid by retailers-
https://www.msn.com/en-au/money/markets/energy-operator-wants-to-remotely-switch-off-rooftop-solar-systems-amid-uncontrolled-growth/ar-BB13pbOi
I also note in April Australia’s wind energy output varied from 62.3% of installed capacity down to 1.3% so I can appreciate the AEMO’s problem dealing with the big picture- https://anero.id/energy/wind-energy/2020/april
Averages can certainly hide a multitude of sins but it would appear that flattening curves is very much in vogue at present.
Thanks Paul. Fantastic coverage of the situation.
Hi Paul, I was on the CSIRO team that produced input reports for their 100% renewables study in 2013. I have several of those still, so let me know if you’d like the main report or any of the input studies. Cheers, Geoff.
Regarding the peak CF trend:
1) Law of large numbers? (larger population means smoother statistics)
2) Increasing geographical diversity (fewer wind farms reach high CF at the same time)
3) There are more partially completed wind farms and wind farm size is getting larger so your aggregate installed capacity value is leading the actual installed capacity by larger and larger amounts? Your graph shows aggregate capacity currently at about 7.4 GW but it is actually only about 6.5 GW as of today. There’s nearly 900 MW of capacity still under construction/commissioning: Dundonnell, Yendon, Coopers Gap, Lincoln Gap,
Mt Gellibrand, Cherry Tree, Granville Harbour, Cattle Hill, Elaine. Most of these appear to be just in the usual process of connecting WTGs are they are completed and maybe some are at “hold points”.
4) More occasions where output is constrained especially in SA.
5) More occasions where wind storms are destroying transmission lines? (or am I imagining this)
Re the reduced ACF of wind farms the relatively recent rule changes, especially for SA, requiring larger specified synchronous generators to run to supply “system strength” may also have led to less “headroom” for wind generation, hence constraints. No idea about effects in other regions though.
Good morning Paul
I am a retired operator from Gladstone P.S.
On line forced cooling has nothing to do with FD/ID fans, it is all to do with boiler pressure and steam saturation temperatures.
Fans are used after the boiler is offline and drained.
Regards
Darryl
Thanks for the clarification, Darryl!
Thanks Andrew, sounds like a great project.
It looks a large amount of the value here is coming from realized spot prices underperforming traded cap premiums. I wonder if it would be possible to simulate outcomes from a quarter where the opposite occurred, such as Q117?
I am looking to a time when some more old coal plants are gone and we start to depend on the wind and sun for continuous power. Then there is the issue of the low points when there is not so much wind about. The aggregate is one of the metrics and it is exciting to see the penetration, especially with the projects that are in the pipeline and the impetus will build with RE investment for recovery. Still, the average is important as well as the aggregate and the low points are important as well because they have to be bridged with batteries and pumped hydro.
Totally agree Tom – the Q1 financial results were largely driven by relatively low levels of volatility compared to what the market was expecting before the start of the quarter. We will win some and lose some over time in this regard. The more important learning was how effective the battery was at providing ‘coverage’ against the baseline of 100% with a financial cap – approaching that levels of volatility will have an impact on this too from quarter to quarter. We were heartened by these results. Although the headline figures aren’t great (mid-fifties), if we’re able to tune out the issues we had with the control algorithm in the two rouge itnervals, the figures would be not far off what you’re able to achieve with a major generation asset (and cap seller) like Wivenhoe. Understanding what % coverage is plausible will help inform how much we can rely on the ‘virtual’ service in lieu of more traditional hedging techniques going forward when trying to prudently build a risk management strategy for each quarter. Of course 5 minute settlement will be a game charger for the effectiveness of batteries in this regard… although will still be faced with the problem of optimising the charge level!
Good read and made even more interesting to me having just read the Hydro Tas articles by Cameron Potter et alia observing that forecasting charge/discarge errors were less important with longer duration storage. I am still unclear how the battery can earn arbitrage and provide the cap service at the same time. Perhaps I’m reading too late at night. A gas plant can generate for the owner’s benefit when price is above SRMC and below the cap price and still be available (assuming gas supply) to supply the cap product. If a battery does the same thing it will sometimes be discharged during the price run up towards $300 and then be unavailable when called on to defend the cap won’t it?
Thanks David. You’re definitely correct that chasing arbitrage with a battery will sometimes leave the owner exposed when needing to defend cap positions due to the relatively short duration storage capacity available. This is where the trading strategy (either manual or algorithmic) is critical to try and get the coverage % as high as possible (but also not lose what may be otherwise good arbitrage spreads). It was interesting to me that even a long duration, manually traded asset like Wivenhoe only achieved a coverage %s in the 70s. The challenges of price forecasting and flow on impact to co-optimising the different services of the battery is discussed in depth in the full report, with a few case study days used to help illustrate these and the pros and cons of different approaches of trying to address.
Thank you David for sharing this analysis.
The caps contract are a financial edge,the BESS like a peaking plant are a physical edge. A physical edge has a capital cost that I think needs to be factored in the analysis to make the correct comparison between the two edges.
Interesting.
My thoughts, coming from having operated several windfarms, is that you can have quite fast and dramatic wind changes associated with fronts moving through. With a front, you often get a change in wind direction as well, and I wonder if the time for the WTG to react to the changed wind direction is affecting this. The WTG take some time to react and yaw to a new heading, with some time delay to ensure the WTG aren’t constantly yawing for each minor gust. With the WTG out of alignment to the wind direction, this also has a significant negative energy output, not allowed for in the forecasts.
You do also see the alignment of fronts across multiple sites, which might account for the same affect across the many sites, including MacArthur. It is was a large front, Lake Bonney would have reduced some time before the other sites.
I enjoyed this article and the webinar, thank you.
I’m aware of the many reports raising the issues that rooftop solar is bringing to the NEM and the SWIS.
Has there been any consideration on what is the
economic limit for semi-scheduled generation? I mean how can a solar farm, like the new one being built in the Western Down of QLD make money when the spot market is heading to zero during the day?
This is partly a rhetorical question because I’m aware that the QLD government has a 10yr PPA for most of the output. It is likely above market price, if so I call it a subsidy. I also would like to know what price the PPA set for the RECs, and if that price reflects a realistic market value.
But there is no slowdown in rooftop solar, so how do private entities make money out of a solar farm?
Over-egged.
I don’t think you need linear control of individual solar systems to get a linear response. A couple of hundred thousand individual systems with different setpoints looks pretty linear at a macro scale.
There is probably enough solar in SA to alter the market price for energy significantly. I’d expect the negative price periods would reduce if you switched off all the rooftop solar.
On the positive side of things, at least wind powered generator output from the two wind farms in QLD was at the higher end of their possible output. This really shows why QLD needs much more in the way of wind farms. Yes there are some big ones under construction, but QLD needs many times what is currently under construction.
Thankfully a bit of basic honesty here – “Home batteries still don’t pay for themselves” – that’s despite the obscenity of taxpayer subsidies, but the fundamental problem that batteries eventually die (probably within 10 years) has been ignored. So has the reality that sophisticated electronic equipment failure is very expensive to rectify. Neither battery replacement or equipment repairs/replacement will be subsidised.
People will be shocked when these realities hit home.
Many other serious flaws in this article e.g. “— then you’ll know that it’s not necessary to have any energy storage at all for renewables to supply the majority of a nation’s electricity” – not even worthy of comment.
Its time the rose-coloured glasses were discarded.
Nuclear is the best option.
I think the author misses the mark. In fact his own website carries a spreadsheet of home battery prices and I believe an examination will show that installed home battery prices haven’t fallen. I would venture a guess that the forecasts made by Morgan Stanley and many others were predicated on an assumption that residential battery prices would fall leading to greater uptake every year. Energia is another, very well respected by me, forecaster who had a similar idea.
A better question that the author might have investigated is given that both auto and utility scale installed lithium battery prices have fallen very significantly in the past four years why is it that home battery prices have been static or in some cases increased?
In my opinion the story is more interesting than the author has made it.I look forward to more articles on the topic here at Wattclarity.
Great to see forecasts being revisited and the reasons for the differences to actuals being analysed. There’s not enough of it, especially done with a sense of humour like Ronald’s. There’s no doubt that most electricity customers are economic rationalists who will only buy a battery system when the financial savings are obvious. But so many of today’s battery owners are looking for supply reliability when the grid goes down, or just the warm, fuzzy feeling of using all or most of their PV generation. The question I ponder is… how many of those people are out there, how quickly will that sub-market grow, and what could make it grow, or shrink?
The case of Callide B2 “off target” +17 MW above dispatch was due to governor frequency response with a drop to 49.64 Hz after Kogan tripped at 08:00 on 9th October 2019. Prior to the frequency disturbance it was spot on its target.
Maybe “Dunkelflaute” will describe the new Zeitgeist. Some may see this with Schadenfreude.
touché, Reinhard.
… but then that’s not German … doh!
Seems to me the capabilities of coal, nuclear, gas and hydro perfectly meet the needs of a modern power system. While wind and solar do not.
If wind and solar were fully exposed to market conditions we wouldn’t need these conversations at all.
Hi Paul,
Maybe this might have something to do with it?
“Neighbours take Bald Hills wind farm to court over health complaints” on 18th Feb 2020 in the Australian
“Victoria wind farm headed to court to fight claims of wind turbine syndrome” on 18th Feb 2020 in RenewEconomy
Could the wind speed be too high for the safe operation of wind turbines in this area?
Hi Dr Mackenzie – I note the submission concluded with changes. In short, did the changes solve the problem for solar and wind, or are there still other changes to be made?
https://www.aemo.com.au/-/media/files/stakeholder_consultation/consultations/nem-consultations/2018/causer-pays/final-determination—causer-pays-consultation.pdf?la=en&hash=1E9B2333C57273E0DC16034A7DA1F5A3
As a useful PS to this article, on Thursday 18th June , Warwick Johnston noted this:
Most houses already have a built-in remote controlled switching device – the hot water service ‘ripple switch’.
Worked well for decades.
The situation with rooftop PV is worse when you consider a black start situation. Imagine having to wait until night time before restarting the grid. Ouch.
Yes until we have a level playing field with electron tenderers to the communal grid being mandated to reasonably guarantee them along with FCAS 24/7/365 or keep them this farce will continue. With up to a third of electricity used to heat water for the home it makes economic sense to use rooftop solar for heating electric mains pressure storage HWS with solar diverter controllers like so- https://www.paladinsolarcontroller.com.au/
Any surplus can then be used to RC heat and cool the home for comfort after nightfall. The unreliables problem is political not technical but it will become increasingly apparent with the demise of coal and with it off peak power pricing.
From a policy perspective, we would want to see any price or control signals from the system or grid operators being set at the consumer front gate. While solar is a challenge at present, the future challenges will resolve around EVs, storage and other in-home devices. The consumer (or their agent) should be the one who decides which appliances respond to these signals. Unless there is a technical reason, the system and grid operators should not be reaching in a operating individual consumer appliances.
Anthony, yes to your sensible points. Central control of millions of individual inverters? There has to be a better way. Some would argue the ‘better way’ was several large generators close to load centres with minimum transmission.
The dilemma is that we are achieving lower grid utilisation by subsidising rooftop PV, while at the same time building more grid by subsidising wind and solar farms, renewable energy zones and interconnectors.
I can’t see a way out of the inverter problem except retro-fits.
But properly designed tariffs could help, especially if a penalty tariff is applied to bigger systems and those on a premium feed-in-tariff.
But if owners opted for a network optimised battery, they’d avoid the penalty tariff and receive a network support tariff.
That would help all consumers by avoiding distribution upgrades.
Then EV chargers – I think you have to put them on the ripple switch anyway.
This has been a weird month for wind power. There have been some extended periods of low wind on each side of some great wind days on the 13-16th and 19th. There were also some periods of very rapidly rising and falling wind power. This must be causing some problems for grid management. Do you think it will get worse as the installed capacity increases or do you think it will settle down?
Paul, if you are looking for another mental model of wind farm output may I suggest you review a recent article I posted on this subject https://www.linkedin.com/pulse/feature-spectrums-bruce-miller/.
Hi Paul, regarding your question about national cabinet, the AEMA and NER vests the authority with energy ministers, rather than first ministers. So if COAG EC were to be supplanted, I presume it would have to be a sub committee of the national cabinet. I doubt first Minister’s would want to be saddled with all the responsibilities of being an energy minister (energy management, derogations etc.). Cheers, Jon
Any attempt to find a recognisable market, in the common sense of that word, in Australian energy is doomed to fail.
It is a house of cards, with additional string and sealing wax, assembled to hide rooms full of legalese and BS in equal parts. Am I right?
Yes, John. But rather this is yet another lesson (for those willing to learn from it – little sign yet – of how regulatory fiddling (including market-sidestepping subsidies) and trying to pursue conflicting objectives have yielded what we deserve. My late-career interests are in mathematics generally of “complex” systems (cryptically: “where life meets reality”) – this is a really beautiful demonstration.
Paul, Welcome to our world. You may do well to have a look at what is happening in Western Australia, a relatively small isolated system now operating with over 1200MW of installed behind the meter PV and no storage. A read of the recently published AEMO Statement of Opportunities (SOO) for WA delves into this in great detail. Now WA you would recall is a capacity / energy market. However we have now reached the point where WAs largest generating unit (350MW) is regularly shut down over weekends simply because it is too big with insufficient turn down to be managed within the residual demand left after Rooftop PV and the ever growing wind capacity (Driven by the tail end bubble of the RET). WA now only has about 1600MW of coal capacity in 7 units and this will shrink by 400 MW over the next 4 years as older units are rendered uneconomic – simply capital reinvestment cannot be justified with their declining capacity factors. But hey, this is what the RET was all about. So fortunately the balance of conventional generation is mainly gas turbines and open cycle at that. And with a capacity market that will fund more there is some chance of keeping the lights on. Apply the same PV / renewables outcome to the energy only market dominated by relatively inflexible coal fired units in the NEM and you are going to have a serious problem. Zero / negative pricing is simply the first symptoms. Worse still, burn the financial viability of the incumbents and expect them to come back and put capital into new (probably GT based) capacity will be an interesting ask, particularly when profit seems to be a dirty word in the electricity industry generally and the expectations of cheaper electricity in the future still seem to abound. I watch with interest. Suffice to say I have a small generator in the shed….because I don’t drink warm beer in the dark for anyone!
The back end acceleration is hardly surprising. Given the RET scheme had a drop dead date this year, and that the obligation was on retailers “to procure” and generally the big retailers are vertically integrated, bringing on new generation at (say) $60-70 / MWhr LRMC that simply canibalises their existing fleet (able to trade at SRMC) why wouldn’t you kick the can down the road as long as you can? And over the years of the RET scheme the technology got better and cheaper.
Many of the really good sites (best wind resource and easiest network access) were probably snapped up first in the process (read South Australia) so lower capacity factors in latter projects may well be expected. Allowing for some seasonal (and maybe annual) variability in wind and it may explain much of it. Might be worth looking again with a couple of years of clean data given many of the projects need to be delivered by years end for RET purposes.
Great article. Excellent explanation on how grid forming inverters help maintain frequency and the difference between FFR and virtual inertia.
Thank you.
If you take ideas to their logical conclusion, and ignore the reasons for our current predicament (ie perfect hindsight), then all generators over a minimum capacity (30MW is convenient) should be scheduled.
Nobody starting a grid from scratch would allow such a high proportion of unscheduled generation capacity on the grid.
But it’s not too late. If all generators above 30MW had to operate as scheduled generators – either by contract with another generator, or by adding technology), the market matching supply to demand would function correctly. Also reduce the LGC price by 20% per year for 5yrs, then you return to an equitable market.
Also, if scheduled generators were proportionately penalised for not meeting dispatch targets, reliability would increase.
And if the shortfall penalty was distributed proportionately among the generators that made up the shortfall, the cost of making up generation shortfalls would fall less on consumers.
Markets are meant to work for consumers.
I can’t help but think that uncontrolled rooftop PV is by far the biggest risk to the NEM. And always increasing risk at that.
I’d love to see more about the wholesale price volatility, particularly around the less and less offers in the 0-300 range, and which price ranges are having the most effect on setting the spot price.
For example, spot price is currently low, but if the proportions of price offers keep changing, will the prices stay low?
Am I right in thinking that volume weighted prices are (a)one more degree of difficulty to compute and (b)more relevant to the business outcomes?
I’m thinking that the generally lower prices for the generally lower load periods (ie night, etc) tend to skew the time weighted prices lower than would volume weighted prices.
Heat stress and all that impressive collaboration I see? It could all do with asking about cooling the past-
https://wattsupwiththat.com/2020/07/10/hottest-day-ever-in-australia-confirmed-bourke-51-7c-3rd-january-1909/
That would be a doozy alright.
Great article. A question about your curtailment calculation, where you estimated that 1.7% of generation might be curtailed under the rule change. Was this calculated using 4s data, or just the 5-minute initial_MW and total_Cleared data? I used 4s data for several sites for a couple of different months and obtained a significantly higher curtailment estimate.
Thanks David. I used the InitialMW data shifted 5 minutes to line up with the corresponding TotalCleared (dispatch level) value. I approximated (as shown in the diagram) the curtailment as a triangle – i.e. (FinalMW – TotalCleared) / 2 * (1/12). It was far too much work to look at the 4 second data!
Thanks for the quick response. Agree it’s a lot of work, hence I’ve only done a handful of generators and a couple of months. But output can vary greatly within a 5 minute period, often ramping above the target. So I’m concerned your curtailment estimate using 5 minute data is considerably below the true value if generation has to remain below the target at any given moment. My estimates are approx 3% for wind and 5% for solar.
The only opportunity for wind and solar to put downward pressure on electricity cost is on wholesale prices during favourable weather.
Wind and solar increase the cost of all other components of the electricity chain.
The correct definition of ‘the problem’, is that there is no market for intermittent electricity. The market matches supply to demand. It does not match demand to weather.
When you phrase ‘the problem’ in these terms, it is clear that all significant generators participating in the market should be able to perform that fundamental function.
Great article Marcelle – especially the analysis refuting the AERs (or COAGs) conclusion of the SA output more than 20 MW during negative prices graph was due to deliberate of generation (it was so dodgy when I read the AER paper that instantly dismissed it but you provided some great data driven reasons), and the reference to the NZ market solution that is a possible pragmatic solution.
Sorry, but I can’t really see the problem – surely the market is simply saying causer pays. You cannot keep putting generation into a power system that cannot be relied upon to “do what it says it is going to do” once it becomes a significant share of the market. And certainly not when ancillary service markets (including LFAS) are simply not reflective of the cost of providing the service. How about instead of looking for free rides all the time that developers fix their limitations by either contracting with others to correct their market position for them, install storage to self regulate their position across trading intervals or even auxiliary generation to clean up after themselves. Yep it is going to cost some money – perhaps the immature conversation around LCOE being the sole comparative metric in the generation space might be about to change – not before time.
Since SMEs are the biggest consumer of NEM supplied electricity is it possible to post the Victorian SME map for June 2020 v 2019 and Apr-June 2020 v 2019.
Although total Victorian consumption is only down by 2 percent think its useful to take into consideration that 4 of the AER’s 10 Expectations announced in late March 2020 will expire on 31st July 2020. In the absence of a new AER Statement of Expectations the expiration of the existing AER expectations is likely to significantly impact on consumption.
eg 4 of the existing AER expectations are:
– Offer all residential and small business customers, including small businesses eligible for the JobKeeper Payment, who indicate they may be in financial stress a payment plan or hardship arrangement, regardless of whether the customer meets the ‘usual’ criteria for that assistance.
– Do not disconnect any residential or small business customers who may be in financial stress (including small businesses eligible for the JobKeeper Payment), without their agreement, before 31 July 2020 and potentially beyond.
– Do not disconnect any large business customer, including businesses eligible for the JobKeeper Payment, without their agreement, before 31 July 2020, and potentially beyond, if that customer is on-selling energy to residential or small business customers (for example, in residential parks or retirement villages).
– Defer referrals of customers to debt collection agencies for recovery actions, or credit default listing until at least 31 July 2020.
The graphs in section A and B showing the increasing volume of $0/MWh bids is very interesting.
Would you consider doing the same trend separately for day-time periods and night-time periods to more directly observe the influence of increasing rooftop PV on bidding behaviour?
Very interesting article Warwick. However, I have to take your “backcasting” of battery revenues with a pinch of salt. Firstly, on a global level the AEMC’s analysis showed that the difference in historical market revenues/costs under 5 or 30 minute settlement was negligible. in other words, 5 minute settlement is only worthwhile if it changes bidding behaviour as compared to 30 minute settlement. in which case you can’t predict future outcomes from historical bidding patterns.
Setting aside the question of the extent to which bidding dynamics change as a direct consequence of 5MS, numerous other factors have changed over the last 8 years such that the past may not be a good guide to the future. Legislation and rule changes have strengthened the regulator’s hand against bidding that is not in “good faith”, which in itself should reduce any gaming. market conditions are also eroding the market power that has been imputed to certain generator portfolios. It’s no surprise that the biggest gains in your analysis are in SA and QLD, given the main allegations of market power abuse have been against AGL in SA, where its TIPS A and B are a significant share of the total capacity and against government owned generators in QLD. In the former case, TIPS A is on its way out (although AGL has a smaller faster plant Barker Inlet that has recently fired up) but more importantly, there are already 3 large batteries with more coming and soon a new interconnector with NSW. In QLD the main change is that the government has instructed its generators to bid differently, which admittedly could easily unwind. For a range of reasons and because 5 minute settlement will remove the incentives for late rebidding in any case, the p1 or p6 spikes you refer to seem less likely to occur.
Even when they do, the arbitrage opportunity will be eroded by new entrants. I don’t know what the equilibrium is, or how fast it will change as the generation mix changes, but there may not be that much new battery capacity needed to make the next battery uneconomic. It didn’t take much new capacity overall to make SA’s FCAS markets competitive again (at least when the state’s not islanded…) after Northern’s exit.
We can also logically expect an increase in DR. how much of that will be due to 5 minute settlement and how much to the new wholesale DR mechanism may never be clear.
Finally, the wholesale market works well when both buyers and sellers have an incentive to hedge out volatility. it may not be a great outcome for the market to have lots of batteries happy to play spot price arbitrage rather than lock in a revenue stream by selling caps. logically this leaves some load on the buy side exposed to those price spikes…
Hi Kieran,
Two of the best traders in the NEM are Harry Hindsight and Felicity Foresight, who can always see the future. So, yes, the backcasting is not a reflection of reality. I don’t believe this is too much of problem in NSW, VIC and QLD where the price spikes likely occurred much of the time at times of peak demand, so whilst you’re uncertain of the pricing in predispatch forecasts, you still likely know when the maximal prices occur. SA is different as price spikes and dips are less predictable, so you are less likely to take full advantage of these. These figures are imperfect but present a rough guide to potential values. You will probably underperform against these, but it’s plausible that you could outperform if you judiciously did some extra charge/discharge cycles.
It’s not surprising that adding the revenues/costs under both 5 and 30-minute settlement makes little difference to the whole market value as the usual change from one dispatch interval to the next is rarely more than 10’s of MW’s. This makes little difference to stable loads or generators where time weighted and volume weighted prices are almost the same. Those however at the margin, who operate for just a short while will be impacted the most such as high power batteries.
I would also agree that much has changed over the last 8 years, including the fact that all the history represents bidding under half hour settlement and for much of that time, no batteries participated in the market. Other factors such as “good faith bidding” seem to have crunched QLD arbitrage values in recent years. Changes in the market cap, value in FCAS and other issues also have an impact.
Predicting the future arbitrage is challenging. It’s difficult enough to forecast future electricity prices. I agree with your suggestion that P1 and P6 spikes will not be as prevalent as we should expect price spikes to be evenly distributed. Arbitrage is not unlike FCAS markets in that in probably does not need more than a few 100MW to begin to erode the market value.
What I’m trying to achieve is to highlight the difference than 5MS can make to batteries and some insight into how different regions and configurations impact the potential return.
regards
Warwick
Thanks for your reply Warwick. I agree the debate about the scale of the arbitrage opportunity doesn’t undermine your basic premise which I think is largely correct. It will be interesting to see the rate of battery deployment over the next few years.
In the meantime I hope your friends Harry and felicity are working in AEMO’s ISP team now…
This is only going to get more and more common as the magnetic south pole shifts further north.
It has already exited Antarctica and is on it’s way towards south western australia.
Thanks Warwick for the interesting perspective and backcasting calculation.
First I have to say that I am not very neutral in the discussion since I work in Business Development for BESS at Wartsila 🙂
We did as well some backcasting simulations over the year 2019 for a larger BESS (based in NSW) and considering the current 30min settlement, and landed on an optimal battery duration between 2C and 1C. Our simulation was considering a merchant application FCAS+Arbitrage, and the purpose was to check what BESS size would have the highest NPV.
I do agree with you below conclusion that the 5MS will further enhance the value of power vs energy.
“One of the key beneficiaries of 5-minute settlement will be high power, short duration batteries”
One question, not related to BESS ; The market price cap (maximum) has been recently updated to $15,000/MWh. Why has the minimum price cap not been updated ?
Cheers,
Loic
Thanks Loic,
I think it is challenging to make the business case for batteries in excess of 2 hours given the pricing I’ve seen unless there are other value streams outside of FCAS and arbitrage requiring a longer duration. On the other hand, whilst your return per MWh is highest for high C-rating batteries, the installed cost per MWh is also higher, so the best NPV is getting the balance right in terms of installed cost and expected return.
I’m not sure why the market floor is still only -$1,000/MWh. I believe a lower price has merit. It’s also interesting to note that whilst some might observe that negative prices help batteries, which is true, the value is still in discharging for a short time for high prices i.e. you’d need 7.5 hours at -$1000/MWh to equal the value of a single half-hour discharging at $15,000 and having a 7 hour battery would have a woeful return compared against a high C-battery.
regards
Warwick
As usual fantastic analysis. I think the point about seasonality is going to be writ very large indeed across the NEM and will be the next shoe to drop. Solaredge inverters are just starting to make their presence felt in australia. I wish I could restrict our household consumption to ~5 MWh. Sometimes I feel like a small industrial company.
Hi Andrew,
Just solely looking at the electricity savings, don’t you think it would be more realistic to assess the payback separately? i.e. you’ve estimated a payback of 7.2 years for only the solar i.e. $8,000/$1118 = 7.2 years and the total system cost/ savings = $18,000/$1438 = 12.5 years, so then the payback on the battery is really net battery cost/annual savings as compared to solar so…$10,000/ $320 = 31.3 years.
This obviously places no value on the potential for UPS or the ability to use your own energy but these items will tend to have a different value for each individual.
Without knowing the exact layout of your house and the regulations in Queensland
1. Would it not be possible to add another 1-1.5 kW of solar panels preferably oriented in a different direction which would be clipped when total output reaches the 5 kW limit of the inverter but would make a useful contribution in winter. It would seem to me that that would do more for your self sufficiency than more battery capacity
2. It must be possible to develop a smarter system to arbitrate between the battery and the HWS, either by allocating alternating 5-10 minute blocks or putting a PWM controller on the HWS so you can regulate the power flowing to it so that depending on the hot water temperature/battery state of charge you can vary the rate of battery charging vs hot water heating. I understand that the closed nature of the software on the various bits and pieces you have now makes this difficult, but it is probably easier than getting the regulatory changes you desire
@David – thanks for your kind words. I was well aware conceptually about impacts of seasonality, but surprised by the magnitude of it in reality once looking at the data.
@Warwick – agree that’s correct when considering the ROI of the battery in isolation. As someone who didn’t already have PV though, I was more interested to compare the two options for the overall ‘system’ as a whole. The PV element certainly helps subsidise the battery part when looking through that lens, but nonetheless is able to provide an overall system that helps achieve my other goals within a not entirely unreasonable payback. As discussed in the article, quite a few things that need to change to make home storage more attractive to more people. Further, first thing I’d advocate to those who already have solar (or perhaps don’t want to spend as much upfront) is to put a hot water controller in – you’ll end up with the bulk of the benefit of a battery with only a fraction of the cost.
@Peter – agree complete about more DC capacity being the solution to help address the issues in winter. Another reader over on LinkedIn kindly pointed me to a document from SolarEdge about the ability to oversize their HD-Wave inverters by 155%, which would give me up to 7.75kWp. I was previously under the impression the limit was 130%. I’m going to look into this further. And to the second part of your comment, whilst that would help ensure maximum bang for back for each interval (if I had spot exposure, which I currently don’t), it doesn’t address primary limitation, which is to be able to discharge and export energy stored in battery that is surplus to be daily needs – eg. if I need 6 kWh on average overnight each day, I’d like to be able to export say 4 KWh into the grid on top of this.
I did the same thing in 2014, but I purchased a hybrid car. It is interesting to see a Queensland report the results of which are roughly what I would expect. Basically, solar PV saves much more than batteries, because the energy saved by a battery is mostly night-time energy.
Grid savings from the 3 kW solar PV system balanced car charging costs, and in the first year compared with the previous year and car, I avoided purchasing 1000 L of petrol, saving $1,300.
I presented two papers on my experience, found at: http://anzasca.net/paper/going-solar-the-ramifications/ and http://anzasca.net/paper/how-big-should-my-battery-be/
Would you consider optimising the charge cycle to better support the grid? Eg charge during the mid day high solar period, and discharge during the peak periods? There are tariff considerations clearly.
The hot water heating period looks good from the network perspective. Discharging the battery at the same time seems obvious from your economic perspective.
Would network complimentary tariffs be something you would take on if the opportunity was presented?
Great article, thanks Andrew. Fantastic analysis and I couldn’t agree more with your clear and logical final thoughts. A few points:
1. Looks like throttling the hot water system in winter to around 2.5 kW and spreading the peak could give you a small quick win if that’s possible, although maybe better to just figure out how to get the controllers to play nice with each other.
2. I suspect the lower-than-advertised battery efficiency just comes down to the old saying “your mileage may vary” (almost literally). Battery efficiency is complex and depends many factors like temperature, charge and discharge rates, depth of discharge, cell degradation, etc. I’m not surprised it’s a few percent lower in your installation than in Tesla’s marketing material, which is probably based on a standardised, more favourable application.
3. Agree with Warwick’s comment that the financial case for the inclusion of the battery doesn’t come out looking great compared with solar alone, and this might be useful analysis to add to the article. However, this is a great demonstration for energy and sustainability geeks of how solar+battery stacks up for energy self-sufficiency in 2020, for those of us who like the story beyond the hard financial numbers. Not to mention that without any changes to technology/costs, the sensible rule/tarriff changes you have proposed would significantly improve the case for adding a battery (and worsen the case for solar alone).
@Ben – this is controllable via the Tesla app, but relies on timers as opposed to the mode I currently have it in, which dynamically adjusts charge rate to only use surplus solar. On balance I prefer the ‘solar self consumption mode’ ion the absence of any exposure to spot price etc. I’m certainly open to any kind of tariff structure that would be a bit ‘smarter’ than the typical offerings out there, although note in a perfect world that if I was able to achieve a higher self-sufficiency % the $/kWh tariff becomes largely irrelevant, but $/day fixed charges do continue to matter a lot.
@Jonathan – thanks for your kind words. Your suggestion about the 2.5kW set off a lightbulb for me – thanks! I realised that the SolarEdge Immersion Controller lets me easily adjust the kW setpoint for the HW system remotely via the app. I’ve now changed it to 2.5kW but starting an hour earlier in the day. Will play with this over the next couple of weeks and look to adjust dynamically as the season changes.
Re: round trip efficiency – our report into the 1MW/2MWh Powerpack system also found lower than advertised round-trip efficiency (which is measured under STC, which are less onerous than the Australian climate!): https://sustainability.uq.edu.au/files/11868/EPBQtyRptq12020.pdf
I’ve had the feedback from a few people that the way I presented the payback for the solar + BESS option was a little misleading – this wasn’t my intention, as to me (who had nothing to start with) the blended payback for either solar alone or solar + battery was the choice I had. I agree batteries in isolation are substantially less compelling, and hence my suggestions at the end. I’ve updated the article where I first posted it (on LinkedIn) to better represent this.
Just remember when there are gaps in reliability with transitioning it’s the regulators wot fail to anticipate and provide insurance wot will get it in the neck-
https://dailycaller.com/2020/08/17/california-blackouts-renewable-energy-california-gavin-newsom/
That’s some raw on target for you and they couldn’t pay me enough to be a transitioning regulator.
Snowtown 1 Wind Farm and the Hallet Wind Farms are Suzlon S88’s, which are older technology turbines. It’s likely they are more sensitive regarding high wind cut out. This would explain the difference between Snowtown 1 and Snowtown North/South which are co-located, but different turbine technologies.
It’s important to note that the issue of High Wind Cut Out has been mitigated in AEMO’s 2016 ECM updates, which require generators to send (in real-time via SCADA) exactly how many turbines are out due to High Wind, and the forecast in AWEFS is updated accordingly. This has been retrospectively configured in some generators, but is mandatory for all new generators.
https://aemo.com.au/-/media/files/stakeholder_consultation/consultations/electricity_consultations/2016/energy-conversion-model-guidelines-consultation—2016-final-report-and-determination.pdf
I just checked my own assumptions, and looks like AEMO rolled back the “Mandatory” HWCO back to “Optional” in the Final Determination. A shame as it’s a simple signal to implement!
https://aemo.com.au/-/media/files/stakeholder_consultation/consultations/electricity_consultations/ecm/oct/energy-conversion-model-guidelines-consultation—2016-second-draft-report-and-determination.pdf
Meanwhile wind production the next day (24 August) is/was abysmal with South Australia producing just 21MW of wind power for the morning peak (6:45am)
This is the problem with renewables, they are unreliable and any claims of peaks and records needs to be considered with the days prior or after the peak which were likely below or well below average.
If you like to have hot coffee in the morning and a hot dinner at night you had better be thankful that there is enough conventional power available because your coffee and your dinner would be cold if we had to depend on the wind and the sun to heat them.
Four factors make it hard for RE to work in Australia. First the frequent and prolonged wind droughts (see June for example) and also WA, SA and Victoria today. Second is the “choke point factor’ that means the sustainability of RE is limited by the lowest points of supply, not the installed capacity, the high points or the average (think about our need for a constant supply of air to our lungs and what happened when we are choking or drowning). Third is the island effect, we have no neighbours to help out if we are short, unlike most places in the developed world. California has rolling blackouts but it will not go completely black because it is not an island. Finally, there is no grid scale storage in sight for the foreseeable future. Please don’t refer to the Tesla batteries as storage at grid scale. And pumped hydro? Believer it when you see it:)
I guess that no matter how much wind and solar are installed, if there is a shortage of dispatchable generation, prices will spike along with risk. That’s the point of the market – to create price signals.
I’m not convinced the price signals are adequate to see dispatchable investment. Is that a reflection on the market? Well… it’s not much of a market that matches supply with weather, instead of supply to demand, by overtly favouring semi-scheduled generators..
What’s the solution?
Not more wind and solar, that’s clear.
Batteries? Handy but expensive, a net load, and can’t do it all.
Snowy 2? For a while maybe, then another big coal plant closes and we get a drought.
Interconnection? That’s been shown on this site to be a flawed hope.
Demand management? I think so. Firstly for demand peaks, eventually for low wind and clouds.
It will probably force wealthier people to invest in home batteries and generators, in order to ride out the rolling blackouts (sorry demand management).
It’s a terrible situation that’s coming, and more people in the industry need to speak up.
It’s almost like all the claims of the ability to have 50% or greater generation from renewables were lies and aren’t possible in the real world.
Meanwhile, no one in their right mind is going to build synchronous generation when the government is literally funding renewables to compete with you.
Maybe the operative sentence is “without fast frequency response”
It’s almost like you didn’t read the AEMO report in question at all, or the AEMO RIS, which spells out exactly how AEMO plans to manage the system above 50% renewables. AEMO’s butt is on the line with system security and reliability and accordingly it manages the system and forecasts the future conservatively – do you really think they’re lying?
And you’re right, no one in their right mind is going to build syncronous generation, they’re going to build batteries to supply FFR, because that’s exactly what AEMO is saying is needed (in the specific situation being discussed here, which is securely managing SA as an island).
Batteries don’t provide inertia. The only way to keep the lights on is to have an oversupply of fossil fuel generation but AEMO won’t say that because they are captured by green zealots.
The way to manage an islanded South Australia is to rebuild the Port Augusta coal plant.
Scheduled demand in SA was also well down, 420MW in the 13:55 interval, which I believe is a new minimum demand (since 1999, excluding 2016 system black). Could be interesting times ahead in the next month or so as rooftop solar generation starts ramping up even further, but before the normal summer higher demand from daytime air conditioning starts taking effect.
Did not look specifically at SA demand, John, but it does appear to have been close to a ‘new record low’ for NEM-wide scheduled demand, as noted here:
https://wattclarity.com.au//articles/2020/08/casestudy-29th-august-2020/
PS, John
Excluding the System Black, the low point in Scheduled Demand (i.e. TOTAL DEMAND) was 360MW at 13:40 on 9th March 2020 as noted here:
https://wattclarity.com.au//articles/2020/03/scheduleddemand-sa-lowestever-exclsystemblack/
Hi Paul,
Can you expand on your comment:
“That’s not to dismiss the financial pain felt by a broad section of the South Australian grid were burnt by high FCAS costs which added a very unpleasant surprise to (including some providers of FCAS services via Demand Response aggregators, we hear).”
Is this referring to this high FCAS lower costs borne by customers in SA, or something else that occurred?
Hi Paul
Was wondering if this was a minimum record for coal generation (black + brown)? even if not quite since market start, at least minimum for the last 10 years, say?
Belatedly here, Kieran:
https://wattclarity.com.au//articles/2020/09/the-slow-decline-of-coal/
Minimum NEM Coal since 01 Nov 2010 was 10,665 MW on Sun 22 Sep 2019 at 12:30 hrs. Last Sat 29 Aug 2020 minimum was 11,062 MW at 13:30 hrs. Another notable minimum was 11,098 MW on Sun 25 May 2014 at 04:05 hrs.
So my understanding is for both regulation FCAS and contigency FCAS, we sell the generation ability, not the real generation. Am I right?
And the FCAS dispatch interval is 5 minutes right? For each 5 minutes interval,regulation fcas and 6s/60s/5minutes contigency FCAS just account once, calculate the revenue and sum them up.
If I am right, what confuse me is that the price difference between regulation FCAS and 5 minutes FCAS is large.These value should be close?
Correct:
1) FCAS is paid for ‘Enablement’ rather than for actual use.
2) Payments made for each discrete 5 minute period
With respect to your question about the difference in price between Regulation FCAS and Contingency 5-min, the answer ultimately comes from the fact that they are different commodities, with different suppliers, and different size of requirement.
Thanks for clarifying Paul
There is one subtlety to the Regulation and 5 Minute Contingency services that shaojing may have been thinking of:
Regulation capability can be used to deliver 5 Minute contingency response, as it is a slow enough service to be dispatched via AGC (unlike 6 second and 60 second Contingency).
Also note that Regulation and 5 Minute Contingency services are not mutually exclusive – a unit on Regulation duty is also able to provide 5 Minute contingency in the same dispatch interval.
AEMO therefore cooptimises delivery of the two services by including Regulation offers in the constraint formulation for 5 Minute Contingency (but not the other way round, as 5 Minute contingency can be offered by generators or loads which can’t provide Regulation – eg loads which offer only Contingency Raise services, or very fast start generators like hydro units which may be off and thus unable to provide Regulation at the start of the relevant dispatch interval, but can start up rapidly to deliver 5 Minute Raise).
Typically, more 5 Minute contingency capability needs to be enabled than the required volume for the Regulation service. This can sometimes mean that the amount of Regulation capability enabled to satisfy the 5 Minute contingency requirement *exceeds* the total amount required for the Regulation service itself – this depends on the volumes and relative prices of Regulation vs 5 Minute Contingency offers. It doesn’t happen that often – perhaps in a few percent of dispatch intervals.
In these circumstances the prices for Regulation and 5 Minute contingency FCAS will be equal, as it is effectively the 5 Minute contingency market that is determining the value of Regulation offers.
In any case, the Regulation FCAS price will always be higher than or equal to the price for 5 Minute contingency, never lower. You can think of this as reflecting the Regulation service being inherently more valuable than the 5 Minute service because the former can substitute for the latter but not vice versa.
Allan
Just notice. really thanks for your help, sorry for bothering you agian.
just a little confused about some other questions.
1.Whether the 3 kinds of contigency raise FCAS are mutually exclusive? For example,I have a 100MW BESS, can I take part in 100mw 6s contigency raise, 100mw 30s contigency raise and 100mw 5min contigency raise?
2.If we can, when emergency happens, we just need supply 100mw, right? Also if we find we can not supply energy?(maybe caused by fault) Can we withdraw the bidding or can we withdraw the FCAS service even though we win the bid?
I think Jonathon has already answered your question 1 below.
Re 2. the response provided to a contingency should be proportional to the frequency deviation. This is explained in AEMO’s Market Ancillary Services Specification documentation. Regarding inability to respond due to a fault, if known about in advance, offers can be withdrawn at any time up to about 90 seconds before commencement of every 5 minute dispatch interval. Once accepted for enablement, offers can’t be withdrawn within a dispatch interval but can be withdrawn or modified for succeeding intervals. “Acceptance” of FCAS and energy offers is not forward-looking beyond the horizon of each 5 minute interval.
I’m interested in the implications for the contingency plans and constraints that might be needed to protect against an interconnector trip during high rooftop PV output.
With only 500 MW demand, and the NSW-SA interconnector rated at 800 MW, that’s a big frequency whack.
Big risk for rooftop solar trips, big risk that existing inertia won’t be enough to catch it.
It is a game of chicken indeed when you look through the other end of the telescope at dinnertime and see the contribution from wind power when there is little or no solar input. The game gets serious during the frequent and prolonged “wind droughts” when the wind farms are producing little or no power, often under 10% of plated capacity and occasionally much lower. Have a look at June and July. It is necessary to take account of those drought periods and also the “choke point” factor, meaning that the grid needs continuous input in the same way that our lungs need a continuous supply of air and we can choke to death very quickly if the supply is arrested by choking or drowning. As long as we have 100% of demand available from conventional sources the grid will not choke but when we lose conventional power capacity then the wind becomes essential and the sustainability of the system will be limited by the lowest level of wind supply, not the average or the plated capacity.
Looking at the NOS the Tarrone 500/132kV transformer has been offline for a week or two – not sure why there is no associated constraint for MacArthur output
Thanks for that, HR10!
WRT responding to ‘responding to big changes in the national electricity market’ I do hope they take this on board as one way to ease the political fallout/pushback with the need to manage the problem of rooftop solar-
https://www.solarquotes.com.au/blog/divert-solar-air-conditioning/
Of particular relevance-
‘..the timing is important as just a couple of years ago, feed-in tariffs were high and curtailment unheard of, so there was no real need to divert power to local loads. But so much household PV is now live (and more is coming), the situation has changed.
Smith said until recently, users have focused on “how much we made from export”.
“If you’re exporting, you’ve got it wrong”, he stated. “The conversation should now be how much I can consume of my own power because I’m managing it better.” ‘
An interesting statistic is to look at the capacity factor for the sector. It is one thing to have output reduced by the retirement of a facility (eg Hazelwood) but as players continue to try and compete for an ever decreasing energy “pie” then the financial viability of plants comes into question. The first thing that gets crossed off is reinvestment in reliability and sustaining capital. Nasty little cycle now starts because the plants become less reliable and hence have lower capacity factors and around we go again. The serious issue on the east coast is the lack of mid merit / peaking plants to fill in as coal falls by the wayside and rooftop PV and wind takes the daytime energy. And I’m still looking for the commercial drivers to get investment in same in an energy only market. All is well until a hot and still evening arrives. AEMO are flagging this – it does happen – look at California.
Thanks Andy
This was obviously only a short article. More detailed statistics (>600 pages of them) are compiled for all generators in our ‘Generator Statistical Digest 2019′:
https://wattclarity.com.au//other-resources/detailed-analysis-and-reports/generator-statistical-digest-2019/
… which we will be seeking to update each year from now.
Paul
Thanks Paul for answering my question! I note that last weekend got down to a minimum of 10871 based on the dispatch target metric, so still not quite record breaking but we have a few more spring weekends to go…
To Andy’s point, i understand that in Europe (e.g. UK, Germany) some older coal plants have been configured to run more mid-merit (carbon pricing and lower fuel price differentials made some of them less competitive v CCGTs), i.e. better able to start up, ramp up at a reasonable rate and then ramp down. It seems to be an article of faith that this is not feasible in Australia, but it’s not clear why not – different turbines, possibly? Or is it just economics – there would be fewer hours in a typical year in which to earn one’s money back?
Kieran, Coal plant response will never get close to an OCGT. Once synchronised and part loaded they can get close to an CCGT in terms of loading rates (they will usually be also limited by their steam cycle). You are right, much can (and is being) done mainly around I & C to improve start up and shut down times, however at the end of the day big, thick walled steel components have limits of heating and cooling rates and hence consumption of fatigue life. Given many are already older plants with somewhat imperfect history of life already consumed (remember,many did not have control systems that modelled life consumption in the 1980’s and before when they were built) so it can be a bit of a raffle. New I & C lets advanced materials modelling be put in to manage life consumption. It gets very expensive when you find a big crack in something at a subsequent overhaul (that you might have to schedule just so you can look – this in itself is expensive). As I say a viscous little circle and it is all economics if the plant cannot earn the revenue to wash its face in an energy only market. Capacity markets can help but the market purists hate these. Not withstanding Govt policy on carbon, I don’t think there are any of the original 4 x 500MW stations left in the UK now. Simply couldn’t get a run. Even Drax (6 x 600MW) survives only as an imported biomass burner. Now there are CCGT plants having trouble because of increasing gas prices.
Given the current amount of storage that is feasible and affordable the monthly balance may be too coarse-grained to show the full extent of the problem. Looking at daily figures there are extended periods in June and July, up to 20 and 30 hours when the wind fleet is delivering less than 10% of installed capacity, down to 2 and 3% for shorter periods. This would not be such an issue if we could run extension cords to neighbours. For instance California has rolling blackouts but they won’t go completely black. In the Californian situation with the conventional power sources run down before the RE replacement is adequate, we would be in real trouble. This looks like a possibility post-Liddell.
Don’t forget that these are record breaking heatwave conditions in California. If Victoria & SA had record breaking heatwave conditions there would be power shortages as happened in the recent past. Even then the weather conditions that caused the rolling black outs in Melbourne were nothing like record breaking. Victoria hasn’t added enough generation and more particularly storage plus demand management to cope with a heatwave even close to record conditions.
Liddell closure is going to be a challenge. NSW is running too late with their new VRE generation plus storage and transmission. The question is how lucky NSW is going to be with weather in the first few years after the closure of Liddell.
My [rejected] comment was obviously an unacceptable comment on the ‘slow decline of coal’….no doubt the foreign-made juggernaut and those making their fortunes out of it just can’t cope with the milch cow consumers supposedly driving the ‘transition’…being permitted to have any chance to have their perspective on it published.
Good for you Paul McCardle…bad for Australia.
Er, ‘metro70’, I appreciate your perspective – however let’s not try to ascribe motivation via conspiracy theory (e.g. ‘just can’t cope with…’) when the facts are simpler.
There is an automated spam filter used on this site to make the site usable, and it appears to have deemed your comment (a long line of text with many capitalised words and other elements that increase the risk it is spam) as such.
Things look even worse when there is finally the recognition that Victoria can’t hide its high gas consumption for residential HWS and space heating. When this energy use is shifted to electricity, even mainly based on heat pump technology, it creates further problems in the winter months.
Of course factor in distribution of power generation assets during the periods of lack of supply and storage, then it looks worse again. The “excess” solar available in QLD in winter can’t all fit through the transmission lines going south to Victoria.
Hydrogen doesn’t greatly help with the seasonality as winter deficit is going to be too great to even use hydrogen with storage at an affordable level.
And then the sun went down and scheduled demand rose like a ducks head just in time for dinner.
All the wind and solar generation in the world won’t help if they can’t produce supply when there is actual demand.
A little confused about a question.
Whether a 100mw/100mwh storage system can take part in all 8 100mw FCAS(100mw reg raise/100mw reg low/100mw 6s contigency raise……). Or for raise/low, the sum of total is 100MW(for example 50MW reg raise/ 30mw 6s/10mw 60s/10mw 300s)?
Hi Shaojing,
It can be a little confusing but most batteries can take part in all services, even all at the same time, depending on the technology being used. You are right that it is the summation of energy targets, regulation enablements and max enablement of all FCAS services must be less than or equal to the max capacity – its the joint ramping constraint.
Paul Mcardle has my details if you want more info.
Cheers
The 8 TWh deficit corresponds well with David Osmond’s “96% renewable” result.
I’m confused about why more QLD wind wont help? It appears to be anti-correlated.
Curious as to what data you’re basing your inference that QLD wind appears anti-correlated, Brad?
In our analysis of over 11 years of weather history for the Generator Report Card 2018 using wind speed readings across all Renewable Energy Zones, we found that (statistically) the correlation co-efficient was close to 0 … i.e. random, and not true anti-correlation.
This has several implications, as noted here:
https://wattclarity.com.au//articles/2019/07/assessing-the-diversity-of-intermittent-wind-and-solar-output-as-one-piece-of-analysis-in-the-generator-report-card/
… and it also tends to reinforce what David is suggesting is needed, in terms of the need for deep seasonal storage – if the model is VRE + Storage, moving forwards.
Good article. I just had a SA device/gadget installed yesterday, so I guess I’m in “time period 1”. Like you, I have the benefit of a high Net Feed-In Tarrif for another 11 months so my goal or value drivers are diff. to those that don’t have this. Not sure if the SA site deals with this yet. I also have a curiosity for more data, but need to temper this appetite with the understanding that more data doesn’t always lead to more knowledge.
Unfortunately this raw annalysis does not explain the uderlying reasons for lack of generation e.g. low wind or cloud or was it that the full capacity of these generators was not commited and why steps were not taken to avoid the problem.
Alexander, you might be misinterpreting what these case studies are about.
These are not so much focused on lack of generation, but on the fact that there is an underperformance compared to what AEMO was expecting 5 minutes before in the dispatch target.
Whilst some instances might be due to challenges with fast-moving cloud, or fast fluctuations in wind, there are other factors as well – as the case studies are progressively revealing.
Paul, is the discrepancy in the second last image perhaps a result of Wivenhoe pumping demand and it being reflected differently in the different values AEMO publishes? I say this because I have noticed before that the impact of this load does not show up in the demand values shown in the popular ‘Pocket NEM’ app, but does if viewing the AEMO NEM web dashboard (ie. the AEMO dashboard demand can often by ~250MW higher than the values in Pocket NEM’. My guess being that Pocket NEM uses a different data set for demand than AEMO’s dashboard does, which them each treating the Wivenhoe load separately.
Interesting to see at dinnertime with solar off duty the wind was down to about 5% of the power supply in the NEM and less in WA. Surely sustainability of RE depends on the minimum level of supply, rather than the installed capacity, the high points or the average! Does that mean that until the whole of the demand can be satisfied by RE at the lowest level of sun and wind supply we will have to maintain 100% conventional power capacity?
Until the rules are corrected there are going to be problems with the balance in the NEM. The delay to the 5 minute rule and the lack of additional rules to reflect the valuable services that batteries can provide has compounded this problem. If these were in place we would be seeing the installation of GWh scale batteries on the NEM.
As a result we have coal fired units operating in QLD and Victoria, which otherwise would be shut down for weeks at a time during these periods of low grid demand as they would be uneconomic. Of course put a price on the pollution the fossil fuel power plants produce and the transition would be even faster.
Demand management of easy high load items like hot water services would make a difference to the balance of the supply and demand, but HWS are boring so nothing has happened.
I was under the impression that Queensland distributors use ripple control to switch in hot water during such low demand periods. Another low-hanging fruit for demand management would be eutectic storage elements in commercial refrigeration (office towers, shopping centres and the like).
At least this Sunday the QLD – NSW connection was able to operate at close to maximum capacity. Other weekends we have seen up to 800MW of RE being curtailed.
If the QLD Government was serious about green house gas reductions, then they would shut down one of their coal fired units during these months of low demand. Doesn’t make a great press conference, so it doesn’t happen. Might up set their union mates as well.
The QNI Interconnector is undergoing a minor upgrade that will continue through to the end of 2021 (according to recent TransGrid tweet). To facilitate the work, the 88 Liddell to Muswellbrook 330 kV Line has been taken out-of-service on a daily (Mon to Sat) and more recently, a 5/6 days continuous basis. This outage results in a lowering of the Qld to NSW export capability by 700-800 MW. These outages for upgrade work and the loss of export capability from Qld (and increase in RE curtailment) is giving us a glimpse into the future with the unstoppable increase in Variable Renewable Energy Sources in Qld. So much generation with no where to go!!
A couple of questions Craig, given that the sun is gone by dinnertime, certainly in winter, what keeps the lights on when the wind is low? Last night the wind contribution was down to 10% of demand across the NEM. Currently the wind supply is down to 5% of total demand across the NEM, zero in Victoria, 1% in NSW and Queensland and 3% in WA.
Second, how many batteries do you envisage to provide for these gaps in wind supply, given that the original big battery at Hornsdale would have lasted the SA grid for five to ten minutes, It is bigger now so it might stretch to 10-15 minutes.
Basing the amount of wind output at 12:30pm during these months of low demand is going to be very misleading as many wind farms could be curtailed due to the low demand and high solar output.
When you look at the levels of wind output when solar has dropped away, it is much higher. Then you need to scale the wind output based on the projected ratios of roughly 50:50 wind and solar. At the moment States like QLD have very little wind installed. There is a 1,000MW wind farm under construction plus a 800MW wind farm likely to proceed. This is will start to balance out the QLD grid.
In terms of storage there is going to be a dramatic increase in battery storage going in over the next 10 years. Based on Tesla’s projections the cost of cells should drop by 50% over the next five years, while production volumes grow dramatically. Already there is a 6,000MWh and 1,400MWh battery under construction in California. In time we will see similar and larger ones built in Australia.
Batteries are relatively quick to deploy and the costs are falling quickly, so the installed capacity will grow very quickly after 2025 and more so again after 2030.
Then of course there is hydro, pumped hydro and as the grid approaches 90% RE there will be other deep storage such as flow batteries, green hydrogen and compressed air.
The biggest problem largely being ignored by the wind and solar advocates is that rooftop solar is detrimental to both the physics of the grid and the economics of large-scale generators. This article highlights one symptom.
The second biggest problem is CfD that reverses the market signal ie lower wholesale is more costly for CfD purchasers, and makes further investment unattractive without a CfD.
After that we have the requirement for additional equipment and services to incentivise (make more costly) the inertia and system strength services that wind and solar cannot provide.
Then we have the additional network infrastructure required to connect new small distributed generation at low utilisation rates.
Also the reducing capacity factor of the entire system which can only increase costs.
Our electricity system is in dire straits – there are ways out of the mess, but who’s listening?
Craig, check out how pumped hydro is going in North Queensland.
https://petercampiononline.wordpress.com/2020/09/28/genex-kidston-pumped-hydro/
Can you find some sustainable pumped hydro schemes around the world? Just a polite question, I am researching the topic:)
It seems to me that if a generator is contracted through CfD, this effectively reverses the effect of spot prices on bidding behaviour.
The CfD rewards generators for low spot prices, which incentivises generators to bid low and keep spot prices down.
But that doesn’t necessarily translate into low electricity prices for customers, and certainly isn’t providing an investment signal (unless you can score a government offtake)…
The added benefit of being semi-scheduled means following a dispatch set point is almost always optional, which is certainly an advantage compared to scheduled generators who can’t sell their output when they’re dispatched down to make room.
I am wondering how useful a droop characteristic would be for solar inverters, as well as home batteries and loads such as (future) electric vehicle charging in these situations (in particular, commercial systems with “zero export” constraints)? Would it reduce the need for inertia from synchronized generation?
It is possible to program inverters to be grid forming (contributing to fault level and system strength), and grid-scale batteries can even offer black start capability e.g. https://www.powermag.com/benefits-of-battery-storage-based-black-start-capability/
A quick search of IEEE finds quite a few papers on control strategies for use in low-voltage connected inverters, which aim to improve rotor angle stability of synchronous machines and transient stability generally.
Would it be correct to say that over performance is a function of:
– semi-scheduled generators not forced to follow dispatch
– CfD rewarding generators for low wholesale prices
– completely controllable
While underperformance is a function of:
– outage or derating
– unfavourable weather
– largely uncontrollable
Therefore the new rule mostly applies to the over performance situation…
I understand the ESB made the ISP ‘actionable’ and therefore the ISP’s lost of projects must be built (or at least studies commenced) by the respective transmission companies.
I’m not a fan of the central planning model – customers would be better served if a state requiring more power, or more load, negotiated with the adjacent states.
Centrally dictating transmission investment reduces accountability. Greater accountability leads to better decisions.
The changes to solar inverters in SA is really too little too late. There needs to be a rapid shift to making hot water services a dynamic load. This alone won’t be enough, but it solves two problems and is one of the only things that can be done quickly and cheaply.
The term ‘transition’ implies an end point, when the ‘transition’ is complete. I’m not holding my breath.
Meanwhile, in a recent interview, Audrey Zibelman literally said that AEMO is full of people who prioritise decarbonisation over consumer benefit. Perhaps that’s part of the problem.
You’d think the CEO of AEMO would have the cred to propose that states hellbent on forcing more rooftop solar into the grid should back off for a while and start turning the ship, before it runs aground.
You would have to have some reservations about the transition when you look at the difference between the RE contribution in the middle of the day yesterday (black coal below 7GW) and the situation at dinnertime with the RE contribution down to 1GW!
Hot water service is too variable a load to be considered a solution.
I don’t understand what you mean by variable. If the HWS are mostly controlled centrally, via providers if need be, then on days of forecast very low demand, the HWS are controlled in such a way that bulk of the load is run during the period(s) of lowest demand. Based on historic use and knowledge of the ambient temperature a very good estimate of the total load and duration can be modelled. Controllers could be managed over 4G or WiFi and even monitor the load current.
In case of emergency on the grid, controller can switch on or off hot water systems.
There is approximately 200MW to 300MW of residential HWS that could be controlled.
Hey Paul
I think you’ll find it was due to NQLD system strength constraints – won’t be in the public data until tomorrow given constraints only have 1 DUID on LHS. Seems more severe than normal – combination of a network outage and crossing one of the demand thresholds I expect.
Yep, wondered about that. Would also explain coincident price spike (i.e. NEMDE factors in loss of generation in the dispatch process).
PS HR10,
More details here today:
https://wattclarity.com.au//articles/2020/10/casestudy-p2-rooftoppv-kills-largesolar/
… with part 3 to come later
If AEMO’s logic hasn’t yet linked their own constraints to their dispatch engine, it probably should. A bit like that time QLD islanded and AEMO were dispatching QLD generators against a NSW frequency…
The insight on flexible load reacting to high prices and exacerbating a system strength problem is a curly one. Some loads also contribute inertia and fault current.
It would be a shame to be trying to save money by reducing load, and being penalised more under user-pays than was saved.
Interesting choice of title for the post Paul – a backhander at the “PV saved the day” stuff sometimes seen on other sites?
In reality it seems that a less drastic formulation of the relevant system strength constraints could avoid this sort of disruptive dispatch outcome. I doubt that the underlying physics of the network is quantized.
Allan
It would seem that more graduated constraint would have been better. This would have resulted in a more gradual transfer of the generation. It would likely have resulted in a small increase in the price, without the massive spike.
This time the overly risk adverse constraints almost caused network instability.
So now that the full picture is available, can we agree that yesterday’s Wattclarity article headline referencing “many Large Solar Farms tripping” was entirely misleading, inaccurate, and unhelpful? Turns out no so solar farms ‘tripped’, rather they were doing exactly what they were supposed to: following their AEMO issued dispatch targets.
With all the commentary about energy transition “villans” on this website… it would seem Wattclarity fell into its own trap yesterday. I would think an apology or correction is in order.
Trial run for when a big storm comes over.
This trend is plaguing our fledgling hydrogen industry. Project proponents, funders and industry advocates are satisfied that LGCs or nebulous renewable PPAs for an electrolyser running 24×7 can be classified as green. By ignoring the time and place of renewable energy supply in our largely coal-dependent grid, you end up producing hydrogen with a much higher carbon-footprint than traditional fossil-fuel sourced hydrogen. This doesn’t mean its the wrong approach, as scaling up grid scale electrolysis is important in delivering a future low carbon hydrogen industry. However, it’s hard not to get fed up with the disingenuous marketing of “green” projects which are really anything but.
Hedging for purchasers has historically meant protection against high prices.
Now the hedged purchaser is in the unfortunate position of being exposed to very low prices, which costs them more, under a CfD.
Is it possible to hedge against both high and low prices at the same time, or has the combination of rampant growth in rooftop PV, uncontrolled semi-scheduled generation, and manipulated CfD purchasers broken the market for good?
The ‘nameplate’ capacity of the HWS is one thing, but there is no way of knowing what it’s ‘availability’ is.
I.e. how many will turn on, how long they will run etc. the nameplate 200MW might be 50MW one day and 199MW the next.
I agree it needs to be done, simple no cost change. But it’s not anywhere near a game changer.
And yet bankable forecasts are made all the time and banks accept some forecast when lending money. Noone can produce bankable forecasts of the oil price, or of share prices. And yet there is a market for such forecasts because the paradigm for valuation requires a forecast of future cash flows. This is not news.
Its also always helpful when looking at surveys to know the sample size.
And finally like management people don’t know what they want or what is need until they see it. . No-one knew they needed a smart phone until they saw one.
Equally there is quite the challenge in designing rules when the technology is changing very quickly. My own view is that the distinction between market facing and regulated business may be more harmful than helpful going forward and that least cost is not always the same as best. Many of us drive cars that are not least cost. How do you incentivise technology lead investment when volumes are declining and still give consumers least cost? It can’t be done. Anyhow enough rabbiting on on Friday afternoon.
At least the solar farms weren’t off target, but this seems like a perverse market outcome where bidding high can lock your competitor out of the market (and would become worse with widespread price-sensitive demand response).
It can be done:
– by having market rules that do not favour any particular technology
– by having market rules that ensure all participants provide the same product
– getting market distorting subsidies out of the picture
– getting non-market government investment out of the picture eg offtakes forcing RE into a saturated market
– by removing the spectre of a carbon price
– by allowing all possible technology to enter the market
– by limiting hedging to a portion of the generator’s capacity
– by enforcing the intent of the National Electricity Obligation at all levels of bureaucracy
– by eliminating central planning for transmission ie remove AEMO from its central planning role
– removing the ESB
Very nice analysis. The questions are heading in the right direction.
I’d ask a couple of them more directly:
– which VRE CfD purchasers are benefiting from very low wholesale prices
– if the market system is in crisis, then why are governments still forcing in new unnecessary supply using VRE offtakes – surely the government has some sort of obligation not to wreck the joint
– same goes for rooftop solar
The questions are good but as you point out everything depends on your reference point.
Mine is the need to decarbonise. If you start with that as the priority then policy will make the money be there. Same as for police or some other essential service.
The second observation is that there is committed and or under construction new investment sufficient to provide more than a further 10% of NEM energy. And I will offer 3:1 odds that there will be more commitments over the next few years. Yet without a policy to incentivise demand the total pie will not be big enough.
Here’s one of the things that jump out to me, Patricia, when I view that chart from the Renewable Integration Study:
https://wattclarity.com.au//articles/2020/10/wind-nemwide-lowpointtrend/
When I first saw the graph, I realised it was the source of the “75% VRE” statement and I thought, “fair enough, the odd half-hour here and there, but that’s going to need a lot of transmission and support services”.
A bit later I realised the ISP scenarios rely on GenCost, which has a built in carbon price, somewhat separating the scenarios from reality, being much more favourable for VRE and batteries than in practise.
Not long after, the NSW and QLD gov’s announced REZ plans a decade in front of the ISP, and now the second major ISP project – EnergyConnect – is going to cost so much more than originally planned it could be canned.
As we speak the NEM investment environment is completely tanked (or not far off), with governments the biggest investors via long term offtakes with VRE.
Before long AEMO will be opening substation feeders to isolate sections with too much rooftop PV, to prevent wider blackouts.
Back to the graph, now I see an increasing number of daytime periods with zero or negative wholesale prices, still no solar output at night or during evening peaks, and coal plant continuing to make room for VRE (except when there isn’t any).
You can probably tell, I’m skeptical The Transition is providing much benefit.
This post touches on so many issues its worth coming back to.
Whether people realise or not there is a big discussion about risk going on.
If revenue is guaranteed it is undeniable that the risk in financing wind and solar is very low, that’s because there is no exposure to fuel costs. Essentially your only risk is construction and breakdown. So then the cost of capital is low and the required price is low. But the risk has been shifted.
So the most fundamental principle of risk is it should be borne by those most able to price it appropriately. In many markets the private sector is the best placed to judge return and risk, that is a fundamental principle of capitalism.
Equally there are society level risks, such as say lung cancer from cigarettes, defence, a sufficiently educated population, that society has decided are best borne by the Government
The fundamental concern about risk is the need to judge the carbon priority. If there was a carbon price, a credible emissions trajectory, the risk of new investment could easily be judged by the private sector. Without that policy, the risk falls back to the Government(s) who signal their support for decarbonisation by providing PPAs without necessarily having to disclose their carbon policy. In this case the Government has better information than the private sector and may therefore be better placed to bear the risk.
There are other factors in that individual wind and solar farms are very small relative to the size of movement and also quite fast to build so that the risk of misjudging the total new supply required is smaller. Still it would be better done by setting a credible target and then letting the private sector work out how to do it. Its that lack of govt policy that is at the root of many of the problems.
Hi Paul, thanks for another insightful case study. I’m looking forward to you unpicking the the ‘Q-BCNE_821’ constraint set. I understand why such a System Strength constraint exists, but I don’t understand why it would trigger an immediate dial down to 0MW for all solar farms. It seems to me that the constraint itself leads to the very concern that it is trying to avoid, that is the rapid loss of a significant portion of variable supply in the region. Why would the constraint not just dial back the solar output to ensure it remains below ‘reasonable’ levels?
I note your comment “Unfortunately, because of the hard ramp on the constraints at the time* this meant hitting the park brake when coasting down the hill” – but I couldn’t find whether the * meant you had some further notes about the reason for the unfortunate hard ramp?
Cheers
Mark
Did this ‘Part 5’ (follow-on) article help answer your question, Mark?
https://wattclarity.com.au//articles/2020/11/casestudy-p5-q-bcne_821/
Well written Ben, thanks.
Many people, even within the energy sector, still don’t realise that the states hold legislative power over energy rather than the Federal Government.
How do you see the network cost debate moving, given that much of the upgrade is prompted by a conscious pursuit of renewable penetration instead of re-use of existing sites/regions for generation?
Unlike past eras, this time it’s not growing demand that is driving the investment need.
Latrobe Valley and Central Qld are good examples where those states have better renewable solutions elsewhere geographically but that prompts greater transmission investment to enable the better low-emission outcome (rather than for eg. replacing coal fired capacity with gas at lower capacity factors in the same area of the network.
How does the different emission reduction targets/timelines by different states’ governments affect this?
Lots to unpack perhaps, but keen to get your thoughts.
NSW was importing energy when a LHS constrain happened…why the constrain happened, I dont know!
A question that ez2view would help you with, Bruno. If I have time, will look and explain further later…
“…Victoria recently passed legislation to empower the state minister to direct the construction of transmission assets outside of the NEM’s cost-benefit based planning framework5. It would clearly be unreasonable, and a moral hazard, for customers in other states to pay for such follies. …”
*Follies* -Excuse me, but, simply because something falls outside a cost-benefit model does not mean there is only cost and no benefit: it can mean the benefits are less than other choices in the ranking at that time, or the benefit is not sufficiently equalised. If Victoria decides to implement something with asymmetric benefit, you are arguing that the other beneficiary path nothing? How about if we ask the questions: are there other models of paying, and are there other cost-benefit models?
Eraring 4 died?
Good post. Confirms what we have all been saying. For this sort of price spike batteries can’t be beat by any other currently available technology. That said I wonder if any demand response was able to take advantage of the short intervals.
In terms of Demand Response, David, there was definitely some – as noted in Part 2 of the Case Study here:
https://wattclarity.com.au//articles/2020/10/casestudy-p2-rooftoppv-kills-largesolar/
Great to have Andrew contribute this ‘Part 4B’ (in a way) in the evolving Case Study.
Readers should also keep in mind that UQ’s St Lucia battery is also not scheduled (i.e. not even registered with AEMO, so able to operate completely independently of the AEMO’s dispatch process – because it is under the AEMO’s 5MW registration threshold for BESS).
These two differences (i.e. size (re: size of market need), and type of participation) are other difference to consider in the considering what happened on the day … especially if thinking through the ‘if I also had a battery, it’s performance would have been X’ extrapolation.
It is interesting that the battery benefits in 2 ways from slower response times of traditional peaking units: from the obvious ability to capture high prices during supply shortfall; and also from “artificially” lower prices due to the slower ramp down response of the traditional units after the VoLL event. Perhaps it should have to pay a “corrected” price for recharging in the latter case.
Very interesting and well written article Andrew.
As you point out, the bidding dynamics and dispatch outcomes that will occur under 5 minute settlement (5MS) could be quite different because the financial incentives for participants will be driven by the 5 minute dispatch price not the 30 minute average trading price. I think one of the current uncertainties with 5MS will be how operators of traditional peaking generation decide to offer and run their capacity – for example they may have to bid “more cautiously” as they won’t be able to rely on “catching up” to 5 minute spike early in a Trading Interval by starting up in the later DIs of that TI.
It’s also worth noting that although limited ramp rates and start up times for peaking (and online thermal) plant probably were a factor in driving the dispatch price to $15,000/MWh when the northern Queensland solar plants were constrained down, it was presumably the very high downward ramp rates of those solar plants that required very sudden upward dispatch of the rest of the generation fleet. If the solar plants had offered slower ramp rates (downwards) to the dispatch engine or if AEMO’s constraints were less binary in formulation, then the price spike probably wouldn’t have occurred. Sometimes very fast fast ramping can be too much of a good thing :).
Allan
“MR ALLAN G O’NEIL” – lol autofill fail
Terrific post, thanks Paul.
Interested readers may also be interested in this working paper: We document the increasing frequency of system strength and transmission constraints binding in the NEM, coincident with increasing VRE penetration in SA and QLD. We also document some evolution to the form and frequency of the the SA system strength constraints binding. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3501336
Scrutiny on how the constraints in NEMDE are decided upon should be healthy – there are real trade-offs to consider when setting these constraints – while a more risk averse approach should of course improve system resilience, these tighter constraints can only increase system-wide generating costs. Consequently, innovations that allow the loosening constraints without impacting reliability probabilities will provide tangible economic benefits.
Wasn’t this caused by 800 MW of Victorian wind turning off at 1pm to avoid negative prices? This turned Victoria from an exporter to NSW to an importer. The current dispatch rules are leading to perverse outcomes.
I don’t expect wind was doing anything other than reacting to dispatch instructions.
There was an almost 2000MW reversal of flows on VIC1-NSW1 required by constraints. Given the sudden excess of generation, VRE was best placed to respond by backing off.
http://nemlog.com.au/nem/ic/VIC1-NSW1/20201116/20201116/
You forgot to mention the Tumut 300 KV line outage at 12:52
AEMO market notice 80022
Was looking at NEMLOG to try and understand why Victoria was importing against the price gradient. It continued for for a few dispatch intervals, as NSW hydro filled in the gap.
I think market notice #80022 may explain what was going on.
「An un-planned outage of Upper Tumut – Stockdill 1 330 kV Line in the NSW region occurred 1252 hrs 16/11/2020
Constraint sets invoked: N-SDUT_1 and N-CNSD_3C
This constraint sets contains equations with the following interconnectors on the LHS.
V-S-MNSP1
VIC1- NSW1″」
Negative load should not be a showstopper. Inverters will simply need to operate as though they are on an AC coupled system. Provide primary frequency response and accept AGC requests to adjust output.
Domestic, commercial and grid-scale batteries, combined with synchronous condensors can provide sufficient inertia and any curtailed VRE are effectively ‘spinning’ reserve.
The difficulty will be in determining the price.
*AC-coupled microgrid.
This problem sounds like a local lack of reactive power. I wonder if it can be improved by?
a) installing a static synchronous compensator (STATCOM).
b) Requiring Voltage Source Converter (VSC) based inverters remain synchronized and provide voltage control, even when not generating active power (currently on trial in the United Kingdom).
c) series compensation capacitors to reduce AC line impedance.
d) additional transmission lines or synchronous condensors to increase local system strength.
e) follow b) but additionally investigate voltage support from (Doubly-Fed Induction Generator) wind farms.
Unfortunately I suspect economic rules like the RIT-T will slow down technical fixes.
For anyone curious about N^^N_NIL3 the constraint is published here http://nemlog.com.au/show/constraint/?k1=N^^N_NIL_3 (an extensive public dataset, no connection to it)
At time of posting it hasn’t bound today.
Interesting article.
Figure 15 shows 16GWh of Storage Buy each day. Is this plan using Snowy 2.0 – It would be a very expensive Battery!
Figures 14 and 15: At 18:00 both 2021 and 2030 seem to require 6GW of coal. I am misunderstanding something?
just gone up another level, started thinking this may be megawattclarity, but stuff it, terrawattclarity Allan! clarity indeed.
Thanks Victor! Bit more of a mouthful.
So Paul, Allan – (maybe even Ben B or Alex W at AEMO) – any thoughts on whether we will get into a round of debate about the merits or otherwise of having an interconnector term with such a small co-efficient, but wide range (much bigger than any generator) on the LHS? It’s above the threshold, but at 0.09287 you don’t get the best relief – and we have already seen similar equations swing the interconnector 1800MW in a TI recently. economics vs physics anyone? Should interconenctors have a different threshold given their range? do these swings cause any other power system issues, now, will they in the future?
Thanks to Allan for a very clear and concise description of despatch constraints. In a well-ordered system the proposed interconnector to SA would hopefully alleviate this constraint.
Time will tell.
P.S. I propose we refer to the area west of Wagga as the ‘Triangle of Tribulation’.
I’d wondered about ‘polygon of pain’, Malcolm.
I never understood why this ever needed to be a “polygon of pain”. From what I can see there is no reason why AEMO could not have operated the Balranald-Buronga line normally-open at Buronga (with a backup auto-close scheme). This would force the power flow away from X5 up through the high-capacity Dederang interconnector instead, potentially avoiding any curtailment. Maybe I am missing something?
Let me guess – retailers are making plenty out of negative prices, so the tariff reform that should follow the duck curve, is lagging… and lagging… until it becomes embarrassing for the retailers and they concede.
In QLD, the government owned DNSPs set peak tariffs between 6am and 9pm. They charge peak rates during periods of negative wholesale prices.
Of course some is hedged, but CFDs incentivise those generators to bid low, which results in very low wholesale prices.
Ben
I don’t think many retailers are enthusiastic about tariff reform – not least because they don’t see any evidence their customers want it. but it’s ridiculous to claim that they are the barrier here. Tariff reform is an orphan policy. Governments, policymakers, consumer reps – no-one has picked up this ball and run with it, and many have opposed it. For example, Victoria banned widespread tariff reform a few years ago. Some DNSPs have taken some steps as you note, but they don’t care about wholesale prices – they are doing it to manage network peaks, which may often overlap with wholesale price spikes but are not exactly coincident. Even they have limited interest – after all they’re on revenue caps so they get the same amount of money either way.
Despite the rhetoric there’s enough competition and innovation in the retail market that business and household customers can access wholesale price signals if they want to (eg via Flow Power or Amber respectively, there may be others).
The storage buy is effectively spilled energy for the most part. Its available for storage but in the end most isn’t used.
Hi – some issues – forces a semi scheduled generator to behave like a scheduled generator when in a semi dispatch interval -a lot of the time the cap will be below the available resource energy capability to limit output. In this scenario an SSG can’t go above or below semi dispatch target due to variable resource
As the amount of VRE increases so will semi dispatch – it’s getting rid of semi scheduled operation through subterfuge All they would need to do is semi dispatch VRE
system security would be the excuse
I’ve just dialled back in after 14 months of retirement and it’s great to know that the neurones haven’t completely turned to mush in that time. Thanks Allan for a great explanation. I’m thinking of using it to dazzle some mates when someone asks why their bills are so high or how the NEM works. I am very happy to now be just a spectator though I do miss the access I used to have to NEM data.
Negative prices are endured by generators for only two reasons: because they have sold PPAs (or similar contracts) to naive buyers who don’t mind paying the strike price in their contract plus the negative price(!) for every MWh generated, or because a generating unit is not flexible enough to turn off and back on again later when prices are attractive. Surely, an increasing prevalence of negative prices is therefore both the symptom and the cure for this problem?
Hi Paul
With regards to the forecast LORs, in 2018 there was a change to incorporate forecast uncertainty in the process for declaring forecast LORs. Refer to the Reserve Level Declaration Guidelines on AEMOs website for details.
Thanks Jack. More to read, to catch up!
Maybe the worst case scenario is now the failure of an interconnector. From my website:
11/12/2020
NSW power supply problems November 2020 (part 2)
https://crudeoilpeak.info/nsw-power-supply-problems-november-2020-part-2
26/11/2020
NSW power supply problems November 2020 (part 1)
https://crudeoilpeak.info/nsw-power-supply-problems-november-2020
Paul
It looks like common factors in all the NSW volatility this week are:
– significant numbers of coal-fired units on outages, I’m unsure to what extent these are planned maintenance vs breakdowns
– transmission constraints between southern NSW and the rest of the state, due to outages of some important 330kV lines – again I’m not sure if these are planned or forced outages.
While these price spikes were occurring, the transmission constraints were forcing flows OUT of southern NSW into Victoria, until “Negative Residue Management” constraints were invoked to wind down generation in the south of NSW and prevent further counterprice flows.
In essence, despite the reduced availability of coal, wind and solar generation mentioned in your posts, there was still sufficient available generation in NSW + imports from Qld to comfortably meet demand, but the transmission constraints from southern NSW meant that not all of this generation could be dispatched.
Given Liddell’s age and decrepitude with forthcoming closure and this unaticipated transformer explosion injuring a worker, should management close the plant forthwith with their responsibility for WHS?
https://lynchmeyer.com.au/legal-updates/directors-found-personally-liable-for-breaches-of-whs-may-face-hefty-penalties-and-jail-time/
Or should they handball the decision to the AEMO formally advising they cannot guarantee worker safety in future as a result of this incident asking the AEMO what should they do about that?
Now You Can Create beautiful, natural sounding Voice-overs – Add Pauses, Inflections and Tone – Make Listeners believe it’s a real human talking. Don’t take our word for it, hear a free demo now. Just send us an email here: nancyfadodd@gmail.com
I scanned the explainer and could not see the difference between Total Demand and Operational Demand. I had assumed Operational demand was the demand from the grid seen in Victoria. This includes net imports but not behind the meter generation.
If Victoria was exporting as it probably was, then Operational demand would be grid generation less net exports, battery charging and pumps while total demand would be Operational Demand + rooftop solar. So the question is how can Total demand be less than Operational demand
For those that have the data, skills, systems, resources, etc to do it – an excellent approach. Well done.
I think all retailers/embedded networks (virtual or otherwise) started with these ideals, the challenge is to make sure it remains efficient, and it’s a challenge to traditional retailers to leverage their strengths to compete across their portfolios.
Is there a public source for the terms of PPAs?
I’m especially interested about the connection point: is it usually the node or do some use the generator connection point as in the ACT reverse auctions?
Hi Dan, good question. There isn’t a public source for the terms of most PPAs as they tend to be closely-held commercial deals. I assume by the connection point you mean the generation figure used for the fixed price payment of the PPA? I’ve seen PPAs that pay the fixed price on generation at the node (the regional reference node RRN) and others at the generator connection point (sometimes referred to as “at the gate”, which isn’t always physically where it is). The difference between the two is on who takes the risk of the MLF (marginal loss factor) changing in subsequent years, which can be a key point of negotiation between the parties.
Thanks a lot for your reply, I do have a follow-up. I have understood from M O’Neil’s article that MLFs scale the spot price that the generators receive, not the output they sell. But what about the actual loss of output between the gate and the node? In a PPA, is the MLF-scaled price applied to the output at the gate or the output minus the physical transmission losses at the node? If it is the latter then it would amount to double counting the impact of transmission losses.
Hi Dan, the settlement generators receive from AEMO is calculated as RRP * MLF * metered generation (at the gate). This is typically referred to as the “floating” leg of a PPA (i.e. the money the generator is swapping with the off-taker). The “fixed” leg (what the off-taker pays to the generator) could be calculated in two ways (typically, but every PPA is different):
– “at the gate” – the off-taker takes the risk of variability in the (RRP * MLF) part, and settles at a fixed price on the metered generation.
– “at the node” – the off-taker takes only the risk of variability in the RRP but not the MLF, so settles at a fixed price on the metered generation * MLF. This isn’t the generation adjusted for physical losses to the node but an allocation of who’s taking the risk on the financial impact of loss factors between the gate and node.
In theory (the theory of marginal pricing) the MLF scales the spot price – but in practice the AEMO settlement calculation is to multiply the three terms together (RRP * MLF * metered generation), and PPAs often split this (when “at the node”) into RRP and (MLF * metered generation). The actual physical loss to the node is irrelevant – all that matters is who’s taking the risk on which components of the AEMO settlement.
Great, thank you.
Thank you.
One thing I don’t understand is how one can reconcile the relatively good position of the Coleambally plant with regards to the X5 constraint with the many headwinds you mentionned about it in your other post, especially the constraint part?
https://wattclarity.com.au//articles/2020/01/gsd2019-allan-howgoodissolarfarming/
Dan
The fact that Coleambally isn’t caught “behind” the X5 constraint doesn’t undo any of the other factors leading to the headwinds described in my earlier article – they remain just as relevant. The X5 coefficient is just another potential headwind that Coleambally avoided on this occasion.
Another point about constraints is that while generators with (larger) negative coefficients are at risk of losing volume when the constraint binds, there’s no corresponding volume or price uplift for generators with positive coefficients(*) – those positive coefficients typically just mean the constraint may be less likely to bind when the relevant generators are running.
Allan
(*) Except where the same generator sits in two different binding constraints but with a negative coefficient in one and a positive coefficient in the other – there it’s theoretically possible for the effect of the positive-coefficient constraint to offset the impact that the negative-coefficient constraint would otherwise have. But this is not guaranteed and very dependent on the specifics of the situation.
Hi Paul and thank you.
Why would the developper that has signed a PPA with an off-taker care about negative prices? I understand that the newer ACT reverse auctions have a -$20/MWh threshold, but are you implying that this is also the case in many other PPAs?
For the earliest wind farms built in the NEM, the initial PPAs (i.e. primarily signed with retailers back then) might not have had negative price clauses as negative prices were not common at the time, so not forseen as a risk.
Most, if not all, PPAs signed in recent years have negative price carve-out clauses where the strike price does not apply, but instead the generator wears the risk.
Sorry to double post but could you please explain :”If you are the project proponent that has some form of off-take arrangement at the RRN […] you are still wearing the risk.”.
My flawed understanding is that MLFs scale the spot price the generator gets, not the output it sells. If I already have an offtaker at a fixed price and if MLFs don’t impact the volume I sell, why do I care about the MLFs?
Danton,
You note ‘My flawed understanding’, and it is indeed flawed – the output is scaled by the MLF to calculate the MWh notionally delivered at the RRN, and it is this number that is used in the vast majority of commercial transactions – such as:
1) For all wholesale energy purchased by AEMO in the spot market; and
2) In most of the PPAs contracted between generator and offtaker (there may be some offtakers who, for whatever reason, have agreed to take the MLF risk … but this would be rare); and
3) Even in terms of the LGCs registered each year for renewable generation.
All of this makes sense on the principle that the MLF reflects losses in getting the product to market.
Thanks.
Thanks a lot
Paul, do you have any insight into the logic AEMO uses to determine when to send out a ‘power system events’ market notice, or not? Large units trip with some regularity, but do not always attract a market notice flagging the event. For sure, an ~830 MW trip is a “very large” trip and significantly unusual, but a frequency drop to ~49.78 or so (as your screenshot suggests) is not particularly unusual or dramatic. Smaller unit trips have caused larger frequency drops in recent times, without prompting AEMO to send a market notice. Any insight into how AEMO decides to trigger notices, or not?
Double-checked with a couple more learned NEM junkies than me before responding …
Yes, the reason this event was market notice worthy was because it was a “non-credible contingency”, not because of the level of the frequency nadir.
Stations are designed to not have common points of failure across units, hence this contingency is “non-credible”. This event will trigger an investigation to ensure its cause is understood and not repeated.
—–
As an aside, even after a credible contingency, a frequency nadir below the frequency operating standard, i.e. 49.5Hz, will trigger a market notice.
These type of demand reductions are unusual enough that you are writing about them. This makes me wonder what was so different this time? One of the many problems is that the information that should be available from every point of presence via communicating meters is not actually available because there aren’t enough communicating meters, not enough software to exploit their information, and maybe not enough sensors. So if it was cloud cover or rain or a substations or two dropping the centralised data system really should have awareness. Cloud cover for instance generally has a pattern and and can be predicted. Its interesting but (i) what was different this time and (ii) we need to exploit the information each point of presence provides.
There was indeed a cloudbank with associated rainfall that passed over parts of SA on Wednesday morning. It was mentioned on local radio as being over the Eyre Penisula about 9 am and arrived over Adelaide in late morning. It was clearly visible on the BoM radar proceeding in an ENE direction. By mid afternoon the sky over Adelaide was complete clear.
Thank you for explaining how this is done.
MLFs aren’t hard to find but is there a table in particular one should use for the half-hour RRP?
Sincerely
https://aemo.com.au/energy-systems/electricity/national-electricity-market-nem/data-nem/aggregated-data
By the way, above what percentage of yearly curtailment (defined as (AVAILABILITY – TOTALCLEARED)/AVAILABILITY taken over a year) would you consider an asset to be problematic?
Hi Dan, it depends! It becomes problematic when the bank and the investors are unhappy with their returns, so this is different for each asset depending on its financing structure and the assumptions made in the business case. Also, what really matters is revenue – so if it’s a simple fixed-price PPA then curtailment is a fair proxy, but if the plant is merchant (spot-exposed) or has negative-price exposure, then the spot price at the time of curtailment matters as well as how much curtailment there is.
If anyone can confirm or otherwise that it was a big cloud that caused the loss of 500 MW of rooftop solar in South Australia last week please post your thoughts. This is a big deal as it represented a sudden 45% increase in forecast demand for the entire state of South Australia.
Looks like NEM frequency went rather out of band (~50.2Hz!) during the scada outage, this might have tripped some solar. I was looking for public data on DUID responses to frequency excursion, but this is also SCADA based.
Hi again Marcelle,
I have tested out what you said: “The calculation can be simplified by aggregating AVAILABILITY and TOTALCLEARED over a time period (such as months), and ignoring the SEMIDISPATCHCAP, as any time the TOTALCLEARED is less than AVAILABILITY (and therefore making a non-zero contribution to the curtailment calculation), the SEMIDISPATCHCAP will be 1.”
Although what you wrote is the case for most generators, the difference between AVAILABILITY and TOTALCLEARED will change if we include the SDC flag or not for some generators (COLEASF1 over 2019 for instance).
Do you have any idea why that might be?
Sincerely
Hi Dan, good question! There’s another complexity with the dispatch target (TOTALCLEARED) – if the ramp-down rate is involved. Each generator specifies a ramp-down rate in its bid (in the case of COLEASF1 in 2019 this appears to be 3 MW/min or 15 MW in 5 mins). The NEMDE dispatch calculates the dispatch target from the AVAILABILITY and constraints and bids, but then allows the step from the INITIALMW (the snapshot of output just before the dispatch calculation is done) down to the TOTALCLEARED to be no more than the ramp rate. In that case there won’t be a SEMIDISPATCHCAP=1 as the TOTALCLEARED is greater than AVAILABILITY not less than it. For most farms this is infrequent enough to be immaterial in a curtailment calculation but it appears for COLEASF1 that on partly-cloudy days where the generation and forecast moves around rapidly that this can happen more often, and so it is necessary to filter for only SEMIDISPATCHCAP=1, to avoid under-estimating the curtailment.
Fascinating figure! Why do we see such a big increase in total capacity bid from 2000 to 2005? Surely renewables aren’t an explanation, since they mostly came on board from 2010.
Demand in the NEM grew, Owen (up until 2009 or so)!
The tax refund offsets nearly half of the (projected) PPA losses, so it could be said that taxpayers are somewhat subsidising the wind industry, through AGL.
There could also be some element of bluff from AGL in an attempt to warn off other market entrants, otherwise known as competition. Most people are fully expecting NSW prices to jump when Liddell closes.
Others in the hole with overpriced PPAs include the ACT, VIC and QLD governments and Snowy because of their exposure to renewable PPAs.
Expect the RE lobby to spin this as a government failure somehow.
SA’s grid is getting even more complicated. And that’s saying something.
The system strength shortfall declared in 2017 was supposedly addressed by the synchronous condensers, but there was a further inertia shortfall declared in 2020.
SA shortfall 2017: https://aemo.com.au/-/media/files/electricity/nem/planning_and_forecasting/ntndp/2017/second_update_to_the_2016_ntndp.pdf?la=en&hash=A9EE910B7DA3C1D88927871630C02B48
SA shortfall 2020
https://www.aemo.com.au/-/media/files/electricity/nem/planning_and_forecasting/Operability/2020/2020-System-Strength-and-Inertia-Report
Separate but related to the shortage of synchronous generation in SA, is the imminent closure of Torrens Island A which is prominent in the combinations of minimum synchronous generator. Then there is the likely cancellation, or at least delay, of the EnergyConnect transmission link to Wagga, and the inability to shut off excess rooftop solar. Tough place to be in energy politics.
I wonder which heavy industry, reliant on large amounts of electricity, is considering South Australia as a potential investment location? I guess there will always be government subsidised, unprofitable, low capacity factor, hydrogen to clap about.
How much does the cost of SA’s upgrades count towards the rest of the NEM’s bills?
Any idea of the effects of the current constraints on the Heywood interconnecter due to be repaired late June?
With no where to go, negative prices are inevitable
Disappointed these Syncons are taking so long. Originally instal was late Jan and now we are out to late April
Installed capacity is meaningless.
All the renewable generation installed won’t benefit anyone when the wind stops blowing at night time.
There are multiple day long wind droughts across the NEM which see aggregate wind farm output falling to less than 500MW.
The choke point is real and we are going to be hit with the consequences of believing installed capacity is something to celebrate.
Ben, very interesting set of facts and figures. On a point of clarification: In the Outlook section you talk about a forecast pipeline of ” between 500 MW and 1 GW per month” but in the charts just below, it looks like those are quarterly figures not monthly? incidentally, you might want to swap the titles of those charts around…
thanks for that Gordon. I am also an Amber customer (I wonder what %age of their customers are energy wonks…?) primarily for interest rather than expecting massive savings, as i know i don’t have a lot of load-shifting tools available to me at the moment. I got on the Amber train a little earlier than you about March and have since congratulated myself on my impeccable timing, as that’s pretty much when the wholesale price started to soften with slightly lower demand and falling gas prices. So you and i have front run the wholesale price falls that most other small customers will only see flow through with the change in VDO from 1 Jan (I believe). My energy wonkiness extends to keeping track of my bills for the last ten years. Thanks to Amber’s model and the lack of price shocks in Victoria since I joined I am paying pretty much the same in nominal terms for both the daily charge and the per KWh rate as when we moved into our home in 2010! see here for more detail: https://boardroomenergy.com.au/2021/01/08/the-mystery-of-the-missing-bill-shock/
Thanks Kieran, nice post. You really nailed the timing. With no exit fees (or Amber wait-lists), in usual years with a more challenging summer I guess the least cost retail customer strategy (in terms of dollars, not hassle) would be to go on to fixed-rate for summer and real-time for March-November…
Gordon, that’s what i thought too and wondered about switching out in december. Glad i didn’t as my January bill had the lowest average wholesale price so far – 2.47c/KWh! my total wholesale costs were less than $5. Of course, i agree that we can’t rely on such benign prices every January!
I must be a bit behind but haven’t come across the acronym VRE before and my search (Ctrl+F) only finds one mention in this article with no real explanation “”VRE curtailment” – reductions in large scale wind and solar output across the NEM due to various factors, including system strength constraints” doesn’t really explain what it is nor where the acronym comes from. I would appreciate an explainer greater than is given under Figure 40 please.
Shane
“VRE” = Variable Renewable Energy. Basically shorthand for wind and solar generators whose output is variable with incident wind or sunlight. Large scale VRE generators in the NEM generally require capability to control output downwards – curtail – when instructed by AEMO.
Gordon, the environmental certificate cost would include (or be entirely made up of, not sure) the Victorian Energy Efficiency Certificate cost, which is borne by the retailer and passed through to customers.
Thanks Kat, I had forgotten about the state-based programs when I wrote that footnote up. I’m also not across the cost-impacts of the scheme relative to the RECs.
Nice article and well articulated thoughts Gordon. I’m also a recent Amber convert (and work with C&I spot pass through). I love the Amber model, but I’ve got a couple of musings around their long term prospects in such a vertically-integrated heavy market (we have a big 3, better than Coles/Woolies or QANTAS/Virgin[/Rex?] right!)
1. Like Kieran I can’t help but feel Amber’s customer base is mostly energy nerds dogfooding our own products or at least indulging our interests. I was somewhat surprised when a non-energy industry friend of mine recently independently raised Amber over a beer saying that he was on the waitlist and keen to start with them. That being said PowerPal has been seeing great engagement from the Victorian community at large, so there’s definitely some appetite for more novel approaches to managing residential consumption and costs.
2. The back end hedging required to meet both the solar minimum FiT and VDO (at least here in Victoria) must have some significant challenges in keeping the fixed costs low.
3. I’ve joked with others that Energy Locals model of outsourcing their retail licence might be the smartest play in the retail sector, but it imposes some significant risks to the fixed costs – already Amber has had to bump the monthly charge to $15, presumably due to points #2 and #3.
4. The ESB seem intent on legislating in favour of the vertically-integrated incumbents (staring you right in the face RRO). Some of these changes could be seriously damaging to the competitiveness of small, spot-exposed and non-generation owning retailers.
The most serious negative figure is Net cash provided by operating activities down 46% to $614 M. That is close to a 50% reduction. This loss of cash profits is reflected in Underlying EBITDA down by 13& to $926M.
This large change in the net cash receipts is the most concerning as it shows the underlying profitability has been significantly reduced.
If AGL’s write downs were simply due to overvalued assets it would not have impacted the cash flow.
As time progresses the cumulative financial impact on the small business sector from sudden Victorian lockdowns will make it harder for AEMO to accurately forecast electricity consumption. Small business is all about “confidence” and small business working capital has been severely eroded with lockdowns. Is it possible to provide an update to this article with a graph of the actual consumption compared to STPASA forecast? AEMO’s 2020 Electricity Statement of Opportunities Figure 57 states that the largest consumer of NEM supplied electricity in ‘Victoria’ is Small and Medium Enterprises. However, in AEMO’s 2020-21 Budget the largest consumer of NEM supplied electricity in the ‘NEM’ is large business at 57%, followed by residential at 28% and small business at 15%. The 2020-21 AEMO budget also states that “consumption of less than 10 MWh per annum is a household”. Identifying your largest customer and understanding their consumption intentions is critical to AEMO’s forecasts and AEMO’s Integrated System Plan.
Shows the need for those sync cons to come online as soon as possible. Also shows the need for the SA Government to move residential hot water to smart controlled loads.
So Syncons + automated demand response as Craig suggests will possibly see zero gas sometime next year if not 2021
This load shedding is caused by the US policy that requires Gas in times of severe cold to be diverted for home heating depriving gas generators of fuel. Gas generators do not store gas on sight. This exact scenario has been predicted by Meredith Joan Angwin in her Book Shorting the Grid.
It is a system problem in that generators are not paid anything for storing alternative fuels like oil on sight. As much as I would like to blame Wind for this Grid failure it is not a problem with renewables as such. It is indicative of poor grid management. Note Nuclear Plants do not suffer from this problem.
https://www.amazon.com/Shorting-Grid-Hidden-Fragility-Electric-ebook/dp/B08KZ51SDP
“It is a system problem in that generators are not paid anything for storing alternative fuels like oil on sight.”
So the FF generators work on just in time like wind and solar but somehow they’re to be scolded?
My take is unless tenderers of electrons to the communal grid can reasonably guarantee them 24/7/365 along with FCAS then they can keep them and use them themselves. But then I’m just a naive consumer of an essential service.
Correct. Meredith Angwin describes this problem. The gas burning generators are capable of burning oil instead. But no one wants to pay them for storing the oil on site. What is worse is that no one seems to be actually responsible for guaranteeing supply. The market system which is run by the Federal Energy Regulatory Commission (FERC) is a failure.
https://www.ferc.gov/
“What is worse is that no one seems to be actually responsible for guaranteeing supply.”
Well ironically that falls to ERCOT and watts in a name eh? Anecdotally there might be a strong mood for change and real answers-
https://wattsupwiththat.com/2021/02/16/hotel-in-austin-texas-sends-letter-to-guests-we-are-out-of-food-use-glowsticks-for-lighting/
Extreme cold can actually kill people.
The outcomes for 2020 have been compounded across the board by Covid impacts on demand.
The prices in QLD were negative much of the time due to QLD > NSW interconnector being restricted during day light hours during low demand periods. Typically there is some work on the lines, but 2020 saw an increase due to the upgrades on the interconnector and related equipment. Some of the activities of the FF plant in QLD needs to be questioned as there are times when more units should be taken off line, even if it means a bit more use of gas. It will be interesting to see the impact of 1GW of wind generation capacity is online.
NSW won’t see many negative prices until it has more VRE. I doubt this threshold will be reached in the next two years.
The ACCC should be investigating the predatory pricing of some of the coal generators. Just because it is difficult or costly for them to ramp their plant, doesn’t mean it is legal to force the prices negative due to excess supply. This is particularly the case in Victoria and QLD where the coal generators have significant market power to fall under this section of the Australian Competition Law.
Craig your last para does not make any sense. Why is it not legal for coal generators to bid negative? Why would it be in their interests to force prices negative? if it’s not legal for them, why is it legal for wind and solar generators to do it? the point of negative prices is to signal excess supply, so sounds like the market working as it’s supposed to. also why would QLD generating companies run less coal and more gas when gas is more expensive?
Anyway good luck getting the ACCC to do anything about the state owned generators in QLD…i mean they’d probably quite like to but ACCC can’t always do a lot about the actions of publicly owned businesses.
Predatory pricing has to do with market power. There aren’t any wind farm operators with enough market share to be able to dominate the market. This is not the case with AGL in Victoria and the QLD government owned generators in QLD. If you do a search for predatory pricing and Woolworths you will see there was a successful case run by the former Trade Practices Commission under the former Trade Practices Act. The current Australian Competition Law has been strengthen in regard to predatory pricing.
Generally the ACCC will only take action if it is in the media or has a large impact on consumers. These cases certainly fit the latter as the dominant generators can make it unprofitable for small VRE to produce power. This discourages further investment in VRE and would extend the time that the coal plants are operated for.
Why run gas? If the QLD Government owned generators shut down a coal fired unit in the low demand and high VRE output months, then most of the time there would be no or very little need to operate the fossil gas plants. Now it might be cheaper for the QLD Government to run excess coal plant and push the VRE generators out of the market, but it isn’t fair, legal, in the long term interests of Queenslanders both in terms of the wholesale electricity rates or the climate.
I fully understand that predatory pricing cases are complex and difficult. The Woolworths one run through the courts, including all the way to the High Court, over a ten year period.
Well, the Queensland government who own the coal power stations there have set a renewable energy target of 50% and set up a new company CleanCo to invest in VRE, and write PPAs with VRE to help them get there. So i’m still not sure why you think they are trying to discourage VRE.
AGL have a single coal plant in Victoria, so they are in no way “dominant”. perhaps you meant to type NSW where they have two…although they are closing one in the next couple of years which is odd behaviour for a “dominant” player.
There are multiple layers of ongoing monitoring of behaviour in the NEM. the AER and the ACCC both have roles in this regard. So it’s not like no-one is looking.
When there’s a lot of VRE, i.e. when the prices are negative, no-one has market power and anyone who bids below 0 is only doing so because they are desperate to be dispatched. VRE do it because they still earn via their PPA, coal plant do it because they can’t afford to go offline and have to restart. That being “costly or difficult” is a perfectly good reason.
I don’t imagine i’m going to change your mind, but just in case anyone else is reading…
some useful facts and figures here Paul. First question should be: is the planning standard designed to deliver reliable power at these extremes? If so, then yes, something has gone wrong, and Texas needs to find out what it is. but if not, and it does sound pretty extreme, then the question is does Texas want to pay for a higher standard of reliability?
on the subject of whether a capacity market would have saved them – well maybe, but a) at a cost and b) only if capacity was certified for operation at these extreme low temps and/or there was sufficient penalty for non-performance.
As for greater interconnection, well maybe but that didn’t help CA in the summer. its neighbours were also experiencing demand peaks and didn’t have enough to spare. i believer the cold snap extends beyond TX…
what about better insulation of buildings (weatherization in the US)? this would certainly help reduce the discomfort of rolling blackouts, but it wouldn’t prevent them alone. Logically it would just shrink the demand curve and the supply side would rebalance accordingly. you still fundamentally have the load duration curve issue to solve – are there sufficient incentives for those last few MW to be available just for those rare extreme weather events?
finally – blame the market. well having a single public utility is no silver bullet – you still have to have the standard, someone has to oversee/enforce it, and the extra kit (more capacity/gas storage/pipeline insulation…) costs money and has to be paid for.
And you don’t think all this incredible mess was entirely foreseeable ?
Well it has raised the political temperature with the finger pointing but I found this defence interesting-
https://www.msn.com/en-au/weather/topstories/fact-check-is-green-energy-to-blame-for-texas-power-outages/ar-BB1dLEIq
“Wind energy provides only about 25 percent of Texas’ total power throughout the year, according to ERCOT data—natural gas sources account for 35 percent—although the turbines tend to generate less power in the winter.
The Texas Tribune reported that only 7 percent of ERCOT’s forecasted winter capacity was expected to come from wind power sources in the state. Meanwhile, around 80 percent of ERCOT’s total winter capacity is generated by natural gas, coal and nuclear power.”
I’m reminded of the cautionary tale with statistics that the average adult has one large breast and a testicle.
It is now several days after this article was posted. I agree with the observations and analysis, including the forward projections included in it.
Congratulations for a broad, well presented summary of the situation which has now pertained in Texas for 6 days and for the balanced conclusions that (a) this is a system problem and (b) electricity is not optional – it is an essential service that warrants higher standards of reliability than, unfortunately, have been achieved in Texas.
Great post, thanks.
Negative prices are ‘predatory’? First I’ve ever heard it explained that way.
Wind and solar are still supported above market rates by PPAs (just ask AGL) and the RET. these generators almost all bid negative just to get dispatched. With governments underwriting RE the market is oversupplied during the day when the overpaid rooftop solar is in full flight.
Currently there is only investment in generation that runs when there is wind and sun. So the energy-only market indicates oversupply during these periods.
Until governments remove the incentives causing oversupply when there’s wind and sun, there will be no investment in anything else.
Perhaps a capacity market would change things, but I’m doubtful. I think we are destined for blackouts
Adelaide, Melbourne, California, Texas, who’s next – Germany, UK?
Hi Warwick,
Very interesting article. I would just have one question regarding how prices are being calculated in both Case 1 and 2 for 30-min settlement:
In Case 1, payment is calculated as follow ($50 x 0 + $40 x 0 + $60 x 0 + $55 x 0 + $60 x 0 + $14,700 x 1)/6 which is different to Case 2 (-$1,000 x 1 + -$1,000 x 1 + $60 x 1 + $55 x 1 + $60 x 1 + $14,700 x 1)/6). As I understand, regardless of when the battery dispatches, settlement price is calculated as the arithmetic average of the 6 dispatch prices. Would you mind clarifying why 2 different calculation methodologies are used here?
Regards,
Antoine
just a coda on this topic following unfortunate events in Texas last week. you may have read about people there on wholesale pass through contracts who are facing bills for $1,000s because the price there (for those lucky enough to have power…) hit the cap of US$9,000/MWh and stayed there for days. Amber customers here in the NEM are not so exposed because the NEM has a rule called the Cumulative Price Threshold (CPT). This is currently $224,600/MWh. If cumulative prices over an extended period (seven days) reach this level, then AEMO effectively overrides the spot market and applies administrative pricing of $300/MWh instead. So while this situation would still be quite a bill shock for Amber customers, sustained exposure of the level seen in texas last week wouldn’t happen in the NEM. PLus if you’re in Victoria, Amber have to cap your total bill over the course of a year at the level of the VDO, so that’s a further protection.
My comments are restricted to apparent parallels between extended weather driven price hikes in Texas compared to those in AUstralia.
Prices going ballistic for sustained period of time driven by extreme weather is less likely in Southern Hemisphere then Northern Hemisphere. Why? The polar vortex of the SH is much more stable than its NH counterpart. Whilst we can get big anomalies stable polar vortex means, they would be expected to be of much shorter duration (intensity?) then our friends up North.
Antoine,
The two calculation methodologies are 5-minute and 30-minute settlement. The key point is that the battery would rationally be dispatched the same for each case (i.e. 1 single high price or both high and low prices in a half hour) yet it results in adverse outcomes especially in case 2. Therefore 5-minute settlement does not penalise a battery, the way half-hour settlement can.
It’s Economics101 that if the ACCC sits on its hands with solar and wind dumping on the communal grid with power at the beck and call of the weather and sunlight then alternative dispatchable power with FCAS will be driven out. Basically we’re giving up reliable large scale hub and spoke generation and distribution for unreliable spaghetti and meatballs. When that fails or looks like it will State Govts step in with the diesel generators to save the day like SA and Tasmania you’ll recall-
https://www.themercury.com.au/news/tasmania/basslink-backup-fast-tracked/news-story/f338dfade76e55495ccdffe18d98f247
It’s politically nicer if you pop in a Unicorn Big Battery for the mainstream media to drool over while you’re fast tracking that Govt intervention of course. My local IGA supermarket in Adelaide installed a permanent large backup diesel generator after the Labor Govt did that. Probably at the behest of their property insurers who mightn’t renew their insurance or else charge prohibitive premiums if they didn’t.
There are three or four factors which most analysis understates which may make even this report end up being too optimistic in the case of coal fired generation.
1. More governments and businesses are focussing on energy efficiency. Even though Italy has three times as much industry as a share of GDP and its major cities are hotter in summer and colder in winter than ours, Italy uses 40% less electricity per capita than us. i.e. we have huge scope to continue reducing demand from the grid which has already fallen from a peak of 210 TWh to less than 190 TWh in the last 10 years.
2. New solar technologies such as bifacial panels, tracking systems and coatings which improve offaxis collection and finally the increasing share of tracking solar vs fixed tilt systems will mean that annual solar output will increase faster than capacity.so a doubling of solar capacity over the next 5 year will probably mean a 250%+ increase in output
3. New wind farms are approaching 50% capacity factors vs an average of around 30% for those installed by 2018 so the next 5 GW of wind will probably supply as much energy as the last 7 GW.
4. The increasing geographic divesity of renewable installations will mitigate the effects of wind lulls in SA by allowing more excess production in NSW and Queensland to flow south.
5 Batteries , syncons, synthetic inertia can provide most FCAS services at very low cost further depressing revenue for coal plants.
6. If all the above proves out and grid demand falls to 165 TWh, wind output doubles and utility solar rises by 250%, even if gas output halves, coal output will fall by closer to 35%
Peter, your points 2 and 3 claim dramatic improvements in capacity factors for solar and wind in only a few years. Do you have any credible sources for these claims?
Similarly is there credible modelling you can point to that supports your claim that “diversity” of output is likely to be meaningful and not impacted by transmission congestion?
Thanks for great article
I suspect that with improving value in heatpumps and higher COP,electricity gas price differentials moving in favour of electricity and a start of a movement to build gas free developments we have reached a tipping point where domestic and commercial gas is not economical. The building of the first all electric hospitals will confirm the possibility of gas free operation fro most commercial and institutional buildings so hopefully the next 5 years will see a significant reduction in gas use for heating while gas use for power generation could fall by 85% from 2017 levels by 2024 unless coal use falls dramatically, it has already fallen 40% since its peak
South Australia has been importing power at breakfast and dinnertime lately, what happens when there is no spare coal power available from Victoria?
What happens on a windless night with no fossil power?
Check out the capacity of big batteries compared with the stream of power that runs 18-30,000MW deep continuously.
Why are they so excited at RenewEconomy?
I just love the justification for intermittent t supplies being based on averages, while in reality, “ 26.54% of total electricity consumed over the year is during periods when there is no wind or solar being provided to CN from its Wind PPA or solar generation assets.”
Unfortunately electricity system design is not based on averages.
Hi Ben
Nearly three quarters of supply being matched without firming or detailed DR is a pretty good start. This is a financial contract though not a power system design. I don’t believe managing grid reliability should be a customer responsibility?
“I don’t believe managing grid reliability should be a customer responsibility?”
It’s a fallacy of composition to believe we can all take that view. As for opportunities with V2G I fail to see why on earth you would want to make any necessary grid firming battery capacity mobile rather than fixed. After all if our vehicle batteries are larger than they need be in order to firm the grid that’s simply more wasted energy lugging them around wouldn’t you think?
“A fallacy of composition” – I’d be interested if you’d like to expand on that. When you fuel up at a petrol station, as long as you have a compliant vehicle, is it your responsibility to ensure there’s fuel supply? What’s the difference with grid supply?
As for V2G (and I’m no expert), no doubt vehicle batteries will be oversized for customer convenience (who really needs to travel 400km without stopping?) but the broader point of v2g I think is vehicle utilisation. Vehicles are usually not required for their primary purpose (driving) 24 hours a day.
When I fill up with diesel or 91-98 octane petrol for that matter I’m guaranteed full measure and a quality standard although the price varies depending on competitive supply and seller costs. I won’t expect the same price at William Creek and Adelaide that’s for sure but you’re free to set up in competition to any of them. What I don’t want is my car to play up down the road and be told that’s the way it is sometimes as the Premier’s nephew has special rights to deliver his special Green brew as much and as often as he likes but the quality is dependent on the weather.
As for being forced to buy an expensive BEV only I won’t be thinking about my duty to save the unreliable grid with its battery sizing anymore than servo sustainability with my ICE choice now. After all it’s my hard earned and I don’t have the luxury of sticking up everyone’s rates to finance my particular preferences be they judicious or profligate. Naturally if there’s a cushy Council job going with a long range Tesla thrown in I’m all ears but then we come back to that fallacy of composition problem.
Whilst local Councils or we as individuals can fiddle around the edges with the communal grid the really big picture problem is State mandated dumping-
https://en.wikipedia.org/wiki/Dumping_(pricing_policy)
Normally the ACCC would be all over such practice by wind and solar but instead we have the silence of lambs.
Thanks Adam – really interesting stuff. What would be even more interesting would be a net load duration curve for the year: i.e. for each half hour (soon to be 5 minutes…) load – own PV output – the scaled output from the wind farm. this would capture your exposure to spot outcomes and address Ben’s point about averages.
Thanks Kieran, I’ll put that together. Also interested when I have time to see the actual firming size required and also to evaluate average pricing during those periods of unserved load – ie from a financial perspective wouldn’t necessarily need to cover spot-exposed periods of low average prices. That’s where looking and covering the early evening exposure will be the first step.
Hi Adam – great article. Couple of questions –
1) How did you define the BAU scenario?
2) “It is important to note that this analysis relates to the delivery of electricity rather than the associated emission reduction which comes from Large-scale Generation Certificates (LGCs).” – is this saying you haven’t factored in the cost of the LGCs in this comparison?
Hi Sean
1 – BAU is based on fixed price contract pricing supplied at time of tender when the PPA was entered into (I.e the contract we otherwise would have entered into).
2. There’s further commentary on the original post, but yes this is a like for comparison on electricity supply, not emission reduction schemes.
was a transformer fire at Torrens island power station
Nice article Adam. I did a similar analysis for the ACT’s 100% renewable target that you may be interested in. I contrast to NSW, the ACT is using a mixture of wind and solar to meet its needs. Perhaps the main headline from this was that 75% of the ACT’s needs could theoretically have been covered by the wind and solar generation without any storage. Kieran, it includes some generation duration curves!
https://sites.google.com/view/100-renewable-act/home
Don’t need a battery, just V2G instead 😉
The right data to make decisions is hard. Converting the data to action is even harder. Pretty hard to do with only 30 minutely data from the connection point. Often only for yesterday too.
I can’t see any comparable EV to the ICE paying for itself at double the price when you can go all electric with solar panels and with a solar diverter controller use the power to heat a storage HWS and any left over to aircondition the home for the homecoming. Solar panels are relatively cheap and so are electric element HWSs so lithium batteries aren’t in the race for energy storage.
As for V2G you go right ahead and pay for larger longer range batteries than you really need for transport if you feel the charitable urge to help stabilize the grid. As a BMW exec was asked when will EV prices come down he responded ‘never never never’ as they can make an ICE car for the price of a battery pack with its BMS. As a consequence any fuel cost savings with an EV is eaten up by the interest on borrowing as nearly 9 out of 10 do with car purchases.
I am in SA and have a solar system. My solar system was curtailed between about 2:45 tp 4 pm NEM time on the 14th March 2021. Mine is not the latest system with agent control, so I assume I was part of the 40MW reduction directed from substations. They must have raised the voltage above 453 volts. Anyone know how much they raised it to?
Sorry I meant 253 volts :–)
Interesting article Adam
Two questions;
1) How would the BAU comparison look if referenced to todays wholesale price – it looks like you have set a BAU baseline based on a market highpoint versus the collapsed wholesale prices of today
2) what controls do you have to ensure Flowpower / Sapphire are not selling the same Mwh of generation to multiple end customers
Funny how all these issues keep happening in states where they are closing coal plants at record pace and replacing them with renewables which unbeknown to many people, don’t actually produce reliable supply of energy.
The answer from those in the know is to just install more renewables because they seem to be unaware that installed capacity does not equal energy output.
So we then move on to interconnectors at the cost of the end user which are more likely to fail due to the highly variable nature of a more renewable/less coal grid.
All this because the government supports renewables and penalises coal plants so no one builds firm capacity generation any more.
It’s only going to get worse as more coal plants close.
You have to then ask, what if it happened in Victoria? If a busbar trip put Loy Yang 500KV out of action, and perhaps basslink with it?
A 2013 transmission line fault in Victoria, triggered a sequence of events not dissimilar to the SA system black, as transient voltage disturbance led to generation tripping, including the AVR on a brown coal unit. Fortunately, only some under voltage load shedding resulted.
https://www.aemo.com.au/-/media/files/electricity/nem/market_notices_and_events/power_system_incident_reports/2013/psior_trip_of_ringwood_rowville_220_kv_transmission_line_multiple_generators_on_29_sept_2013.pdf
Hazelwood was still operating back then. Would be interesting to see how the situation would go in the current VIC market.
I imagine it would be much worse due to the expansion of renewables and closure of coal plants.
Did you even read the fucking article?
In case reading comprehension isn’t your strong point one of the key contributing factors was a significant shortage of DISPATCHABLE THERMAL GENERATION. But no, you’re right, some more coal plants will solve complex system resilience right away. Just build them and watch all the historical problems of the NEM design (and the state electricity commissions that preceded it) melt away.
Hell maybe we can export your simple and genius strategy to other grids around the world (hi Texas, you need more gas), along with lots of beautiful black Hunter coal.
How come we never had these problems before the expansion of renewables? The cause of the problem is the expansion of renewables which then means no one invests in reliable fossil fuel plant.
The “historical problems” of the NEM design are only problems if you decide to wholesale change the entire generation mix and rely on large scale transmission from unreliable sources.
Yes, my Enphase recording system shows a big drop to 10:30am to 12:45 pm Adelaide time. AT 1pm it shot back up from 800W to 4kW
Congratulations on taking in this type of analysis. It is refreshing.
It must also be possible to assess each solar farm earnings against an assumed PPA strike price, to gauge the ongoing effectiveness of the CfD as a hedging strategy for purchasers.
I presume a similar effort will be made on wind farms?
Why do people insist on using the $0.00c format. It’s confusing. Are we talking dollars or cents here. It’s either $0.00 or 0.00c but it can’t be both $0.00c.
It’s not that hard to write one or the other format. Plus it takes up extra characters.
Great work Adam.
This kind of analysis is crucial as we move the world towards True Zero (as opposed to Nett Zero) carbon. In the past 6 months we’ve seen Google, Microsoft and IBM all commit to this idea – 100% renewable energy procurement without relying on offsets of certificates. In Europe there’s even a as standards push (energytag.org) towards making RECs completely time based, and it’s what Enosi’s Powertracer is designed to do as well.
I see a seismic shift is underway-
https://www.msn.com/en-au/news/australia/seismic-shift-underway-of-electricity-grid/ar-BB1eUQK3
Interesting article. In 2019 I mentioned to an energy bigwig that gas was being displaced by wind and solar and he heartily disagreed. The data says otherwise.
In the DDPS section, it should be noted that Origin are also gas developers, so their fuel cost is lower than market. Similarly for Arrow (Braemar 2) and QGC (Condamine).
I think a lot of confusion about the role of gas in the NEM arises from failing to distinguish between sources of energy and dispatchable capacity. The data clearly shows that gas-fired generation’s contribution as a significant energy source has fallen and because of its inherent variable cost disadvantage to renewables and coal, that is not going to change.
The more open question is around the role of gas-fired generation providing (relatively) firm capacity for relatively short periods. In this role only small amounts of energy (as a proportion of total production / consumption) are required so fuel costs are much less important and the economic question turns to which technologies – storage, demand response, gas or even liquid fuelled generation, transmission enhancements etc – provide the best combination of fixed capacity costs, flexibility, and supply duration. My view is that it’s premature to be talking about replacing most of the NEM’s current gas fleet in this role in the near term, but it’s much less clear how much additional gas-fired plant will be economically justifiable.
What the AEMO approach does not do is allow end users that have a weather driven demand (eg shopping centres) to provide WDR. These sites are ideally suited to provide WDR when it is most needed, but are excluded from providing WDR
Absolutely agree, David. The 20 per cent cap on adjustments will prevent many temperature sensitive loads from participating, which is unfortunate because these are exactly the kinds of loads you want to be offering WDR when it’s hot and supply/demand is tight. AEMO seems resigned to its current approach for market start, but I’m hopeful they will open the market to a wider range of load types soon after.
Confused. There must be a structural reason that NSW is consistently top of pack?
Capacity factors are only going to decline further as we get more clouds due to more cosmic rays.
We are already at the highest concentration of cosmic rays in recorded history and it keeps climbing.
From what I can see the emergency rating from PowerLink failed over to the static summer rating, which is generally quite conservative
This was discussed in this ‘Part 4’ article:
https://wattclarity.com.au//articles/2021/04/24jan2021-scadaoutage-casestudy-part4/
I don’t think its so much VRE.
Isn’t it more “semi-scheduled generation” VS scheduled generation ?
Semi-scheduled generation is the low-cost energy,
yet the system needs scheduled generation. It’s the system energy.
(Especially when they get paid the same price)
Isn’t it time we put the onus on the operator to provide the system with system energy.
Shouldn’t the strategy be all DUIDs register capacity as scheduled.
The operators should take cares of the “managing around” bits. If they choose VRE,
then they need to provide the “managing around” bits thru hybrid generation or partnerships with
other providers.
100MW of Generation should be 100MW of generation. Not something less.
That would be completely the wrong approach, if we want to balance supply and demand at least cost (including environmental cost)!
100 per cent of demand is non-scheduled but system operators have found ways to manage its uncertainty and variability for at least as long as we’ve had large scale grids. Non-scheduled or semi-scheduled variable generation does not create a new kind of problem as far as capacity adequacy goes.
Of course any system will always need a large volume of fully dispatchable capacity, but it certainly doesn’t need that type of capacity exclusively, nor must any fixed or minimum proportion of energy over time come from that source. And you’re further incorrect to state that all forms of generation are paid the same price, in the NEM at least – there are dozens of analyses around, some of them on this site, demonstrating that the average spot price earned by different forms of NEM generation over almost any given time period is very different for fully dispatchable vs VRE plant, with “peaking” plant earning the highest averages and VRE typically the lowest.
I’d suggest you read the Grattan Institute report released this week (https://grattan.edu.au/report/go-for-net-zero/) for a balanced discussion and analysis of how the “managing around” task might work in a high VRE system.
Too bad the max output is worthless.
The minimum reliable output (choke point) is what matters and that is usually in the sub 2% (sub 300MW) range.
Just 2 weeks ago on 31 March wind power generation across the NEM dropped to just 268MW out of an installed capacity of 8100MW.
Is this just a case of having 20/20 hindsight now, but not having it at the time? i.e. actual charge/discharge will never be perfect to maximise price arbitrage and you will always get caught out simply through not being able to predict future events perfectly (i.e. who’s the marginal generator, who’s next, who’s holding back capacity, credible contingency events which appear out of nowhere, non-credible events more so, demand forecasting errors due to e.g. cloud cover etc., etc.)……. of course post event you can go back and dissect what happened and perhaps build up a library of better responses, but it will never be perfect simply because the future, even the near future, is to always some extent, unknown……
Also, are these battery responses somewhat influenced by what contracts they have in place and what is being played at any given time by the owners/retailers to meet e.g. some contracted obligation? Or are all batteries, as the article suggests, simply working in the spot market maximising arbitrage opportunities (i.e. in the visible part of the NEM)?
Also, how many batteries (in MW and Mwh) can connect before they eat their own lunch? i.e. before they remove the spot price highs and lows so effectively, they remove their main source of income. And how much influence will shifting domestic loads (e.g. pool pumps, air-conditioning, EV charging etc.) have in this space (reducing the requirement for storage) noting this is also happening at the same time as batteries are being rolled out.
Superb article Carl; some non-trivial optimisation problems in there!
“Prediction is difficult particularly of the future” Neils Bohr
It’s interesting to think about lower FCAS as an “opportunity cost” to a generator i.e. getting paid for FCAS means not getting paid for energy. For batteries, things look a little better, given the energy constraint. Another big question is how different bid stacks of larger generators, and thus dispatch price outcomes will be under 5 minute settlement.
Breaking a few records recently with renewables due to the weather I see-
“Persistently low electricity prices below their cost of generation required AEMO to direct South Australian gas-powered generations on for system security for a record 70 per cent of the quarter,”
https://www.msn.com/en-au/money/markets/solar-panels-cooler-summer-drive-power-prices-into-negative-territory-in-south-australia/ar-BB1g8v17
We have known since Paul Miskelly’s paper in 2012 that there are always going to be prolonged and widespread wind droughts when high pressure systems sit over SE Australia.
We also know that on windless nights there will be next to no RE generated.
When there is no spare wind anywhere it will not help to have more transmission lines.
We know that the so-called big batteries do not store power at grid scale.
Pumped hydro costs the earth and wastes up to 40% of the power that is fed into the system.
Join the dots:)
Looking at the situation in SA (the wind-leader,) checking at breakfast and dinnertime – the peaks of demand coinciding with little or no solar, we find SA almost always importing from Victoria (with help from the battery of the nation quite often.) When the wind is low, look at the amount of gas used in SA and the coal plus gas in Victoria. Then contemplate the situation when a big chunk of Victorian coal capacity gone, bearing in mind that no amount of extra wind and solar capacity is any use on windless nights, or even windless breakfast and dinnertimes.
Stop celebrating big wind periods and focus on the lows, like the low points of a flood levee where the water will penetrate regardless of the high points of the levee or the average height. It is the low points of the wind supply that kill the grid, regardless of the installed wind and solar capacity. Build as much as you like if you want to displace coal power, RE can displace it but not replace it.
Have a look at SA and Victoria this morning, they are back in wind drought territory (less than 10% of plated capacity).
Actually Paul knows all this and he has been dropping hints for years, but he doesn’t want to frighten the horses and he is waiting to see how long it takes the RenewEconomy people to work it out for themselves:)
Here are the periods of maximum and minimum rolling 1,2,3 and 7 day NEM Wind means for the last year .. http://nemlog.com.au/nlog/nem-wind-summary/
That Capacity Factor (CF) should be one of the most important considerations here.
While there is a total Nameplate now of 8703MW, when CF is taken into account, that means that Nameplate equates to a power generation of only 2610MW
Over the last 135 weeks (a bit more than two and a half years now) that long term CF is 29.6%, roughly the same as it was 135 weeks ago.
In that time there has been a number of increases to that Nameplate, adding more and more ‘better technology turbines into the grid.
In the most recent 12 Months, there have been four increases to that Nameplate, from 6960MW up to that total now of 8703MW, so an increase of 1743MW.
However, that rolling CF for the most recent 12 Month to this date is only 29.2% in fact, so with the addition of better technology, the CF has actually gone down.
As to the weather situation, it’s relatively simple to work out when wind generation will be low. 65% of all wind plants (Nameplate (around) 5700MW of that 8700MW) are located in that area of South East South Australia and Central Western Victoria. With the approach of those large High Pressure systems, and then very slowly passing over that area, it’s easy to see when wind generation will be low, sometimes falling by anything up to 3000MW in hours.
Conversely, that in between times between those Highs and the following lows, when those isobars are closer together, hence higher winds, you can see large and sudden losses of power, ranging between 500MW and 1500MW in around the time spacing of ten to 15 minutes. Those sudden losses are due in fact to those newer technology plants, taller towers, larger nacelles with larger generators, and considerably longer blades. In those high wind situations, vast numbers of those newer tech turbines automatically turn off, so the blades, (higher tip speed) are not destroyed.
Those sudden losses of power can be seen by looking at those wind graphs, and you can see numbers of those ‘spikes’ occurring in succession, off, on, off, on, etc.
And no, I am also not a Meteorologist, just an observer with a bacground in the electrical trade who puts two and two together, having been doing this for so long now.
That CF figure should be the most important figure in all of this.
Tony.
The only thing being ‘ruled out’ is talking about mothballing, or closing down, the Stanwell generators.
QLD total annual demand is about 60 TWh per year. The 50% RET means 30 TWh has to come from wind and solar. Currently about 8 TWh come from wind and solar. The government requires at least another 22 TWh of wind and solar.
In 2019/20, gas generated 5 TWh and coal 46 TWh.
To get wind and solar up to 30 TWh, and keep the 5 TWh of gas, requires 20 TWh of coal to be displaced by wind and solar.
Stanwell operates about 3000 MW of coal generation, delivering 18 TWh in 2019/20.
To make the 50% target, Stanwell has to close all its coal power stations.
But I’d put money on Stanwell building a wind or solar farm, or at least entering more long-term contracts to underwrite wind and solar.
100% of QLD wind farms are underwritten by the state, and 30% of solar farms. That public ownership rate will continue, and likely increase, as the deadline for the 50% RET draws nearer.
This gov has free reign until at least 2024, with no senate to reign it in. We’ll see a big public spending splurge shortly on wind and solar, plus the transmission necessary to connect it all.
Wholesale is negative more often, demand is not increasing. There is no incentive for private investment.
The existing public-owned generators will run at a loss, and the new RE will run on over-priced fixed price contracts. Subsidies both ways, plus ongoing subsidies for rooftop solar, plus public spending on transmission, plus distribution costs rising.
So much to look forward to.
Working from the consumer back, all consumers want is reliable, available & affordable energy.
In the absence of empirical evidence proving the case against CO2, I believe all Australia therefore needs is a simple, easy to understand, sensible Energy Policy that’s;
• Technology neutral (eg fossils fuels, hydro, the ‘Unreliables’ of wind & solar, Bio-mass, batteries including the nuclear option – ALL on the table),
• NO subsidies ie a level playing field.
• Generators choose their own technology in a free, fully transparent, competitive market place where market forces prevail.
• Generators simply contract in advance (auction?) executing a firm commitment to deliver guaranteed power 100% 24h/day to meet forecast demand (by the likes of the AEMO) for the next day/week/month market needs at their BEST competitive price(s).
• Substantial financial penalties to apply for ANY generators failure to meet contractual commitments in accord with clearly defined 100% QOS standards (not withstanding force majeure in the unlikely event of natural disasters eg earth quakes, floods & bushfires etc).
If those T’s & C’s don’t appeal to the power generating industry, tough. Just revert the power generation industry back to whence it came ie respective State Governments & be done with it.
“If those T’s & C’s don’t appeal to the power generating industry, tough.”
But that’s the point. The problem is not technical but political because you’re outlining the level playing field we as consumers needed all along with this ‘energy transition’ as it’s euphemistically described. Instead we have State sponsored dumping by solar and wind driving out dispatchable coal with a Catch 22 situation evolving.
Normally the ACCC would be all over such market behaviour like a rash but they know their politics and hence their stony silence. To implement the level playing you propose is the only logical and rational position. However somebody has to break it to all the voters with rooftop solar they’ve been led up the garden path and their highly correlated and fickle reactive electrons are no longer wanted by their equally generating neighbours. If they don’t like it tough but that’s the tough political message no politician wants to be the bearer of. So we all continue ‘transitioning’.
What exactly are we transitioning to in the absence of telling it like it is? Well that’s continually trying to disprove a fundamental axiom of engineering. Namely that you can’t construct a reliable system from unreliable componentry and count me a skeptic that we’ll manage to disprove it.
a wealth of knowledge but the presentation and embedded logic is very hard to follow, for me at least. In providing the information, please don’t assume the reader knows all the background to your concise points.
I know a bunch of wind farms schedule their annual maintenance shutdowns at this time of year (in preparation for the more lucrative winter months I guess). Could explain some of the low overall output that you’ve seen recently.
Could the QNI upgrade works which have restricted daytime exports in late 2020 be a factor in the reduced Tarong revenue in that period?
Instead of “the super-secret financial support being provided by the Victorian Government to EnergyAustralia”, why not be above board and pay them to close coal fired stations down provided they use the funding to support the development of renewable energy?
It’s not a question of paying coal to close down but a question of timing as they’re going one by one anyway because the deck’s been stacked against any reinvestment returns. They’re simply running on sticky tape and string now to eke the last returns from their stranded assets. As they go it leaves a gaping hole with dispatchability and that’s one thing politicians understand. Or rather they understand the punters won’t be happy chilling in the dark with warming fridges etc.
Suddenly a unicorn Tesla big battery appears pronto to take the spotlight off the diesel gennys guzzling copious tankers full of refined fossil fuels. What that’s really exposing (along with backhanders to coal to stay afloat) is the dire lack of storage to make wind and solar dispatchable here and now. That’s the inherited problem can that’s being kicked down the road politically. No matter the clever technical folk at the AEMO have it all in hand as they can work miracles can’t they?
Hardly surprising. One of the most difficult engineering challenges is to wind down older coal fired plant ahead of retirement. It involves the application of much good engineering judgement to get right as there will be emerging risks known to those trying to operate the plant. Historic performance is not necessarily indicative of future performance where a particular issue may have been recently identified. No one operating with a commercial mindset is going to retire a pristine plant – maintenance spend is sensibly rationed at the expense of reliability as the plant winds down over several years. The trick is to do it safely. There are big bits of high pressure and temperature kit in a power station and some of it heavy bits spinning at high speed. A healthy dose of conservatism is justified.
The real risk here is that the economics of all coal units around Australia is being seriously damaged by renewables causing their rapidly declining capacity factors. That squeezes available maintenance money and that will inevitably play out as reduced reliability. Someone had better start thinking about how the market signals for firming capacity…..and soon.
Interesting development but my first thought was Tesla owners beware of Greeks bearing gifts-
https://www.agl.com.au/get-connected/electric-vehicles/smart-charging-trial
Meanwhile I note at the global level some battery enthusiasts are concerned about scarcity for the first time as well-
https://www.msn.com/en-au/news/uknews/insufficient-recycling-of-rare-metals-could-hinder-climate-efforts-experts-warn/ar-BB1gzmnW
FFS is not inertia but batteries and grid forming inverters can and do provide synthetic inertia. By far the biggest shortfall in all the ISP and related work is a clear understanding and plan for how to run the grid using inverters as will be required when coal generation goes away. It is disturbing that we haven’t done the appropriate engineering studies to plan for this not merely foreseeable but seemingly inevitable event.
FFS is not inertia but in some ways it is better. If inertia is thought of as stored energy then it can be shown that virtual or synthetic inertia can actually contribute much more energy per MW in two seconds than synchronous inertia can. If there is about 30 GW nameplate of gas hydro and coal turbines spinning they will have about 240,000 MWs of inertia If the minimum safe frequency on the gridis 49 Hz then available energy is 1-(49/50^2) x 240,000 MWs = 9,500 MWs and that is all.
In two seconds even allowing for a 140ms ramp time 5 GW of batteries can supply as much energy as 30GW of rotating machinery. However if the loss of generation takes more than than two seconds to recover the batteries go on supplying energy whereas the inertia is completely exhausted. In fact if the frequency stays low for more than 10 seconds or so then trips will start across the system. Thus some of the initial generator response is required to replace the inertia.
Even if a minute is allowed for frequency to return to say 49.5 Hz that means new supply has to supply 9,500/2/60 MWs = 80 MW for frequency restoration let alone replacement of lost supply to customers.
Wind and solar that are constrained can also rapidly increase output and inductances along the grid also provide a stabilisation. Hydro plants can be run dry to provide inertia and by 2028 Kidston and Snowy II will be providing another 8-10,000 MWs
If over as little as 5 years, half the coal plants departed the grid that would reduce inertia by about 100 MWs. From the above it can be seen that 2.5GW of batteries will do the job even without pumped hydro or virtual inertia on wind and solar farms.
As plans have been announced for some 7GW of grid batteries plus somewhere around 0.5-1GW of VPPs even if 1/3rd of those plans come to fruition, it is unlikely we will be short of “inertia”.whether real or virtual
Thanks for the quick writeup Paul
This was a whole-of-mainland high price event, with interconnectors unconstrained between the four regions.
The $14,950/MWh price in DI 18:00 was being set by EA’s Jeeralang units in Victoria with plenty of spare capacity in other regions but offered at prices that were in effect higher than this level when translated to Victoria after accounting for inter-regional marginal loss factors.
Overall I think your comment that there was a reasonable amount of NEM-wide reserve (5,400 Available capacity over Total Demand) is fairly pertinent – this capacity was simply being offered at five digit prices, showing that bidding dynamics, in addition to physical supply demand factors, was a major driver for the event.
Allan
Allan makes a good point about availability. It may be misleading if there’s spare capacity interstate along with some extra slack in interconnectors.
Oh sorry ignore that – you did mention it. But I do think a mention of interconnectors is relevant.
Trying to understand this a little more deeply. So there was capacity at interconnectors, and there was plenty (well enough) supply, but perhaps this supply was from a small number of participants who worked out they could set prices at high levels without being undercut? (I’m being careful here….)?
My laymen thoughts:
#1) Having less actual demand than the 30min prediction could lead to a skew to higher frequency.
#2) Having less demand makes it easier to keep the frequency above 50Hz and keep it close to 50Hz.
#3) The software control of inverters can change output without changing frequency. More generators using inverters (solar&wind) and less using inertia (coal,gas,hydro) could mean the supply minus demand differences are having less effect on frequency. Or overuse of faster FCAS.
#4) NT had a problem with the software & parameters of their grid connected mix of generators where the FCAS capacity was used too quickly, a lack of frequency reduction failed to properly signal correct load shedding, and a few other issues. The frequency increased during a time of demand>supply so some generation was curtailed until the overloaded battery supply could be sustained, this drop from batteries & the grid was unprepared for a sudden loss of battery et. al. generation. They now say they’ve fixed the control software glitches but it could happen again elsewhere in the country if these rare cases are not tested & searched for.
#5) Just because it looks better (tight around 50Hz) while everything is close to “average”, it doesn’t mean it will be better (follow same pattern to be close to 50Hz) during extreme demand & sudden changes compared to 2 years ago.
Thank you for your efforts to analyse the Australian energy market but unfortunately I’m not a big market player (just a little home user). Knowing my kWh from my megawatts, I hear these claims of green energy nirvana with scepticism. If I had the detail from AEMO etc. you (Global-roam/WattClarity) have I could work out modelling of the supply & demand requiring storage & increased supply (roundtrip losses when using storage) to make it work under different conditions/mixes (gas vs battery vs pumped hydro). During each day and between seasons.
Would you say electrical grids are still run within a minute of disaster and always have been? We would need distributed storage with megawatts to meet more than peak demand and with enough GWh storage to meet natural shortfalls & absorb the peaks (not as simple to calculate as it sounds, as Paul McArdle already knows). 100% green is not a problem of just technology and some of it’s cheaper but “whether or when we have the economics, materials & time to do it ALL when competing with the rest of the world for the same limited resources”.
I think you are trying to analyse this to death guys. The % variations you are seeing here is “noise”. The NEM PFR deadband is nominally +-15 mHz – that says the machines governor action does nothing until is falls outside that band. All very nice, but….that is just a crazy setting. Simply too tight and will wear governing systems out and on thermal machines send control system oscillations all the way back through the boiler. Hydro machines are even worse in this small signal area because of the mass of water in the turbine runner that needs to be either accelerated or decelerated. There is a lot more to tuning a governor than just deadband. Gain and reset action will determine how quickly they attempt to correct the error. To add to that not all machines will be in pristine condition all of the time , depending upon where they are in their maintenance cycle. This will mean their will be mechanical hysteresis in their systems also. So all I am saying is a few % here and there is to be expected. This distribution looks pretty good with an effective deadband of about +-50mHz – about as good as I would expect on a system where the dominant frequency regulator is large thermal machines. It might get better if there were lots of big inverter connected batteries around but we aren’t there yet. Generally on falling load I’d expect to see high frequency and rising load, low frequency. Operationally pretty normal and hugely better than before PFR was introduced as you are observing. Simply all machines are now obligated to contribute to correcting frequency(as I might add historically – pre markets – they always did). The other wild card is where an additional generator synchronises (or desynchronises) . Most need a block load or unload their last few MW very quickly when coming off to prevent chilling of metals from excessive throttling. All will represent frequency disturbance. Also, don’t forget that the inertia of the power system is not only the number of synchronous generators on at the time but also the number of induction motors driving loads in the power system at the time. It will therefore change as industry starts up and shuts down each day – the same time generators are loading and unloading in response. All of this modelling is what system modellers spend their lives doing using very powerful modelling software.
Is there any literature on the original decision to relax the primary frequency control deadbands, when that originally occurred?
WA has an island grid known as the SWIS. SEN modelling for 90% RE on an hourly basis using 2014 weather data (a low year for wind out of the previous decade) shows us that while there are winter wind droughts that can strike for ten days in a fourteen day period, it is manageable and even with these wind droughts, storage and overbuild still can provide 90% affordably, and 100% for a little bit more cost (summer is not an issue due to PV).
I’d post an image of a summer and winter four week period, but I can’t. Here’s a tweet you can look at.
https://twitter.com/SENAustralia/status/1396860795309285383?s=20
I should add the wind correlation in WA wind farm locations on the SWIS is pretty high. Batteries and PHES really helps, but only goes so far (say 2 hours for batteries today, 4 hours in 5 years time and 10-14 hours for PHES in WA)
Hi Paul,
Your link to Part 1 of the explainer about todays outage is resulting in a 404. Thought others might be looking for your insight as well.
https://wattclarity.com.au//articles/2021/05/loss-of-2000mw-of-demand-in-queensland-on-tuesday-25th-may/
WattClarity is my go to for all the energy market updates 🙂
Thanks,
Carol
oops, fixed.
Why did the H24 Calvale 275Kv Transmission lines trip 22minutes after the fire event?
Significant power system event – 25/05/2021
AEMO ELECTRICITY MARKET NOTICE.
Significant power system event – 25/05/2021
At 1406 hrs 25/05/2021 a significant power system event occurred.
Description of event – Multiple transmission lines tripped
Transmission element(s) tripped – All 275kV transmission lines out of H24 Calvale
Region Qld
Updates will be provided as additional information becomes available.
Why do you keep focusing on the maximum production when it is a pointless measure?
What matters is the minimum reliable output (choke point).
From 7am on 17 May to 11pm on 18 May wind generation in the NEM never reached above 1800MW despite there being 8500MW of installed capacity.
Paul,
It may not be generally known that most modern generators are hydrogen cooled so it is possible that it was a hydrogen explosion that was reported as preceding the trip (I believe). I think that is unlikely as there were no casualties, though this may just be a matter of luck (everyone at smoko). Another source of explosion are pipes containing steam which hopefully would only have been a small line e.g. a drain or instrumentation line. Loss of an instrumentation line may effectively cause a false signal to the turbine control systems leading to a trip. All speculation of course at this stage but any explosion in a power station is a worry.
you need to check the days of the week and dates because I think some were incorrect
Looking at the diagram it must’ve got very close to splitting Queensland into two electrical islands. It will be interesting to find what caused the tripping of all circuits out of Calvale.
Exactly right Philip. Understandably it’s the incident at Callide C that has attracted a lot of attention, but it was the Calvale trip and its apparent impacts on other generators that had the most widespread effects. It’s not at all clear whether the Calvale trip was the direct result of what happened at Callide C or some other cause.
Hi Allan, Nice to see some details. In what looks a related incident, Wivenhoe was pumping at the time of the trip of the Callide units, and from 5 minute resolution it looks to me that that pumping load tripped pretty much around then. Perhaps that helped to stabilise that first trip. Maybe that was the low hanging fruit to arrest a frequency collapse.
I look forward to your further exploration of this event.
Cheers
Controversially one might question whether the repair cost is justified in a 50% and going higher renewable generation world. If, as is usually the case, insurance pays then I guess it’s a non issue, but Callie C unit 4 replacement will confirm the death sentence of some other no doubt less efficient unit.
The problem is that coal loving QLD is no where near 50% RE and won’t be for a long time. Indeed, QLD is struggling to maintain their 20% RE target with all the VRE curtailment that keeps occurring in the State. This failure does allow the QLD Government a bit of a politically safe out in terms of closing a unit that isn’t really required. In terms of insurance, I would be very surprised if the failure would be fully insured as there is typically a significant “excess” to be paid. Given the decision to repair the unit will be a political one, the economics of the situation don’t greatly come into the decision, even though they should.
Those measurements certainly point to loss of synchronization (est. about 3.8 cycles different at the point of QLD nadir, based on pixel counting your chart). 4 second data also record zero flows on QNI. It would be interesting to know what was going on with voltages, in relation to the transmission line trips. Also loss of synchronization between NSW1 and QLD1 may have been preceded by collapsing voltage on the NSW side, triggering UVLS rather than UFLS.
The AEMO report on the “double islanding” event notes changes were made to the protection scheme “The protection equipment on the Dumaresq – Bulli Creek 330 kV lines operated as designed, although the protection design did not cater for the circumstances of this event. The auto-reclose control for the loss of a single phase of both circuits required a successful synchronism check before reclosing the lines after the fault had cleared. Powerlink and TransGrid have since modified the control logic by reducing the reclose time and not requiring the synchronism check prior to re-closing the CBs.”
Earth lasers’ plasma shield CAN prevent a devastating global blackout/all nuclear plants’ blasting by asteroid explosion (as in Tunguska-1908 & Chelyabinsk-2013) or solar storm hit! NINE times near-miss extinction so far: 1972, 1989, 2003, 2006, 2012, 2017, 2019, 2020, 2021 https://GlobalBlackoutPrevention.wordpress.com
I remember the islanding that had the QLD generators being given AGC set points based on NSW frequency, while trying to deal with local frequency going in the opposite direction.
Thought AEMO would have fixed that by now.
AGC probably wasn’t an issue here as the two systems appear to have synchronised and reconnected within 15-20 seconds, which is a remarkable outcome – fantastic recovery of frequency in Qld after the under-frequency load shedding, and some very clever controls on QNI. The recovery in Qld must have largely been due to primary frequency response from the Queensland units remaining online. While lots of things went wrong in this event, some very important things went right.
This is a very interesting graph. On your chart of reserve margins, is it possible to point to:
a) when actual blackouts have occurred due to supply shortages
b) generator closures
c) droughts
Also, it’s intuitive that all reserve margin is dispatchable generation – is that a correct assumption?
Also also 😂, how does the reserve margin correlate to inertia levels, or AEMO system strength constraints?
Thanks, Allan, for your commentary and to the people who have put together all the protection settings across the network.
Just 5 days later total wind farm production in the NEM fell below 400MW for almost the entire afternoon. It hit a low of 110MW for almost an hour…….
I had a look for diagrams and from an old AEMO incident report it seems that Calvale is a breaker and a half substation, so even a bus fault or a CB fail shouldn’t have resulted in the loss of all circuits from Calvale.
Diagram on page 5 of (https://www.aemo.com.au/-/media/files/electricity/nem/market_notices_and_events/power_system_incident_reports/2014/psior_trip_of_calvale_no2_275kv_bus_and_feeders_8810_and_8811.pdf?la=en&hash=6B85AA15D5204B73462FAE1AF7867965)
Philip – those details are getting beyond my area of knowledge, but there’s some good discussion of just this issue by Winodh Jayewardene on LinkedIn: https://www.linkedin.com/posts/activity-6804036620938047488–zFt
If the media buzz is to be believed, and it is a turbine rotor ripping out through the casing, the parts alone to fix it are the better part of a year to source. And it’ll be weeks (at a minimum) before there’s enough failure investigation to understand precisely what needs to be ordered, and whether you can simply do ‘like for like’, or do you have a design failure (and then what happens with the sister units C3 and Tarong North). And even within all of that, the insurance claim is going to be pretty messy.
Two statements which stand out to me as defining the event.
“At 1334 hrs on 25 May 2021, Callide C4 stopped generating but did not disconnect from the power system.”
“At 1340 hrs, CS informed AEMO of a possible turbine hall fire at Callide C.”
The Toshiba technology is simply a GE licensed design. And there are a lot of people with that license. And specifically, this turbine is simply rated for a temperature and pressure, and the forging capability would be the same as any other thermal (like solar thermal, or a garbage burner, or . . .) turbo-generator.
Well said Jeremy. The article actually says they are stopping coal fired power plant sales, not turbine sales. Toshiba used to do boilers also. There is a huge fleet of Toshiba machines out there that will need to be supported for many years to come, so bashing out another one is no issue. And if they won’t Doosan will – another GE licensee. The big issue will be lead time. If they have wound back their order book they wont have an inventory of big forgings to make shafts out of – and they need at least one from the bit lying on the floor. I suspect there will some very sad insurance underwriters just now. There are a lot of people commenting on this that really have no idea the step by step process to sorting this out and that it is going to take quite a long time. Having been directly involved in the rehabilitation of a lesser magnitude, near death experience with a machine many years ago I can but express my sympathies to the people involved and the no doubt sleepless nights and long hours this is causing them.
Paul Paul Paul, when you say ‘the transition has one or two decades to go”, you imply an end to The Transition!
We all know there is no end, because who in their right mind thinks we’ll be powered by intermittent dispersed low density low efficiency generation?
There is no “one or two decades of transition” with wind and solar – it’s a transition in perpetuity. Transition implies an end point. You’ve assumed 20yrs.
But how many wind turbine sites are refurbished with new turbines after 20yrs? How many solar farms? Batteries?
A sane transition would be from one technology to another technology that delivers improvements in the major facets of energy delivery:
– density
– efficiency
– cost
– footprint
– mining (number of holes in the ground)
– waste
– pollution
– manufacturing, refining, transport
– transmission
Wind and solar improve very few of these facets of energy delivery, and actually make some worse.
Ben, Ben, Ben, [kiddo], since you did not care to provide any numbers or analysis to support your assertions about what matters and how technologies compare, perhaps I can help you out. I’ll do one, then you do one.
Let’s take density/footprint, and compare Liddell [1GW] with the Forest Wind project [1.2GW]
Liddell itself, not accounting for the coal mines, etc takes up about 2 km square, which isn’t usable for any other purpose. Forest Wind’s footprint will consist of 226 30-m circular foundations, adding up to 0.16 km square. All access roads already exist. There will be temporary extra space required for construction, but that can go back to forest afterwards. See https://www.forestwind.com.au/project-overview Therefore this wind farm, conservatively, is FIFTEEN TIMES more footprint efficient than a coal fired thermoelectric power station. This is generally the case for wind farms – while they are spread over a large area, their footprint is small, and the ground space in between is generally fully usable for farming or conservation.
Your turn.
Jaz, that’s more spin than a wind turbine, which isn’t hard…😂.
Seriously though, there are serious problems being baked into the system, and very few signs of anybody arresting the decline.
Thanks Marcelle, handy diagram.
Callide B1 has come back online about 16.45 Wednesday and is slowly ramping up.
Thanks Allan for dissecting the elephant’s toe nails. This consumer advocate found it fascinating since 5 minute settlement is rapidly approaching. Volatility in wholesale prices can flow through to retail power bills. Since QEUN’s April 2021 survey of over 200 businesses in the Cairns region found over 60% had a medium concern about the payment of their power bills (20% had a very large concern), then are consumers already struggling with power bills about to struggle with fluctuating power bills caused by volatile wholesale market?
Fair question Jennifer. As you know wholesale energy costs are a proportion of consumer bills, and that proportion depends quite a bit on location (network costs), customer size, retail on-costs etc. Most customers won’t be directly exposed to short-term spot market volatility but trends in wholesale prices flow through to forward contracts and to retail pricing with a time lag – the length of this lag can be quite appreciable, again depending on customer specifics. The first post in this little series (https://wattclarity.com.au//articles/2021/06/eating-the-nems-q2-elephant-an-appetizer/) has some views of the impact of recent volatility on average and forward contract prices. For Queensland, average and forward prices are now well above where they were six or even twelve months ago, but still well below the levels prevailing over most of 2017 – 2019. The energy price component of consumer bills for many customers, eg those on tariffs set ~12 months ago would still reflect some impact of the high prices prevailing in that period so I don’t think the current volatility by itself should lead to a *jump* in that component, but obviously if it extends into subsequent quarters that could change. What’s probable is that is that signficant falls in that component which seemed on the cards just a few months ago are now less likely.
The photos of the site show a catastrophic failure starting at the HP end with the steam pushing the I
P stage back into the LP stage and alternator shearing the main shaft that was ultimately spat out. These turbines have multiple temperature and vibration sensors to shut them down if things go pear shaped. Do we know why the shutdown failed
Thanks Allan. If volatility increases then this could be priced into retail offers which could be more problematic for small retailers with small generation portfolios, this could impact retail competition. With the COVID19 pandemic continuing and no real idea as to when the international border will open, many businesses who are at present benefiting from lower wholesale/retail prices will have little ability to suddenly pay more if volatility is priced into retail offers in the future. Volatility is also enhanced by Jurisdictional Scheme Charges – a significant charge (eg 3.5%) that can without warning be imposed on consumers’ power bills. The COVID19 recovery of small business – the largest employer in Australia – is fragile eg Reserve Bank of Australia’s May 2021 charts has lending to small business as flat lining and business investment is continuing to fall despite the extension of generous tax write-offs (agriculture is bucking the trend). Therefore downward wholesale/retail prices are most welcome but a sudden spike or a consistent rise in wholesale prices would impede the COVID19 recovery. Since energy costs to over 700,000 business and residential customers in regional Queensland represent 35% of a power bill and 75% of the energy cost is the wholesale price, we are very interested to see if 5 minute settlement will increase volatility and therefore be priced into retail offers.
Hi people,
A BESS provides much faster response to system events then tradition gen sets, as there is more delay in goveners response and engine speed pick up etc.
A BESS will determine a frequency change in 10ms has it clocks the time a a voltage signal crossing the zero point, which occurs ever 10ms for 50Hz, plus a small processing delay, so realistically say around 20ms response. On the system we see response in under 200ms, which stops a blackout event. This seen on small systems, less than 10MW, using an ABB BESS.
Through the grapevine – in other words just a rumour. The destroyed unit did not have a self starting capability and so the generator could also function as a motor by feeding back off the grid. It has been suggested that a lubrication failure one end was fighting a generator that was acting as a motor due to a software/hardware failure allowing the grid to feed back into it.
The Network Outage Schedule from the AEMO website yesterday shows outages of the Callide C- Calvale 275 kV lines 853 and 854 and the associated circuit breakers at Calvale from 25th May until 7th June.
Correction: the Callide C/Calvale network outages are from 25th May until 2nd July.
So much for lots of little guys helping out with the demise of dispatchable coal generation-
https://reneweconomy.com.au/did-aemo-just-kill-off-the-market-for-battery-based-virtual-power-plants/
That’s the consensus Green dream and you can sense already they’ll see the AEMO’s stance as somewhat of a conspiracy by big generation and the regulator.
With solar FIT rates dropping continuously with a glut of rooftop solar and no forseeable economic case for home batteries there’s only one viable option for the little people going forward-
https://www.solarquotes.com.au/hot-water/pv-diverters/
They might also store solar energy in the home with the right gear-
https://www.solarquotes.com.au/blog/divert-solar-air-conditioning/
Nothing surer than the demise of coal coupled with all those prospective EVs charging at night peak grid power rates will be after sundown.
These winter peaks will only rise as more houses, and to a lesser extent commercial buildings, switch to reverse cycle air conditioning for their primary source of heating. This will be most obvious in Victoria.
Equally important is the completion date of the 800 MW EnergyConnect between NSW and SA.
It seems that within a 12M period the following occurs:
– NSW loses 1000 MW of 24/7 baseload from Liddell
– SA loses 600 MW (?) of dispatchable gas generation from Torrens
– NSW sees 800 MW of new load from SA via EnergyConnect
With NSW the most connected pricing region in the NEM, any price increases in NSW will spread to the other states.
The future looks good for NSW coal, gas and hydro in the medium term.
Longer term it’s likely a few more semi-scheduled RE power stations get built as wholesale prices are buoyed by gas and hydro (note the recent government intervention in these technologies). But I’m still doubtful too many will be built without state government contracts.
The growth of rooftop solar will certainly suppress daytime prices until FCAS costs overtake wholesale costs.
That clearly sets the scene for batteries to make money from arbitrage and FFR, hence the interest and hype.
Interestingly the market for batteries and pumped hydro pretty much destroys any market for green hydrogen production using low-cost ‘spilled’ power.
If a generator is mothballed, can the Notice of Closure of 42 months include the time period the generator is mothballed?
Given the extent of the damage even that is optimistic. Big question why did the unit’s many safety features not shut it down. The only explanation i can offer is that the high pressure inlet valve failed. Has anyone got more information or a better theory
Hard to comprehend why they’d bother fitting a new turbine when coal has to go. Unless of course transitioning gas is really setting power prices and they’re expected to keep on rising.
The value of electricity always lies with demander/consumer. Up to any free entrant supplier to decide if they can compete with other like suppliers to meet that value with whatever costs they face. But the complexity lies in sum of consumer values at different times and likewise the costs of the various suppliers at those same times including frequency and voltage of said power of course.
Too much autocorrelation of values between consumers and costs between suppliers within each sector but they’re way out of sync with each other and hence those volatile prices reflecting that.
Does this address Villian #8 for both scheduled and semi-generators?
4.1.1 Publishing individual unit available capacity and PASA availability
The proposed rule includes a new requirement for AEMO to publish the available capacity and PASA availability of individual scheduled and semi-scheduled generating units20, scheduled loads, scheduled network services and wholesale demand response units.
It clearly opens up opportunities for retiming hot water, demand response and thermal storage if not batteries. Wind developers will have to think about areas that are out of sync with the bulk of existing wind farms rather than maximum annual energy production Gippsland comes to mind, offfhsore or nearshore in southern NSW. Some existing wind farms may be repowered with taller towers and larger rotors without increasing nominal capacity but significantly increasing output in low wind conditions.
The 5 minute rule may also reduce the incentive to game the system at the high price end whether it will help at low prices is another mater.
Both C3 and C4 have some shared and/or linked systems (fire systems among others). Fully segregating those systems so that C4 can have carte blanche for the next 18 months is never easy . . . essentially going back to a state last properly seen in the late 90s, when the two units were being built.
Thanks for the clarity, Jeremy! I’ve never actually been onsite at Callide.
Gullen Range, Collector, Biala, and Crookwell 2 wind farms all dove to zero at the start of the price spike. So did Gullen Solar Farm. A total of about 300 MW whilst there was a NEGRES constraint on the VIC-NSW interconnector. Nearby wind farms Cullerin Range and Gunning did not change their output. The DUIDs that went off line are all on a pair of 330 kV Transgrid transmission lines to/from Yass BSP. Maybe something happened at Yass BSP?
Might help explain?
https://wattclarity.com.au//articles/2021/07/14july-nsw-morning-spike-explore/
Same thing just happened again at 18:00 on Thursday 16 July, only this time the price spike happened in NSW and also in QLD. Again those same windfarms went to zero watts.
…and another huge swing on the VIC-NSW interconnector
Another dip for the NSW southern highlands wind farms at 19:00 this evening. 500 MW in total. Is Transgrid testing SmartWires as part of the VNI upgrade? Has it something to do with the commissioning of Stockdill BSP?
So how much will we need to flatten out the bumps and fill in the slumps?
https://www.energetics.com.au/insights/knowledge-centres/large-scale-storage-tracker
This is something we’re 90% complete in directly exploring for an Appendix in GenInsights21, as noted here:
https://wattclarity.com.au//articles/2021/04/announcing-development-of-geninsights21-preorder-open/
… hint = it’s a lot!
“ We need a federal package to help rollout this infrastructure because nobody wants to handle the risk. Really this is the only successful way we’ve done nation building infrastructure projects before. All the historical generation zones relied on government funding to build out several redundant lines at once. The model used in the past has been to build it and socialise the costs later once the benefits are established. The states can’t do it alone, and most of the value of new transmission is diffused across the entire NEM not just in local pockets.”
Tom, I enjoyed this write up, it is largely measured and sane. However the paragraph quoted above misses a couple of important points.
You mentioned risk, and later you mentioned ‘white elephants’. But transmission infrastructure receive guaranteed returns based on the value of the asset, regardless of its utilisation. The only risk appears to be changes to the regulated rate of return.
You mentioned that past practise has been for governments to build the transmission and socialise the cost. That was true when it was necessary for governments to build this infrastructure. That was the case as electricity demand was increasing and remote locations were connected to the grid. But in most regions the industry is now privatised, and electricity demand is not increasing.
You mentioned ‘nation building’. It is incorrect to label transmission projects, that do not connect new locations or large projects as ‘nation building’. Expanding the size and cost of infrastructure, for no increased demand, lowers the utilisation rate of that infrastructure – this is a net reduction in efficiency of the overall system. Reliability and availability can reduce utilisation, but that’s far removed from building infrastructure purely to connect a preferred (by some) technology for reasons unrelated to supplying electricity (emissions reduction).
I agree on the direction of your analysis, which argues for an multi-faceted assessment of transmissions n investment. Good job.
Orchestrated criticism of anything that might delay renewable nirvana? No way!
The fact that an organisation titled ‘Energy Security Board’ even exists is a living example of how the market has been manipulated to achieve poor outcomes.
The options paper described how the market redesign would incorporate subsidies and targets for renewables, as if this is now ‘the new normal’… a phrase I’m beginning to dislike!
I cannot see how the NEM can achieve the best outcome for customers when the market design is becoming more and more a central planning behemoth, catering to lobby groups and political whimsy.
Security – as if.
I think Ben underestimates the professionalism of the team that Audrey Zibelman recruited before she handed over to another proven performer in clean Energy implementation.
Nothing too surprising here. Metal fires are extremely difficult to extinguish. They can’t be smothered because they produce their own oxygen and hence self sustain. All you can do is either deluge them with enough water to cool them to a point where the temperatures fall below that necessary for auto ignition (a pretty big ask for such a densely packed module) or let it burn out – which is what is happening here. Simple message – real estate is cheap in Australia – put space around each module to prevent spread from the radiant heat. So next time someone suggests putting one of these things inside your house…..show them the photo. If something (anything) goes wrong simply the energy release is rapid and disproportionately high compared with usual combustible materials and your house will catch fire. By all means have your battery, just don’t put it in or near your (or anyone else’s) house.
I wonder how many investors are backing away from renewable investments…
Regarding the home battery situation, I recall that a draft Australian standard recommended home batteries be installed in a bunker style housing, separate to the house. The RE lobby got that removed.
On these batteries, in general the strategy will be to let the bad one burn out and try to keep the rest cool. Makes you think about the risk of having them scattered throughout the distribution network, in suburbs…
It would appear that fire barriers between the individual batteries may be worth considering. From the photographs it appears that only one battery caught fire and then damaged the adjacent unit. Substation transformers present the same problem, they can occasionally catch fire or even explode. As a result they are separated by blast walls so that the damage cannot spread to adjacent units
Tesla will figure out what went wrong and then fix it.
There is a bit tension as to the safer form of lithium battery technology-
https://www.energy-storage.news/news/lfp-vs-nmc-not-a-silver-bullet-solution-for-safety
Although fixed battery providers like Redflow may have a point that if transport is to be batteryfied it won’t make much sense to use up light weight lithium for fixed battery backup for the grid. OTOH increasing demand for lithium resources should make that clear fairly shortly via the price mechanism with the rush to EVs.
Medupi: If air was admitted whilst there was still hydrogen in the alternator then it was a pressure vessel and thus an internal explosion has occurred. Hence the explosive damage to the alternator.
Callide: The unit was in service, thus it would have contained pure hydrogen. Failure of the seal oil system would have allowed an external explosion when the escaping hydrogen mixed with air. Hence the alternator in the photo is intact but shows signs of burning at both ends.
Medupi: The Damage is confined to the Generator and surroundings. The end covers are blown OFF and Excitation Gear is damaged but since the unit was OFF LINE and stationery there appears to be little damage to the LP Turbines and the structure of the Turbogenerator set is largely intact. The generator is probably fixable because it was not running. There is quite a bit of Building damage, more than at Callide C4, mainly because the Generator is approximately twice the size.
In comparison with Callide C4 where it is clear that the shaft is destroyed and the turbines severely damaged seems to indicate that the Callide unit had a catastrophic Turbine failure followed by a hydrogen explosion after the Generator Casing was ruptured.
The AEMO Incident report on Callide C4 makes interesting reading, particularly Section 3, page 10, in summary:
– prior to failure Callide C4 was at 278 MW
– 13:34 it ceased generating but did NOT disconnect
– 13:40 AEMO were advised by Callide of a possible turbine fire
– 13:44 Callide C4 absorbing 50 MW and 300 MVars (275Kv voltage healthy)
– 14:06 Multiple 275Kv CB’s opened causing large blackouts
https://aemo.com.au/-/media/files/electricity/nem/market_notices_and_events/power_system_incident_reports/2021/preliminary-report–trip-of-multiple-generators-and-lines-in-queensland-and-associated-underfrequenc.pdf
The problem was they basically ran out of capacity – everything that was available was on and full load – the only plant left to come on was a cold coal fired unit or a cold CCGT. Can’t start those up in the hour’s warning.
The generation data shows the windpower significantly dropped in the period 1500 to 16:30. That and weed blocking the screens of a hydro just after 18:05 seemed to have tipped it over the edge.
Was there a hydrogen explosion at Callide at all? Or did the GE of the LP turbine destroy itself during the motoring, possibly without lube oil? The damage done is not what one would expect from a hydrogen fire
Probably a lube oil fire and hydrogen explosion due to leaks following turbine damage. From the damage at each end of the LP cylinder and holes in the roof, I would have said an it was an overspeed causing the turbine last stage blades to fail but there don’t seem to be holes in the LP casing. Maybe there was a loss of lube oil while the generator remained synchronized as you have suggested and this caused the turbine to want to stop while the generator wanted to keep it going and this led to the shaft rupturing?
Since the mid 1990’s NZ population has grown from 3.3M to today’s 5.2M. There has been no new generating capacity, while Watermelon politicians press for greater intermittency of supply and demand the populous drive a ‘trick car or walk (except them), and the NIMBYs block the development of new generating resources. 60% of NZ power (once all domestic consumption) derived from renewables, chiefly hyrdo. NZ has UNEP divested itself of gas and coal investment and exploration, and is currently bearing down on one of its main sources of revenue, agriculture. The virtue signalling Watermelons doom NZ to a future of division, destitution, depopulation, deindustrialisation, delinquency, despair … and hunger.
NZ wind generation accounts for an intermittent, unreliable meager 5%.
Have you sent an email to AEMO?
Yes, we have had a conversation, and AEMO is investigating.
Hey Paul – thanks for this. Do you know (or can you explain the difference between) “Operational Demand” and “Demand and nonschedgen”. I actually thought they were analogues of each other – but checking out the data, that doesn’t appear to be the case! Cheers, Mick
Mick – there are many wrinkles to demand definitions, even the ones AEMO uses. If you hunt around you might find exactly how AEMO measures Operational Demand. I think the biggest difference between Operational Demand and the MMS DEMAND_AND_NONSCHEDGEN field is that the latter includes supply to Scheduled Loads (pumps and batteries) but Operational Demand excludes them. Allan.
This W/C article has some good references to AEMO’s definitions towards the end:
https://wattclarity.com.au//articles/2018/04/an-explainer-about-electricity-demand-take-1/
Thanks for that Allan – Make sense. Much appreciated!
The published information on the motoring shows firstly that the machine did not overspeed and secondly, there was major resistance to the turning at grid speed. It is very unlikely it could have motored if there had been a hydrogen explosion. And it seems to have motored until the line was disconnected at the substation.
There is other information around on primary and secondary causes but nothing formal.
All that has been published so far hints at operation as an induction motor. Unofficial speculation hints at loss of steam, excitation, lubricating oil, and power to protection systems occuring simultaneously.
Windage loss will heat the turbine, circulating currents heat the generator, and loss of oil destroys the bearings.
Next week: “Trapezoid of troubles”
Some big issues coming, no doubt. I wonder at the priorities of the ‘influential’ people, and what’s more important to them – more RE no matter what, or a working system… because we are fast approaching disaster.
We currently have an energy ‘security’ board – because clearly the market rule maker (AEMC), rule regulator (AER) and operator (AEMO) aren’t capable of maintaining the ‘security’ of the system.
It’s not that difficult.
1. Restrict the semi-scheduled category to plant below 30 MW nameplate capacity.
2. Limit PPAs to 20% of a plant’s annual output.
3. Constrain rooftop solar to consumption only (unless a battery is configured to complement the grid)
Some great charts there Carl – thanks!
One possible reason for the delay in coming up with something clear and meaningful is that the task they have set for themselves, or the task that they are expected to perform, is impossible.
They are supposed to come up with a plan to proceed towards net zero with all possible speed by building more RE capacity plus [think of a number] GW of dispatchable firming capacity from batteries, gas and pumped hydro. Can we get enough gas? believe pumped hydro when you see it and forget about batteries for getting through windless nights (do the arithmetic).
We are supposed to end up with less emissions, cheaper power and a more reliable grid. Tell that to the Germans.
And don’t mention the environmental impact, the impending avalanche of toxic waste and the extra reliable power required for the great leap forward into electric vehicles.
But don’t laugh, this is serious!
Coal has to go, one way or another, although the Government it also has to stay, one way or another, as long as necessary. How long as that?
With the right Energy Policy in place, I don’t care whether Generators elect to use; wind, water wheel, tidal power, solar-PV, Bio-fuels, fossil fuels, batteries, nuclear, or even pedal power, if that’s what turns them on.
Provided they don’t expect to receive a subsidy from taxpayers to underwrite their fantasy.
In brief, I’m of the view that all Australia, and or any other country, needs is a sensible Energy Policy in keeping with market forces that;
• Is technology agnostic;
• Has clearly defined QOS standards eg 24/7 power available, reliable & affordable;
• Facilitates / encourages strong competition between generators to deliver their best price per GWh for guaranteed power output eg 100% (force majeure permitting);
• Provides no subsidies & repeals past legislation that provides for same (eg in Australia the RET, LRET, Safeguard Mechanism); and
• Importantly, imposes substantial financial penalties for any failure to meet contracted QOS obligations for an essential service.
Easy.
What was keeping the lights burning in SA and Victoria on Sunday evening?
Hi Paul: Two years ago I prepared a report for a State Government and highlighted the issue of what I called strategic energy capacity when %VRE exceeds ~ 90%. Also the fact that wind energy provides greater value than solar energy because it generates after dark – but still there are times of low supply. The 200 page report was about challenges for the power system transitioning to 2050.
The other thing I pointed out (relating to this article) is is that economic transmission capacity will still provide significant constraints in Australia with a large network (km) to be able to share VRE capacity across the distances. Simply the fact that 1000km of DDCT 330kV transmission line costs over $2.5Bn (800MW per circuit). Even if you take $2.5Bn/1600MW the unit cost is about $1.5M per MW. Hence you can build local wind and solar excess capacity at less cost. Strategic energy capacity would also be less than this taking into account it is say 96% reliable compared with remote generation from distant VRE which may also be low hence less reliable – your data will provide me some insight into this level of reliability over the quarter for consideration of this future issue.
Many thanks for such great articles in WattClarity!!
On the wind output record, that might require some pretty high demand to turn up – otherwise output will be limited by local demand + export capacity less required synchronous generation, as well as wind self-curtailing on negative spot prices. So the record may last a few months yet.
will be interesting to see to what extent directions on GPG reduce. I’m interested to note that “no changes to the SA generator combinations will be made at this stage”. This implies that there won’t be a significant reduction at this point, or have i misunderstood?
And if so, is this just because AEMO/electranet are in testing phase and changes will be made in due course when testing is complete?
Because otherwise the syncons don’t seem like great value…
The plan as announced is to allow the network to run with the machines for several weeks to monitor, then make adjustments to constraints (you are here) and monitor performance for a couple of weeks then repeat until fully commissioned. Updating the minimum synchronous generation constraints is the final stage of the plan.
AEMO’s program (here) is to reduce the minimum synchronous requirements (to 2 x TIPS B or equivalent) about 6-12 weeks after commissioning of the first two syncons, which occurred a couple of weeks ago. So all going well in this interim period, we’ll see reduced gas directions in around 4-8 weeks.
This change is Stage 2 of a 3 stage process which is described in AEMO Market Notice 89336. Generator combinations will be changed in Stage 3.
An easy-to-miss rider in the AEMO notice: the current limit will apply for Davenport synchronous condenser or 275kV line outages – there’s one such line outage running until the evening of Sep 14 so the old limit effectively remains in place until then.
Hi paul
Great work as usual. Suggest you may want to do a piece soon comparing how well different ageing plants are going at developing more flexibility under these conditions. In particular which ones are successfully getting to the lowest mingens for extended periods.
On 16 December 2020 Watt Clarity with the permission of David Leitch and Ben Willacy reproduced their article “Wind and Solar transactions: what to buy in the new year sales”. The article contained a table which specified the type of revenue (Power Purchase Agreement or merchant). It further stated the remaining lifetime of the contract based on a notional life of 25 years for wind and solar farms. Would be great to see this table updated together with what public companies are reporting as their asset life. One of the major investors in Australia in its Interim 2021 Results on 30 July 2021 reported “Depreciation and amortization expense was almost stable (up €0.2 million), as the mechanical increase reflecting growth in assets in operation was offset by the positive impact of extending the useful life of wind and solar farms from 25 to 30 years. Are other public companies reporting a 30 year asset life? AEMO use an economic life of 25 years (period of time required for the project to recover build costs) and a technical life of 30 years (period of time the project is operational before life extension refurbishment or retirement).In AEMO’s 2020 Inputs and Assumptions Workbook the retirement costs of single axis tracking solar farm was $21,7366/MW and a wind farm $10,883/MW (real 2020 $). AEMO did not specify retirement costs for solar and wind farms in its 2021 Inputs and Assumptions Workbook but the workbook did update retirement costs for coal and gas generators. The first wind farm in Queensland was built in 2020 and AEMO has it retiring in 2025 – its a baby compared to today’s wind farms ie 20 turbines with a capacity of 12 MW.
Why does MT PASA only appear to apply to generation plant and not to transmission? Many of the times of high prices are because of interstate transmission outages, much of which is planned months ahead but doesn’t seem to be visible to market partipants. There are also intrastate transmission limitations that restrict how some dispatchable plant can bid or contribute to a capacity market. Until Humelink is built there are limitations on how Lower Tumut (1800 MW), Uranquinty (664 MW) and the Maoneng battery (200 MW) can access the Sydney market. How would a capacity market allocate the limited transmission capacity to each of these generators?
The Network Outage Schedule is available on the AEMO website covering a period of 13 months in advance. Search on the AEMO website for “Network Outage Schedule” and “Planned Electricity Network Outages”. The MT (and ST) PASA processes use a model of the entire NEM including generation, demand and transmission. AEMO uses information from the Network Outage Schedule as an input to the PASA process.
On the flip-side of this article, what is the forecast peak load for SA this coming summer? Noting peak during a heat wave will happen at about 7pm when solar is gone and it is potentially still and calm weather across SA. Without NPS, Playford and now all of TIPSA (so minus some 1260 MW), does SA have enough conventional generation plus interconnection capacity (assuming there are no problems elsewhere like in Western Vic and something can come through from Murraylink)? Has any shortfall been flagged by AEMO?
Since most wind and solar power stations are contracted, and any uncontracted gets REC, I expect the headroom necessary to increase output on demand would usually cost more than the available FCAS income. The lost income from capacity not earning will add up over time. Just ask any coal-fired power generator…
Great info, Paul.
“Trying to understand how the current rules are legal in a country with strong consumer protection laws does my head in.” In theory only one extra regulations should be required as the existing consumer laws do cover all these aspects. The problem is the need for the laws to be enforced. The ACCC typically won’t get involved in small or minor cases. This is supposed to be handled by the under resourced state consumer affairs bodies. Solar PV systems are a complex system with variable output depending on the conditions. This makes it difficult for an educated customer determine if a system is performing as specified. The vast majority of customers don’t fall into that category.
The one area that current laws don’t cover is the phoenixing of businesses or operators.
Even even 20-storey blocks of home units are turning up with fatal construction faults. It is stretching credibility to say that we live in “a country with strong consumer protection laws”.
This seems to me to be eminently sensible that all power plants including wind and solar should run with a bit of headroom to cope with system disturbances. On a 5 minute dispatch cycle the headroom could be provided by capacitors or flywheels at much lower cost than batteries. That would be probably be more financially acceptable than limiting production. Perhaps Synthetic Inertia or simply overrating generator capacity compared to the connection will also help.
If all generators do have some near instantaneous reserve then the cost will be widely spread and therefore have little impact o any generator
These changes are long overdue and most welcome. However, the predatory signing up practices are as big an issue as the quality of a solar system. Consumers also need to know that solar systems are not set and forget ie all solar systems need regular maintenance in a way that does not void their warranty… this should be explained before a customer is signed up.
And directly related to this article, what happens when SA eventually occasionally/more regularly drops into negative demand territory due to the amount of non-scheduled embedded solar PV and then SA islands? How is SA run by AEMO as a stable island?
great note. Thanks.
May be interesting to split it up by DUID/Owner/Fuel type
It is well overdue that someone publicly identifies that the option that “states go it alone” is a real possibility.
One of the 49 recommendations accepted as part of the 2017 Dr Finkel led Review into the Future Security of the National Electricity Market – Blueprint for the Future was Recommendation 7.3 which stipulated that:
By mid-2018, COAG leaders should agree to a new Australian Energy Market Agreement that recommits all parties to:
• Taking a nationally consistent approach to energy policy that recognises Australia’s
commitment in Paris to reduce emissions and governments’ commitment to align efforts
to meet this target with energy market frameworks.
• Notifying the COAG Energy Council if they propose to take a unilateral action that falls
within the scope of Australian Energy Market Agreement prior to taking the action.
• Within 28 days of notification, the Energy Security Board will provide advice to the
COAG Energy Council on the impacts of the proposed action taking into account the
objectives of Australian Energy Market Agreement.
Realise COAG Energy Council has been rebadged but it is effectively the same entity without the open communication to stakeholders. Maybe COAG members are using the rebadging as an excuse for missing by 3 years the mid-2018 timeframe for agreeing to a new Australian Energy Market Agreement. States going it alone is a real possibility.
This is a very thorough report almost as thorough as the ISP and coming to similar conclusions particularly if one uses the ISP step change scenario, which is actually my base case.
When ITK started formally forecasting electricity prices a few years ago and said our modelling was different to other approaches in that we just assumed there would be enough policy to get to 50% renewable by 2030 it wasn’t taken all that seriously. And our modelling probably still isn’t although I take it seriously myself. But the point is 50% renewable by 2030 (which includes hydro) is now pretty much base case.
The forecasts in these models are done to meet the reliability standard, and yet they produce reasonable electricity prices and none of them suggest you need nuclear or anything similar. That is one reason why I am so supportive of Matt Kean and the NSW approach. It is based on sound modelling and a proper business and execution approach. And it is done in the open with feedback and consultation at every stage.
Of course as every epidemiologist and many others will point out, modelling is a very error prone process and any good policy maker will not rely solely on models . Still it has to be done and its great to see this effort from Transgrid.
The problem with optimistic projections of RE penetration is that we are only just approaching the point where it matters whether the sun shines or the wind blows. When we lose some more conventional capacity the problem of windless nights will become real and we will have to confront the full cost of firming that has been carried by the coal power stations up to date. See the story of the frog and the scorpion!
https://newcatallaxy.blog/2021/09/25/the-real-cost-of-unreliable-energy/
Is any work being done by anyone to determine how many MW/MW.h of batteries can be connected in e.g. SA before they eat their own arbitrage lunch, eradicating the very pool price volatility that drove their investment decision in the first place i.e. through there being too many connected all trying to draw a load when there is no load (and when pool prices are/were generally lower) and conversely all trying to be generators when generation is scarce (and pool prices are/were generally higher)?
Didn’t notice the power filter function! By the way, using all volumes, effective as-bid availability showed the change in bid stack from solar and wind plants in a clear way too.
Nice that this article helped you discover that one, Adrian. Remember the individualised ‘Widget Guide’ accessible under the ‘?’ button in the top right of the widget might provide other ideas for how to use that widget!
Not so wild in SA it turned out. High prices only due to lack of reserve in QLD in the afternoon. Perhaps some maintenance work was delayed by weather?
Great idea. Any options for a video link for those trapped in the Garden State?
Did they start synthesizing 5 minute offers from 30 minutes, or start accepting 5 minute offers around these times? Are the sensitivities part of the same file, or separate? I believe they are incorporating FSIP into the P5min but I’m not sure that has occurred as yet – would require an additional solve per interval.
Nothing from AEMO yet?
Alas have not really progressed much, in understanding what’s going on, HR10 …
Any idea how much the QLD tax payers lost on Saturday due to the poor management of the operation of the QLD Government owned generators?
Without knowing how much of that power is under contract it is difficult to determine, but operating at negative $1,000 per MWh for hours at a time must have cost millions of dollars.
It is actually a very clear case of market power abuse, with the near complete removal of all competing generators priced out of the market.
We knew that the intermittent providers can displace coal but they can’t replace it. SA was importing coal power until after breakfast and it is importing now, thanks to the efforts of the much maligned brown coalers that are running over 90% of capacity while the wind is delivering 2% of Victorian generation.
Exciting times!
Would it not be more profitable for the State-owned generation companies to shut down one or more coal generation units until demand picks up in mid summer, and sell the coal on the spot market, which is now extremely high?
What is the point of achieving that amount of penetration for a short time on the weekend? When the sun goes down and the wind fails, as it often does for hours at a time, SA depends on local gas and coal power from Victoria.
A chain is only as strong as its weakest link and the effectiveness of a flood protection barrier is limited by the lowest point, regardless of the high parts. Windless nights are the weak link in the chain, like the gap in the levee.
How hard is that to grasp? Don’t mention pumped hydro storage, when Snowy2.0 comes up at immense financial and environmental cost it will only partly replace Liddell. Big batteries to cover windless nights? You have got to be joking!
“How hard is that to grasp?”
Well if you have the obvious answers Rafe there’s plenty of well paid opportunities going begging it seems-
https://reneweconomy.com.au/architect-of-integrated-system-plan-to-leave-aemo-as-staff-exodus-continues/
“Don’t mention pumped hydro storage”
There’s not much worth mentioning in SA anyway-
https://reneweconomy.com.au/one-pumped-hydro-project-still-standing-as-south-australia-heads-to-100pct-wind-and-solar/
Rafe, I don’t think the flood protection barrier analogy works. In that case, if the flood exceeds the lowest parts of the levee, everything on the other side is compromised: a literal tipping point. So, as you say, the only thing that matters is how high the lowest points are.
That doesn’t apply here. Occasionally resorting to gas or even coal is still a huge improvement over constantly using it. “What is the point?” Not burning all that gas when the wind *is* blowing or the sun *is* shining. That’s the point.
Your analogy would only make sense if the entire system were to somehow collapse if gas or coal were ever used, which isn’t the case.
More importantly, what has it cost us in higher electricity costs by not implementing 5 minute settlement earlier? I don’t just mean the delayed implementation, I mean that it should have been brought in at least 5 years earlier than that.
We should be looking at what it is costing us by not moving to 1 minute settlement. I don’t think we are there yet, but it needs to be watched. The NEM grid now has the ability to convert almost all of these 5 minute price spikes, to 1 minute price spikes with grid scale batteries.
I note the AEMC are doing their best to deal with the big picture problem-
https://www.msn.com/en-au/money/news/kill-the-viability-big-batteries-to-lose-out-from-electricity-grid-rule-change/ar-AARgprt
Essentially with RE we should have started out with a level playing field for all tenderers of electrons to the communal grid. Namely reasonably guarantee them 24/7/365 along with FCAS (ie short of unforeseen mechanical breakdown) or you can keep them and use them yourself.
That only leaves the political bunfight over carbon taxing penalties for fossil fuels but now we have the worst of both worlds the AEMC has to grapple with. What’s more with the switch to EVs for transport we’re seeing scarce light-weight lithium battery resources being squandered on fixed batteries for temporary grid firming as that’s where the cream is. I wish the Regulators well with it all when we really need rational statesmanship.
way to go Paul. Good effort focussing.
Fortunately most very low wind events happen during the day when as you say solar is increasingly covering them. There are some occurring at night which leave a 15 GW gap between hydro + wind and net demand so even if wind is quadrupled over the next 10 years and 3-4 GW of pumped hydro comes on line there is still a fair gap to fill. However moving a lot of non time sensitive demand from nighttime offpeak tariffs which were designed to keep coal plants going to daytime solar sponge tariffs will fill a substantial part of that gap
The era of fixed times for non-time sensitive loads should be dead and buried. The cost of smart control systems are low now and we should be taking advantage of this. This means a much better match between production of RE and loads.
The problem of low-wind periods is never going to go away because when the CP is down to 2 or 3% that is practically zero compared with the demand in the grid regardless of installed capacity. This was clear years ago, as soon as we knew that the 2012 study of wind droughts across the NEM published by Paul Miskelly was not going to be superseded by more recent data. Prolonged wind droughts can occur between sunset and sunrise as well and it doesn’t matter if these events are rare, because until we have several more Snowy schemes or ample nuclear power the lights will go out bigtime sooner or later, and how often do we want that to happen?
The crux of the problem is the “gap in the flood protection wall” effect. Build the high parts of the wall higher and increase the average height but the water will still come in through the gaps.
https://www.riteon.org.au/netzero-casualties/#217
The AER rule change request posted by AEMC today (22 December 2021) may assist to shine a light on why Queensland is seeing such insane wholesale prices compared to the rest of the NEM eg the year-to-date FY22 wholesale electricity price in Queensland is $95.88/MWh which is nearly 50% more than NSW at $65.13 and 3 times more than Tasmania at $27.68/MWh.
The AER’s usual report on individual high price events is no longer fit for purpose and this has been identified by the AER in its point 2 below “The timing and format of reporting, to enable the AER to look at prices more holistically e.g. grouping prices in where there is an observed behavioural trend.”
Posted by AEMC today:
Dear Ms Collyer
Re: Rule change proposal – AER reporting under clause 3.13.7 of the National
Electricity Rules
Please find attached a rule change request proposing amendments to clause 3.13.7 of the
National Electricity Rules (the rules), which relate to AER reporting obligations on significant
price variations and high price events in the National Electricity Market (NEM).
Currently, clause 3.13.7 of the rules places a number of wholesale market reporting
requirements on the AER:
• Reporting quarterly on ‘significant price variations’.
• Reporting on any market condition the AEMC or ACCC asks the AER to report on, if the
AER agrees.
• Reporting on energy 30-minute prices above $5,000/MWh.
• Reporting on ancillary service 30-minute prices above $5,000/MW.
While it is important that the rules require the AER to analyse and report on significant
wholesale price outcomes in the NEM, the AER considers that the current requirements are
overly prescriptive and no longer fit for purpose.
The AER’s proposal is to replace the current rule with a principles-based reporting
framework, supported by AER guidance. The intent is to replace the current clause 3.13.7
with a new, less prescriptive framework for reporting on significant price outcomes. This
would provide flexibility on:
• The prices the AER will report on, to enable the AER to adjust to changing market
conditions without a further rule change.
• The timing and format of reporting, to enable the AER to look at prices more holistically
e.g. grouping prices in where there is an observed behavioural trend.
• The factors the AER must consider in the reports, to tailor reporting to the particulars of
the price events.
Increasing flexibility in this way would ensure the AER’s reporting provides more insightful
analysis into market outcomes and behaviour, supporting the achievement of the National
Electricity Objective.
The AER requests that the AEMC expedite this proposed rule change. The AER considers
this proposed rule change is non-controversial, as it relates specifically to AER reporting
requirements and it is unlikely to have a significant effect on the wholesale electricity market
or the interconnected national electricity system.
Finally, I note the work done at a staff level between the AER and AEMC including meetings
and guidance on the principles-based approach to drafting the amendments. I would like to
thank the AEMC staff for their assistance in developing this rule change proposal.
For any further information please contact Scott Johnston on 08 8213 3426.
Yours sincerely
Justin Oliver
Board Member
Australian Energy Regulator
Sent by email on: 21.12.2021
Do we know any more details from the in-house report? Curious to know the chain of events that led to catastrophe and how other plant can avoid them.
Very Good Article on very note worthy topic.
The statement that 22,000MW of coal fired power will be closed down by 2050 strikes me as not terribly remarkable given that probably all of these stations would be at least 10, and more likely 20 years, old. The usual life of such plant is 40 or sometimes 50 years so little to remark on there i.e. I don’t see it as a collapse so much as regular replacement. Its just that in the current circumstances they will be replaced by non-coal and probably non-fossil fuelled plant. The task is to ensure that the replacement plant is reliable 24/7.
“AEMO should provide more explanation as to why thermal coal plants are not assumed to operate more flexibly”
Yes the EU and UK certainly got more flexible with their coal generators this winter and nuclear along with gas got a whole greener in outlook as well. Flexibility is certainly in vogue at present.
BNGSF1 came back around 15:00, after a short run at ~14:30 while BNGSF2 was around 17:00. I can not see a market notice for non conformance, so I assume the action was either intentional by the plant operator, or requested by AEMO/TNSP.
Queensland has now cracked $100/MWh as the year-to-date 2021-22 average wholesale electricity price.
To put this in perspective Queensland is around a 50% premium on the NSW wholesale price of $67.94/MWh and nearly 3 times the wholesale price of Victoria and Tasmania ($38.77 and $37.00).
As a consumer advocate we know this is disastrous for small business which are the biggest consumer of grid supplied electricity and the largest employer of Aussies (nearly 5 million Aussies). Affordable electricity is one of the keys to Australia’s Covid19 recovery. Main stream media hasn’t even acknowledged this sad milestone.
I’ll be provocative here Paul and say that your 86% / 14% split is very creative accounting. It’s capacity that counts, not number of units. On that measure something more like 75% / 25% which is not great for the height of summer.
Totally true, Allan … but the #units calc was just quicker to do in the process of a busy day!
Why is direct link sending energy into NSW? Sometimes these constraints create what appear to be very strange outcomes.
https://nemweb.com.au/Reports/Current/Market_Notice/NEMITWEB1_MKTNOTICE_20220201.R94395
An Actual LOR2 condition has been declared under clause 4.8.4(b) of the National Electricity Rules for the QLD region from 1700 hrs.
The forecast capacity reserve requirement is 443 MW.
The minimum capacity reserve available is 77 MW.
Ok. This pains me, but you need to add the GT26, most thermally efficient major generation station absence generating 325MW over the last two days too. I cannot bring myself to name the station. What is the state government trying to do. Get themselves voted out on a Eastern seaboard blackout?
‘It’s all happening….’
These would be the management costs they speak of would they?
https://www.msn.com/en-au/money/markets/aemo-calls-for-funding-hike-as-renewable-energy-levels-send-management-costs-soaring/ar-AATkOir
Can you do a similar explainer about why so much energy from NSW is being exported into Victoria against a price gradient? For example at 16:10 NEM time on Sunday 6 February 624 MW was exported from NSW with a high price to Vic with a low price (according to the AEMO website). This is now occurring quite often, and it’s a much larger quantity than we would expect than just from the far western NSW solar farms.
Good observations Malcolm. This is almost certainly to do with the N^^N_NIL_3 or X5 constraint that featured in a couple of articles here a while ago (https://wattclarity.com.au//articles/2020/12/casestudy-followup-x5-constraint/).
Essentially that constraint can choose between curtailing generation in SW NSW by 1 MW or swinging flows between Vic and NSW by 10 MW as long as the price difference is less than about $100/MWh. So we see plenty of situations like the one you highlighted.
But I agree it would probably be useful to summarise the impacts it has been having, and why, in a follow-up piece sometime, now that it’s been in operation for a while.
Allan.
Is there some publicly available way to view compliance with a Negawatt target?
Hi MK,
Unfortunately, to my knowledge, no. The official calculation of the response – actual consumption relative to baseline consumption – is done for settlement after the event so it would be private to the Participant.
A good question, MK
Further to Linton’s comment, this was one of the concerns (there were a number!) that I raised when this Rule Change was being considered:
1) On 5th March 2019:
https://wattclarity.com.au//articles/2019/03/some-thoughts-about-demand-response-in-parallel-with-aemc-deliberations/
2) On 21st October 2019:
https://wattclarity.com.au//articles/2019/10/all-aboard-the-negawatt-express/
In short, transparency in the NEM has been reduced as a result of this rule being implemented.
Paul
I imagine the original uptick was coincident with P5 sensitivities, although I would have thought they would be run separately… On a side note these were paused for a period because the demand scenarios were being applied incorrectly.
A couple of weeks ago there were some quite bad performance impacts – some PD runs were missed entirely! I’m not sure how actively AEMO monitors – it persisted for some time until reported, worryingly. They investigated and things improved.
I did have a look at the # of binding constraints and didn’t appear to see much correlation (indeed there didn’t seem to be that many in the first place).
I know that were going to add FSIP into the P5 formulation, which would involve a commitment and pricing solve… The other thing that comes to mind is BassLink switching term?
I find it extraordinary that just as SA proudly announces work beginning on the NSW interconnector we get news of Eraring closing much earlier as well as NSW announcing big plans to fast track even more local renewables. Something does not compute with the purported benefits-
“Project EnergyConnect will be a freeway for renewable energy produced in South Australia and exported to New South Wales, reducing electricity prices and fuelling economic growth in both states. Importantly, it will strengthen the power grid, so South Australia can securely add more renewable energy in line with the state’s push for net zero emissions by 2050.”
https://energymining.sa.gov.au/latest_updates/work_begins_on_sa-nsw_interconnector
I’m not sure what the future of solar and wind holds but it would appear the ethos of Covid State border controls is catching on with renewables in general-
https://reneweconomy.com.au/bob-brown-foundation-slams-marinus-link-as-isp-white-elephant/
Hey you lot pay for your own expensive backyard grid firming and no mooching on ours! Seems the NEM is about to live in interesting times as Origin’s announcement on Eraring has concentrated some minds around interconnectivity.
I didn’t think the weather conditions were at all exceptional for this time of the year in SE QLD where most of the demand was. Yes, I am including humidity. I am not saying it wasn’t challenging weather conditions, I am just saying this normal for this time of the year. The poor design and insulation of most of the houses in the SE, combined with climate change is going to make things worse in SE QLD.
Of course the unbalanced and uncompetitive grid in QLD just makes things worse. If QLD had a similar amount of wind based generation as it does solar, then prices would be so much lower.
Has AEMO made any comment?
AEMO don’t seem very interested in fixing problems with their data and data systems that are public facing. I have raised several issues with errors in archive files and received no response, nor has the situation been rectified.
Due to the nature of AEMO, there doesn’t appear to be anyone to make an official complaint to regarding these issues.
“Again this week I am wondering whether it’s possible for us all to collectively cool the ideological barrage and really think through what we’re expecting to happen as we move together through this energy transition….”
It does appear to be a case of ‘in lithium they trust’ with transitioning to weather dependent power generation and yet some obvious storm clouds appear with that-
https://premium.goauto.com.au/ev-fires-become-hot-issue/
as the lawyers weigh in too-
https://premium.goauto.com.au/ev-fires-what-the-lawyers-say/
Presumably why I keep hearing all this hydrogen hype all the time. All I know is I wouldn’t be much use at all heading up the AEMO with more questions than answers for those at the coalface. Not an easy task with universal interest in the outcome and as politicians know full well- only the impotent are pure.
I find it interesting, from an electricity grid perspective, that the last man standing, the intact third connection was the Bass Link cable.
“Some traffic was re-routed over the Basslink cable, so Telstra’s mobile and some other services remained online, along with TasmaNet.” according to https://www.itnews.com.au/news/digital-tasmania-lobbies-gov-to-support-new-cable-builds-576790
How strange the grid market operator has to suspend service due to non grid based cables being taken out.
Thanks for the analysis Linton. I too have been wondering about the high consumption considering flooding impacted thousands of residential and business customers in the Southeast. I put it down to the Queensland demand was high as the rest of Queensland is suffering from heatwave conditions – one Cairns businessman described it as the worst summer for heat and humidity since 2006 and Cyclone Larry. The other is there are about 2.1 million residential customers in Queensland of which about 1.5 million are in the Southeast. Similarly, there are about 230,000 small businesses in Queensland and about 140,000 in the Southeast. Energex has done an amazing job getting power reconnected so not sure the amount without power was too high and anyone with power would have been using it for cleaning and air-conditioning as it was hot in the Southeast too. You would definitely need a good nights sleep after a day of cleaning and getting businesses back working so air-con would be a must. Also the rest of Queensland actually consumes more electricity than the Southeast due to mining and heavy industry and this was unaffected by the floods.
This is very interesting, I am wondering is there a part c where the possible solution to what is obviously a countrywide situation at certain times?
IS the speed of drop off capable of being forecasted and could it allow for an alternate power source?
Russ Christ effect, well at least cloud location compared to their shade, also depends on the height of the cloud and its thickness. I did some modelling and comparison to cloud on the Gold Coast and found it was extremely difficult to match cloud conditions based on satellite pictures to output of my PV system. Part of the problem was the low resolution and frequency of the satellite pictures. This was particularly the case when the cloud was broken into cells as distinct from a solid wall.
I did try using ground based photography of clouds, but the topography defeated me in this regard.
This event needs to be put in context though. The odds of such extreme rainfall in a particular location might be 1 in 500 years, but the odds of 80% of this level occurring some where in this region of QLD and NSW is much higher. We only need to look at 2017 and 2011 for similar events. Sure the rainfall levels might have been different and the locations slightly different, but the results are not that different. Major flood damage in this region.
I think there needs to be a better way of stating the statistical probability to the region as clearly 1 in 500 years or even 1 in 100 years, doesn’t convey the risk.
What happened to the interconnectors during this period?
They didn’t even average 10% of their capacity. Given there was fore warning about these conditions, how was it possible that they were not brought back online? I can only assume it was to do with the floods, but that is a worrying sign as flooding events are going to become more frequent and more extreme.
Unfortunately there was almost no wind based generation during this period. I haven’t checked for constraints, so I assume it was just that the conditions weren’t suitable. This is going to make solving this sort of event very hard in a near 100% RE grid. It is going to take many levers to solve this sort of event, more demand management, much more energy efficiency, a little PHS and huge amounts of battery.
Craig, we’ve had this discussion elsewhere, but at the risk of repeating myself it’s absolutely pointless to talk about the notional maximum capacity of interconnectors – typically that would mean thermal capacity of the lines etc under steady state conditions – when looking at their actual capability to transfer power over long distances in a real power system.
The key limiting factors on what interconnectors can actually transfer can often be nothing to do with that thermal capacity but with *transient* electrical phenomena that occur if there is a fault or trip leading to sudden changes in flow and/or impedance and/or phase angle along the interconnection path. In the case of QNI and Directlink, there is a very long distance between major generators in NSW and load centres in Queensland, which makes this interconnection quite susceptible to voltage collapse if a fault were to occur with flows out of NSW exceeding certain thresholds.
AEMO uses a large number of constraints to manage these risks – which can arise on trips of different generators in Queensland, lines in NSW and so on. The limits imposed by these constraints are quite dynamic and I wouldn’t pretend to understand all the factors involved. The bottom line is that it is not at all unusual to see net exports from NSW to Qld limited to only a few hundred MW at times of high demand in Queensland.
Recent work on upgrading points along the QNI flow path – involving both lines and new equipment at terminal stations – should see some increase in capability later this year. But that capability will almost NEVER approach the underlying thermal capacity of the transmission lines themselves.
On March 8 there was an outage impact on flows (apparently a Static Var Compensator at Lismore or Armidale), but it still only reduced transfer capacity relative to system normal (nil outage) conditions by something like 100 MW.
Origin is planning to run its gas fleet as baseload (Uranquinty is a big OCGT station), which will be needed when Eraring shuts down of course. Higher wholesale power prices will be a nice reward for them.
Uranquinty at full output will cover almost half of Eraring’s usual output,
But when you rely on your peakers to provide baseload, what supplies peak demand?
Great article. Rules generally produce unintended consequences and unforeseen behavior.
Thanks Jonathon. Great background on the CFP arrangements to help understand forecasting behaviour.
Perhaps a naive question… but are the deliberate deviations necessarily bad behaviour?
From your description, it sounds like there is systematic error in AEMO forecasting that e.g. underestimates NEM demand on a typical afternoon, hence dipatches insufficient generation, hence under-frequency conditions. If a solar or wind generator anticipates this and responds by intentionally under-estimating generation, leaving some in reserve, then they would be benefitting from addressing that error. Sounds like a good thing??
Of course it would be preferable to fix the forecast errors in the first place or perhaps create a form of frequency regulation that VRE can formally particpate in (beyond contingency). But in the meantime, perhaps this self-serve proxy is useful??
There are massive difference between the 2021 GSOO predictions and the 2022 predictions.
To believe the predictions in the 2022 version, you’d have to believe all AEMO’s past predictions are way off the mark.
Just last week Frank Calabria stated he’d be running Australia’s largest gas turbine fleet for periods of “days and weeks” ie baseload gas, once Eraring closes.
At the same time Liddell is closing, so NSW is down about 3,000 MW of baseload and adding 800 MW of peak demand from SA when EnergyConnect comes online around the same time.
That’s gas for power generation.
To follow the AEMO prediction for domestic and commercial, you’d need to immediately start changing out a lot of perfectly fine gas equipment for electric equipment, all over TAS, VIC, NSW and QLD.
The ‘most likely’ scenario doesn’t seem all that likely.
It was good. Thanks for taking the time.
I have been reading the ESOO with following questions in mind:
(1) Will there be power shortages as aging coal plants are closed?
(2) Where is the modelling for storage of renewable energy over daily, weekly and seasonal cycles?
(3) How many Snowy2 schemes do we need for mass storage of renewable energy?
(4) Will there be enough power for 100s of apartment towers, skyscrapers (without solar panels) and other power hungry infrastructure governments are planning?
(5) How will additional demand from electric vehicles be accommodated?
(6) As we are hitting peak conventional gas on the east coast, how will that impact on gas peaking plants?
(7) How does the speed of oil/gas decline and global warming determine the amount and sequencing of investments in the power sector?
Matt
Do you think the ESOO has answered any of these questions for you? or even helped you answer them?
Great articles – thanks for these resources.
I have a burning question to the more complex cases. Assuming that the Increase field captures the marginal increase in energy for the ENOF and the marginal decrease in energy for the LDOF type entries (for the extra 1 MW), and then there is FCAS (Increase sums to 0 MW), what exactly does the Increase field capture for the tbslack (tie-break) type entries? And for MNSP type ones (Basslink maybe – although Basslink shows up as non-scheduled in AEMO’s registration list (https://www.aemo.com.au/-/media/Files/Electricity/NEM/Participant_Information/NEM-Registration-and-Exemption-List.xls))?
Good spotting. Wonder what it was?
Just a tidbit in case it’s useful. Millmerran has a coal ash dam shutdown starting today (https://www.seek.com.au/job/56745397?type=standout#sol=8b6f7600ba668dcefd72cd3f1e97237f0a4c695f) and has been trucking (!) coal down from Callide for a couple of weeks (https://www.banana.qld.gov.au/news/article/375/increased-heavy-vehicle-traffic-in-the-area). Could be related to that generator.
Nice pick-up, in terms of Millmerran’s coal issues – which I have flagged here:
https://wattclarity.com.au//articles/2022/05/03may-trucking-coal-from-callide-to-millmerran-as-a-result-of-the-la-nina-rain-event/
Thanks for that!
Oh, also, a cryptic comment about one of the contributors here:
https://wattclarity.com.au//articles/2022/05/03may-price-volatility-in-qld-and-nsw/
Paul – The lack of the non-mandatory rebid fields is probably a very deliberate decision by at least some participants (not saying this applies to Milmerran). What’s the benefit to the participant of populating those fields? If they put data in and are assessed for compliance or something else by the AER they may have to show that the numbers are accurate, which is probably not straightforward in at least some cases – “awareness” and “decisions” can emerge over time, not necessarily in an instant – and they are also on a hiding to nothing if they made any errors or omissions. So strategically it may make very good sense not to provide that data given it’s not required. So perhaps I differ slightly on what’s ‘Best Practice’.
I suspect, Allan (based on the absence of these fields for many DUIDs) that a number of participants have a view similar to yours.
Imports into SA via Heywood are severely constrained across Thursday. This combined with the low wind forecast is the cause of the forecast LORs. If Heywood imports were at their normal level there likely would not be any forecast LORs (even with the low wind forecast), although it would be close to a forecast LOR1.
Thanks, Jack, for highlighting!
Also relevant is that Pelican Point has completely offline for over two weeks. This could be for maintenance or possibly for establishing a gas connection to the neighbouring Snapper Point generator which is being commissioned using diesel fuel.
Thanks Alex!
Many of the SA peakers have been running this morning peak but Barker Inlet has been offline and Torrens Island limited to 80MW instead of up to 400MW in recent days. Pelican is still offline. Currently SA/VIC interconnectors are very severely constrained.
Hi Paul, it would be great to have a look at the generation profiles and bids that have resulted in the prices.
We’ve had a few occasions in the last week where there has been over 150MW liquid fuel at these times. Would not be surprised if the generators who routinely bid at $-999MW/h have probably been impacted by calm & cloudy days and those who can turn on are wanting to make some cash to make up for the all too frequent periods or even whole days of negative regional prices.
Yes calm and cloudy in Adelaide Friday and diesel generating 127MW and I noticed it as high as 157MW around 2:15 in the arvo yesterday Thursday. Is diesel cheaper than gas or just available?
Here’s a puzzle. Friday night tonight and around 3:50pm total SA output is 1704MW with 564 gas 478 wind and the rest solar when diesel kicks in as high as 173MW as solar shrinks to zero around 5pm and both battery storage comes on from SA and Vic in that time. SA diesel continues down as low as 115MW until it drops out around 8pm and by then wind has picked up. At one stage gas reaches 910 with wind the rest for 1717 total and wind is picking up while all batteries have dropped off.
By 9 pm SA price is around $300 Vic $320 and NSW $340 and Vic is feeding NSW suffering a max shortfall around 1000MW maxm and Vic is in turn being fed by SA and Tas. By this stage SA wind is 1436 and gas 338 in SA and at one stage all States batteries are supplying. The main surprise for me was to see SA diesel running for around 4 hours and clearly competitive with gas for the period.
Whereas earlier in the week when most of the diesel generators were in action, on Friday the diesel output was only from Snapper, which was at 120MW. This is the highest I have seen it in the commissioning process. Pelican Point ramped up from outage during the afternoon reaching 450MW before going offline at 9pm. Barker Inlet returned from outage this afternoon.
11:35 am Monday morning and the prices per MW in round figures are showing Qld-$525 NSW-$130 Vic-$120 Tas $112 while SA is NEGATIVE $95 yet SA power consumers pay the highest power prices. Clearly power retailers factor in lack of predictability in their business models and something for easterners to look forward to as they follow SA’s lead with solar and wind penetration.
How AEMO people deal with that lot beats me but it doesn’t seem like an optimal career choice if you like the quiet life.
Since we are low on coal capacity lately I thought that the experience during low-wind periods would be interesting. Gas probably picked up some of the slack but I expected that hydro would be the cheaper option and it seemed to be the case looking at the Aneroid Energy plots. Maybe crude compared with the analysis here but there was clearly a difference in hydro from month to month, expressed as a larger number of spikes over 60% capacity for all hydro. I appreciate the analysis is better without Tasmania.
For some time I have wondered how long hydro would last if it is required to fill a serious gap in supply.
Drawing an arbitrary line at 10% of installed capacity for wind drought, it has been barely more than that all day.
Still the message from the wind records is clear, the retirement of conventional power has gone as far as it can until grid-scale storage is on deck, on time and on budget, as you signalled years ago. Still waiting:)
RE can displace coal power but not replace it. https://www.riteon.org.au/netzero-casualties/#217
South Australia has demonstrated the point, they import power whenever the wind is low, never mind full-on droughts like yesterday. Currently they are importing and 70% of local generation is gas.
The NewWatch widget is the clearest indicator, a glance in the morning and evening is enough to see the unreality of the green dream. Not to mention the way things have gone in Britain and Germany.
Who is going to tell Albo and Chris Bowen:)
Apologies that this is somewhat off the exact topic yet still on vital subject of wind generation.
Wind generation plateau for 4 years in South Australia
http://www.warwickhughes.com/blog/?p=6921
What has caused wind generation percentage to plateau for 2018, 2019, 2020, 2021 in South Australia?
Aggregate wind production in South Australia has been limited (i.e. ‘constrained down’) due to various forms of the ‘System Strength’ constraint in recent years, Wazz.
Recently Marcelle provided this illustration of how the pain is shared around when constraints like this bind:
https://wattclarity.com.au//articles/2022/05/who-gets-to-run-when-everyone-bids-the-same-a-crash-course-in-disorderly-bidding-and-tie-breaking/
… and Allan previously wrote about how the syncons partially alleviate this issue:
https://wattclarity.com.au//articles/2021/02/new-sources-of-strength/
More articles on system strength are tagged here:
https://wattclarity.com.au//articles/tag/system-strength/
Hope this helps
Thanks for explaining that Paul – I tried searching the subject at various search engines but got nowhere. I must have been using the wrong words. I wonder what the implications are for this in other States and further down the path to “Net-Zero by 2050”
As an ex coal fired power engineer, I could always justify turning off the unit to fix something. There was always a reason to turn a unit off. So I would pull out my risk assessment, and say that the risk of keeping the unit online was unacceptable, and that repairs are to commence once the unit has cooled sufficiently (48-72 hours), and that owing to concerns about the complexity of the repair, it will be done on day shift only. I could conceivably keep a unit offline for a week or more in that scenario.
That only buys you enough time to get past the immediate challenge, but to get short run marginal costs that high, you’d have to be exposed directly to very high fuel prices (and presumably water prices for water cooled units). Which in the current environment I would see as gas only.
Generators can apply for compensation if their SRMC exceeds the APC: https://www.aemc.gov.au/market-reviews-advice/review-compensation-guidelines-include-wholesale-demand-response
PS Ryan … thanks for noting, you’ll see more here this evening
https://wattclarity.com.au//articles/2022/06/10june-qld-cumulativeprice-above-1100000/
It will take a while to digest, but I very much appreciate the detail and the effort.
I’m sure if I had the software I could answer more of these questions myself.
Alas, I am stuck with NEMWEB, Excel, and a whole lot of mouse clicking 😀.
Cheers
Ben
Would be good Paul if you added a colour legend for the various States in the NEM. I make them – NSW Turquoise – Qld pale purple – SA Red – Tas Green – Vic Brown.
Thanks for Wattclarity info over years.
Sorry, fixed – left it off by mistake
Tas hydro has no competitors –
is minor in size compared to NEM –
Tasmania pop is ~525,000,
TasHydro is Tassie Govt owned like TasNetworks –
is in horrific debt –
must have Departments of staff whose task it is trying to sucker the Feds to invest in their 2 oversold projects –
the new Marinus Link cross Bass St. copy of their BassLink disaster –
plus their Battery of the Nation version of pumped hydro copying Snowy 2.0 –
I hope the Feds keep right away from investing any taxpayer $’s in above.
Let TasHydro find private enterprise to invest in their great projects.
Oh!! almost forgot my opinion of the answer to my question in thread headline –
The outer galaxy $15,100 per MW/Hr prob due to TassieHydro book-keepers having to balance columns in some spreadsheet.
There is a very good blog TasFinTalk who in Feb had an article on Basslink –
Sunday, 13 February 2022
Basslink: Turning a blind eye to its lessons
https://tasfintalk.blogspot.com/2022/02/basslink-turning-blind-eye-to-its.html
I think the overwhelmingly obvious thing the current situatio illustrates is that you have to build new capacity for energy and power before the old stuff goes out of service.
To me it is incredible that Frank Calabria is talking about restoring coal supply and is closing a generatorbut is not contracting or building any new wind or solar even though every model shows that lots more of it is needed. What does he think, someone is going to give hime 5 or 6 GW of wind or solar the day Eraring closes so he can keep his retail customer loads? The same for AGL and EA with Yallourn.
As you can see it makes me grumpy.
Nobody with any honest insights believes The Transition (to replace coal with wind and solar) is realistic – either in the realm of physics or economics.
The faint technical possibility of a very high % RE system is dwarfed by the massive cost and the impracticalities of storage, transmission, inertia, fault level, over supply, negative wholesale prices, land use – just one of those challenges is a spanner in the works. Combined they are insurmountable.
At what point does the community of energy professionals decide to speak up and tell the truth?
Tristan Edis states that RE output is not down, comparing May 2021 and May 2022. What he has omitted is that there was 3 GW of rooftop solar installed in 2021, and a bunch of large-scale wind and solar too. In the context of much higher RE capacity than 2021, the same MWh output in 2022 is in reality a lower capacity factor in 2022. This is in surprising because the weather has not been favourable. The weather is also partly responsible for some coal plant scarcity because of mine output constrained by water issues.
Tristan acknowledges that gas generator output is not much different to previous years. In fact some gas generators are price takers, being dispatched at bids of zero dollars.
Gas and diesel peakers, while responsible for short duration price spikes, are not the problem either. This is evident by the lengthy periods of wholesale prices sitting around $300/MWh. This is not a high average price caused by 15min afternoon spikes.
Consider this scenario: a coal-fired power station, with fuel constrained by a partly flooded mine, offers a larger portion of their output at high prices ($200-$500/MWh) to avoid being dispatched, to conserve fuel, unless other generating capacity is short. Concurrently, there are several units on planned outages and a couple of unplanned outages. All of a sudden, baseload generation is scarce, and that higher-priced capacity gets dispatched.
Continuously high prices are no mystery.
As for gas, imagine if a state thousands of km from the bountiful QLD gas fields banned their own local gas development and failed to upgrade pipeline capacity… also imagine that with an early onset of winter (damn that global warming), gas demand for heating suddenly increased … and the pipes are already at capacity.
When demand outstrips supply, you get price increases. Again, no mystery.
No shortage either.
Closing coal-fired power stations and expecting gas, wind and solar to fill the gap cheaply is an outrageous fallacy.
The knee jerk reaction to slap new taxes and levys on coal and gas producers is exactly the wrong strategy. At a time when high prices and some high profits can be expected to stimulate new investment, new development and therefore new supply (and jobs and taxes and prosperity), eliminating those profits by imposing arbitrary levys is a sure way to push these companies out if this country while doing nothing to increase supply and therefore doing nothing to lower prices.
I have to say that high energy prices and coal generation scarcity is exactly what the RE lobby wants.
I can’t imagine why anybody is surprised at the results.
I think most energy realists expected the crisis in the power system to become obvious due to rolling blackouts.
In the northern hemisphere the crisis arrived earlier than expected on account of the war and this may be a factor here although Ben Beattie did not need to refer to the war to explain the situation.
The underlying issue is the fact that the green transition is not sustainable due to the combination of wnd droughts (zero RE on windless nights regardless of installed capacity), the need to meet 100% of demand all the time, and the lack of feasible/ affordable grid-scale storage.
There is also the matter of cost – we are stuck with a hybrid system because we can’t allow the coalers to close until we have storage or nuclear power, the transmission lines and interconnections will cost several arms and legs, and everything containing rare earths, especially lithium and copper, will be more expensive. Please, no more talk about the price of batteries plummeting:)
It’s hard to comprehend why an analysis such as this did not even once mention the nuclear alternative.
Our kids are being indoctrinated with the global warming message even in kindergarten, more so in primary and secondary school, and the process finally being completed in university. As grown ups these naive people haven’t seen the logical alternative for it is rarely discussed, banned by legislation, and terrifies the uninformed.
Reversing this deeply rooted belief will entail a herculean effort at re-education.
In the meantime we will all suffer from a failed power grid that must surely open the eyes of government and an enraged community shaken from its torpor.
Emissions free nuclear power – the ultimate means for providing a viable, plentiful, and cheap supply of electricity.
Paul: I think the local contingency raise FCAS constraints, as well as the negative residue management constraint on Qld-NSW flows, were still fundamentally the drivers for these Qld events this morning – they were all binding in this DI, and as noted in Marcelle’s article (https://wattclarity.com.au//articles/2022/06/interconnector-constraints-and-limits-an-explainer-and-a-few-traps/) AEMO data only lists one constraint as setting an interconnector limit even when there are multiple constraints binding. The only reason N^^Q_ARTW_B1 bound was because NEMDE was trying to increase northward flows to satisfy the NRM constraint. That in turn was only binding (violating in fact) because the local FCAS constraints had previously been pushing counterprice flows southwards.
… so I was tripped up on those same tripwires Marcelle had written about coincidentally today!
Gas exporters were supposed to source their gas from coal seam gas, and that hasn’t happened because they’ve been allowed to buy fields like Arrow previously domestically focused. 10% of their sales are spot – exporter source, and can readily be stopped. In the UK, they know war profiteering when they see it, hence a windfall tax. The US has also declared war powers as well. Not surprising given the kind of mess we’re seeing here.
Thanks for your thoughts and the info.
That “symphony” reminds me of Schnittke music. Here is a suitable example https://youtu.be/N1U4tN_ZUt0
Other contender is Karlheinz Stockhausen: Helikopter-Streichquartett for a quartet of string players and four helicopters https://youtu.be/7ykQFrL0X74
You get the idea:)
“It also seems rather odd that the gas industry finds it so difficult to supply a level of gas power generation in 2022 that is not much different to what has occurred historically.”
The gas industry has been extremely vocal in the past few years that we’re facing a gas supply shortfall (with specific emphasis on this year and the next). The RE lobby responded to these calls by saying that total gas demand destruction was imminent (not exactly seeing that right now) and no new supplies were needed.
“What appears to be the real reason for high prices is that both gas and coal miners are charging very high prices for their products because that’s what they can sell them for into international markets.”
We were seeing 100 TJ/day of gas sent north from the Moomba Sydney Pipeline in the beginning of the year which is in line with what they’ve done historically so the LNG producers weren’t pulling up more domestic gas than usual. Because domestic prices have been so high recently, we’ve witnessed a negative chill spread (domestic prices higher than export prices) and yet prices still remain high domestically. Shouldn’t the greedy LNG producers completely cut off their exports and pump everything locally according to your logic?
By the way, OCGT gas generators (the bulk of them in the NEM) act as peakers and not as baseload. Therefore, calculating their total GWh and claiming that it isn’t high compared to previous years is completely misleading. The commercial driver for peakers is based on spot prices, their economics are not tied to long baseload runs. Furthermore, it drives me nuts when people do not appreciate how important those gas peakers have been during this crisis. They were absolutely critical in helping tame prices during the morning and evening peaks. Something that “free” wind and solar isn’t too good at.
Where to begin. How about, simply, Western Australia.
Anyone who wants to talk power and price, and supply, please look at Western Australia’s last 10 years.
It simple what they have done, it was not hard, in fact it was easy. People can make up what ever crap they want to explain the eastern sea board. But it was simply planned poorly.
“…one would hope to see availability exceeding 80-85% from a coal generation fleet performing well”. I thought the capacity factor for coal was 50-80%, so 71% is about average, especially for aging infrastructure.
Dave – thanks, but you might be thinking of actual capacity factor which measures energy actually dispatched relative to nominal capability, and will always be lower than available capacity factor because of market offloading – generators don’t always operate at their available maximum and we see plenty of swing on the NEM coal fleet. But available CF is a more appropriate metric to consider generation reliability. Allan.
Great article Allan. One cant help but observe the coincidence of outages and high coal prices at the start of winter although I have no doubt coal availability at the mine mouth and rail availability to move it to a power station are also down. I do wonder whether the bayswater issues will further impact agl profitability.
Sorry for the very belated comment Paul but having done some similar analysis recently on coal units for Q2 2022 I came to this post. I can’t see that there would have been much material value or meaning to inclusion of semi-scheduled generation in these datasets, so I think the rule change was sufficient. While including S/S capacity might have shown some small variations in maximum capability of wind and solar farms (eg known maintenance on a few wind turbines here and there), or commissioning profiles on newer farms, these variations would be dwarfed by uncertainty in future sun and wind levels, so information on maximum output of any wind or solar farm, let alone the whole fleet would be of very limited value. The STPASA forecast availability datasets based on ~1 week’s weather forecast, and also incorporating known capacity restrictions, seems as much as one could or should expect.
A price cap, in any market, tends to be the last step before outright physical shortage emerges.
That basic pattern has been at the heart of physical shortfalls overseas in various countries. Generation shutting down for purely financial reasons with the end result being physical shortfalls.
Same applies to gas since the recent price capping simply thwarts a shift to diesel, thus holding gas consumption higher and depleting storage more rapidly.
We are in interesting times indeed.
RERT now in QLD too.
tomorrow morning and evening? 14th?
I wonder if AEMO’s own data centre has been load shed? Seems impossible to load Market notices or price data from either Telstra or AAPT connections at present.
My guess is more equivalent of a denial-of-service attack, with a massive upswing in people interested in such arcane concepts…
Hi Paul,
The last 7 days sum of spot prices for all states except for QLD is well below the $1,359,100 CPT. How can the CPT be exceeded when the sum of that last 7 days of spot prices is at or below $1,000,000?
Note to other readers … we answered this directly with Alex one-to-one.
As consumer advocate we are keen to see network tariff reform include more emphasis on load controlled tariffs. Around 900 MW in Queensland but it could be more.
Thanks Paul, great service to the community!
PS I am doing a list of our world class windwatchers and you will have to be on the list with Tony From Oz, Paul Miskelly and Mike O’Ceirin (Spasmodic Energy.)
http://www.spasmodicenergy.com/Pages/Home.aspx
Looking at strange events, reserve (dispatched generation less load and export) has been negative at times in Vic and NSW, while generators appeared to operate off target (IC flow suggesting denand was met). YWPS unit 4 started up yesterday afternoon, then tripped right near peak demand.
Anyone know why basslink is constrained to zero in the Tas to Vic direction?
So the chair thinks being forced to sell at $300/mwhr when your inputs are >$1,000/mwhr is not “reasonable cause” to stop offering. Is this guy straight out of communist russia?
Paul, it’s interesting that some have commented that generators have withdrawn capacity rather than be dispatched at a loss making price due to the cost of input gas. Whilst there is a mechanism through the AEMC to claw back costs, I’d expect the prospect of losing millions of dollars per week and needing the cashflow to manage this whilst waiting (months?) for a determination on your compensation may not be viewed as particularly attractive…
Paul, congratulations on the work that you do. I was looking for information on why Swanbank E has been out of service and when it’s expected to become available again. Bingo.
A couple of other questions I can’t yet find answers to are:
1. Where can I find information on each of the grid-connected wind turbines that shows their unavailability due to mechanical or electrical failures/faults? Individually, their forced outage doesn’t make much difference, but collectively their outages may be contributing as much to grid integrity as the coal-fired and gas-fired generating units.
2. Where can I find information on each of the grid-connected BESS units to see how they’re performing and when they’re showing early signs of needing replacement? Just knowing how many charge-recharge cycles each installation has done, would be a leading indicator. I’ve noticed Hornsdale on numerous occasions doing more than one cycle per day. On current usage, what is the forecast date for the original batteries to be replaced.
Report in the Guardian that the transformer was not energised at the time of ignition. Begs the question of why it caught on fire in the first place.
Hello, can you explain what snowy 1 has contributed during the current crises. My understanding is that SH 1 was designed to step-in to supply during such crises.
I have not heard a peep from SH to say it was on duty. Or in response to the white elephant claims by media commentators.
Thi is very unsatisfactory. Thanks Bill
This leads to distrust in what snowy 2 will do
Bill
Whilst we (the general population) might have intense interest in what’s happening inside the wholesale market during this crisis, and so might think that the participants in the market should be doing more to talk about what they are doing, it’s important to keep in mind that their first duty is to do whatever they can to help AEMO keep the lights on … which might mean their communications come a distant second.
It’s unfortunate that some take silence (most likely just because of busyness) to leading to … ‘distrust in what snowy 2 will do’.
In terms of what Snowy Hydro has been doing, coincidentally Ben Kefford from KPMG has posted some interesting analysis here on LinkedIn:
https://www.linkedin.com/feed/update/urn:li:activity:6944212641888243713/
… that might satisfy some of your appetite.
Paul
but a change of pricing ….
Thanks!
Just now: https://www.aemo.com.au/newsroom/media-release/aemo-moves-to-lift-market-suspension
“or feel free to add as comments below”
https://reneweconomy.com.au/aemo-suspends-new-wind-and-solar-projects-as-it-battles-to-deal-with-market-crisis/
https://theconversation.com/the-national-electricity-market-is-a-failed-1990s-experiment-its-time-the-grid-returned-to-public-hands-185418
So…at a time at which in-market participants were only entitled to $300/MWh and/or their direct costs, out-of-market resources could earn around $20,000/MWh. Does this feel right?
Hello Paul,
How about its also been pretty cold for broad areas for long periods, low winds, the snowy hydro not running properly, high gas costs, high coal costs, water sodden coal mines restricting supply, gas supply tight, very run down coal plants, and I am sure gas plants, poor planning, bit of gaming the bidding system?, and i am sure a bunch of others i have missed. All in all a fair amount has gone the wrong way all at the same time.
We may learn some lessons from another energy sector: oil.
Peak oilers have long been thinking what will happen when the global oil peak occurs. There is the theory of a Seneca Cliff after the peak in which the whole system declines because everything depends on a smooth supply of oil.
The first global peak we had in 2005 (as predicted by Deffeyes) which caused the US recession in 2007 and the global oil price shock in 2008. High Chinese oil demand for the Olympic games also played a role. At the same time banks were so imprudent to have invested in car and therefore oil dependent US suburbia just as the peak was happening.
High oil prices give a strong signal to motorists to economize the use of fuel with fuel prices at the pump following oil prices after 2 weeks or so. If demand destruction happens in all countries then oil prices will fall again. If these prices are too low for oil companies to make money then oil production may suddenly drop, potentially below levels needed for the safe functioning of the world economy. This did not happen yet because of US shale oil but this oil has astronomical decline rates and will also peak.
Drawing parallels to the power sector, capacity of dispatchable generation may have peaked (including from coal depletion in mines near Eraring and Liddell/Bayswater) whereby this is a function of the price. We then have a cold snap with higher demand (similar to China’s 2008 Olympic Games). All the while Australian governments were driving up evening peak power demand by increasing immigration over 15 years, requiring hundreds of new power hungry apartment towers (similar to expanding US suburbia). The surge in renewables is similar in function to the new frontier of US shale oil and occasional/unpredictable recoveries of oil production from unreliable suppliers like Libya, Nigeria etc. and new fields.
The big difference between oil and power:
In oil markets, the feedback loop for lower demand via fuel prices at the pump is comparatively short as mentioned above. On the fuel supply side, oil companies are sitting on inventories and have a certain flexibility in case crude production is short. So this gives at least a short to medium term window to balance supply and demand.
In power markets, however, generators must balance supply and demand instantaneously while the feed-back loop to lower demand is up to 3 months when the next power bill arrives. On the supply side, generation depends on variable renewables without sufficient pumped storage to provide dispatchable spinning reserve.
So in the power market, there is a higher chance that the system collapses (=load shedding)
Interested in responses to this.
Looking at this rather unfortunate state of affairs from the other side of the country, I suggest the main drivers are fairly obvious. There were a few coal fired units out for maintenance then a few more were out of service from forced outages – either plant related or fuel supply related. Oh and lets not forget that there didn’t seem to be a lot of wind around for a few days. Now there is some gas fired gas turbine plant around but nothing like the proportion of the fleet we have here in WA. But that only works if you have contracted fuel supplies and hence a sensible price. Given most OCGT’s will have a heat rate of between 10 & 12 GJ/MWhr then if you cap your price at $300/MWhr then as soon as your gas price exceeds $25-$30/GJ you are out of the money before you have even paid for your starting costs on the unit. Spot gas prices that I saw exceeded this so clearly there is little contracted gas reserved for these peakers. Liquid fuels are even worse.
This is one of the serious limitations of an energy only market and why Western Australia doesn’t have one. The NEM used to have lots of spare capacity but this has been hollowed out over the last decade or so as older stations have been retired and a market regularly trading at SRMC results in little incentive to invest in sustaining capital. I have heard much nonsense about “aging coal fired power stations” – coal fired power stations do age but are like Grandad’s axe. A new head or handle and it is still Grandad’s axe. Unfortunately when the economics of reinvestment have been so damaged by the hollowing out of daytime energy served the plants become unreliable. Their eventual retirement (whenever that might be) is also a bit of an are as no one retires a pristine plant – the glide slope is always tricky.
This is where a capacity and energy market works. It guarantees a revenue stream, but it has to come with strings attached seeing at the end of the day the punters (via the retailers) are paying. There has to be firm fuel supply arrangements in place – if the generators want to be paid there are no excuses. Seeing the capacity market pays for some of the fixed costs of the plant then the market price caps get pulled right in. In WA they have always been $300/MWhr based on a GT burning liquids – probably about half the number it needs to be right now but the point is numbers like $15,000/MWhr just break the little guys who have little chance of hedging with a generator.
Capacity markets work – they simply need to be carefully designed and calibrated (in terms of $$) to get the correct outcome and prevent gaming by people whose business model is only capacity payments. They can also provide good drivers for preferentially locating (say) a wind farm where it delivers energy best aligned to the daily load profile. Ultimate energy produced might be less but the capacity payment keeps the selection whole.
The NEM is certainly creating some volatility alright-
https://www.youtube.com/watch?v=CSkFPA81Uws
That’s not mincing words and calls for a response from the CSIRO and AEMO.
Paul, 2 things I appreciate from you:
1. Trying your best to stay neutral and wholistic
2. Summarizing hours’ and days’ worth of research into a short, concise and structured article
3. And sharing for free. Well with some subtle marketing but who wouldn’t!
Ok I can’t count but you get the idea.
More information on this here Port Augusta Renewable Energy Park Oscillations on 23 June 2022 – What really happened?
https://www.linkedin.com/pulse/port-augusta-renewable-energy-park-oscillations-23-june-jayewardene
Very interesting and important, suggesting that it may be winter wind droughts that will set the amount of storage required, rather than the maybe more widely studied wind lulls in summer, in association with heatwave induced demand spikes, which often in the past didn’t cause problems because they fell on public holidays.
Solar has a daily drought of course, but must play a role during daylight hours in replenishing the storage, and preventing it from becoming fully drained.
“We look forward to being part of the discussion on what a robust, reliable, low emissions, and lowest total system cost NEM might look like.” Good luck with that, there must be a huge tendency to bury the bad news of possible future problems.
A good read Dan. As someone with a background in electronics I was aware of the implications of I2R lossses ,but not across the Jargon.
Sun Metals lodged its 5MS rule change request on 4 December 2015.
This graph is a classic example of everything that is wrong with the uncontrolled operation of intermittent generation. There is something like 500MW/hr being added at the whim of the wind from 6pm on Sat at the very time when system load will be declining. This would force other generators to ramp down and / or decommit. So please don’t complain when the reverse happens and suddenly thermal plant is expected to restart at short notice with no guarantee of recovering start up costs. How is the large gas turbine fleet needed to deal with this going to be incentivised and where is the fuel coming from? (or for that matter the deep storage. ) Forget the big Coalies – they simply won’t be able to play like this as the problem gets worse – exactly what you are seeing with older plant like Liddell & Yallourn that will simply exit.
Hello Andrew,
The whole point of the market is to get coal plants to exit currently. If the greens had gotten into power they had planned to simply start shutting them, not even have them exit.
And not in any order or age either, it seemed rather random.
I have an issue with this article.
Wind (excuse the pun) back just a few days to 14th July and during the evening peak wind production was a paltry 600MW. This is less than 10% of rated capacity, supplying a measly 2% of the NEM demand.
The fact that wind output can vary so much over the period of just a few days is a far more news worthy and important issue than reporting a once in year outlier in maximum wind generation.
Michael
There are a number of article (like the following) on WattClarity that also point out the lows:
https://wattclarity.com.au//articles/2022/05/27may-short-term-wind-lull/
It’s becoming increasingly the case that both extremes need to be recognised, and the averages become increasingly just that … average. This has also been noted a number of times here.
Paul
As an electrical engineer who repairs electronics for a living, I can tell you that it is mainly the temperature of the device (or rather, inside it) and hence the temperature of the components that negatively affects its life.
Specifically, in Switched-Mode Power Supplies, the Capacitors are rated in hours at a particular temperature, for example, 2000 HRS @85 deg C. If the temperature in the device is 85 degrees, then it can be expected that the capacitor will be past its useful life in 2000 hours. This is under a year if the light is on for 8 hours a day.
However, if the temperature is reduced significantly, then the capacitor will last much longer.
The amount of Switched Mode Power Supplies I have repaired with failed capacitors is countless, and in larger devices it is usually a small 10uF 50v cap that goes, and this is used to filter the power to the SMPS control chip.
Whenever I replace them I try to install 10000 HR @ 105 deg C rated capacitors for longer life, if available in the size and voltage required.
In an LED downlight, some are direct plug in 12v versions, while others have their own 240v SMPS to enable dimming, and the old transformer style power supply is replaced.
However, it’s usually the direct plug in 240v ones (I.e bayonet cap / edison screw) that fail because they have to pack in the power supply, the LED with its heatsink on top, all into a small area with barely if any ventilation!
At home, we have had a few seperate power supply type downlights fail, however they were just pushed into the roof space, and both failed ones happened to be covered in insulation – further evidence that keeping the temperature down is key to long life of any device that relies on electrolytic capacitors.
Most mains rated equipment generally has a safety margin for incoming voltage, and although a higher voltage will also increase aging in the capacitors, the heat is the main cause of rapid aging.
Ameo, is like a drunk, drugged person at the wheel of a car they have been saying they are getting serviced for 2 years, they have neglected paying rego and insurance and are now about to go around a corner and the headlights have failed!!!!!!!
(Seriously can it be possible that they let gas get down to those levels then have to step in at the last second claiming its anyone else fault but their own for not actually having anything planned, its mind boggling)
Unless I’m reading the chart incorrectly it looks like there were liquids in there too which would be setting the price. Thermal plant all off and / or cold so no way to recover quickly.
One big one is missing, and should be in bold and at number 1.
Seriously poor planning, on the eastern half of Australia.
Unless the residential market is required to use wholesale linked rates, there is little incentive for households to change their use patterns or invest in reducing their consumption through better efficiency such as insulation. Without this change to wholesale linked rates, we are all going to pay more due to greater storage and transmission requirements.
Craig, What sort of storage do you have in mind, considering that Snowy2.0 is a long way off and batteries do not have grid-scale capacity?
Due to the fact which Paul and others have noted regarding the intractably low production at the bottom end of the wind scale we have probably gone as far as we can go to get rid of coal until there is a radically new storage vehicle available or nuclear power is on deck.
https://spectator.com.au/2022/07/energy-policy-where-parallel-universes-are-set-to-collide/
Digging in to this one, looks like generation at DUIDs CHYTWF1, STOCKYD1, MUWAWF1 and MUWAWF2 were suddenly interrupted. Do not know if these were the result of constraints binding (i.e. dispatched off), or there was some kind of fault in the area.
Some DUIDs were on the LHS of the bound/violated FCAS constraints. Without double-checking, the ones you list sound like the correct ones.
This may also have been a participant response to the extreme FCAS raise prices. Generators pay for raise FCAS and a number of them will bid their plants down or off when FCAS raise prices spike.
Is another major wind drought forecast?
There was an epic wind drought in several states over the weekend, running into Monday and lifting on Tuesday, hard on the heels of record high winds.
That’s intermittent energy!
https://spectator.com.au/2022/07/energy-policy-where-parallel-universes-are-set-to-collide/
I’ve lived in this area for 50 years and this has been the coldest winter ever
I watched this as it happened, and then wrote about it the following day at the blog site I contribute to. I have been keeping all the AEMO coverage area wind generation data on a daily basis for the last four years in my daily Posts, and for the year prior to that all the (AEMO coverage area) electrical power generation data from every source for a year and a half, and then for the year prior to that, all the Base Load data, also on a daily basis.
This was a failure of (ONLY) wind plants and Solar Plants. (not rooftop solar, just the solar power plants) The initial fall at 11.35AM saw an immediate loss of 3600MW, and then over the following 45 minute further losses to take the total loss to 4700MW. Keep in mind here that this was ONLY those wind and solar plants. Now, while you might think that 4700MW is not all that much, at the time of failure that amounted to a loss of 17% of overall total power consumption across the AEMO, so this was a significant failure.
At the same time as the failure began, Hydro came ‘to the rescue’, as it always does when there is a failure of any power Unit, because hydro can come on line almost immediately. The plants at Tumut Three (pumped hydro) Upper Tumut (Tumut One and Two) and Murray (Murray One and Two) immediately put 2450MW into the grid to replace the missing wind and solar power. As you are aware these (some of them older than 50/60 years) Units are smaller Units, around 150 to 250MW, so there was around 12 to 15 Units coming on line here. Also at that same time, EVERY black coal fired Unit (15 Units in NSW and 22 Units in Qld) also ever so slightly RAMPED UP (and hey, haven’t we been told ad infinitum that they cant ramp, despite doing exactly that twice every single day by between 4500 and 5500MW) and in the next 10 Minutes, those coal fired Units also added an extra 650MW to the grid to all but cover that initial loss from wind and solar plants.
Note here especially, that the point of failure and the point in time of rescue were exactly the same, 11.35AM.
It was almost 2 hours before wind and solar recovered to their level at the time of failure, but this was the failure of just wind and solar plants.
There is more text, and images showing the failure on load curve graophs at the following link
https://papundits.wordpress.com/2022/08/11/australian-daily-wind-power-generation-data-wednesday-10-august-2022/
I hope this helps.
Anton Lang
Anton
I’ve only skimmed what you wrote at the link in your comment – but it does seem that you have misinterpreted ‘the reason why‘ the wind and solar farms collectively dropped in output. It was not a ‘failure’, in that sense – though the underlying software glitch would seem to be a failure of other processes.
Coincidentally I’ll be posting to explain this here on WattClarity in the next day or so, so stay tuned…
Paul
Paul, and thanks for the quick response here.
Using the same word semantics you used, I didn’t ‘skim’ your article. I read it all.
At no point did you even mention the failure of ONLY those two renewables as the end result of what you found.
You mentioned the Queensland situation, with no explanation.
You mentioned the rise in hydro. (without even going to check why)
You mentioned the spike in the cost of generation, also without checking why.
All of that is down to the immediate loss of 17% of power, and the extreme short notice of finding replacement power ….. OUTSIDE of the bidding process, hence enormously expensive, hence the spike in prices.
That 17% of power was 4700MW, and that’s the equivalent of six to nine of those large scale coal fired Units. If that many coal fired Units dropped off line, even for this same reason, then you would have been shouting it from the rooftops. A failure of delivery of power ACTUALLY IS a failure.
You say that I may have misinterpreted the reason why, but it seems you have also misinterpreted things as well, perhaps more than I have.
I look forward toy your coming article, ….. a week after I wrote about it. Feel free to use anything from your skim, but hey, you might credit me if you do, eh!
Anton Lang.
The one big omission in the AEMO report is the absence of a review of the Administered Price which at $300/MWh was set some 20 years ago based on the cost of energy from distillate fired gas turbines. If this price was higher there may have been less intervention needed by AEMO and much simpler financial settlements. Alternatively a dynamically calculated administered price could possibly be calculated each time.
Alinta’s submitted a rule change on this, Bob
https://wattclarity.com.au//articles/2022/07/alinta-rulechangeproposal-raise-apc/
If a school kid was looking at the graphs, thy would say diesel is the go!!!!!.
I really did not think it was still in the system, but it is, and its in SA I am thinking.
Hi John, your thinking is right. There were two diesel generators that contributed to that group, both in SA.
It’d be interesting to estimate the additional CO2 emitted due to that NEMDE software error!
What is the definition of a wind “drought” ie how many days constitute a wind drought?
Not sure if there’s a (widely accepted) ‘official’ definition … but would think it would need to be over a longer period than what happened on Friday 26th August.
Something like what happened in June 2017, for instance:
https://wattclarity.com.au//articles/2017/06/wheres-the-wind-gone-nem-wide-wind-farm-operation-lowest-in-5-years-maybe-ever-on-like-for-like-basis/
It is rare for all regions outside of FNQ to have very low wind conditions, but it does happen. It really does highlight the poor progress with wind development in QLD compared to other regions. While additional capacity in SE QLD wouldn’t have helped this time, certainly more in FNQ, NQ and to a lesser extent central QLD would have helped.
Paul,
surely you must know why the output was so low on that day, surely!
Having keep all the data now for more than seven years, I saw pretty early on why these low power times happen, due to those large High Pressure weather systems hovering over Southern Australia.
Here’s my own article for that day, posted the morning after it happened. (https://papundits.wordpress.com/2022/08/27/australian-daily-wind-power-generation-data-friday-26-august-2022/)
Here you have almost two thirds of total wind Capacity in the one area, South East South Australia, and Central Western Victoria, and every time those Large High Pressure weather systems come across, the bottom falls out of wind generation as they hover over that area. You only need note the synoptic chart, which I might add is always shown at the Aneroid site, so you can quite easily see it as it happens, and even as they approach,, you can tell when it’s going to be low.
And umm, it’s not just an isolated thing.
TWO YEARS ago now, I researched the matter, for the earlier time frame of 800 days, so a little less than three years, and in those 800 days, there were 270 times when wind generation failed. All of that information is shown at the following link. (https://papundits.wordpress.com/2020/11/30/wind-power-generation-intermittency-its-worse-than-you-think-it-is-introduction/)
Maybe this looks like I’m linking to my own Posts to drum up readers. Nothing could be further from the truth as this is from two years back now.
Perhaps this may be of some assistance to you.
Anton Lang.
Paul Miskelly published a paper in the scientific literature in 2012 describing the prolonged periods of very low wind that occur across the whole NEM when high pressure systems linger. It was hoped that a wider spread of windmills would even out the supply but Morgan found the same thing some years later and Mike O’Ceirin looked at a decade of date up to 2021 and found the same thing.
Complete failure of due diligence on the wind supply.
Disaster looms.
https://spectator.com.au/2022/07/energy-policy-where-parallel-universes-are-set-to-collide/
Have a look at Britain and Germany.
We need more Battery’s or bigger Battery’s.
The root of the problem is obvious as we speak, there is a nation-wide wind drought which means in an hour or two there will be next to no RE across the nation and no amount of additional installed capacity will make a scrap of difference.
At present we have no grid-scale storage from pumped hydro or batteries and there is none in sight in the foreseeable future. The absurdly hyped “big batteries” are only useful for FACs, not carrying baseload through wind droughts.
The so-called transition to green energy has gone as far as it can go until nuclear power is up and running. It is time for the public and the politicians to face the facts and commentators in the industry should be explaining this.
Great Article.. “winter evening periods that will be increasingly critical”. The calm cloudy winters which may last for months, like 2010, will outlast any possible storage even the great 350 GWh of Snowy2. It will probably happen again before 2030. I think “Success is by no means guaranteed” is a positive spin.
The resignation of Snowy CEO Paul Broad symbolized the whole problem. In an interview (min 4)
https://www.abc.net.au/newcastle/programs/saturday-breakfast/paul-broad/101379138
he referred to the Energy Security Board’s June 22 report which said every year one Snowy 2 has to be built
A number of factors were probably contributing. Murraylink has been offline for several days and Heywood was severely constrained at the time of your snapshot. Pelican Point has had only one GT running for about a week but was completely offline since 0130. Solar was depressed late morning because of a weather front which has led to transmission line warning market notices for SA. Many of the gas peakers were called on during the day.
Just on the discussion of PASA Availability:
“there’s also a significant volume available to AEMO, but not bid into the NEMDE price forming process (which might be for various reasons):
i.e. 4,444MW (PASA Avail) – 1,411MW (bid into NEMDE at 00:30) = 3,033MW (available, but not priced)”
Most of this difference (about 2,400 MW) relates to large scale wind and solar, for which PASA availability is usually shown as facility capacity before taking into account available wind and sun. However in most cases the difference between PASA and bid (market) availability for semi-scheduled plant is obviously not available to AEMO or to anyone else, because actual output is limited by wind and sun conditions.
Interesting read but much of it I don’t agree with.
1 – The article makes a tacit admission that long term PPAs are a marvellous way underwrite what would otherwise ne highly rsiky investments in renewable generation, but as a means o reducing emissions are in fact a bit of a scam.. (I agree)
2 – The sentence “Once implemented, liable entities (retailers) will need to invest in the roll out of not just renewables, but the storage and DR to flatten it out across the day, across seasons and come flood or drought. Where they don’t invest themselves…..” pretty much gives the game away. Just an admition that generators will no longer be obliged to supply stable and reliable power, and users will just be supplied with what ever power is available from intermittant renewables, and the huge costs associated with dealing with such intermittancy (such as storage, firming or DR) will need to be borne by the retailers and users themselves.
Well retailers are not obliged to remain in a market or operate at a loss, so all I can see happening is most energy retailers exiting the market and investing their finite resources elsewhere, leaving relevant state governments holding the can in yet another energy market thought bubble…
Thanks for sharing your ideas, James.
Back in May 2019 when we released the Generator Report Card 2018, we wrote about what we saw as an emerging schism between:
Service #1) ‘Anytime/Anywhere Energy’, a name we coined to describe the dominant form of support in use for renewable generation – i.e. not focused on the time, or the place, of its production.
Service #2) ‘Keeping the Lights on Services’, a name we coined to describe all of the other services that need to supplied over the short term and the long term to keep the electricity grid operational.
We then expanded on this Schism in GenInsights21 late in 2021.
You spoke about your ideas for a ‘Three Stage Transition’ back at an Australian Institute of Energy event earlier this year, and it sounded like you were talking about ideas that could potentially address some of the limitations of ‘the Schism’ above. Hence they are worth sharing on WattClarity here today.
Paul
James. Great article – I like the simplicity. Can you clarrify how you see storage being treated within this framework? Is storage granted a REC time stamped with the hour it is dispatched and obligated to buy a REC time stamped with the hour of charging? Or are you imagining that storage is a way that a customer is able to shift the time profile of its load (and, hence, move its REC liability to a cheaper hour)?
I don’t think this would be at all simple to implement or manage. Spot prices deal with attempting to value energy when it is most needed already, so this would add to complexity. It’s also unclear what hours are we talking about i.e. which of the 8760 per year do we plan to average? Are there summer/winter/weekly/weekend hours that are different? How are they forecast in advance? What if the market fails to meet an hourly REC target, is there another penalty? Can hourly REC’s be produced in another region? This may also dissuade low cost renewables with storage in preference to high cost renewables producing in the right hour…it makes you want to revert to a carbon price where the externality of pollution is priced and has a spot market that values the scarcity of energy supply.
I think its somewhat accepted now that the term for wind and solar combined is variable renewable energy [VRE]
More contentiously given inter state contstraints there is still value to be had in understanding the position of each region individually.This comment applies as much to Tasmania and Victoria as to Queensland.
Even leaving aside the interconnection limits issue it remains the case that each State has its own policy, for instance QLD has of course 50% renewable by 2030 and this will/should/might impact on investment decisions.
Too many acronyms, David – had a mental blank and forgot about calling Wind and Solar VRE.
Understand your defence of region-based metrics. They have a place, but some overuse/misuse is why I noted “risks becoming…”.
Great summary. Thanks Dan & team
Thanks Jeremy!
Do you have a cost stack graph by region by year from June 2018?
Hi Jennifer, I haven’t done that analysis but you would be able to extract the data to calculate this using NEMreview.
Useful summary of the different elements, thanks. One important distinction between the spot price revenue and everything else is that buyers of energy (retailers/large users) can hedge spot prices. So unless they have rashly left themselves exposed, the net cost to them of energy is likely to be well below the $6.52 bn.
By comparison, the everything else figure of $250m (plus successful claims through the AEMC process) cannot be hedged. While some FCAS costs are part of normal market charges, most of the rest is unusual and will be a sting in the tail for whoever ultimately bears these costs. Who that is is not entirely clear. Roughly speaking, i’d expect that many C&I customers would be on contracts that allowed the retailer to pass through such costs, whereas small customers wouldn’t be (and so won’t directly bear these costs). Whether retailers build in some contingency for similar future events in their market offers in the future, i don’t know. Happy to be told i’m wrong about all this if anyone knows better.
would be interested to know how much the $250m works out as a $/MWh figure. I would guess it’s in the region of a $10-$15/MWh uplift on top of energy costs. That is not insignificant.
It’s interesting that if we were to include the demand drawn by dispatchable loads (hydro pumps and battery recharging), which are excluded from measures like “Operational Demand” and “Total Demand”, then yesterday’s event wasn’t the lowest demand we’ve seen NEM wide or in New South Wales – in both cases 17th October 2021 would remain the record low day. Shows that large scale storage is playing more of a role.
A very useful piece of analysis. Obviously, the main contributor was high spot prices, it would be interesting to know what composed the $6.62 Bn price outcome. Higher demand, plant outages and higher gas prices were the dominant effects, the impact of gas prices seems to be less important unless gas producers were forcing gas bids to be higher by a substantial factor.
The question of what happens next year in my mind prices will be lower but gas prices would still remain high. However, all the planned outages would not go ahead and if demand was lower then market total revenue are likely to be a lot lower.
One might also note, although I doubt if many readers here need the heads up, for solar farms the difference between AC (inverter) and DC (panel) capacity. Solar farms generally have an Inverter Load Factor [ILF] greater than 100% because it is better economics.
How does this affect the activation of RERT contracts?
Looks like the Operational Demand minimum for Vic may have dropped from around 2,330 MW to 2,260 MW.
… or 2,285MW for the half hour ending 13:00?
Perish the thought that journalists or anyone else would want to spread fake news but it is easy to inflate the contribution of humungous batteries by measuring them in MW so we can be impressed by a 500MW battery that is going to substitute for a 2000MW coal power station. Or the total MW of all the big batteries on the ground at present or in the pipeline.
Of course giving the MW number to indicate the amount of storage is like giving the length of an area without the breadth.
We just continue to have a disgraceful level of energy literacy in this country that can be traced back to a couple of generations of people who had little exposure to basic science and physics at high school. It allows much of the evangelical spin put into the public domain about big batteries and how everything can be replaced by windmills and solar panels to be swallowed hook line and sinker by an unknowing populous – oh, and how it will all be cheaper than it is today. If we want to decarbonise then the sooner the facts get out there the better so we can have a sensible conversation about cost and how it will be funded.
Paul,
Great articles. I am on the pessimistic side as well.
I hope there are clever people working on supplying reliable cheap power to NEM.
It is disingenuous to report “new generation and firming capacity” in one number, do we assume that is 12GW of firmed generation during all months, or for 4 hours.
Boston et al (2022)* modeled using 10 x Snowy2 schemes storage and still had significant blackout during Winter 2010.
Interestingly the wind drought June 2016 you mentioned does not show up in their modelling.
Would it be possible to only run Coal power from mid-May to July?
Ref: https://wattclarity.com.au//articles/2022/06/seasons-of-low-wind-and-solar-output-will-influence-the-shape-of-the-nem/
Assumes Snowy 2.0 supply 2GW of capacity but 350GWh of storage.
Of course one should be talking in MWh but nut just MWh produced but worst case. So if Loy Yang A produced 16,000 GWh in a normal year we would need enough wind and solar to produce 16,000 GWh in a poor year or about 17,500 GWh at nominal capacity factors. However, capacity factors are increasing due to better controls, better technology etc. So in 2025 an optimally designed solar farm with bifacial panels etc might average 28% and similarly wind farms with taller towers and large rotors should manage 44%. So, to produce 17,500 GWh the system needs 3.2 GW of new wind and 2 GW of new utility solar. In practice behind the meter solar is still the cheapest power so it will continue to expand even if the Victorian government reduces incentives. So by 2035 at least 20% of demand will be provided behind the meter so to replace Loy Yang A you will need about 2.6 GW of onshore wind and 1.8GW of utility solar. At last count between restricted capacity of plants working up and new plants committed Victoria already had 2.8GW of wind scheduled to be online by 2026 and at least 800 MW of large scale solar for an expected addition with new rooftop solar of about 15,000 GWh/y. So that will easily take care of the production from Yallourn 8,500 GWh/y and 40% of LYA by 2026. Therefore, there are nine years to find the balance. If Star of the South is built, it alone should supply 9,000-9,500 GWh per year and ongoing additions of 600MW of btm solar will provide another 6-7,000 GWh. In sum if the only new large scale renewable plant that is built after currently financed plants are completed is Star of the South, Victoria will add 30,000 GWh/y of capacity and close 24,000 GWh. Now it is possible that no offshore will be built, but the idea that there will be little or no new large scale renewables contracted in Victoria for 13 years is completely unrealistic.
Appreciate the careful analysis here Paul.
The problem with trying to replace coal power with wind and solar is that on windless nights wind and solar deliver nothing. End of story.
https://spectator.com.au/2022/07/energy-policy-where-parallel-universes-are-set-to-collide/
The real game in town is finding a way to keep the coal fires burning until nuclear is up and running. Watch that space!
Britain and Germany are demonstrating the alternative.
This may pose a problem or small delay for callide.
Callide Power Station’s C4 generator stuck at Gladstone’s Port as CS Energy and TMR seek alternate route amid unsuitable weight bearing bridges
Roads and bridges deemed unsuitable to carry heavy loads are being blamed for a delay in transporting a generator to Callide Power Station to replace one that blew up, cutting power to thousands of Qld homes. Here’s what we know.
Nilsson Jones The Courier Mail
October 10, 2022 – 2:06PM
A vital replacement generator destined for the Callide Power Station is stranded at Gladstone’s Port as unsuitable bridges have left authorities “scrambling” to find an alternative route.
A 250 tonne generator set to replace the c4 Unit at Callide Power Station — following its destruction during an explosion in May 2021 that impacted thousands of Queensland customers — remains at the Port of Gladstone as the access bridges have insufficient load capacity to enable transport to Biloela.
Flynn MP, Colin Boyce confirmed to The Observer on Monday morning the generator had been unloaded at the Port but transport issues meant CS Energy, which operates the power station, and the Department of Transport and Main Roads were planning to relocate the generator further along the Port.
Mr Boyce said authorities were planning to load the generator onto a barge and move it further north along the Port to Fisherman’s Landing, about 17km north near Yarwun.
“The problem here in Gladstone is we have a port access issue that has been here for many years,” Mr Boyce said.
“The generator is sitting on the wharf but we can’t move it anywhere because of the bridge and road infrastructure which is not suitable to carry 250 tonnes of generator across it.
“Everyone has known about this for a long time and now they’re all scrambling, wondering how they’re going to get it up to Biloela.”
Mr Boyce said decision would have to be made regarding which route would then get the generator to Biloela which “itself had its own problems” with roads and bridges on the corridors leading to the Callide Power Station.
Thanks John … relevant here:
https://wattclarity.com.au//articles/2022/10/12oct-callidec4-rts-unchanged/
Very interesting and thorough analysis of contributing factors. I wonder if the prolonged period of low Solar generation prior to the June event, led to a draw-down of low marginal cost storage supply (ie. Big Batteries and Snowy 2.0), which then led to the perverse withdrawal of bidding to ensure a more ‘certain’ compensation regime when ‘directed’. This in turn led to profit maximising bid behaviour.
It seems that if we want to prevent further ‘inevitable’, but relatively unpredictable similar events. Considerable investment needs to be made in storage.
As a pro-sumer, it seems shocking to me that the regulatory agencies largely ignore Rooftop Solar DER in capacity planning, incentives, etc. and seem only to be considering ‘centralised’ options for planning, storage and transmission.
Further, this seems a perverse and unintended consequence of not pricing emissions, nor kWh, but only Watts.
It would seem to me that like the internet, a more distributed and standards based solution is the most economic, scaleable, resilient, and reliable. One might even argue it will be more scaleable and lower cost to build.
This will indeed be a very interesting incident report to read when it eventually appears more for the things that went right. It would appear Tassie had a very real “near death” experience. The simultaneous loss of what really is the backbone of its North / South bulk transmission system must have come very close to causing separation of the island into two parts. The line faults themselves (let me guess, a double circuit tower line) will have cause a voltage depression and hence loss of some load instantly, but clearly a major automatic underfrequency load shed has occurred probably as a result of the loss of the significant Basslink input to Tassie. It appears that the Hydro machines rode through and responded under governor action as intended (primary response). Hydro units by their physics will immediately reduce export power (accelerating the water mass in the runner) for an increase governor response hence perhaps the small decrease in frequency after the UFLS recovery but then got on with it. Nice job! Must have been a bit hairy for the operators I suspect.
Andrew, I agree that the system did well to survive what was reported to be a tower failure (due to landslip?) on the main N-S 220kV double circuit line. The large scale load shedding may have been primarily operation of the FCSPS which trips specific large loads on trip of Basslink when importing (rather than frequency-driven), but may have involved some UFLS as well. Allan.
The reason for the “plug being pulled on curtailment” (double negative!) is that a prior outage on the Heywood interconnector to Victoria was completed ahead of schedule at 1025 (AEMO Market Notice 102382). This outage was the underlying cause for heightened security risk levels in SA that prompted AEMO to consider PV curtailment.
This is ideological nonsense based on neo-classical economic models. The simplest way to price electricity is to to take the average cost-plus a standard rate of return of return and retquire providers to supply on those terms on annual contracts. The facilities should be publicly owned. The idea that these are necessarily inefficient is just based on ideological observations by Hayek. The idea of letting providers opt out of supply is absurd for provision of a necessary social good. The idea of marginal cost determining price is absurd for a good where there are few providers. The “rent” for such a policy is claimed by private providers under NEM and with a high marginal cost for gas it means that provision is made at enormous system wide cost. My belief is that a courageous government needs to drop the ideological nonsense of neoliberalism and r
go back to public provision of socially necessary goods. Whatever the ‘inefficiencies” of this, they are nothing compared with the inefficiencies of having a small number of retailers, with each supporting a superstructure of trading administration supplying complexly different energy plans to the public.
Good luck with building any form of temporary 220kV transmission line in this terrain. There is a reason why even the 110kV is on steel towers. The real big one will be the loss of stable ground to build foundations in. I suggest this will take a while to sort out.
The different frequency for north vs. south Tasmania may reflect an oscillation between the 2 generation blocks especially if the lost circuits reulted in a higher impedance link between the 2 areas as well as different inertias and governor responses.
Loy Yang A2 is ramping this morning (Tuesday).
Nearly there, Alex, as now noted here:
https://wattclarity.com.au//articles/2022/10/29oct-lya2-false-restart-now-eta-31oct/
The frequency in Tasmania is always separated from the mainland because there’s only a DC link, it’s not synchronised to the rest of the NEM.
Excellent article. Would like more info and wondering how to contact author
Quite a lot of negative prices. Means renewables getting a hit.
My comment about a possible frequency oscillation within Tasmania assumes that BassLink was still out of service given that it tripped during the initial fault disturbance. Another possibilty is the timing of any automatic reclose of the 2 faulted transmission lines.
Interesting work… we have also published work on this that might be of interest.
Boston, A., & Bongers, G. (2021). MEGS: Modelling Energy and Grid Services to Explore Decarbonisation of Power Systems at Lowest Total System Cost. Energy Strategy Reviews, 38. https://doi.org/10.1016/j.esr.2021.100709
A pity the Griffith report talks about “average”, electricity is all about the instantaneous. Average wind and solar will not give a stable electricity grid.
So we have over 8GW of installed wind capacity across the entire eastern seaboard but we can see total output drop to just 250MW and stay low for quite some time.
This seems to fly in the face of the assjmptions that adding more wind farms will increase the minimum ouput during a wind drought
Not necessarily, because new wind turbines perform much better in low winds than older designs, however it is true to say that even with five times as much wind capacity and better turbine designs and wider distribution eg offshore and eastern Victoria, minimum wind output at the time would probably have been only about 2 GW, however as is hinted at in the article if the NEM has 4 times as much solar which is pretty much certain, solar would have supplied between 120 and 200% of the demand between 7:30 and 4:30
Hello Peter,
It seems to me that solar is the way to go, and forget wind. Things that move are much harder to look after and much more costly.
At best the capability for solar is 12 hours, often much less.
Solar can only work in conjunction with significant storage which ends up being way more expensive than thermal plant generation.
Hello V,
Thermal Plant generation is not allowed out in time, it must be from what is deemed Renewable energy. Solar and Bats or solar, bats, and hydro or Solar,bat,hydro, and hydrogen. Is what will end up as the used power supply. Wind just has too much maintaining. There is also Nuke, but too many don’t like it
This will be celebrated by the RE community but it is very seriously misleading, like the new Renewable Penetration tab on the AEMO data dashboard.
The green transition is just a massive waste of money until there is some way to lift the delivery of RE on windless nights. As long as these gaps in RE supply persist we will need all the coal and gas capacity that we have at present to keep the lights on.
Pity about that, because it means that prematurely connecting intermittent providers to the grid will be seen as the biggest policy blunder in our history.
It is interesting comparing the 2 papers: Boston & Bongers 2022 and Gilmore 2022.
I have used both MERRA and NCEP RR2 in hindcast studies. Recently I tend to use the ERA5 data, it is finer resolution and contains a 100m wind for wind studies, a height more consistent with modern turbines.
There are a few comparisons papers of MERRA, RR2 and ERA5: https://asr.copernicus.org/articles/17/63/2020/
However both these papers seem to produce reasonable calibrations against recent measurements.
For storage, I prefer the Bucket Model approach taken by Boston.
This clearly show the curtailment issue in Spring/Early Summer and blackouts in Winter with their scenario of using 10 x Snowy 2.0 schemes storage.
Boston do not address Ramping which may be a problem with to many 2GW/350GWh Snowy2’s.
The shorter period of modelling by Boston may not be representative. The 42 years of Gilmore is much preferred.
The 15-year used by Boston doesn’t show if how unusual Winter 2010 Dunkelflaute was.
Interesting both papers note that 17 May 2010 was a bad day. Boston shows this continued for the whole winter and the storage never recovered.
Boston model assumes perfect transmission. Tasmania was not included in Gilmore’s work, it’s pumped hydro will be critical storage.
Boston discussion for the use of offshore wind farms with their time-shift of wind generation and increase CF was very useful.
We need more of these long term studies done to ensure the reliability of the NEM.
Looking at the curtailment are we reaching peak Solar?? I think storage has become the main issue.
I am certain clever people in AEMO are looking at all these issues.
Dave Duncalf
Thanks Kate, Brilliant summation as always.
Bravo Kate – I’m sorry I missed this at All-Energy!
I believe you are destined to be eternally disappointed, because the bureaucrats are in charge now, and they won’t let go without a fight. They have targets to meet you know.
“Where has all our engineering gone?” – indeed.
We often hear from credible experts on scientific, medical, legal, economic, and even defence topics. Yet when it comes to a sensible and informed discussion on powering our national electricity grid, the best the media can manage is politically loaded partisan mantras and slogans, overlaid (or is it buried) within the religious imperative of respecting climate change as undeniable. Who must take responsibility for this monumental ignorance of the community at large.
Can confirm that the SA network voltage was raised for approx. 4 and 1/2 hours just after 12.30pm.
Our monitoring systems alarmed with a voltage increase of around 200v on each phase
Re supply authorities ramping up voltage to trip inverters and Jim’s comment of of a very large increase at his place. I’m wondering if that increase in voltage that Jim commented on was natural. Don’t forget that at the household end of the grid, each suburb has a final distribution transformer, stepping down (usually) 11KV reticulation to 240/415volts. If the energy of all the suburban panels at midday can’t be used up by a load, then a higher voltage will naturally occur. 240v rising to 270v is not out of the question (and this will trip inverters), but 200 extra volts seems very large and would fry many household appliances. Also I find it hard to understand how the SA power authority would easily be able to increase voltage. Most final suburban transformers can only increase voltage with a manual change-over of their output tappings. It’s also very dangerous to increase voltage on a grid scale as things will burn out. Lastly voltage levels are set by the AS 3000 Wiring Rules, and must be no greater than around 10% of normal.
The administered price cap applies to all FCAS commodities not just Lower Reg.
Thanks for the correction Lisa. I will add in a note to the article.
Thanks Dan. Good analysis, but I’m not sure I’d agree with the headlines. These frequency traces seems remarkably good compared to previous separations. It would be worth comparing to reports on them in AEMO reports.
Separation circumstances are always different, but I suspect we are seeing the beneficial effect of fast acting inverter equipment.
Thanks Ben. I will post a follow-up when the AEMO publishes a report on the incident, and try to compare to previous separations for a full picture.
Hi Paul,
Your comment “the Murraylink interconnector is still operating, but it is a DC interconnector so there’s a limit to the benefit it can supply to South Australia”
Why is a DC interconnector limited benefit?? Any reference would be appreciated.
Thanks
A DC connector can not provide frequency control services. It could provide some MWs, but otherwise SA’s frequency was completely separated from the rest of the NEM, same way Tasmania’s frequency is always separate. Tasmania’s grid is already more unstable than the NEMs.
Don’t have any references, just my guess.
I look forward for your review on the day in particular the constraints and synchronous generation dispatched during the day. It was surprising to see Dry Creek peakers and diesel thrown into the mix.
My pv has been constrained to internal demand only until thursday when it was completely cut off to zero.
This was frustrating as obviously generators had a good day at my and the publics expense.
No market operating and no feel demonstrated by the powers in charge to allow home pv to cover in home use. This was a planned action to limit pv should have been done with more finesse than it was as the circumstances of too much non scheduled will only increase in frequency (no pun intended!)
Imagine supporting the development of an energy sector where you literally have thousands of megawatts of supply just evaporate or appear out of thin air and thinking this is somehow a good solution.
We are going to get more and more frequency events creating islanding like SA saw last week because of the unreliability of renewables.
Wonder how both models performed on the night of Wednesday 7 September 2022 when South Australia “lost” 433 MW of wind generation that didn’t arrive as forecast at 11 pm.
Dan that is really insightful
Thanks for covering this Dan. Just noting that the Reliability Panel has completed the 2022 Reliability Standard and Settings Review and recommended an APC of $500/MWh for the period 1 July 2025 to 30 June 2028 (along with recommended changes to the MPC and CPT).
The Reliability Panel will now need to submit a rule change request to the AEMC in order to implement the recommended changes to the settings. The AEMC would consider any proposed changes through the usual rule change process, allowing further opportunities for stakeholder input and consultation.
Thanks for the additional information Ryan
One thing I don’t understand is how the Hornsdale wind units can provide FCAS services. That seems to be a contradiction in terms. Gas and batteries make sense, but wind units? Can you explain?
Hi Paul, the three Hornsdale wind units are registered to provide six of the eight FCAS services (Lower 6 Sec and Raise 6 Sec contingency the exceptions). The document below describes the trial process that was undertaken in order to allow them to do so:
https://aemo.com.au/-/media/files/electricity/nem/strategic-partnerships/2018/hwf2-fcas-trial-paper.pdf?la=en&hash=9C9EFF402D42FD6170A0967AE877BE11
Any wind or solar farm that has suitable inverters and control systems can register to provide FCAS. Currently my understanding is that they are only providing a service by quickly cutting their output, but having an output like a big resistor to dump load into, could also provide a similar service.
However there is no reason that a wind or solar farm couldn’t de rate their output, thus holding some capacity in reserve to deliver immediately. To date I don’t think AEMO is registering this later form of capacity as it is considered more risky than having a battery operating behind the inverter.
Thank you for sharing. Newbie here.
AEMO forecasts a need for “the equivalent of up to 40 large synchronous condensers” provided for by a “range of technologies” to sustain 100% instantaenous penetration of renewable energy. There are just 4 synchrons in SA today. I haven’t seen this mentioned anywhere. What sort of technologies could provide this service? Does this imply massive investment in synchrons?
Paul Hyslop’s article like many others, including unfortunately some of my own, focusses on the 6300 kcal/kg coal price which is (i) extremely high relative to history and (ii) very visible because there is a futures market. However it is not I have read a highly traded contract and possibly not a good reflection of the lower spec 5500 kcal coal that forms the bulk of sales to Asia and which is, in an even less washed form, what is mostly used in the domestic market at least in NSW. The lower spec coal was recently “only” US$136/ or say A$200 and at 23 gj/t is $8.7/gj and at a heat rate of 9.5 gets you to a fuel SRMC of “just” A$82.6 although that is pre transport. Using the “proposed” and no doubt soon to be legislated cap A$125/t and assuming probably incorrectly the same number of GJ/t gets you to say about A$55/MWh + transport and other costs. Not sure what freight rates are at the moment
It’s not hard to see where this is all going. The AER et al are now operating beyond their core function, by interfering to promote wind and solar. Regardless of one’s position on the reasons, the outcome is entirely predictable – increasing market intervention until the government owns it all.
The media release is incredible, with three ‘recommendations’ including:
– contract monitoring
– increased regulation of coal-fired generators
– where governments have choice over their investments or underwriting, consideration should be given to actions to diversify the operation of dispatchable generation
The AER admits that hydro is setting high prices more often. But governments own the hydro! There’s no price caps on hydro being mentioned!
Let’s look at NSW post-Liddell. With 1300 MW needed to be supplied by the remaining coal-fired generators, they need more coal. But if it’s not already contracted, who’s going to sell it to them under the price cap?
The only way that will happen is if Matt Kean intervenes again and takes the power to curtail exports for local use. That’s a coal reservation.
If not, then the power will come from gas. But Bowen stuffed that up too, so the feds will need to implement the DGRM.
In both scenarios – forced export curtailment – the producers will be compensated. By taxpayers. So we pay anyway, but this time in national debt and less opportunity as resource companies say ‘no thanks’ and gradually pull out. Which is exactly what the Greens and co want.
As private companies start making noises about offloading projects, and politicians realise these power stations and mines and gas projects are needed to keep the lights on, politicians will spend public money to nationalise it.
Which is exactly what the AER has recommended ie diversify government investments into retiring coal to keep it online – Yallourn is already there, Eraring will be next. I’ll be surprised if Liddell doesn’t get a lifeline.
It’s a horror movie, with no end in sight.
High prices have been a function of several things, mostly fuel and generation scarcity. The AER’s price-setter data shows hydro is increasing wholesale prices far more than coal and even gas. Hydro is not affected by Putin as far as I can tell.
Why the focus on coal, apart from ideology and the narrative?
A list of Basslink failures would be great!
Given the distance of Cressy from Heywood and Portland the SA island resulting from the 31/01/2020 incident should probably be listed as the SA & Western Victoria island which raises the question of whether SA was able to supply the smelters at Portland and what were the possible/ likely consequences if not. Was SA reliant on renewable sources during the event and how did they perform? What fossil fuel sources were available to assist and how did they perform?
SA was, with the assistance of one fossil gas based generator in Victoria, able to cover the needs of SA and the very significant load of the Portland smelter. Due to the more limited options and the conservative approach of AEMO, many wind and solar farms were direct to shut down. At the time of the event the Hornsdale Power Reserve (Tesla Big Battery) was smaller and not capable of operating as a grid forming inverter (synthetic inertia). Nor where any wind or solar farms. The SA grid now has much more grid stability capacity from at least two batteries and one wind farm, so when SA separated recently far more wind and solar farms were able to keep operating. There is another large grid forming inverter based battery under construction in SA and others planned, so in the future there will be little to no need for fossil fuel based generation in SA to provide grid stability services even during a separation event.
Thanks Dan for hunting down all these transmission fails into one reference.
Re Basslink – I think I read at the TasFinTalk blog that Basslink has never paid its way.
The NEM currently operates in an old style grid where the timing signals for the phase angle and frequency is transmitted via AC transmission lines. This makes sense when the grid is dominated by large synchronous generation like coal and gas fired generators. However it is not the most accurate and creates greater risks when separation events occur. For years now it has been more accurate to have phase angle and frequency coordinated across large grids using atomic or similar reference clocks to drive grid forming inverters. Such as system has no issue with regard to frequency control when separation events occur. The cost of these highly accurate clocks and referencing systems have come down significantly to where they are less than a few thousand dollars, so can be installed at all sites in the NEM with grid forming inverters.
Well done Kogan Creek operations team, a cold start after a long outage is always a challenge!
My recollection of the Callide C4 debacle is that both dedicated and the Backup shared 120V DC supply busses failed (or were turned Off). All super-critical plant control functions were inoperable. Critically, the turbine could not be disconnected from the Grid. (The critical OCB is in a substation about a kilometre remote from the turbine.) Unlubricated bearings then overheated and seized, progressively increasing the load drawn from grid until the Grid collapsed some 45 minutes later. A truly magnificent balls-up!
ABC news website had article on 26th May 2022 entitled “One year on from Callide Power Station fire, answers are still being sought”. This mentioned the report on the forensic investigation by Dr Sean Brady would be likely be finalised later in 2022.
All very clever and all very well, but what happens when the millions of roof-top solar installations reach “end of life”?
Is anyone suggesting that taxpayers will again be on the hook to subsidise their replacement – as was vital in the original take up phase?
I take issue with the premise that “they might be able to set “dynamic” limits on households for the import and export of energy.”
From this simple idea and assuming that both electricity providers and governments are not passive actors in this scenario. It won’t be long before you get a text message that ‘your electricity use has exceeded the limit for this time of day, you will be fined $XXX, if it continues further action will be taken.’
This is a neat way to ensure that our electricity network will be protected due to huge fluctuations in supply.
It looked to me as if Wivenhoe pushed QLD over the cumulative price threshold. Interesting since there has been lots of rain so water is not scarce.
I suspect pumping costs were high, therefore the station bid its output high to maintain a satisfactory arbitrage.
Spells danger for a system reliant on storage.
Are you speaking about Sunday 12th June 2022, which was the last time QLD exceeded the CPT … or some other time?
If so, not sure what that has to do with compensation for Gladstone PS over the coming year?
Any News on the report into the failure event?
Nothing yet under ‘Media Releases’ on the CS Energy site, Bruce … will be keenly read when there!
On the “uprating” of C4 to 466MW, it is possible to do with some quite noticeable loss of efficiency by bypassing quite a few of the heaters, which is what was done with C3. Effectively running the unit in permanent “overload”. Given the likely remaining (economic) life of the units, pushing them hard for what time remains probably makes sense.
Are there going to be unforeseen ramifications from the triggering of the RRO in South Australia from 12 January to 13 March 2026?? One South Australian MLO generator has about 900 MW of gas-fired generation of which about 40% are Open cycle gas turbines (more expensive electricity than combined cycle gas turbines) and on current advice to AEMO their OCGTs will close in 2030. Of the MLO’s OCGTs about half run on gas and the rest on diesel. How can a South Australian MLO with OCGTs due to close in 2030 offer contracts for March Quarter 2026 with uncertainty over gas supply and gas prices and Project Energy Connect due to be operational in July 2026? Will the $19.5 million deal between the South Australian Government and the other South Australian MLO be enough to keep the lights on in South Australia?
At the risk of adding more complexity, a better way of assessing curtailment or economic offloading is to just consider the difference between forecast availability and target output. If there is no forced curtailment, or offloading due to offer price, then target output will equal availability. Differences between target and actual for whatever reason – poor forecast or something else – shouldn’t go into calculating curtailment or economic offloading.
Thanks Allan!
Thanks for your insights and additional analysis Paul! These numbers came from our (ANU Battery Storage and Grid Integration Program) submission to the ESB’s “Transmission Access Reform Directions paper”. The analysis was done by my colleagues Jack Sorensen, Marnie Shaw, and Alix Zeibell.
https://bsgip.com/news-events/news/curtailment-due-to-congestion-whats-the-current-state-of-play/
If we followed Allan O’Neil’s methodology of assessing curtailment as the difference between forecast availability and target output wonder what the result would be for these two solar farms.
Imagine a summer where green hydrogen vessels are lined up at east coast ports and electrolysers need to operate all day every day to supply the export green hydrogen. Politicians are speaking as though this will be Australia’s reality pre-2030. Add to this demand from EVs and desalination plants and what will be the NEM peak demand? How much pumped hydro and batteries will be operating pre-2030? The alerts may be coming in thick and fast and not just in summer months.
Much of the green hydrogen and related products are likely to be off grid. That is even if they are partly connected to the NEM, the bulk of the energy will generated and consumed outside of the NEM. I notice that the QLD owned transmission companies have been trying to at least obtain a small cut of this potentially large pie. I expect that we may see private transmission lines in a few locations to avoid some of the costs and regulation related to the NEM.
Having had experience with a similar structural failure of a mechanical draft cooling tower many years ago the timeline does not surprise me too much. Even though the tower is not that old as these things go, the more they will have looked the more cracked and rotten timber they will have found (yes these things are big timber structures) and trying to remove / replace / reinforce damaged timbers in a very constrained space says it is a very slow process. Particularly when you are coming at it from a standing start. If you knew this was coming at you and you had a complete kit of timber to hand it is easier, cheaper & quicker to pretty much demolish and rebuild the timber bits and install new fill, but as a break down job it’s just hard work.
It was a very nice article, but nowhere did you actually state *what* the constraint was. It would be nice to know the simple root cause.
As noted in the article, Michael, the ‘N>NIL_94T’ constraint equation was one of the most significant ones. Obviously with the ez2view software we can drill in further … but that would be a separate (and longer) article.
Bit ironic that on 3rd Feb this year consumers could be facing yet another very expensive RERT bill. On 1st Feb last year AEMO forecast a Queensland demand over 10,000 MW and it cost Queensland consumers a whopping $50 million for the period 1700 hours to 2130 hours.
I am not sure why the demand is forecast to be so high. Yes it will be hot and humid, but nothing that is even close to a record. It just looks to be on the higher side of average for SE QLD at this time of the year, not very pleasant.
RERT used to cost less than $5 million per event but now it can be more than 10 times that for a single RERT event…. all ultimately paid for by consumers. This is all happening before the first coal generator retires in Queensland and BEFORE desalination becomes Southeast Queensland’s new source of drinking water. Imagine being dependent on desalination (liquid electricity) for your water supply. Probably why Dubai doesn’t run on solar!
Tube leak as in high pressure steam boiler tube? Or something less nasty?
The weather conditions didn’t seem anything like a record, so it would have been disappointing if a record Market Demand had occurred. The Market Demand was only 3% or so short of the record. There are so many variables in the mix for these records, weather, changes in work patterns (more people working at home), fewer international students and lower tourist numbers than pre Covid.
There does appear to have been a better level of self consumption of rooftop solar than there was under similar conditions say four years ago. This is a good result for the grid as a whole. Further improvements would be obtained if all residential customers were moved to a wholesale linked tariff.
Maybe also customers are getting smarter and using simple timers and manual interventions to minimise peak demand. Peak grid demand this summer has been 29.8GW well down on previous years
It would be interesting to know if, and by how much, Energex triggered the demand management on residential air conditioners. There would be significant number of air conditioners in SE QLD that Energex can ramp down the compressors on. This program was/is an opt in program with an up front incentive paid.
Hi Craig and Peter. Thanks for the comments. It appears Energex’s PeakSmart program was indeed activated on 3 Feb. Participating AC units were capped to operate at 50%. This would have definitely helped.
Hi Paul, you are right. This is obviously an expanding frequency oscillation although luckily it does not break the +/- 0.15Hz deviation. A few observations: 1) disturbance always exists in the system, but usually well damped. However, with the high load, the damping is often weak. The system does not look like well damped in this case especially around 13 o’clock. It would be interesting to see the voltage profile. Usually they are decoupled, but not in the case of weak damping. 2) The skew to the lower frequency indicates in average the load is increasing, thus although it happens in the mid of the day, the rooftop solar should not be scolded for this expanding oscillation. It is more likely that the disturbance from ramping up the outputs of the generators and even switching in some generators both introduce disturbances and can cause this oscillation. After the transients of ramping up the generation, the system should settle down with much smaller deviation. 3) The oscillation looks like has a period of 10-15minutes (say 4-5 cycles between 11am -12pm). Does that time scale also roughly conform to the time scale of ramping up the generation and the AGC? Highly likely. Anyway, thanks for sharing and it would be interesting to use these observations to validate the models used for system frequency studies, but of course hard as the non-transparency of the control model, which is fully understandable.
Jin – you may be interested in a deeper dive into this I posted here yesterday
https://wattclarity.com.au//articles/2023/02/investigating-last-fridays-frequent-frequency-swings/
Allan
Thanks for the analysis Linton. A quick query for you and the other smart folks at WattClarity. It seems that the spot price peaked at $1,297/MWh, well below the market price cap (MPC). And this was only for one dispatch interval (DI), before RERT was activated. During the period of RERT activation, the spot price barely topped $300/MWh. This implies there was in fact more supply available which surely have been a lot cheaper than RERT. I appreciate AEMO has to make the call on RERT in advance and that we don’t know yet how much they paid. But is this the lowest price (outside of market suspension or CPT situations) at which RERT has been activated?
Kieran on a very quick scan it looks like there was a RERT event with lower outturn prices in Vic on 30 November 2017, and another on 30 Dec 2019. Quite possibly there are others as well. As you point out AEMO has to make the call in advance.
Pricing under RERT uses intervention methodology and doesn’t perfectly reflect what might have happened without RERT being called – I’m sure that’s not an issue last Friday, but may have been on those earlier episodes I mentioned.
It would be interesting to know if in the most recent case AEMO had other RERT options ready to go with shorter lead times, which they were able to avoid activating when demand fell short of forecast. The 21 MW dispatched last Friday seems hardly worth bothering with by itself.
Regards
Allan
Thanks for your detective work Allan!
South Australia, today 9th Feb approx. 7pm AEMO-time, spike of $15,500/MWh. Should we be surprised?
Coincidentally, as noted here, Scott:
https://wattclarity.com.au//articles/2023/02/09feb-spot-volatility-and-an-actual-lor1-in-south-australia/
In terms of ‘should we be surprised’, there’s several levels to your question.
In our widely read GenInsights21 report we illustrated how there were a number of drivers towards increasing volatility into the future. Hence the answer’s ‘no’.
To a great extent this seems like a truism: the cheapest generation isn’t available at the most expensive times, but that’s precisely why those are the most expensive times.
Pretty interesting outcome last Friday in QLD I thought. I was wondering if some of the demand degradation might be driven by tariff 31 and tariff 33 whereby retailers can turn down residential load via heat-pumps and other energy intensive devices during the peak?
SA and VIC prices are mostly hard to work out. SA has had the diesel gen sets pumping out over 300Mw for well over an hour yet the price is only $500. One would think to get any kind of return on diesel gen sets you would need a bit more than $500.
Also of note, almost everyday SA have gas running, yet the price is often very negative. They must have cheap gas somewhere. Maybe they have found a new kind of Gas turbine and finance system?
SA prices are currently $4000. All fossil have been online.
Gas running when prices negative are because of AEMO system security requirements to have two synchronous generators in SA. The minimum output from gas is 80MW.
Hello Alex,
So does that mean SA does not operate like all other markets and gas because it has to run does not get costed in.
As I understand it, the last bid for the market set the price, this cannot be happening in SA unless gas is not bidding a lot of the time.
The cost of FCAS (grid stability services) is not factored into the wholesale spot prices, but is applied through the additional costs similar to the transmission and distribution costs. The amount of standing fossil gas operating in SA has decreased due to the relatively new synchronous condensers providing that service at a lower cost. In the future there will be no requirement for standing fossil gas generators to provide inertia to the SA grid as this will be provided at a lower cost by big batteries and the EnergyConnect interconnector to NSW.
So the cost of something that is deemed to be needed is not costed into the cost to run the grid in SA, yet we are fed thats its wholesale price is much less than anywhere else. Hmmmmm.
The SA Big Battery was not helping much. It took me a while to find it in the plot (top right purple).
SA was importing 600MW (23%) at the time.
The Big Battery (HPR) is mainly there to provide grid stability services, so if one of the other generators falls over or the interconnector fails, then it can keep the system stable while that generator or transmission line with the problem either comes back online or another generator fills the gap.
When the additional big batteries come online there will be additional capacity in the system to provide more battery supply during evening peaks.
The diesels should achieve about 36-38% efficiency so 3.9 kWh of generation/litre at a tax free price of $1.40 that is 35c/kWh or $350/MWh + $10-15 in other variable costs, $500 is a good contribution to overheads
Hi Peter,
I agree, if you owned it and you got that for it every day for a few hours maybe, but when it sits around for long periods and only runs every now and then, $500 is not what it needs. Very miss priced, not reflecting the actual cost to have such a thing, in your energy mix.
On Thursday there were periods where the diesel generators were receiving more than $4,000 per MWh. The standing capital costs of diesel generators aren’t that high, nor are the maintenance costs. It cost is all about the fuel. It doesn’t take long at $4,000 per MWh to cover the standing costs for the year.
Hello Craig,
I would have thought things should get run to make money?. Are these Genset’s there for simply a backup and to make sure SA has power?.
If that is the case its yet another thing thats very different about how the price in SA is being arrived at.
A lot of this talk feels like re-arranging the deckchairs on the Titanic because there appear to be three principles that constitute something like an Iron Triangle of Power Supply.
1. The grid needs continuous input to meet 100% of the demand.
2. Nights with little or no wind break the continuity of wind and solar input.
3. For the foreseeable future there will be no effective way to store enough power to bridge the gaps.
Conclusion: We need to keep enough conventional power, mostly coal power, to meet the highest levels of demand at dinnertimes in high summer and deep winter, until we have nuclear power on deck.
In the meantime people need to appreciate that wind and solar can DISPLACE coal but they can’t REPLACE it.
https://www.flickerpower.com/index.php/search/categories/renewables/21-7-intermittent-solar-and-wind-power-can-displace-coal-but-cannot-replace-it
The rate of exit from coal is not accelerated by increasing penetration on good wind and solar days, it is limited by the lowest level of output on nights with little or no wind, as a convoy travels at the speed of the slowest vessel, the water penetrates the levee at the lowest point, a chain is only as strong as the weakest link and stock get out of the yard through gaps in the fence even if the rest of the fence is built to the sky.
South Australia has proven you can replace coal with renewables. But getting rid of gas is more problematic.
On the NEM, existing hydro is also available to help meet demand during times of low wind & solar.
You might want to look at AEMO’s ISP to see how we can get nearly all our electricity from renewables. Or you can look at my own simulations here:
https://reneweconomy.com.au/a-near-100-per-cent-renewables-grid-is-well-within-reach-and-with-little-storage/
South Australia that has to import fossil fuel power from Victoria to keep the lights on and when they can’t they have a week long blackout?
Not a great example.
The concerns are over-blown. Does anyone seriously think parts of Australia will be facing blackouts in the 2020s and 2030s?
I mean seriously. Of course not.
This is overblown fear-laden rhetoric based on nothing.
Both Tasmania and South Australia had a week long outage last year and things are only going to get worse as fossil fuel generation is removed from the sytem and replaced with unreliable renewables.
SA did suffer a transmission outage due to a downed tower near Tailem Bend but this did not lead to a supply outage.
Funny how these outages are becoming more common as we demonise domestic fossil fuel generation and rely on importing fossil fuel power over significant distances from other states.
https://www.abc.net.au/news/2022-11-14/sa-blackouts-compared-to-2016-electricity-cut/101649566
Why was Market Demand at a 9 year high while Operational Demand only hit a 3 year high?
I think the reason will be those non-scheduled wind farms that now have to bid and be dispatched like their semi-scheduled cousins. That change meant that their output is now counted as contributing to meeting Market Demand; previously their output effectively subtracted from the grid demand met by market-dispatched sources.
In other words there has been a basis change in the last 18 months that increases the measured Market Demand.
On the other hand the Operational Demand and Demand and Non-Scheduled Generation measures have always included demand met by these non-scheduled wind farms.
Good pickup, thanks Allan!
Guessing this should say “winter 2023”
This does leave (at least) winter 2003 where Liddell will be gone and this new gas-fired dispatchable capacity will not be available. Currently, AEMO’s MTPASA medium-term supply-demand outlook is not highlighting any reliability gap over this period, noting that the ez2view chart below does not explicitly include import capability into New South Wales via the Queensland and Victorian interconnections:
So now let’s take out the 650MW being contributed by the two running Liddell units for next summer and you are in LOR2 territory……and if the contingency then is loss of the interconnector to Qld and things start to get a bit dark. How come there are other, more modern and efficient plants (BW & MP) off in the first week of March? This is starting to look third world stuff. This slow-moving train wreck will not be arrested quickly either. Ironically, the last time NSW was on the cusp (and actually in) this sort of disaster over summer was in the 1970’s when the (then) near new units at Liddell started having generator failures. Market price signals only work when there are some generators left to respond to them!
None of this is terribly surprising. If the asset managers in the stations with announced closing dates are doing their jobs they will have started rationing expenditure on maintenance (both routine and particularly sustaining capital) so that the units are kept safe but will perform a graceful bellyflop on their last day of operation. This means you can expect increasing unavailability and particularly forced outage and derating where you start to increasingly operate on run to failure / breakdown maintenance. If you have no interest in the longevity of your units then this is the cheapest way of doing things. Just guard against catastrophic failures that can kill people and off you go. People outside the industry often don’t understand that a cheap little component can quite easily stop one of these big units. Given that these units have maintenance cycles measured in years and this is the reason you cannot easily change direction once you start going down this slippery slope. Plant engineers have been doing this for decades when they move to retire old power stations – so they know how to do it. What is different this time around is the process is being undertaken across many units simultaneously – many of which would normally have had many decades of operating life but simply they have had their revenue and hence commercial viability destroyed. Inevitably will end in an untidy outcome.
Further to Andy’s comment NSW has nominally 9,500 MW of coal after accounting for the derating of the remaining three Liddell units and yet peak coal output was 7,017 MW. It also has 2 GW of gas which only managed 1.2 GW, not to mention 4.1 GW of hydro which maxed out at 2.1 GW. Combined output from 15.6 GW of dispatchable capacity did not exceed 12.1 GW, and 11.5GW of coal and gas maxed out at 8.3GW so much for the “reliable coal and gas”
Announcement now on CS Energy website of the delays in return to service due to rebuilding of both C3 and C4 cooling towers.
You could look at https://electricinsights.co.uk/#/dashboard?&_k=ykdb3l which shows the UK still uses coal most of the time. Technically it still has 4GW connected to the grid but output has not exceeded about 1.6 GW for about 18 months. In 2021 the UK ran for more than 5,000 hours without coal, but even though annual coal share has continued to decline, http://www.mygridgb.co.uk the period with zero coal have actually fallen over the last couple of years, https://www.mygridgb.co.uk/coal-tracker/
It would be helpful if you could provide or point to a simple but clearer explanation of the issues discussed in 13.2 re “hybrid” facilities. For example why would a renewable-battery combination want to be semi-scheduled?
Great charts on the curtailment/constraint side of things. In the USA based on NREL work it seems the DC/AC ratio is known as the Inverter Load Factor (ILF).
With about 35TWh/y of renewables committed including rooftop solar over the next 4 years 19 TWh/y of coal retiring and probably 2-3GW of storage in front of and behind the meter and storage/pumped hydro sucking up negative priced coal and solar power at midday and dumping into the $130/MWh evening market, what role is left for gas. If gas generation is half what it is now in 2026/27 it will be amazing.
In the meantime, heatpumps for hot water and space heating combined with gradually increasing energy efficiency will accelerate, so AEMO’s 2% per year decline in other gas demand will be looked on with wistful fondness by the gas companies in a few years’ time.
Paul
It looks like a significant driver of the volatility (in addition to high demands and generation outages in NSW and Qld) were constraints on the New South Wales 132kV network which limited flows from the south on the main 330kV route from southern NSW & Victoria. Check out the N>>NIL_998 and N>NIL_999 system normal constraints for more detail. Hard to say without deeper analysis but could have been a “tail-wagging-dog” or “gateway” effect where limits on a peripheral flow path throttled the adjoining higher capacity path (but I don’t know by how much). If so, another argument for getting the transmission system in better shape to deal with changes in where generation is concentrated.
Allan
Thanks Allan
You’ll note I have referenced your comment in the article ‘Reviewing what happened in NSW and QLD on Thursday 16th March 2023 (with Actual LOR2, Volatility and Reserve Trader almost triggered, etc) … part 1’.
Will explore VIC1-NSW1 and these two constraints (and others) in a subsequent article.
Paul
There was a spike to $1000 in SA at 0410-0415 as well, obviously unrelated but an interesting time for a price spike nonetheless.
A lot of focus is rightly placed on high demand situations; my question is whether there is value for “dispatchable demand” in high generation/low demand situations?
I have developed a manufacturing device that converts electricity to plant food; it is profitable to operate even if on only 2 hours a week on average, if the electricity coming in is under $0.02/kWh. It can turn on/off in the 5-10 second time frame.
Is this type of capability of value for a grid operator, or I should just focus on being a consumer?
Hi Dan, as a longtime resident of rural Timor-Leste i am familiar with the unreliability of power, especially the daily blackouts in the rainy season (could be 1 hour, 4 hours or even on occasion 4 days)
For rural communities such as Atauro (or the coffee farming households/ hamlets, villages in the mountains) are you able to suggest any affordable and sustainable solutions?
Hi Peter thanks for reaching out, and it’s great to read about your work at Timor Mountain.
As for your question, I am unaware and unsure of the Timor-Leste government’s plans to improve or augment the transmission/distribution network in the immediate future. So it *may* be the case that consumer-led or community-led solutions are the answer.
In the short-term, lead-acid batteries or other forms of UPS – are ways to increase realibility. We met with one hotel owner who was doing this effectively to have 24/7 power on Atauro.
In the longer-term, perhaps community-based offgrid solar + battery projects are an efficient way to pool resources together to provide reliable and sustainable electrcity supply to individual villages.
Obviously, a lack of capital and a lack of skilled electricians/engineers are major barriers. We have a number of students at UQ who have shown interest in doing further research into this, if you’d like me to keep you updated then please shoot me a message at daniel.lee3@uqconnect.edu.au
Dan
c1ue, under the Wholesale Demand Response Mechanism you can be paid to turn off. It’s also possible for smaller consumers to get wholesale-linked pricing through retailers like Amber (so you can turn your plant on and off depending on the current wholesale price) or very large consumers to directly participate in the wholesale market as a dispatchable load.
as requested – posting coming from
https://www.linkedin.com/posts/kate-summers-aus_market-suspended-in-nsw-due-to-service-interruption-activity-7043440656442126336-P6_b
I think there’s possibly a bigger problem (albeit related to your root cause suspicion).
The question in my mind is what is the impact of rebidding (esp. as the interval decision point/s approach).
It’s an issue that is also seen, with much bigger numbers, in wagering in general but especially totalisators (TABs) as the “jump” approaches there is both an increase in information updates (approx dividends) and number of bets … and rapidly increases the transaction rate (it’s referred to as an avalanche and it doesn’t do it justice especially on big gallop races!).
Efforts to manage “collations” via (sub) aggregation results and new challenges (and ones that the pro-punters hate because they want (must) bet as late as possible with the most information)!
It’s only my opinion (thoughtful/considered but nonetheless still an opinion) that the market (especially the consumer) would be better served if “bids” (rebids) had a total limit or at least a cost to them (escalating as EoI approaches)
Without targetted research it’s difficult to be certain but imho it reasonable to assume that generators (the predominant bidders) are trying to maximise the despatch price. Unrestricted bidding facilitates that!
Kevin,
Thanks – I have looked at this type of setup in Texas.
What I am wondering though, is just how well this will work as curtailment amounts and %time negative/zero electricity prices increase due to duck curve mismatches with solar PV/wind generation?
Large scale industrial users of electricity – their biz models and tech simply don’t work in an intermittent usage setup.
So I guess my question is whether there is value to having dispatchable industrial demand at scale vs. simply turning off plants/generation.
I do know Tesla is trying something along those lines using coordinated Powerwall recharging, but I really doubt just how much dialable demand can be aggregated this way.
There was an outage on a Victorian 220kV transmission line (one of the Thomastown ones) earlier overnight. This implemented a constraint on the Vic-NSW lines. I wonder if that was at play, constraining down Vic-NSW transfer?
Clearly one lesson that should be learnt is that “Summer” doesn’t finish on the 1st of March. It shouldn’t be assumed that scheduled maintenance can start on the 1st of March.
There needs to be greater focus and reporting by AEMO on the performance of interconnectors and major transmission lines. Too often we have seen major supply problems which would have been far less critical, or not even occur had there not been constraints on transmission.
It would be interesting to know what residential demand controls were used in QLD during these periods, such how many and how by how much were residential air conditioners ramped down.
These events show the need for almost all residential consumers to be moved to wholesale spot price linked tariffs, so residential consumers are incentivised to reduce their consumption during these periods of extreme demand.
The problem does not seem to be capacity of the interconnects themselves, after all VN1 has been upgraded and one of the justifications of the Victorian big battery was that it would allow an increase of 150-200MW of Victorian power to NSW by stabilizing the transmission link.
Despite these upgrades imports from Victoria were constrained to zero on the 16th of March and total net imports from Victoria over the last 90 days were about 500 GWh, yet way back in 2010 NSW imported 1,950 GWh in Q4 from Victoria. Further in 2010 NSW prices were only 21% higher than Victoria, now they are almost double, so the cost of transmission can’t be the issue.
How come capacity has been increased 20-30%, the price differential is as high as it has ever been and yet flows are down by two thirds. Methinks either incredibly poor grid management by Transgrid or some seriously dodgy constraint equations in the NEM dispatch engine
From The Australian’s 24-Mar piece:
‘The joint venture partner in control of one of Queensland’s major power stations has collapsed after a dispute over future funding for the troubled asset. Deloitte turnaround and restructuring partners Grant Sparks and Richard Hughes have been appointed voluntary administrators over a number of IG Energy Group entities, which own half of the troubled Callide C coal-fired power station near Biloela.’
[Editor’s Note – the rest removed so we don’t breach copyright]
From the Financial Review’s 27-Mar piece:
‘State-owned energy company CS Energy is considering buying out the remaining stake in the Callide C coal-fired power station after its joint venture partner, [**Genuity**], went into voluntary administration on Friday.
The move is seen as a way to secure the long-term future of the power station which has been beset with problems since a mysterious explosion at its C4 unit caused havoc on Queensland’s power grid in May 2021.’
[Editor’s Note – the rest removed so we don’t breach copyright]
[2nd Editor’s Note – note the Editor’s Note above in the article itself about the corrected name for the JV partner]
Excellent article Dan. Another consideration is what did the developer or owner assume for all these factors when building their business case or arriving at the price they paid for the project. Whatever the effect of each factor, if it turns out operationally to be ‘better’ than assumed, then the project could still be regarded as successful.
As a mere mechanical engineer, a lot of this is barely understood, but my simple question is. Why did we spend $42m upgrading VNI and in the last quarter Victoria exported 350GWh to NSW yet in Q1 2015 we were able to export 1,500 GWh. Back then the price differential was $7/MWh-this year $43/MWh.
This seems to need more than tweaking a couple of constraints in a constraint matrix. It would seem that either the money was spent in entirely the wrong place and Transgrid is being paid some $4m/y for adding costs to consumer bills or that the whole system model needs to be rebuilt
Peter – very reasonable question, no-one who looks at the current situation with VNI thinks it’s an acceptable outcome, but unfortunately it reflects the current market design and getting agreement to change things like transmission access and pricing has proven nearly intractable. If there was a “NEM benevolent dictator” this particular problem could probably be fixed tomorrow – but no-one is in charge of the NEM and the number of stakeholders with wildly different positions on things makes changing the status quo very very slow and painful. Allan.
Could you please explain the difference between market demand and operational demand?
The difference is mostly due to demand met by “non-scheduled” generators who do not bid their output into the spot market. “Operational” demand includes this component, but “market” demand is exclusively the amount which is supplied by generators through the spot market.
The more telling information is what the demand is in the sunset to two hours afterwards. That is the size of the duck curve which the grid operators will need to ramp up for each day.
Yes, when we have extremely low demand and high renewables (like over the Easter weekend), we’re likely to see little change (fewer renewables constrained off). But that was never the concern. The concern is every shoulder and evening peak since we will have 600-960 MW of bids lost. And it doesn’t have to be 16-March levels of demand for it be translate to higher spot prices.
It should be relatively simple to understand that prices will rise as we saw with Victorian prices after the closure of Hazelwood.
You can’t remove continuous generation capacity and expect a more volatile system to provide the same reliability and thus the same pricing.
It’s only going to get worse as Queensland progresses with it’s coal plant closures.
So Project Energyconnect is adding 800MW of transmission capacity between Buronga and Darlington Point – presumably this will result in decomissioning this section of the X5 line and replacing the X5 constraints with new, looser ones?
A bit of uncertainty is hardly surprising. Will be all about getting the last of the coal yard cleaned up, as many coal bunkers cleaned out as safely possible and maybe even some slops out of the fuel oil tanks used up. Get rid of as much as possible rather than trying to clean it up as part of the decommissioning process.
Good luck for winter guys. Taking another GW of dispatchable generating capacity out of what was already an uncomfortably tight winter last year…..
Why would anybody be “popping champagne corks”?
No doubt the demolition will start the day after shutdown – to ensure there is no way back.
Surely prudence dictates a period of (perhaps 12 months) care-and-maintenance before declaring that the entire grid is stable and maintainable.
What’s interesting about the NSW LOR2 for the day after Liddell closes is the time of the day of the LOR ie 1400 to 1500 hours.
That is a rather bland reference to a period where the NEM would have gone black in the absence of remaining coal burners with the wind practically absent from Friday evening through the weekend into today.
At sunrise on Saturday morning the wind was delivering 6% of demand at CP 11%, in the mid afternoon the wind was delivering 4% of demand and RE was 33%, at sunset the wind delivered 3% at CP 8%, Sunday morning it was 5% at CP 10%, Sunday evening 3% at 6%CP, sunrise this morning, 3% at CP 6%.
It all over Red Rover! RE can displace coal, but not replace it.
https://www.flickerpower.com/index.php/search/categories/renewables/21-7-intermittent-solar-and-wind-power-can-displace-coal-but-cannot-replace-it
The rate of exit from coal is not accelerated by increasing penetration on good days, it is limited by the lowest level of output on nights with little or no wind, as a convoy travels at the speed of the slowest vessel, the water penetrates the levee at the lowest point, a chain is only as strong as the weakest link and stock get out of the yard through gaps even if the rest of the fence is built to the sky.
As long as periods with effectively zero solar and wind power persist, 100% backup from conventional power will still be required, assuming that we want security of supply. This means that we will have to keep burning coal until nuclear power is on deck.
Four pieces from The Spectator.
A simple demonstration that wind power will not work in Australia due to the need for continuous input to the grid, the interruption of wind power by wind droughts, and the lack of storage at the scale required to bridge the gaps.
https://www.spectator.com.au/2023/02/the-iron-triangle-of-energy-realism/
On wind droughts and lack of storage in Australia. When the next coal power station closes, every wind drought will threaten the power supply and prolonged wind droughts will be potentially catastrophic.
https://spectator.com.au/2022/07/energy-policy-where-parallel-universes-are-set-to-collide/
The vital importance of “wind literacy.” Wind droughts have been recognized in Australia for a long time but defective advice (neglecting the droughts) resulted in the the worst policy blunder in our history. This is connecting subsidised and mandated intermittent wind and solar power to the grid.
https://www.spectator.com.au/2022/11/the-voice-of-energy-realism/
How did the meteorologists hide the wind droughts?
https://www.spectator.com.au/2023/04/the-endless-wind-drought-crippling-renewables/?
Rafe your comments are valid. Looking at the AEMO dashboard it appeared NSW compensated for the wind drought by using extra hydro on the evening 30/4. But this can’t go on forever given we are at the end of La Nina.
A well designed and operated electricity generation and distribution system will supply customers with a cheap, reliable, and sturdy supply 24/7 365 days a year.
We’ve already had such systems in Australia – large, long life, affordable, coal fuelled base load generators, backed up by quick start, hydro and gas fired generators – more than capable of taking us through the typical peaks of the 24 hour cycle.
Incredibly, we are seeing these “dirty, old, clunkers” being demolished well before life expiry, in favour of thousands of individual small scale “renewable” entities, requiring billions of dollars of taxpayer subsidies, and as the above extensive analysis shows, not quite being up to the vital role they are supposed to fulfil.
It’s worse than that, because the much shorter life of these multitudes of minuscule power generators come with repeated and grossly expensive replacement and no doubt more and more taxpayer subsidies.
When will the blinding reality, that large scale nuclear reactors with long lifetimes – dropped in at locations close to the heavy capacity transmission lines already in existence – are far and away the only plausible way to power our nation into the future.
Smaller SMR units can readily be dropped in at other easy connection points i.e. close to heavy industry, and with existing hydro and gas units used to back-up the system up at times of peak demand.
This article treats our nation’s vital electricity supply system as some sort of grand experiment – let’s keep toying about with renewables, irrespective of the cost – and keep on avoiding the reality that there really is a better way to meet this need.
In addition to the multiple failures of the push to abandon reliable (fossil fuel) generation you pointed out, the pundits claimed renewables would be cheaper. However that claim only existed because the first batch of renewables did not need additional transmission infrastructure.
…or expensive storage initially-
https://www.msn.com/en-au/news/other/you-can-t-close-eraring-former-snowy-hydro-ceo/ar-AA1aFGFm
The fallacy of composition is Econ101. Trying to disprove a fundamental axiom of engineering that you can’t build reliable systems from unreliable componentry would rate alongside that.
What a fantastic summary! Thankyou Tristan!
Yesterday afternoon at 3pm wind output across the NEM was 6300MW at 60% capacity factor.
This morning it is 420MW at 4-5% capacity factor.
The situation is even worse in SA with output below 20MW (between 4am and 8:30am) at a capacity factor of less than 1%
This reflects on the lack of wind based generation in NE NSW and more particularly QLD. If QLD had the 6GW+ of wind based generation capacity that it should have by now, then there would be far more consistency in wind based generation output across the NEM as the cold fronts sweep across the country.
Wind generation isn’t a good option for QLD due to it’s higher latitudes.
Maybe SA should build a new coal plant.
Pithy, but unhelpful in my humble opinion.
In progress – another interconnector from SA to NSW
In progress – NSW REZ
Build more wind in QLD, build additional interconnection to NSW possibly – good for the whole NEM as it diversifies away from localised geography.
Why take a state-by-state approach to the NEM which is, ultimately, larger than a single state?
Great reporting: market design flaws AND lack of dispatchable power will lead to an expensive rush for batteries which will fail…dire consequences!
What batteries have failed? Batteries are more reliable than coal or fossil gas generators.
Once the 100GWh pa MegaPack factory comes online in Shanghai we are going to see a significant increase in the storage deployed in Australia, along with lower pricing.
Is this related to recent unavailability of Torrens Island B2 and B4?
Issues re Kiamal are dealt with in the following AEMO Power System Operating Incident Report:
06/09/2022
Trip of Red Cliffs Kiamal 220 kV line, Kiamal synchronous condenser, Buronga No. 2 and No. 3 synchronous condensers
Date of incident: 1 March 2022
580.72 KB
A major contributing factor was the lack of transmission capacity between Victoria and SA. Transmission capacity was down by at least 600W, which would have had a major impact on the price.
Price is hardly surprising seeing you got into liquids to keep the lights on. look forward to it as reliance on intermittent generation increases. Unless called up well in advance cold thermal machines won’t get there in time and they won’t be there eventually because their business model is broken and they are all bleeding money because of daytime solar. Little business case for GTs without a capacity market. Then their is the small matter of how do you contract for fuel (gas) when you have no idea when and how much you need. Can be fixed with storage (both electricity for short term and gas for long term) but the punters are going to have to pay at the end of the day. But the experts tell us that renewables are cheaper …..really? This is going to end in tears. Looking at the spot market right now, dare I say the missing 1000MW from Liddell might already be showing up in pricing.
ditching the boat before reaching the shore.
My experience in Government and the Industry, across a range of infrastructure sectors, has been (and continues it seems) that politician-pushed programs (rather than technically/economically/soundly-managed programs) so often disappoint.
Great find! I Hope this gets fixed soon and in the future the constraints will get updated automatically to remove unused terms.
Why did small-scale solar drop off vertically at around 1600 hours but not large-scale solar??
Great article along with Allan’s and others but it gets to the same point. There are too many constraints. Better modelling might lead to better price forecasts but the actual outcomes for consumers won’t change until some of the constraints are alleviated.
Good thought David. The upside of good models is the ability to prioritise which constraints should be removed and the economic benefit of such a project. Similarly the benefits of DR can be better identified.
Thanks to the author for demonstrating that not only is the grid being structurally weakened by Heywood and subsequent closures and additions, but that this has always been the expected outcome.
This knowledge is essential if we are to have informed discussions about our energy future. Thanks for wording it in non-technical language with minimum jargon.
My take-away: Dunkelflaute is getting progressively worse and our defences against it are progressively weakening.
Lunch time Sunday in Victoria 24 MW wind and at 8.40 am on Monday 2,028 MW of wind ie a swing of 2,000 MW in wind output in less than 24 hours. This is not the way to run an industrialised economy or maintain reasonable living standards. And did anyone notice Battery Of the Nation in Tasmania in AEMO’s database has now been reduced to 1,500 MW which just happens to match Marinus Link (Stage 1, 2a and 3a was removed). With all the Snowy 2.0 cost overruns and delays where is all the long duration storage coming from?
I suggest that this issue was compounded or perhaps even primarily caused by poor investment confidence over the past 10 years. Now with the confidence turned upwards with new targets and incentives, it takes 7+ years to developed PHES projects while it takes 2+ years for solar and wind projects. We did need Snowy II to come on by 2026 but we didn’t need all that capacity at this point. Seems like we will need many more higher capacity battery storage projects between now until 2029?
Dunkelflautes are not getting worse, they just become a serious matter when conventional power capacity runs down to the point where wind power matters, as in Britain and Germany.
Hardly a surprise surely. The real elephant in the room is how (who?) is this going to be paid for. All of the modelling I have seen says you cannot sensibly build enough storage to get through the quite common, seasonal, wind and solar droughts that occur in the NEM (and indeed here in the west also). To attempt to overbuild wind and solar to compensate simply sees these become dispatched generators and hence they no longer have the privilege of free spilling to market. This will make network curtailments look like a picnic. They can’t (won’t) take market risk on this (project becomes unbankable) so who does? Simply their LRMC becomes much higher based on what a proponent reasonably thinks they can get to market. Ditto storage. Unless storage gets some sort of capacity payment (that has to be funded) then their whole business case will assume frequent extreme energy market swings to get an arbitrage. High risk, so who invests? But I thought all this renewable stuff was going to be cheaper? The higher the renewables penetration the less dispatchable generation survives and with the current mindset (that the only answer is more renewables) the overbuild will become worse, capacity factors decline and the system levelised costs go stupid. I suppose it all stops when you get to VOLL. Reminds me of a wise engineering lecturer that once taught me “in real life there is no such thing as the infinite busbar” – we can’t just keep doing this and somehow assume it will all hang together! Trouble is it is up to proponents to decide if they have a commercially viable project and I challenge the rigor going into some of the analysis sitting behind these. Wasted money is wasted money and it doesn’t really matter if it is public or private – it is opportunity lost.
Paul, Thank you for the great waterfall plot.
How far into the future could it be reliably extrapolated?
Kuri Kuri, so 375MW in 12 months time and another 375MW in 18 months, still 2500MW short.
Rafe: Dunkelflautes, well at least wind speed, will get worse with Climate Change! Ref: https://e360.yale.edu/features/global-stilling-is-climate-change-slowing-the-worlds-wind
There are a few concerns here about investor risk and some suggestions that the business case for storage is somehow different when it is not.
In the NEM, the business case for generation (including storage) of all types ultimately relies on favourable spot prices. What favourable spot prices look like is different for generators fuelled exogenously (eg coal, gas) than for those fuelled endogenously (ie battery/PHES or similar). To invest, a business case for the former wants spot prices to be high on average if they run all the time (coal) or many short periods of high price spikes for peaking units. This is well understood.
What appears to concern people is that the business case for the latter relies on spot prices to be highly variable – this might seem new (even though PHES has been around for decades?) but the indicator of profitability (future spot prices) is the same. In all cases, generation (and storage) investment business cases rely on forecasts/expectations of future spot prices as they determine which technology is likely to be most profitable.
In relation to the comment above “* note that one thing that might potentially change this to some extent might be when we start to see results for increasing diversity of wind farms across Queensland … however that’s a strong ‘might’, as other analysis we’ve performed (such as Appendix 27 within GenInsights21) suggests that this is unlikely.”
On April Fools day this year there was zero wind in Queensland for 3 hours and on 2nd April zero wind for 5 hours. Queensland wind farms are located from Mt Emerald west of Cairns to Coopers Gap west of Toowoomba. Therefore from the north to the south of Queensland there was no wind. Chemical batteries generally last for 2 to 4 hours and in the situation of no wind across a state could be discharged quicker. Then there is the instance of 204 MW wind output for the 5 states in the NEM which means even diversity of wind farm location across the NEM does not solve problem of not enough generation to meet demand. PS 1st and 2nd April 2023 was a weekend but in the future short and long wind droughts will occur during the working week and that severely impacts our economy, jobs and food security.
Allan great analysis. I believe it’s not going to be good ending.
“Heard about the guy who fell off a skyscraper? On his way down past each floor, he kept saying to reassure himself: So far so good… so far so good… so far so good. How you fall doesn’t matter. It’s how you land!” 🙂
The NSW volatility events on 17 and 20 May and on 16 March were characterised by constrained VNI flows and high (2000 MW) Tumut hydro output. Is the VNI constraint causing the high hydro output or is the high hydro output causing the VNI constraint? Is there a lack of capacity in the Snowy 330 kV network?
Under and over deliveries to target may be manageable with the NEM’s existing 19 GW of large-scale solar and wind. However, imagine the impact when solar and wind farms reach 44 GW in 2030 and 141 GW as per AEMO’s 2022 ISP. Then what if Australia becomes a global green hydrogen superpower as per the Bloomberg/BNEF report yesterday and large scale solar and wind reaches 812 GW by 2050???
$300 per MWH, price cap?. I thought this was revised a little higher?.
Hardly surprising pricing given you are into liquids again. Low sulphur diesel is around $1 per litre (excise free) these days (a lot lower than 12 months ago) so putting this through an old 13,000MJ/MWhr OCGT somewhere gives you about $380 / MWhr just for the fuel. There are some very subtle high level messages from lack of reserve. This is going to end badly people. Liddell is gone, Callide 3 & 4 still off (and how much more could you get south into NSW anyhow?) and it’s not even cold yet. With the announced retirements in coal, sensible asset managers will be rationing expenditure on those as they move to retirement, so it is not unreasonable to expect reliability to decline. Without a capacity market I can’t see people queuing up to build OCGTs (even on liquids) to at least keep the lights on. This is completely analogous to the risk Swiss cheese model. Keep retiring dispatchable capacity and you are removing another layer of diversity and defence. Sooner or later the holes will line up. In the years of high growth here in WA I went through a summer of less than 10% reserve plant margin and it is a very uncomfortable (and expensive) place to be. Clearly, I must be missing something!
Looks like we didn’t have to wait long to see the impacts from the closure of Liddel.
Allan’s article from 15 May was just a touch early.
While the NEM is a complex machine and it’s very rare that there’s ever a single ‘cause’ for any particular outcome, it could be argued that the systemic destruction of reliable coal fired power plants and removal of a material volume of dispatchable generation is the driver.
Notice how most of the price spikes are in the early evening, the head of the duck curve as it were, when solar (and to an extent wind) is less productive, and people are demanding power for cooking (and heating)
It might have been cold, but hardly peak load stuff. Cold overnight temperatures usually result in high daily energy, not necessarily load. (Dare I say the generators make some money!) What has changed is a not insignificant lump of relatively cheap SRMC generation (Liddell) removed from the market, so it all shuffles up the cost curve. Most energy in the NEM is (in reality) settled by bilateral trading arrangements – particularly so for the vertically integrated gentailers. Retailers in unhedged wholesale positions got scorched last winter you may recall. Now the energy served by Liddell is simply being supplied by more expensive sources on occasions when wind (and solar) are missing or the peak demand goes a bit higher and a higher cost generator bids in and settles the market. Your previous remarks about pricing when liquids appear highlight this. Nothing to see here – your new normal.
We almost need to have capacity adjusted records to understand how good were the actual wind conditions at the time.
The reversal of flow on VNI and the increase of 1464 MW in NSW generation appears to result from the dispatch of ~1400MW of Tumut3 at that time. Upper Limit was already generating 600 MW. AEMO published a document on 3 November 2017 https://www.aemo.com.au/-/media/Files/Electricity/NEM/Security_and_Reliability/Congestion-Information/2017/Interconnector-Capabilities.pdf which notes (Section 2.3) that the nominal capacity of VNI is highly dependent on the Snowy generators’ output. This seems to indicate that the Snowy transmission capacity is not sufficient to simultaneously support high VNI flows and high hydro output.
PS more here, Alex:
https://wattclarity.com.au//articles/2023/06/18june-nn_dtkv_2-constraint-equation/
The concern here is the time of the IRPM ie after the sun has set. So when the chemical batteries are discharged in a few hours and there is low overnight wind then what? The 2 GW Snowy 2.0, 2 GW Borumba Pumped Hydro and 5 GW Pioneer-Burdekin Pumped Hydro will be post 2030. The NEM is desperately short of LONG duration energy storage that can operate for days (or weeks) if need be. The IPCC have said that climate change could cause global stilling in the decades to come. Australia’s east coast is one area that could be impacted by reduced wind speeds. A 10 percent drop in wind speeds could drop wind generation by 30 percent. Will be interesting to see how the proponents of these massive pumped hydros intend to return the water to the top reservoir if the IPCC are right and global stilling does eventuate. The NEM has to be resilient to climate change which requires a mix of generation. Geothermal according to IRENA is 0.5% of global renewables but could it be more. Japan and Korea sit on the Pacific Rim of Fire and Queensland has potential that has so far resulted in failed geothermal projects in western Queensland.
It seems that solar sets the price effectively until the next bid kicks in at around and is capped by thermals at $300
If a coal plant mothballs a unit for the shoulder season, is that considered curtailment?
If so, the MWh involved are potentially much more than wind and solar.
No, and that really isn’t happening yet. The shoulder season is typically when units are taken off line for maintenance.
It would be interesting to see on these days of very high curtailment if there was a corresponding restriction on one or more of the transmission links between regions such as QNI, VNI, Heywood or Basslink. It is relatively rare to have both excellent wind and solar conditions across all regions. Today was one of those exceptions.
Is there any way to find out how much curtailment was happening?
Yes Dave there are some methods to get an estimate:
https://wattclarity.com.au//articles/2020/06/not-as-simple-as-it-appears-estimating-curtailment-of-renewable-generation/
But I have not had the time this week to explore.. Will follow up with some analysis if time permits.
How can you publish an article on this now when we had record LOW wind generation this week resulting in high prices across the NEM. Especially of note is the non-existent wind in SA, compounded by interconnector constraints, resulting in some of the highest prices we’ve seen in the region in recent months.
This is great news but what is the implication? One thing that sticks out is that total NEM wind gen was well under 1 GW on Monday. Which is a +7.5 GW variance within 5 days. Relatively speaking, that is the equivalent of saying that we lost 2.5 Eraring stations on Monday. What does that mean for system security? There’s all this fanfare for difficult to predict peak wind days, but no mention of times when wind gen is miniscule (which happened five days earlier).
The sub-1 GW wind gen on Monday was at its worst during the evening shoulder period where there is no solar and rapidly rising demand resulting in a measly 15% wind contribution to the NEM. The rest of the supply coming from coal, hydro and gas.
The interesting metric isn’t the absolute maximum of minimum but the % of the nameplate. The min % is far more important than the max. The recent Net Zero Australia plan reckons we need 400-500 GW of wind+solar to power what is just a 25 GW system (+ reserves). That’s a 16 to 1 overbuild ratio.
Generator profits are a combination of quantity and price. Recently truck loads of avocados were found at the local dump as there was no market. People were outraged. However, there were simply too many avocados at that point of time. It would most useful to have the price when wind output was 1 GW compared to when the output was over 8 GW last Friday evening. How many wind farms have a Power Purchase Agreement and how many have a PPA for the expected life of the wind farm?
Don’t forget most of the more recent wind farm builds were motivated by the old RET and the associated penalties for retailers. Business case for the development was driven by this – energy value was almost a bonus. Not normal PPAs as we would know them.
It certainly seems like, under some constraint situations, TUMUT3 is able to ramp up and take advantage of high NSW prices, pushing out cheaper Victorian generation even to the point of causing counter-price flows into Victoria.
The last graph is an eye opener. Presumably due to wind and solar so it’s only going to get worse. How can negative wholesale pricing be transferred to the retail sector?
Good charts thanks as always Dan. I wonder if it’s time to update the threshold for the incidence of spike prices graph from $300. In think this year is being dominated by prices in the $300-$400 range which are no longer reflective of a spike.
One approach to maintain historical consistency could be to use an inflating threshold.
All the best.
“Giles Parkinson wrote on RenewEconomy about ‘Vales Point coal closure date pushed back four years in latest threat to renewables transition’.”
That is one of the most nonsensical statements I have ever read. He actually used the word, “threat.” Isn’t this the same guy that insists that wind, solar and battery are cheaper than fossil fuel generators? Why is worried then? They’ll just get priced out of the market and the threat of dispatchable 24/7 gen will be vanquished.
Take a high level view and nothing terribly is surprising here. Volatility and slow uptick in average pricing started in 2017 with the retirement of Hazelwood. As wind and solar render the economics of traditional generation unviable and more exit the market both volatility and average price will get worse as holes are plugged with higher SRMC sources (gas and diesel). The “fat” of spare generation is just about gone. Throw in the outage at Callide (though this is tempered by transmission constraints) and the trend is clear. The WA situation is also clear. 200MW of coal was retired prior to the summer of 2022/23 and the coal industry in WA (a non export commodity) lurches towards collapse on shrinking sales volumes, there was more energy produced from gas than coal. (WA already was second only to SA in terms of gas for generation). Indeed for several months it has become almost normal to see diesel burnt on weekday evenings to maintain reserve requirements. WA has a capacity and energy market with regulated domestic tariffs so the price volatility that would otherwise occur is well hidden in the red ink in the Government owned retailer’s books. The actual volume of energy traded in the STEM in WA (the number you have interpreted here as the wholesale price) is proportionally very small.
Inertia is best understood as short term energy storage which is used to absorb/release energy in response to changes in the supply-demand balance. It is not an objective in itself, it is a tool to maintain stability of voltage and frequency.
It should also be remembered that synchronous machines can’t actually deliver much of their stored energy. For example, if the maximum frequency excursion is limited to 0.5 Hz Then a rotating machine can only absorb or supply 1-(49.5/50)^2 of its inertia so a 500 MW steam turbine with an Inertia Constant of 9 or 4,500 MWs of inertia can only convert about 90 MWs of energy. If the frequency change occurs over two seconds it is an average power of 45 MW. A 50 MW battery could supply both the power and the same energy, but if the frequency decline persists, then there is no more inertia available and the governor response firstly has to replace the lost inertia of the steam turbine i.e. return the 90MWs and then start to make up for the shortage of power in the system. The battery is just adding power and energy back to the system from the get-go. Without going into the boring details, for the first few seconds gas turbines are even worse because they have around half the inertia of a steam turbine and engine power drops if speed (frequency) falls so the governor response has to be even more vigorous.
If generation can respond fast enough the system needs very little inertia, in fact inertia can become a hindrance. Many years ago, in machine tool design it was considered necessary that the inertia of the motor should be at least 3-4 times the inertia of the load so to accelerate the load a torque of 5-6 times the nominal load torque had to be applied. Now it has been realized that if the control system can detect errors quickly enough and the power amplifier has a very high rate of response on the power curve, lighter more accurate systems can be built with low inertia motors with as little as 1/10th of the inertia of the load because now total torque is only 1.1 times the load torque so less energy is required to correct errors.
The same is true in reciprocating engine design where engines 50-80 years ago had huge flywheels to stop them over-speeding if load fell or stalling if load increased abruptly, now with high resolution speed and torque sensors, micro second control of fuel injection and low inertia flywheels, diesel gensets, sportscars etc can ramp power up or down almost an order of magnitude faster than their 1950’s predecessors, so better load matching with less inertia
Similarly in a power system if the output of a battery can be ramped up or down by full capacity in 150 ms then it can contribute/soak up far more energy than the inertia of a steam or gas turbine. Some years ago, Eire Grid determined that a 10 MW battery contributed as much to grid stability as a 100 MW gas turbine. Partially curtailing or reducing curtailment of a solar farm can have a similar effect and because wind turbines are asynchronous they can contribute much more of their physical inertia while still maintaining frequency and of course in the case of over-frequency the inverters on wind or solar farms can reduce power safely far faster than thermal generators. Even reciprocating thermal power plants which have relatively low inertia can partially make up for that by having very high power ramp rates as long as the gain on the governor system is set appropriately.
Then there is the load inertia to consider, large fans pumps etc and the inductance of the transmission and distribution system all contribute to system stability. Now many modern motors are connected via VFDs so that decouples their inertia but other systems such as fans or pumps have the added benefit that the load drops as the square or even higher power of frequency, so not only does the system gain the mechanical inertia, but a useful reduction/increase in load.
When all of the above is considered and control schemes are reconfigured appropriately it is likely that in the future, the system frequency will be maintained at or better than current levels with less than 1/10th of the inertia now thought to be necessary
Peter, I’d like to introduce you to my electric-arc furnace.
We’re creating a grid that is no longer fit for industry-purpose.
Although your mention of nuclear was not a central part of this article, it still bothers me that its legitimacy is brushed under the carpet and completley ignored. I’m not a big nuclear power advocate but the arguments presented here are weak. The volume of actual nuclear waste products are miniscule compared to the impending toxic solar panel and wind turbine waste entering the global electronic waste scene. Yes, nuclear plants just boil water, but this straightforward design is why they are operational for well over 60 years. You can replace control panels and steam generators with relative ease and keep these stations running for decade after decade. And if you’re concerned with nuclear sources getting into the wrong hands, why don’t you advocate against the use of them in hospitals which are under far less security than a nuclear power plant.
Regarding the minimum demand issues, that really is one of a million issues we’re faced in the future. What is even more pressing is the abolition of all forms of dispatchable generation except for hydro and perhaps batteries if they ever develop their capacity and throughput. Minimum demand days are scary but so are MAXIMUM (scheduled) demand days when we have coinciding wind and solar droughts (days and even weeks in duration).
I stopped reading when I reached the claim “It was good to hear Chris Bowen articulate that nuclear is a distraction …”.
Bowen is a megalomaniac, clearly totally out of his depth and the most incompetent energy minister Australia has ever produced.
We are heading to a grid disaster and Bowen is front and centre of the fiasco.
@Rob
Says a lot about the author’s credentials.
I am bemused at how near the beginning, nuclear power is dismissed as merely ‘boiling water’, yet by the end the author is calling for “clean steam for electricity to fill the dunkelflaute”. Er, hello?
Even odder is the claim that nuclear power plants have been ’weaponised’ in Ukraine. What sort of weapon has no measurable impact whatsoever? *Fear* of such may be a different matter, but very little of this is well founded. In any case, what on earth has this got to do with Australian scenarios?
SunCable guy, you realise nuclear energy uses steam turbines, right? A central point in the trends this article highlights is minimum demand crashing below zero MW in almost all regions, and the need for flexible generation to work with that. A solution to the problem that twice a day demand will hit crazy lows, but then at two other points in the same 24 hour period it will be very high. Steam turbines don’t like to ramp, and definitely not multiple times a day every day. Simply bringing nuclear into the discussion completely misses the point of the challenges Australia’s grid is facing. Its like saying “delivery trucks find it hard to navigate CBD streets… No worries, aeroplanes can carry heaps of parcels. Let’s use them to make CBD deliveries”.
To quote the article, “How prepared are we for coincide minimum or negative demands that are forecast to occur within 4 or 5 years?” installing generators that require gigawatt level baseloads is pretty much the worst thing you can do to prepare for that.
Australia’s problem is very different to Americas or Europe’s, where they run large interconnected grids. Got too much power? No worries, send it across a border. Not something we can do. The argument for nuclear distracts the point that Australia’s grid issues are not a debate about fuel, we have all the fuels we could imagine. its a debate about generator capability, and how we can balance loads and generation to meet the intra-day demand curve that is getting harder and harder for a typical power system to support. Which is central to a point that Kate has long advocated, that is the need for good system control.
There’s a very good reason Kate is always invited to talk at such events, and that Engineers Australia made her a fellow. There are very few minds as sharp as hers when it comes to the challenges of Australia’s power system.
I invite Mr Farley to review his statements that gas turbines supply less inertia than an equivalent steam turbine, and that mechanical inertia is increased by the use of variable speed drives.
@HowGoodAreRenewables
Plumetting demand periods are EXCACERBATED by your net zero targets. You’re creating a problem by replacing dispatchable generation with renewables (that distort prices due to their LGC-creation) and then blame dispatchables who follow a different set of rules in the NEM. The solution to the problem is to REPLACE wind and solar with dispatchable generation such as nuclear. Don’t try and confuse the issue. Kate blasted nuclear from the introduction as an inferior source of generation in general – way before she discussed the minimum demand problems.
This speech by Kate is just an admission that the system she heavily advocated has serious issues with no solutions. But that doesn’t stop her from pushing those ideas.
Could you ask Kate whether her minimum demand models that go beyond 2050 (a joke to anyone in the investment/commercial industry) is based on Australia constructing 10,000km of transmission lines? How about the problem of realising the monumental task of constructing that much infrastructure? Did Kate assume that all those REZs will magically send their electrons to the load centres and give us those forecast minimum demand days?
I don’t care about how many speeches she makes. She’s one of many people who aren’t involved in the trading/commercial space who make out-of-touch calls with the energy industry.
Keen to understand how LGC-accredited renewable energy reduces demand… i think you’ll find you’re confusing rooftop solar there, which has nothing to do with large scale renewables. LGC accredited generators are suppliers of electricity… literally impossible for them to impact demand. So with that, i again challenge the idea that you can solve a ramp issue with steam turbines. Remove every renewable generator in the NEM… demand is still hitting zero, and you’re still having to implement a 24-hour shutdown procedure on your steam turbine.
Supply is just one side of the story, the other side is changing loads. go back a few decades and all large loads were also spinning machines more or less directly connected to the network. big industrial motors and the like. Now days, they’re more and more connected by VSD’s, and in general loads are far more digitised. That means there is less and less demand-side inertia. AEMO has already started responding to this with changes to its FCAS procurement, so it is very much real. You need smarter, faster reacting generation to meet the demands of evolving loads. How is a steam turbine going to help in AEMO’s new one-second FCAS market? Hot tip… it can’t.
As for the minimum demand forecasts, you can thank AEMO for those. they’re pretty much lifted from the ESOO, as anyone intimately familiar with network planning would have picked up. And hopefully i don’t need to point out that the ESOO is one of the key documents used in future planning for the energy network. Sorry if that’s funny to your investment bro’s, but its reality.
And to say she’s not familiar with trading and commercial.. mate, at least check her linkedin before running your mouth on things like that, and take a moment to understand the core of what the article is about. Regardless of whatever fuel you use, you need network control. Kate’s core points that she often shares with industry is the importance of proper power system control, how Australia’s energy market falls short, and the problems it creates. Forget fuel for a moment, if you don’t have good power system control schemes, you’re not going to have a stable market and you’ll be writing policy that will not has positive outcomes. Many of us in industry will fondly remember being told we need to add synchronous condensers to everything we build… look how that worked out.
What happened to all the big batteries? Down to 100KW by 14:00
So we have one solution that involves paying $11 billion (plus $5b for Snowy2?) and another that costs $6 billion. This is for transmission to supply renewable energy that costs $50 per MWh plus a RET equivalent of $50 MWh. Not discussed is an alternative that would see no transmission costs but augementation of coal based electricity at $50 per MWh
This project only gets worse, it’s now in the courts of Brisbane – no surprises after its history
https://www.courts.qld.gov.au/__external/CourtsLawList/Brisbane.htm
——–
Editor’s Note: CANADIAN SOLAR CONSTRUCTION (AUSTRALIA) PTY LTD -V- RE OAKEY PTY LTD in August 2023
From the VEPC report:
“The Australian Energy Market Operator (AEMO’s) modelling for the VNI-West Project Assessment Conclusion Report (PACR) shows that AEMO expects that for the next four years the wind farms in Western Victoria will waste up to 40% per year of the electricity that they are able to produce because the transmission system is not able to accommodate their production. The picture is no better with solar farms in the Murray River REZ expected to waste up to 40% per year of their production potential over the next eight years.”
Who exactly is to blame for this? A bunch of international renewables developers failed to model power flows properly when making their investment decisions and now the Australian public have to pay for it with the garbage proposals from the VEPC report? The RETs aren’t doing it? Everyone of us is already rewarding them by paying for environmental certs which we all see at the bottom of our electricity invoices.
Where is this magical $50/MWh coal coming from. Finance, Depreciation and overheads on a new brown coal plant will be about $70/MWh O&M $20-25 and coal $12-20
The curtailment situation is complicated. In the past AEMO was not allowed tell new aspirants if anyone else had applied to build plants near a particular substation because it was “Commercial in Confidence”. Also, it was probably felt that a public access system would encourage a gold rush situation where people just “pegged claims” and held genuine developers to ransom. Net result, too much capacity built in the wrong places because the sun and or wind resource looked good
Minor issue but if they don’t deliver power they don’t get LRECs
“We know, because of rigorous research and analysis conducted by AEMO and the CSIRO, the lowest cost replacement for this coal generation is renewable energy from the sun and the wind – backed up by batteries, gas and hydro to smooth the bumps in production.” [AEMO, from the AEMO (initial) Response linked above.]
The roles of AEMO and CSIRO in arriving at a conclusion that “the lowest cost replacement power…” has returned as a very hot issue, with seemingly both expert groups blaming the other for errors or omissions.
I am eagerly watching progress, eg in newspapers and Twitter. No longer is it only models at 30 paces. It is experts at 30 paces. I have no expert opinion to offer here – I am an interested bystander, but my impression is that there is a capital cost hole that must be addressed in the period pre-2030 and the arbitrary exclusion of nuclear power options is enigmatic.
Peter Farley
I wonder how coal plants in China manage since they too would be up for $120/MWh. Fact is that a coal plant that was allowed to contract in the absence of subsidised wind/solar would operate at ~ 90% availability. Fuel $15 (If it was brown coal fuel would cost <$5/MWh); capex $20; O&M $12
Hi, I just will like to understand how AEMO determine whether the information from SCADA is reliable. Thank you.
Hi Stephen, The SCADA signal should include a quality flag that would indicate the quality to AEMO. We could anticipate AEMO also includes some additional logic, e.g. to check for stuck readings and or readings outside a reasonable range, but I don’t think I’ve seen these documented explicitly.
Thanks Linton!
Using the MaxAvail as a means to avoid negative prices is inappropriate. If your intention is to turn off when prices are too low then simply include your minimum price in your bid. Including a price in your bid that is lower than your real minimum price is misleading and could/should be found to be in breach of the rules.
Hi Greg, I’m not a lawyer so would always defer Participants to seek their own legal advice where appropriate. However, I can offer that the AEMO Operational Forecasting and Dispatch Handbook for Wind and Solar generators indicates in S2.4.4 that the use of MaxAvail is ok, subject to the requirement that bids must not be misleading (NER clause 3.8.22A) and rebids must be consistent with the requirements in NER clause 3.8.22. The other good reference is the AER Semi-Scheduled Generator Compliance Bulletin and Checklist but that hasn’t been updated since 2022, as far as I can tell so can’t yet reflect good-practice use of MaxAvail in this regard, since the change on 7 August 2023.
New coal is a non-starter on the grounds that it’s complete and utter lunacy to contemplate building a new machine to turn large amounts of brown coal into even greater amounts of carbon dioxide.
Will power supplies be sufficient if current high rates of immigration continue (i.e. 235 k pa) as envisaged in the recent Intergenerational Report?
Interconnectors seem very constrained. Continuing planned outages on SA transmission lines?
It’s worth pointing out that there are two quite different “high temperature” risks being discussed here:
1/ Wind speeds may be low at the time of peak demand on hot days (typically 6pm-9pm).
2/ Wind turbine generator capability can degrade at high temperatures (the “mystery DUID” chart in Paul’s post).
The enhanced modelling by AEMO deals with the first type of risk: it has identified that wind speeds and therefore wind generation output might be significantly lower in a few scenarios (“reference weather years”) than was previously incorporated into ESOO forecasts.
It actually appears as though most of the difference relates to modelled wind speeds and output for a single peak demand day in a single reference year – 3 January 2015. See Figures 24 & 25 of the ESOO.
The second type of risk – generator derating at very high temperatures – applies to some degree to almost all forms of generation, and AEMO uses data provided by generator participants in modelling this – see Section 3.2 and Figure 18. Wind has the highest drop in capacity from nameplate levels under “summer peak” temperatures (above 37 degrees). However I don’t think that this derating or AEMO’s approach to modelling it has changed in this ESOO.
It’s also plausible that in the case of wind generation, this type of derating doesn’t matter much for reliability modelling – it’s low wind *speeds* that are the problem. If wind speeds are moderate to high, losing some %age of output due to heat isn’t likely to cause a reliability problem.
It is know that the generation from inverters cause these types of problems. There are IEEE papers on it. Damas et al did a good review on it in 2020. They identified a major source as from HVDC converters as well as series=connected capacitors. One wonders the part the switchyard at Redcliffs plays in this, particularly when SA is exporting.
Thanks Linton
Interested in the extended -$350 prices on Sunday 3rd. The circumstances and who set them?
While I agree with your sentiment that the reactions are over the top, the fact of the matter is that we rarely if ever had generation shortfalls before the ramp up of renewables.
If the energy market is unable to meet the demand from customers, then the energy grid is not meeting its core requirement and we should be reconsidering a new plan to ensure generation supply can meet demand.
Saying it’s okay because we will just shut down large energy users is a bandaid solution. We need chemotherapy to stop the spread of the cancer through the NEM.
Something I found surprising in these tables was the relative effectiveness of short storage in NSW, with 4610 MWh as 2h BESSs said to be equivalent to 4043 MWh as 4h BESS in improving reliability.
This doesn’t make any sense to me. All else being equal, 1 MWh of 2h BESS capacity should be able to do everything that the same storage in a 4h BESS can do and more. Is there something I’m missing that would explain how these numbers can be internally consistent?
Good pickup Tom, I agree that it doesn’t seem logical at all that more MWh of 2 hour storage would be required to match the contribution of 4 hour storage. I wonder if this is some artefact of the way AEMO’s modelling dispatches storage units.
Hi Tom, I noticed the same thing. The low relative effectiveness of the 2hr BESS must be driven by the derating factors applied, as described on p108. 2-hr batteries have the MWh capacity reduced by 25%, 4-hr batteries by 10%. As you notice above, the results above demonstrate that these deratings are much too harsh.
This “game play” of energy battery arbitrage is pure speculation at this point since it is not happening right now. You are only guessing what battery traders may be capable of doing in the future since we are seeing very little of it now in the other NEM regions. The current fleet of operational batteries are still playing the FCAS game even as the revenue pool shrinks further and further with each new battery commissioning. Furthermore, we can safely say at this point that battery “energy contributions” in Queensland are nil.
To say that intra-day arbitrage is easier to manage than very long-duration arbitrage (i.e. up to a year in duration) is not necessarily correct either. A battery operator has a smaller margin of error when trading on an intra-day scale since their arbitrage period is spread across fewer 5-minute dispatch intervals. If a battery trader forecasts low prices in the next dispatch interval and it happens to be much higher, then they have purchased electricity at a higher price and reduced their arbitrage revenue significantly.
On a yearly scale, we actually do know what months give us lower prices and which months give us higher prices. E.g. it is no mystery that summer demand is higher and prices are higher as a result.
The energy (or MWh) of these batteries are tiny which makes arbitrage difficult on the “sell side” and this is compounded by smaller time frames (e.g. intraday arbitrage). I.e. how do you decide when to sell your miniscule volume of energy that you bought earlier in the day? Do you sell it all during the evening shoulder period when prices are sensitive and risk selling when actual prices are too low? Or do you wait until demand hits its absolute peak later on? My point is that the small energy of batteries makes it harder to sell during the forecast high price periods.
I hate to burst your bubble, but energy arbitrage on a short-scale only works if you can predict the market perfectly. And this is obviously not the case. And this leads to a sobering conclusion with batteries. Batteries are only capable of arbitrage over the short-term which has all these major aforementioned problems.
There is a mental confusion in the terminology. There is no need for a one hour battery to discharge over an hour, it can run at a quarter power for four hours or full power for one 5 minute bidding cycle and then ramp down over successive cycles over 6-8 hours if prices fall.
It can sometimes charge at night even buying power at a positive price that is still less than morning peak. It won’t have two full cycles /day but it may have 1.2-1.5.
The beauty of batteries vs pumped hydro is that it is easy to increase the duration as demand for longer duration increases and as the cost of storage falls. Pumped hydro is pretty much stuck once it is built.
If for the foreseeable future we retain the existing gas and hydro fleets and triple solar and wind output, the peak output from new backup/storage will be around 10 GW less whatever demand response is implemented, and for all but a few hundred hours per year wind/solar/hydro/gas will meet demand. So, in the medium term to 2030 or later, we should be focusing on backup power rather than energy. In the current environment it is difficult to see batteries with more than 1.5-2 hours duration being economically viable because on all but a few days per year, the batteries will run at far less than peak power and their output will therefore be spread over the entire evening peak.
The next limitation on storage is excess wind and solar capacity. All reliable grids have significant excess capacity both in terms of peak power but also annual generation. In 2008 the coal and gas capacity on the NEM could have comfortably provided 280 TWh, but in fact provided 195 TWh. If we build enough wind and solar to provide 280 TWh at normal capacity factors then on seven days out of ten there will be excess energy and hydro will handle most of the peaking demand, new storage will be active for only a few hours and will rarely need to run at anything like peak power. We can also see that if you examine the seasonal variation in renewable share on OpenNEM against the running average renewable share, that there is not much variation, typically winter renewable share is about 90% of the preceding summer renewable share. If there is 280/195 i.e. 44% excess capacity then seasonal variations will easily be covered.
Consequently depending on the cost of long term storage it is quite likely that we can easily exceed 83% renewable grid with no more than 10 GW/25-35 GWh of storage
Over 40% capacity factor certainly seems optimistic even for latest-technology onshore wind.
That said, analysis like this based on NEM market data is going to yield an “apples vs pears” comparison for several reasons:
– the existing fleet is up to 20+ years old (with its underlying design parameters even older) so we would obviously not expect that fleet to deliver to new-build levels, not even closely. Today’s electric cars probably perform a lot better than whatever you could buy five or even three years ago. I’m not an engineer but understand that there have been numerous improvements to turbine technology and windfarm modelling / layout methods, as well as in site selection, that should yield better performance under the same wind regime.
– registered fleet capacity always runs ahead of actual achievable maximum output while there are farms being built, commissioned, and extended. To allow for this you need to go through the data and manually back out farms (or adjust maximum capacities) for those still in the commissioning / ramping up stages.
– economic offloading and constraint-driven curtailment will drop fleet capacity factor a bit if the measure is based on actual output; availability is a better numerator if the aim is to compare technical performance before these impacts
Finally, I very much doubt that curtailment or offloading would materially affect minimum coincident output levels across any significant group of farms let alone the whole NEM fleet – minimums will be driven by widespread still conditions in which case curtailment / offloading are almost certainly going to be neglible. On the other hand maximum output levels will almost certainly by affected by these factors.
Thanks for picking up on some shortcomings in my quick explanations, Allan … as always.
It’s a valid and worthwhile analysis Paul, and you’re aware of the caveats, but just pointing out for other readers why its results can’t be directly compared with new build capacity factor estimates.
Who cares about monthly wind capacity factors anyway. A monthly figure totally hides one/two/three-day wind lulls where capacity factors are absolutely abysmal (<500 MW total NEM wind gen over evening peaks) and we're faced with actual LOR2 conditions. A purported high capacity factor month can completely hide severe price spike days that took place over that time period due to very low wind gen.
@Allan O'Neil
The bulk of the NEM wind farms are already located in the best wind resource locations in the NEM which is the VIC/SA border. Are the technological breakthroughs in wind the same ones where we just build bigger blades without any conception of their longevity and maintenance costs?
Good question about managing risks of contingency events tripping off SA rooftop PV Paul. There are various system normal constraints which limit flows from Victoria into SA via Heywood under conditions of high PV generation in South Australia for just this reason – to leave headroom for suddenly increased flows if a large block of rooftop PV were to trip because of some disturbance in SA – for example the V::S_NIL_MAXG_1 and _MAXG_2 transient stability constraints. But on a low demand day like today when flows are exports out of SA, these constraints wouldn’t bind as there will be sufficient headroom for imports after a contingency anyway.
Alfred Hill makes a good point about averages, it is the low points that cause the problems.
Also the best sites are taken and if Germany is any indicator the capacity factor falls away over time due to ageing machines, best spots taken and interference by overcrowding.
https://www.theregister.com/2023/09/04/microsoft_australia_outage_incident_report/
The Register has some detail from a Microsoft report, which confirms the power issue occured during the thunderstorms. It notes that only 3 staff were on site to try and get 5 tripped chillers and 2 backup units up and running. Only one restarted successfully.
The loss of cooling created a cascade of problems. “Storage hardware damaged by the data hall temperatures “required extensive troubleshooting” but Microsoft’s diagnostic tools could not find relevant data because the storage servers were down.”
It does raise a question of whether the “public cloud” is a suitable place to house critical infrastructure.
Good illustration of the flexibility of large-scale solar (and no doubt wind) responding – via the price mechanism – to the inflexibility of its small-scale rooftop PV cousin.
It’s worth pointing out that *which* large scale generators reduce output in these events isn’t directed by AEMO for security or similar reasons – which is a common implication of the term “curtailment”.
What’s happening is normal market-based dispatch, driven by the prices at which generators have offered their output. We don’t say that there has been “curtailment” of coal-fired or other conventional generators in these events when they bid their output down to mingen, so I’m not sure why we should label large scale solar or wind bidding and responding to market prices as “curtailment” either.
If AEMO had needed to step in and have distribution companies reduce rooftop PV exports or trip inverters offline for stability reasons, that would certainly constitute “curtailment”.
Thanks once again, Allan
Reminds me that on 9th May you wrote ‘Renewable curtailment – forced and not quite so forced’ that explains more…
Might help some readers here, as well.
It will be very interesting to see how this impacts Tasmania’s hydro capacity as I understand their dam levels are already quite low and Marinus Link is not yet in place to provide redundency for Basslink.
Was the trip this morning associated with the strike or is it something else? Nothing in MTPASA from Saturday.
Might help, Mike:
https://wattclarity.com.au//articles/2023/09/27sept-kogancreek-part3/
Allan, at Squadron we use ‘economic curtailment’ and ‘grid curtailment’ to distinguish between a market response and a binding constraint equation. We VRE are used to semi-dispatch caps ‘curtailing’ our generation and that does not discriminate between a market response or a bidding constraint. We also don’t get to withhold our fuel until prices rise, so I guess it ‘feels’ more like a curtailment.
Just offering some context… Nothing wrong with your comments 😉
One small observation: The battery seems to have been charging.
Now consider recent reports that new apartment buildings must have fast charging facilities installed. Quite apart from the cost of the batteries and their connection, my experience (long ago but not forgotten) in designing fire-resistant buildings is that is a highly technical matter. Combine this with the frequency and severity of large battery fires, it’s time that the community became educated on these matters.
Quite frankly, the idea of large batteries in parked cars and/or chargers in basements of highrise buildings is, in my estimation, somewhat scary.
John
Can you explain what data you are looking at as the basis for saying ‘The battery seems to have been charging‘.
The AEMO market data I am looking at suggests it was last charging at 12:45 (which is what we’d expect – i.e. when prices are low, in the middle of the day).
Paul
Saturday 30th Sept appears some rooftop solar owners in Adelaide were being curtailed-
https://forums.whirlpool.net.au/thread/36yz02x6
(not my 6kW 3 phase producing 39kWhrs for the day)
A bright sunny day hitting 30 plus degrees but homes were still too cool for aircon so you could easily see why in some areas.
Received this from SAPN –
Thanks for that piece of context, Genevieve
Hi team, I love all your articles, despite not always understanding 100% of the terminology. I was wondering, is there any tool available, that a casual observer like myself could see live bids into the NEM for wholesale prices? I’d be curious to see for example: “Eraring – Black Coal – 5MW @ $120 / MWh”, and every other generator listed out with the same detail. I’m sure it’s probably a dumb question, and it’s probably nowhere near that simple, but if such a tool does exist, i’d love to use it! I’m eyeing off your “Nemreview” tool’s website, and thought perhaps this is where you get all your fantastic info from.
Thanks in Advance!
I don’t understand why I, as a customer of a PV sales company, needs to deal directly with Goodwe, my inverter’s manufacturer, for guarantee issues and/or the installer, whose contact details I never received.
My commercial arrangement is with the seller.
Until now, I believed that the seller is, under federal consumer law, able to suggest that I communicate with his suppliers and installers, but that required my prior permission.
Any comments?
Re that Whirlpool post ‘smokiedawson’ posted today Sunday 8th Oct at 4:40pm the following-
“Mine went off again today 1030am – 1230pm….so annoying and no notification. At least it reconnected.”
Checking my Goodwe 6kW 3 phase inverter tonight showed exactly the same and it’s the first time I’ve noticed throttling since a new install in December 2022. I still managed to generate 38.3 kWhrs today with 6.64kW nameplate panels and I was heading for a perfect bell curve output at 4.6kW until 10:30 then zero for 2 hours then on again at 5.87kW at 12:30. It’s not the Goodwe inverter problem here as I’m on TOU with those 2 white Dalek communicators on top of the power box
I should add I’m all electric with Origin and 2 TOU meters one for general 3 phase and the other for CL1 for HWS and hence the 2 Dalek communicators presumably.
yes its very frustrating, and I find the lack of transparency and communication not acceptable. Trouble is, I think all the complaining will make no difference at all, as sapn do as they please.
There’s no lack of transparency from those tasked with dancing on political eggshells whilst struggling technically with the growing fallacy of composition problem-
https://reneweconomy.com.au/solar-switch-off-a-must-for-all-states-says-aemo-to-control-seven-erarings-of-rooftop-pv/
You just have to read between the power lines.
Would be interesting to see what is being offered by coal units for the lower (runback) service. Prices seem to spike around middle of the day when solar is at maximum implying coal units are sitting at minimum load with nowhere to go backwards. Implies to do so would involve stinks like removing a mill from service incurring coasts in expensive auxiliary fuels to do so, or even decommitting a unit that will then be needed in the evening in the stack. Western Australia has been here for a while now. Runback is a bigger issue than spinning reserve.
Thanks for this update. I too was very surprised by the “return to service” statements.
Paul
Limited coal supplies are not a good thing, but probably less of a concern for summer reliability than in winter (cf last year’s winter energy crisis).
Summer demand and price peaks tend to be of much shorter duration, and average demand levels are lower in summer than winter. Provided the power station manages its coal stockpile so that it’s able to operate both units at full capacity over peak demand hours during periods of hot weather and higher demand (which don’t extend over weeks and weeks) – and that coal deliveries don’t fall to very low rates – then I’d be surprised if this leads to reliability isses this summer. But I’ve been wrong before.
Allan
This seems to ignore NSWs increased import capacity and declining grid demand. Not only have VNI and QNI been upgraded, but in the past when Liddell was operating, southbound capacity for imports was partially crowded out. With PEC on line and generous generation surpluses in Victoria and SA, NSWs practical import capacity will increase by 1-1.5 GW
Since 2019 grid demand in NSW has fallen by almost 6% and ongoing energy efficiency and behind the meter generation and storage will probably continue the decline. While annual demand is not directly related to the annual system peak, they do follow similar trends.
While AEMO is rightly conservative, it seems to me that this forecast is overly so.
What did Stanwell get fined for the setting for going into Islanding mode that made the recovery from this event much shorter, let alone if the state did indeed get sent black, having it still operable would have been much faster than the old AVON GT at Tarong whining into action or WIVENHOE running out of water due to some failed attempt to start from black again. Still don’t see why CSEnergy continue to get such favorable treatment.
What was the last official word for C3 RTS?
“A decline in intermittency is predicted in regions of Eastern Australia by 2099.”
How can anyone take these types of conclusions seriously? Making forecasts that far into the future with modelling that utilises highly sensitive inputs is suspect.
“Around Townsville” lol you mean almost the entire of NQ, it went all the way out to Julia Creek, Cloncurry, south to Collinsville in the Whitsundays. It is a major city so not all of consumption would have been major industrial users. Either way it was really a non event, had it happened during AM or PM peak hour then possibly it could have been different.
There’s an ABC article online at https://www.abc.net.au/news/2023-10-23/power-prices-to-stay-higher-despite-wholesale-market-slump/103004170.
I recall about 20 years ago that the manager of one of the MacGen stations I worked at, bid on FCAS / black start ability seeing it as easy ‘money for jam’. He eventually had it pointed out to him just how costly it would be should the event occur so quickly withdrew that bid.
PEC was purported to provide energy security to SA.In fact on most days it will drive up prices in SA because higher prices in NSW than SA will drive output from SA to NSW.
However on high renewable days in southern NSW, capacity on PEC will be mostly taken up by Riverina solar and Western slopes wind.
With storage being installed in SA and increasing surpluses of cheaper Victorian brown coal it is difficult to see any more than trivial amounts of NSW energy flowing to SA.
You are mixing up energy security with spot prices. PEC will result in greater energy security for SA as it will provide an additional transmission path including inertia. This will further reduce the need for the operation of fossil gas based generators just to provide inertia and fault services.
In addition with PEC in place it will allow more RE to be in the SA grid should there be a failure on the Heywood interconnector such as when the 500kV tower was knocked down in Victoria. This fault significantly increased the total cost of electricity in SA due to reduction in RE in the grid and the increased need for additional grid services to be provided.
The peak wind conditions in SA are largely out of phase with those on the Western slopes, so exports from SA will be good during those periods.
The capacity to handle solar in all markets when solar generation is good is going to be a problem, however when conditions are good in eastern or northern NSW and poor in SA, energy will be able to flow into SA. The same applies for wind.
I agree that in the next few years the scope for cheap energy to come from NSW to SA will be very limited, however as NSW builds out its RE capacity the situation will change.
@Craig Fryer I didn’t realise how simple the solution was! Just build 900km of 330kV lines! Easy and cheap as! I’m sure this dirt-cheap endeavour will be completed by July 2026 and we won’t experience any delays or cost blowouts with our trusty Australian workforce. Furthermore, I’m sure our energy situation will be perfectly fine from now until then.
“The peak wind conditions in SA are largely out of phase with those on the Western slopes.”
Patently false. The same high-pressure systems constantly sweep across SA and into the Western slopes at the same time and give us those nice NEM-wide wind droughts.
“when conditions are good in eastern or northern NSW and poor in SA, energy will be able to flow into SA. The same applies for wind.”
PEC connects to Wagga Wagga which is as far south-west of NSW as it gets. What makes you think generation in northern NSW will easily make its way into SA?
“As NSW build outs its RE capacity”
You mean that genius idea of “over-building?” Which effectively means committing to large-scale infrastructure financing without a return on investment? But maybe it’ll come by once all that transmission is fitted out, right?
The reason why there SA isn’t a net exporter is due to the lack of transmission capacity in NW Victoria going back towards Melbourne. When RE conditions are good in SA and Victoria, energy in NW Victoria flows into SA as some of the RE producers in NW Victoria seem to be able to supply the grid at more negative prices than those in SA. This may be due to older contracts.
The end result is a net reduction of 300MW export capacity from SA to Victoria. This has resulted in considerable reduction in RE exports from SA to Victoria. When transmission capacity out of NW Victoria towards Melbourne is increased, SA will become a net exporter. Potentially EnergyConnect will come online prior to this and may result in SA become a net exporter earlier.
The real limitation here is on northward flows out of NW Vic and SW NSW – as Peter pointed out above. The physical solution for that set of constraints is Humelink.
Might be interesting to look at market prices between SA / VIC / NSW on the same graph. I suspect that once much of the constrained on dispatch in SA ceased courtesy of synch comps etc and the price of gas (including start up cost) into the SA stations simply the RE was sopped up at home and shortfalls met by low SRMC coal out of Vic (or Tas). The market doing what the market does.
As Andy says, while transmission constraints played a part, it is mainly the price difference between Victorian brown coal and SA gas. In 2019/20 when gas was cheap in SA and coal output was limited by outages in Victoria then SA was a net exporter. As soon as Victorian renewable output increased to the point that there was spare coal-fired capacity, then Victoria became a net exporter again.
However over the last 5 weeks SA has become a net exporter while averaging 83% renewables, so it seems that transmission constraints can be managed at least some of the time.
As for energy security SA would have been far better served by a similar investment in renewables/storage.
Thanks Dan
I think you should be able to adjust for (2) and (3). (E.g. QED reports do). I expect this would be material and give some support to the improving technology claims.
Thanks Ben. Yes if I have time in future I will loop back and examine how availability factors have been evolving.
You can access the CSIRO data here.
https://data.csiro.au/collection/csiro%3A44228v10
the select “files” tab at the top – then select final 22/23 xls.
CF is detailed on sheet “Apx Table B.8&9”
you can access the capacity factor discussion by searching for “D.2.1” in the 2022/23 Gencost report
Would be good if Joel Gilmore could provide a link to his just released working paper on this subject. Doesn’t appear to be on Griffith Uni website.
Joel Gilmore’s working paper can be found by going to his Linkedin where it is available for free via ResearchGate.
As a less drastic solution than reconfiguring the NEM into a nodal system, how about enacting a higher market price floor for renewable (or inverter-based, or non-slow start…) generators? Snowy Hydro has lodged a rule change request with the AEMC to limit semi-scheduled plant to a floor of $-100: https://www.aemc.gov.au/rule-changes/addressing-access-risks-dispatchable-resources
I’m not sure if that proposal is progressing or what discussions are occurring around it, but at a glance it seems a reasonable approach to limit the impact of out-of-order bidding on interregional transfers.
In a system without any or many slow-start generators it would certainly be preferable to have a significantly higher MFP than -$1,000/MWh. Variations in offer price floors to deal with priority access are also something that ESB / AEMC has been considering – not sure where that has got to, perhaps the Snowy request has got wrapped up in that process. Others have suggested changes to threshholds for interconnectors going into the LHS of constraints. There’s a slightly ad-hoc feel to all of it.
Still waiting for somebody to explain how massively expanding wind and solar capacity, and the transmission to connect it all, is expected to reduce consumer electricity costs.
Allan’s detailed look at settlement residues and constraints is a great explainer of just one largely hidden additional cost imposed on consumers directly from the renewables push, but looking ahead 5yrs it is clear the situation must get far worse.
Surely there’s an explainer somewhere by now…
This is a great article. When I read the nsw energy infrastructure report I gained the impression that this part of the network would be upgraded via:
“South West REZ transfer capacity is delivered through ISP projects. The cost has not been included here as this capacity will be delivered
through Project EnergyConnect, HumeLink and the Wagga Wagga to Dinawan section of VNI West.”
In your opinion Allan, will these projects relieve that constraint? I note via EIS filings that there are a number of large prospective wind projects located East of Balranald that will connect to Energy Connect. Surely there must be a way for that power to get to the load?
Thanks David. As far as projects relieving the constraints, I only considered Energy Connect, as that is actually being built. Humelink and VNI West still seem way too far from breaking ground to think about as providing relief anytime soon. It’s possible EnergyConnect will worsen these specific constraints as it will bring in more flows from west of Wagga and as far as I can see do nothing to improve capability into the rest of NSW.
Thank Allan.
I personally see Humelink as a project that will be built and more or less on time. Around $600 m of early works capital is already either committed or spent and approved by the AER. In my opinion it’s something that developers and existing stakeholders will find it profitable to take a view on. If you don’t think it will happen on time there are lots of implications. If you do think it will happen (and the target date is 2026) then that’s 2.5 GW of more power that can get to Sydney/Newcastle.
You’re probably closer to it than I am then David. I just hear about lots of landowners and others arguing for it to be built anywhere but on their land or underground or not at all and wonder how it’s possibly going to happen on time, even without the construction delays and cost blowouts that seem to beset just about any major engineering project these days.
At some point in Australia’s energy future, reality and common sense will kick in and everyone of us will at that stage realise we have been saddled with the most glorious of stuff ups in our nation’s history.
That is, we will have destroyed a fully functional, sturdy, secure, reliable, and cheap means of meeting our fundamental need for huge amounts of electricity.
It’s all been predicated on the theory that man made climate change is destroying our Earth and the only way to address this concern is by purchasing and installing horrendous amounts of costly short lifespan “renewables” and batteries. We now have a privatised national electricity grid, in which hundreds of entities all expect to make fat profits via a huge and bewildering accounting system – extensively outlined above – that few people could comprehend.
How can it be that we are destroying healthy coal fired generating plant with great urgency, whilst elsewhere in the world, new long life coal fired plant is being built at a far greater rate than that at which we demolish our own.
A sensible nation (if it truly must reduce its emissions) should have been doing so by building large scale nuclear powered generators at the various sites, where coal fired generation is reaching end-of-life, and operational transmission lines and already exist.
We would be seeking to build small modular reactors at these sites where growing demand for minerals processing and other heavy industry development could be encouraged by cheap electricity prices and the minimised need for new and lengthy transmission lines.
Of course, such a national electricity grid would have been best provided by full government ownership and run by the best qualified commissioners and electrical and mechanical engineers and executives the nation could enlist.
We must presume that such a coterie of altruistic Australian professionals will always be waiting patiently in the wings.
If not, we can only shudder at the potential disaster now looming ever closer.
Interesting comments Rob. I find it bemusing that you seem to have enormous faith in technologies and technical subject matter with which you’re evidently personally comfortable (“healthy (!) coal fired generation”, large scale nuclear, small modular reactors), but prepared to dismiss in such offhand terms other technologies and equally well-grounded areas of scientific and technical expertise (“horrendous amounts of costly short lifespan “renewables” and batteries”; “theory that man made climate change is destroying the earth”; “huge and bewildering accounting system”).
I think Rob’s concerned that the efforts (by state and federal governments) to chase emissions reduction targets using our electricity system is not going to achieve any measurable emissions reduction, but will certainly deindustrialise our economy while it industrialises the countryside.
The only conceivable benefit for doing this at all is if we end up with very low cost electricity and the associated benefits.
I’d very much appreciate some insights into how you think all this is meant to result in low cost electricity for consumers. I think that’s a fair enough question.
Common sense surely dictates that if it actually shuts down in August 2025, the Eraring Power Station should be kept in some form of “care and maintenance” state.
It will be a massive tragedy if the first action following closure was to be the spectacular dropping of its huge chimneys – something that coal’s detractors typically find absolutely delightful.
In the potential event that Energy Minister Bowen’s fanciful meddling backfires on us all, Eraring would become a vital asset whilst new options were needing to be brought forward.
This is only going to get worse with the introduction of the EnergyConect loop constraint
https://www.transgrid.com.au/media-publications/news-articles/more-than-meets-the-eye-to-complex-transformer-convoy
Might not be such a problem after all. Five phase shifting transformers to go in at Buronga, will allow adjustment of power flow on energyconnect independent of dispatch.
If I may reply to the above response to my earlier contribution:
My reference to “healthy coal fired generation plant” was in reference to coal fired plant (that still has many years of valuable service life), being demolished prematurely. Perhaps a bad choice of words, but we taxpayers are spending billions of dollars on renewables and the essential collateral equipment required to connect them to the grid, while we should be dragging the last bit of useful service life out of our existing plant.
Furthermore, can there be any doubt that horrendous amounts of taxpayer cash are required to facilitate the introduction of intermittent renewable generation into an already functional system and then deal with the reality that this “well grounded area of scientific expertise” comes with short life spans and more horrendous expense to cyclically rebuild it all every 20 years or so.
As for the porridge of cost and profit sharing amongst the hundreds of corporate entities now enmeshed in the grid, I defy anybody to explain how the incredibly complicated financial processes are working for the benefit of all consumers.
Timely reminder that market pricing is supposed to be an investment signal as well as a dispatch signal.
The mistakes baked into our critical electricity system will require significant effort (and guts) to undo.
In SA there are only 4 solar farms that are 100+ MW. The others are owned by SA Water, the largest of which is 25 MW.
Utility solar was and should be the third leg of the stool in the NEM> there is no way that utility solar can be generated and delivered, billed etc for the same cost as rooftop solar.
In effect the battle will be for utility solar + a lot of storage say 60-70% R MW/250% R MWh (R being the rated power) vs wind with 15-20% R/.5-1 R MWh storage for the remaining 50% of the energy after behind the meter generation and storage, hydro and waste to energy take the other 50%
It’s really the joint effect of non-price responsive rooftop PV crashing into the flexibility floor (minimum stable loading) of coal fired generation that’s driving this. With a different synchronous generation mix the issue would be less severe (unless the synchronous stuff was nuclear, in which case it would be worse).
It’s also important to understand that the Vic & SA results also reflect the endemic constraint issues on VNI not just the “wind correlation penalty”.
Thanks Allan!
Your first point was the reason for the asterisk at the top.
Good point about VIC1-NSW1, which you recently wrote about in ‘What’s happening around Wagga?’.
This aligns well with an AEMO webinar a couple of years ago by Jack Fox, explaining that the weather is now the largest source of forecasting error.
Hi Dan: I have done some simple modelling using the Griffith University 42 year historical wind and solar production work versus the ISP 2022 Plan and the CSIRO and for storage requirements.
I found the ISP 2022 storage requirements more reasonable but still only just (my finding it covers up to 1 in 20 year events). Thank you for your analysis – very interesting. I looked at Figure 14 in the Griffith Study and it appears this event is less than the worst “1 in 42 year event” for 2 days of low VRE output. Figure 14 compares % output to average VRE output.
You can also see in this screenshot the VIC-NSW interconnector running counter-price. It’s been doing that all day – I saw over 1GW of counter-price flow earlier – with intervening periods of constraint when the NRM_NSW1_VIC1 constraint set has been invoked due to the accumulation of negative settlement residues.
I am not sure the costs of inadequate transmission from Wagga north-eastward are being accurately captured. Possibly (probably?) the need to activate RERT today would have been avoided if that south-westerly generation was available to the Sydney load centre.
Yep, noted in this subsequent article:
https://wattclarity.com.au//articles/2023/12/14dec-nsw-demand-13031mw/
————-
PS have added more detail in the subsequent article ‘Disallowed flow north on VIC1-NSW1 another significant factor in tight supply-demand in NSW + QLD’
It seems these constraints in southern NSW have reduced flow in both directions in the past. I seem to remember when supplies in Victoria were very tight that there was no flow from NSW. Maybe Transgrid should concentrate on fixing those issues rather than splurging on VNIWest and even perhaps Humelink
I would say that rather than being “disallowed”, what was happening here was simply a case of insufficient transmission capacity between the Snowy region and Sydney to carry all the available southern NSW generation + additional flows from Victoria. Once Snowy bid in its 2400+ MW from Lower and Upper Tumut, Origin 400+ MW at Uranquinty, plus hundreds of MW of solar generating in the area, there was no northward transmission capacity left to carry anything from Victoria. The N>>NIL_4 constraint (on one of the 330kV lines out of Yass) bound solidly, implying that any MW from Vic would have been at the expense of a matching quantity from a southern NSW generator – a zero sum game.
The only fix for this sort of limitation is augmentation of the main Snowy to Sydney flow path.
From the previous article on the issue, it sounds like there is additional transmission capacity that can’t be used due to the risk of a trip on one of the 500kV lines. Should this occur the “excess” energy would flow into the 330kV lines, which wouldn’t end well.
I suspect that the installation of a utility scale battery of sufficient capacity to provide fail over capacity should the 500kV line trip, would unleash additional transmission capacity. The Victorian Big Battery provides a similar service during part of the year for the VNI.
I can see this been a lawyer’s picnic.
Paul, you continue to provide the most erudite and accurate analysis of all things NEM coupled with in-depth common-sense assessment of the relevant issues. Thank you. Greg
If I recall, this is a side effect of snowy hydro. Some of it is alloVic, some in NSW. Ramping up the hydro in NSW results in some power flow into Vic and vice versa. It might better if the snowy had its own economic region again.
*some is allocated to Vic, some allocated to NSW despite being phyically located on the same transmission network. Comment was truncated by some error.
Similar problems exist with NSW to Victoria flows. During hot weather a few years ago, I was shocked to see ~1000MW flowing into NSW against the price. In this case, part of the problem was Melbourne’s HV network (constrained at South Morang). I suspect we need to add phase shifters or some other means of regulating real power on those 132kV parallel paths, possibly the same on the EnergyConnect loop.
As to comments that these problems are new, they are not. The supply-side issues are from the design of VNI by then state and federal governments. The big change on the demand side is population growth in the urban heat island of Western Sydney.
AEMO’s on the day forecasts for Victorian rooftop PV maximum output have grown from 2,730 MW (forecast early this morning) to a latest forecast of 3,099 MW. Currently running at about 2,900 MW.
I’d say that’s the biggest factor in the falling market demand forecasts you’ve highlighted Paul.
Umm … the more interesting “records” will be curtailment? … in order to maintain demand generation balance ?
For what its worth, I’m putting my money on the Callide U4 operators making a switching error when doing a test on the battery bank. Things just cascaded from there.
The problem occurs a lot more often than people think. I know of at least two other similar incidents this century. Problem exacerbated when they went from shaft driven to AC driven oil pumps.
Great summary Alan! Interesting times, particularly with the Victorian emergency solar backstop applying to small scale generation from mid this year.
Still waiting for somebody to explain how wind and solar reduce retail electricity prices.
I feel obliged to point out that a state’s demand met almost entirely by behind the meter solar sees the transmission grid at very low utilisation. But that grid is not paid for by utilisation, it’s paid for by guaranteed returns on the amount of stuff it owns. So it costs the same whether you use it or not.
And since the grid is growing, AND the grid operators now have responsibility for network stability, the costs can only increase.
For some reason nobody talks about it.
It would seem that safe minimum demand was calculated when most generating units were typically 400-700 MW and had response rates of <1% per minute. The loss of one generator or transmission line when only eight were online would be a significant disruption, which would probably trigger further shutdowns, just as the 400 MW loss of Callide C4 led to the loss of 2,300 MW of generation as other units tripped off to protect themselves from instability.
Now however, not only batteries and the newly installed syncons but curtailed wind and solar facilities can respond within milli-seconds. In addition, the tightening of primary frequency control requirements on conventional generators means that gas, coal and hydro plants also respond more quickly to disturbances. Thus, the minimum safe level of generation will gradually decline in Victoria, then NSW and Queensland, just as it has declined from 250MW to 80 MW in SA.
In fact, it is clear that SA can already run on zero gas for hours if not days at a time but the risk of a mistake both in short term load shedding and the long-term political risk to the whole transition are not worth the benefit of saving 80 MW of gas generation.
Similarly, because across Victoria and SA it appears around 600-800 MW of wind and solar was being curtailed and in Tasmania and Victoria 4,300 MW of hydro was only running at less than 500 MW while batteries were actually drawing power, it is quite possible that Victoria/SA/Tasmania could have run at 100% renewables for at least four hours if Victoria's coal plants were flexible enough.
Thanks Peter
I don’t think the minimum demand issue is to do with loss of a large unit – Victoria is strongly interconnected to three other regions and frequency control wouldn’t be the problem. I’m still inclined to think it’s something a bit more arcane like voltage control, or Victorian AUFLS becoming ineffective because many distribution feeders would be carrying only small net loads with relatively low demand but high rooftop PV generation.
Allan
Sorry but this outage schedule looks like nothing more than minimum spend to keep a 1970’s power station struggling along – hardly a significant recapitalization which is what would be needed were the plant to be campaigned in the longer term. A read of the annual report says it all. Red ink in Australia and the second para on page 3 says it all. Major investments will be in “flexible plant” not old colliers. Hardly surprising. The asset managers are doing exactly what they should be doing – spend nothing they don’t need to on an old plant gliding to retirement. Keep it safe and sort of reliable – the trick is to get the balance right between maintenance spend and losses in the wholesale market from unreliability. Expect forced outage rates to go up the closer they get to retirement. And they won’t be the only owners juggling these balls in the air. Best hope they don’t all drop them at the same time.
Thanks for sharing your thoughts, Andy, and apologies for any confusion inadvertently caused.
The above line of thinking followed what’s noted on p3 of the CLP report, specifically:
… so will watch with interest to see the extent to which operational performance at Yallourn is restored.
Bit confused by the 2023 timelines in this Watt Clarity article.
The Energy Safe Victoria Annual Report is for 2021-22 year and states Cohuna Solar Farm started commercial operations in 2021 and “on 18 October 2021, three minor grass fires were reported at the farm that originated from the solar panel rotational control boxes. These boxes allow the solar panels to track the movement of the sun. We issued a direction to disconnect the tracker boxes as a result of initial investigations carried out after the fire. Further investigation uncovered other issues with the T4 connectors connecting the solar panels and with the T7 connectors connecting the fuse boxes to the direct current harness cables. We issued a second direction to stop the use of T4 connectors. This meant that the farm could not operate.
We worked with manufacturers Enel and Soltec to rectify the issue and ensure the farm was safe to be operated. The T4 direction remained in place until a solution was implemented. As of 30 June 2022,
the direction for the tracker controller boxes was still in place. The supplier has re-designed the product and is in the process of getting it safety tested to show compliance to the relevant standard and
that it is safe to be connected to the supply.”
AEMO databases have Cohuna Solar Farm moving from in-commissioning to in-service in mid-2023.
The Watt Clarity article has 5 outages in its first 6 months of commercial operation including a ESV direction to disconnect equipment on 8 December 2023 following a grass fire in November 2023.
Are the year dates mixed up or was the problem not resolved in the first place? Which means the unresolved problem caused the November 2023 fire which could have had serious implications for surrounding properties.
It seems you are confused about dates, Jennifer.
You say …
… but to be clear I have not specifically looked at when commercial operations started.
You say …
… but to be clear the chart at the top only looks at calendar 2023 (so starting 14 months after the first fire reported in the ESV 2021-22 Annual Report.
Finally you speak of …
… but readers must be clear that I have no knowledge, and am not commenting, about whether the fires in Oct 2021 and Nov 2023 are related or not.
Hi Paul My comments are not a criticism of your excellent article but complimentary of your digging for further information on what is going on with Cohuna Solar Farm and whether the issue is visible to the market and AEMO. My concern is AEMO’s Generation Information Page has Cohuna Solar Farm going into service in mid-2023 which is after the concerns of the ESV reported a year earlier have supposedly been addressed – but have they been addressed? as ESV directed Cohuna SF to cease operations 6 months after the “in service” date for what looks to be the same issue.
According to AEMO’s database on expected closure dates Cohuna SF will close in 2045.
Are the issues experienced by the “single axis tracking” Cohuna SF isolated to Cohuna or are other SAT solar farms experiencing similar issues.
Its unclear as the ESV report states “we also issued a safety alert to warn other solar farms about the importance of maintaining electrical installations and clearing vegetation to minimise the risk of
fire. An owner or operator of a solar farm should have a plan in place that minimises the risk of fire.”
It would be great to know if the issue is design or maintenance related. Design means more solar farms could be at risk of fires.
As an active rural fire fighter who has recently worked alongside Victorian fire fighters in Far North Queensland (yes only a couple months ago FNQ had fires now we’ve had torrential rain and floods), I sincerely hope there is a fire management plan for all solar farms as rural firies in Queensland cannot attend structural fires or fight fires under power lines.
“an elevated risk of insufficient demand to maintain a secure operating state”
Shouldn’t that be
“an elevated risk of over supply to maintain a secure operating state”
Do the curtailed wind and solar farms get paid for production loss?
Hi Peter – I think the AEMO wording is appropriate. Potential oversupply is readily managed – at worst AEMO could direct generation off; on the other hand issues like voltage control or impairment of AUFLS capacity are related to low operational demand, not excess large scale supply.
Wind and solar was not curtailed in most cases – they simply offered their output to the market at prices not low enough to get dispatched. Nor would they want to have been dispatched at -$400/MWh. Generation curtailment (resulting from network constraints) does not get compensated in the NEM in any case.
Allan
Jonathan, you arguably should have taken Peter to task for besmirching your excellent article.
Peter: inertia is not kinetic energy. You need to go back to the source, Isaac Newton.
Inertia is resistance to velocity change. A T/G has the same inertia when it is at rest as when it is at 2,970 rpm (49.5 Hz) or 3,000 rpm and the same whether it is generating 50 or 500 MW.
You also appear to be confusing energy and power. In your example, when the frequency started to drop the T/G was delivering 500 MW. At 49.5 Hz, if its AVR took no action, it would be delivering ~490 MW – the drop due to the drop in voltage due to the drop in speed/frequency.
If the drop in frequency occurred in 2 seconds the T/G delivered ~990 MWs of energy to the system (in those 2 seconds) not 90 MWs. How much of this came from a decline in the T/G’s kinetic energy is irrelevant.
Yes, a 50 MW battery only takes a tiny amount of time to start delivering 50 MW, and if we ignore that short delay and any tiny power decline, the battery delivers 100 MWs in the 2 seconds. To deliver the same amount of energy in the time interval you need a 495 MW battery.
But that’s also irrelevant. This is about RoCoF and the inertia of the T/G was in play immediately.
The system is in need of inertia to slow the rate of RoCoF and of power to reverse the decline.
To be harsh, if the decline was caused by the 500 MW tripping a bit more than a 50 MW battery will be needed to halt and reverse the decline.
The battery is good for the secondary response and can give that service from a quiescent state. Whereas the T/G had to be operating and then, if designed to contribute to frequency control, could deliver another 50 MW in a minute simply by AVR and governor action (yes 50 MW of so called “overload”).
Fortunately the boffins in AEMO appear unswayed by your arguments. AEMO’s 2023 system inertia report agrees batteries can contribute to solving the problem of declining inertia, so can syncons particularly if they have flywheels as in SA, and grid forming inverters.
CSIRO has again chosen not to receive expert advice on nuclear costs for their draft GenCost 2023-24 report. Aurecon has again provided expert analysis of all technologies except nuclear. The last time that CSIRO obtained expert advice was from GHD in 2018. For the latest report, CSIRO has attempted to prove by their own analysis that nuclear is too expensive to consider by quoting the cancelled NuScale SMR for Utah Associated Municipal Power Systems (UAMPS) Carbon Free Power project (CFFP) but their analysis is misleading, for example the project was cancelled when the LCOE increased from USD58 to USD89 (at 0.7 exchange rate used by CSIRO = AUD127) but CSIRO quote an LCOE of AUD636. I suggest that CSIRO will need to clarify their analysis for the final report.
Who still measures the spread of grass fires in acres?
Why not hectares?
And then there is this from the IEA :
(Link at https://www.iea.org/reports/projected-costs-of-generating-electricity-2020)
“Projected Costs of Generating Electricity 2020”
Nuclear thus remains the dispatchable low-carbon technology with the lowest expected costs in 2025. Only large hydro reservoirs can provide a similar contribution at comparable costs but remain highly dependent on the natural endowments of individual countries. Compared to fossil fuel-based generation, nuclear plants are expected to be more affordable than coal-fired plants. While gas-based combined-cycle gas turbines (CCGTs) are competitive in some regions, their LCOE very much depend on the prices for natural gas and carbon emissions in individual regions. Electricity produced from nuclear long-term operation (LTO) by lifetime extension is highly competitive and remains not only the least cost option for low-carbon generation – when compared to building new power plants – but for all power generation across the board.
The International Energy Agency’s next report, due in 2025, will likely be very sobering reading.
How does the bidding fixed load works? I presume it is required for plant restoration times. will the generator be exposed to any market price risks when bided fixed load?
Stacked under the bids down at -$1,000/MWh in the NEMDE dispatch process, Selva … easiest to think about it as a ‘must run, for technical reasons … I will take any price’ bid.
Commonly used in testing scenarios, and on other occasions like this.
A HPBV is an “optional extra” on a unit intended to permit the unit to survive a major load rejection without tripping the boiler. Basically HP steam is bypassed directly to the unit condenser rather than through the turbine. Not intended for extended use but to manage transients and also assist in start up. Now it is interesting that it is Stanwell 1 – the first unit of what was then Australia’s first heavily automated coal fired station. As such it probably went through lots of start up , shut down and testing cycles during commissioning. Creep is a function of temperature/ pressure and time whereas fatigue can have a few causes. This one is likely to be a result of rapid temperature changes to the pipework around the valve when it opens in anger. Probably has a bit more accumulated use from being unit 1. Probably found an issue when performing metallurgical inspections during their outage (eg a crack or similar). This will now be being managed by making sure the HPBV does not need to operate by (maybe) running at less than full boiler pressure (hence reduced output) and limiting governor action (constant load) until they can get replacement pipework manufactured and get it glued in (no small undertaking). Sounds like good maintenance and engineering at work. Pleasing to see the guys are on top of their plant.
Thanks for helping readers out with some more context, Andy.
Along the same lines, worth reminding readers that Stanwell has TTHL (trip to house load) capability:
1) Which (if my scratchy memory serves me correctly) would be using HP bypass
2) Tested regularly as part of their black start contract, but also used ‘in anger’ on occasions, such as following Callide C4 catastrophe on 25th May 2021 as Allan noted here afterwards.
Paul
Dan
Great article very interesting.
Looks like most will exceed their design life but the fall from ~28% to ~20% in CF may accelerate their replacement with larger turbines. Their break even time is longer than I expected and curtailment will probably increase. It would be interesting to see a similar analysis on Solar..
Maybe they were testing whether the inverter is flammable, in which case the lodged rebid reason is entirely appropriate (and the test successful).
But seriously, since when do we demand that rebid reasons be 100% specific about the root cause of the plant issue, especially when the operators need to diagnose, triage, and rebid very quickly? That rebid reason is fine and no one in the market is worse off for its un-specificity.
When VBB and Bouldercome experienced fires, Wattclarity didn’t publish articles admonishing them for not using the word “fire” in their rebid reasons, so why the consternation this time?
Understand the unscheduled 12 MW Windy Hill Wind Farm in Far North Queensland was not selected for your excellent article. However, it could be the trail blazer for what happens when a wind farm is decommissioned as it was commissioned in 2000, three years earlier than the oldest WF in this article.
Also as far as I’m aware there isn’t a wind turbine fire database. Think this should be mandatory as any fire caused by a wind turbine could have have catastrophic impacts, especially if the wind farm is on forested ridge difficult for firefighters to fight. It may also give investors some insight as to what turbines are best suited to the Australian climate.
My submission to the CSIRO pointed out that they should review nuclear technology costs in their entirety (instead of exclusively SMRs, still under development), and that the coal costs (greenfield, AUSC, Japan coal costs) are not representative of a project that would be built in Australia.
If I cared enough to go further, I’d point out that they use the high performing wind and solar as the average, then assume future projects will be both cheaper and better performing (higher capacity factor), while reality shows the opposite (especially with wind) and that over-build must increase curtailment.
The CSIRO should be embarrassed.
Realise your list is curated but would have been interesting to see analysis of the Toora Wind Farm – it predates most of these by at least a year, though is smaller (21MW).
The original Albany Wind Farm (Western Australia) was commissioned in October 2001, after ten years of planning. The wind farm has the capacity to produce 80 per cent of the electricity requirements of Albany. Originally commissioned in 2001 the farm was the largest of its kind in Australia.
courtesy of wikipedia
Pleasing to see commentary and encouragement of understanding of bidding practice, and in particular, uderstanding that plant dispatch directed on the basis of simple variable (e.g. fuel) costs mdels cannot efficently sustain operations. A generator’s minimum bid is the price below which the generator will be losing value. A loss of cash by bidding below fuel cost in one interval may however be anticipated as a risk investment in being ready to generate more profitably in the next or a later interval. Conversely a bid above the minimum runs the risk of missing out a profitable dispatch. To manage these risks, the generator needs to anticipate the system demands and the likely behavior of competing sources.
Good to see commentary and encouragement of understanding of bidding practice. In particular, plant dispatch directed on the basis of simple (e.g. fuel) costs cannot efficently sustain operations. A generator’s minimum bid is the price below which the generator will be losing value. A loss of cash by bidding below fuel cost in one interval may however be anticipated as a risk investment in being ready to generate more profitably in the next or a later interval. Conversely a bid above the minimum runs the risk of missing out a profitable dispatch. To manage these risks, the generator needs to anticipate the system demands and the likely behavior of competing sources.
I look foraward to seeing the next instalment
My god they are hopeless, some crappy weather delays them? how weak and pathetic.
The weather forecast for the SE QLD doesn’t look that bad or abnormal for this time of year. That still doesn’t mean the QLD grid isn’t going to have troubles. Their lack of utility scale batteries and wind based generation really hurts them at this time of the year.
I’m not sure what question you are trying to answer?
Being ~30MWdc they have multiple PCUs/Inverter stations potentially available, and depending on their reticulation arrangement they can shuffle their collector feeders to continue their 5MW (notifiable exemption status likely) export for hot commissioning on the non-burnt units. And their commissioning test plan would likely be rotating PCUs for that hot commissioning.
I assume you are only trying to ping them for the rebid reason wording, in which case it might still be accurate enough… at least as accurate when FF generators have ‘maintenance’ during low $ periods etc?
Thanks for the details, Sarah … but just for the avoidance of doubt (re ‘only trying to ping them’) it’s not my intention to ping anyone in my articles … just seeking to understand more.
If the NSW is so confident it doesn’t need Eraring then the answer is simple – the NSW Government should bear all RERT costs. Just one RERT event in Queensland on 1 February 2022 which was only 331 MW for a few hours on one afternoon cost Queensland consumers $50.1 million. The SA Black System event which occurred almost at the close of a working day cost SA businesses hundreds of millions of $. How much are NSW considering paying Eraring to stay open?
One little correction Paul – the high price in NSW wasn’t *because* of the forced southward flow on VNI. It was the outcome of a complex constraint situation that was both forcing flows out of Queensland into NSW and setting high prices in those two states, with a further side effect of the same constraints being the flow out of NSW into Victoria.
we just had a storm and rain come over in logan, temperatures dropping fast
Meanwhile we have evidence of Energex throttling home energy usage in South East Queensland and loss of power in Moreton Bay Council regions (load shedding?) as solar came off for the afternoon.
So it seems supply was not able to actually keep up with what the consumer was demanding.
Hi V, thanks for highlighting those aspects. The ‘throttling home energy usage’ would have likely been seen by those with home devices (e.g. AC units) participating in demand response programs e.g. ‘Peaksmart’. But I’m aware that there is some consternation from affected renters who don’t own said appliance and didn’t sign up to the program themselves.
There was no load shedding or out-of-market direction to reduce demand at the state-level by the market operator and all market generators appear to have remained online. It is likely the local outages you reference were in the distribution network. Definitely appears the distribution network struggled and in the absence of these outages demand could have been even higher.
Imagine the Queensland demand with an export green hydrogen industry in regional Queensland and the additional demand of desalination in Southeast Queensland. The additional loads won’t have much flexibility to ramp up and down for any extended period of time. I’m assuming green hydrogen will need offtake agreements for project financing.
The Gold Coast Desalination Plant has been operational since November 2023. The December 2023 Seqwater Water Security Status Report states the “Gold Coast Desalination Plant (GCDP) production was maximised during the month of December, in order to support the SEQ Water Grid and slow the decline of SEQ water storages”.
Imagine being dependent on the wind to blow for urban water supplies during a wind drought. The IPCC is forecasting a global still of winds and Australia is a country that the IPCC say will have reduced wind speeds.
The SEQ Water Grid has increased to 72.6% (ie above the 70% drought measure threshold) so assume the Gold Coast Desalination Plant was not operational yesterday and did not add to the record over 11,000 MW peak demand in Queensland.
Always hard to talk about dam building when the media is full of floods and cyclones but unlike solar and wind farms, dams for hydro and water supply take years to plan and build.
We need water to survive and affordable and reliable energy to thrive.
You can have both please check out the proposed 1,000 MW pumped hydro Koombooloomba Energy and Water Project (that can operate in low or no wind and solar) on the Queensland Electricity Users Facebook page.
Update – Its the Queensland Electricity Users NETWORK page for details on the proposed 1,000 MW pumped hydro Koombooloomba Energy and Water Project (that can operate in low or no wind and solar).
The project would secure the water supplies for Cairns for the next 100 years (Cairns needs to build a dam for water supply by 2035).
If the proposed 2,000 MW Borumba Pumped Hydro was reduced to its original 1,000 MW, water could be supplied to the SEQ Water Grid reducing the size of the new SEQ $4 to 8 billion desalination plant that is currently undergoing a detailed business study along with a detailed business case for an expansion of the Gold Coast Desalination Plant.
Energy and water are joined at the hip – water to survive and affordable and reliable energy to thrive.
Hi Dan, Bouldercombe is a 50MW/100MWh unit, not 150MWh.
Good spot Andrew – thanks, have corrected now.
If I am reading the AEM Dashboard correctly, Queensland gets a lot of power from NSW via the Interconnect?
And that is after NSW buys it from Victoria who seems to be selling to NSW, SA and Tas.
So if NSW have an LOR2 would that mean they are unable to onsell power to Queensland?
How about Tarong Unit 4 – that seems to have come off, but not as planned!?
Following Alex’ comment, Mick, this looks like a ‘trip to house load’ (i.e. TTHL) test … in order to qualify for SRAS (System Restart Ancillary Services) contracts.
Would potentially be able to see more tomorrow, with ‘next day public’ bids.
See further in the article ‘Record of ‘trip to house load’ (TTHL) at Tarong unit 4 yesterday’, Mick and Alex
So why the massive power disruptions from 22nd to 23rd in SE QLD? No explanation from Energex whose website updates are useless, only telling how many customers are without power and no advise as to how long estimated to fix.
Hi Baz,
It’s clear there were many distribution-level outages, many unplanned and likely driven by the weather and demand circumstances on the day. The supply side – by which I mean electricity generators that supply the electricity market – and the high-voltage transmission network appeared to fare quite well.
Peak dispatchable supply occurred at 6:45 when wind and solar were supplying just 245 MW. 7.9 GW of available coal was supplying 6.7 GW and 3.4 GW of gas, diesel and biomass supplied 2.8 GW and 330 MW of batteries were supplying 125 MW so it appears that the system was well in hand
If this happens next year Qld will have Barron Gorge and probably Kidston 330 MW of hydro, at least 300 MW of batteries and of course Callide C. In total, almost 1.5 GW more dispatchable supply. With some 1.5 GW of wind and more than 1.5 GW of solar coming on line over the next 18~24 months, even if it is only running at 10% still a useful contribution. So assuming Queensland coal manages a bit better than 75% capacity next year there should be no problem
Tarong 4 reduced to 5% output for ~2 hrs but then ramped up to full power from midday.
Batteries and pumped hydro are not generators – they store and release, using more electricity to charge/pump than what they discharge. Sadly batteries and pumped hydro seem to be continuously referred to as some type of dispatchable generation.
Absent coal generation and with the uncertain future of gas generation, batteries plus the Kidston, Borumba and massive Pioneer Burdekin Pumped Hydros will be reliant on excess solar and wind to recharge.
Hopefully they can recharge before the next peak demand period.
However, this could be a massive problem in a prolonged wind drought which could go for days or weeks or a heatwave that goes for 2 or 3 days.
Hopefully the sun will always come up but the wind doesn’t always blow…. and batteries and the pumped hydro will be hoping for the overnight wind to recharge.
But in the future Queensland could have new overnight demand in the form of a green hydrogen export industry and desalination plants in Southeast Queensland.
The evidence does not support your concerns. To have a reliable 95%+ renewable grid Queensland needs 3-4 times as much solar and 6-8 times as much wind and about as much storage as Victoria is planning.
Thus between 9 and 4 wind and solar would have been not only supplying all demand but most of the existing “offpeak” demand and still putting 6-8 GW into storage. At6:45 when coal output peaked, wind and solar would still have been supplying 3-4 GW,hydro and storage 6-8 GW. On a few occasions per year, gas might supply 20% of peak demand 10-15% of daily demand and 2-5% of annual demand
Hello Peter,
what about in a very serious drought?. I think you may need to revise your gas a touch higher?.
Interesting to note that the power outages were claimed to be due to storms but Moreton Bay Regional Council didn’t even experience rain and had a large portion of the power outages.
Also all that solar and wind and it produces nothing when demand is peaking.
Look at the chart in the previous article – PV was generating a significant proportion at the time of peak market demand, and even more at the time of peak underlying demand (which was of course higher, but also earlier and continued for longer).
RERT contract entered into but may not be activated or dispatched.
As stated in your articles this is a long weekend but Queensland is also without thousands of customers due to Cyclone Kirrily that impacted the Townsville region on Thursday night.
The cyclone knocked out 66,000 customers but as of Saturday 9.00 am 20% had been restored with most expected to be restored by Sunday night (great effort by Ergon staff).
However, this means that on a long weekend and thousands without aircon, Queensland did not have sufficient scheduled and semi-scheduled generation to meet demand.
Remarkable that ‘Market Demand’ in QLD today (Sat 27th Jan 2024) would have been a new all-time maximum for QLD … except for what happened on Monday 22nd January 2024!
https://wattclarity.com.au//articles/2024/01/27jan-qld-marketdemand-10210mw/
Don’t you think you could have eleborated just a little on what the dew point actually is and why it’s such an important metric in this context? Just c.c-ing someone else’s Twitter post doesn’t really amount to an article.
With all the extra renewables there seems to be no problem. The sooner the coal fired and gas power stations are shut down the better.$1000 MWh would seem to be unsustainable.
Looks weird. Perhaps something off with the LOR, or OUTAGE_LRC STPASA runs? Big hole in unconstrained capacity for the 1st Feb.
How does a home user get these “Alert Price” messages? Thanks!
As the price has increased over the years there would be some reduction in demand at the margins as some people will economies. That would moderate the increase in demand that we expect from population growth.
Another type of demand management is the closure of firms on the margin of survival in energy-intensive industries. The result is deindustrialization that is happening bigtime in Britain and Germany!
https://newcatallaxy.blog/2023/07/18/lets-all-follow-germany/
What a mess – and given that the population is growing out of control and there doesn’t appear to be any believable figures for the additional generation (firmed?) necessary for charging more and more EVs, “system blacks” appear unavoidable.
I agree with the conclusions regarding the RRO, but the first section of this article misrepresents history.
Once Northern PS committed to closing, there was only one generator that could have feasibly supplied that large customer load. Realising this, that generator sought to extort the customer as much as possible, demanding a price well above market rates. The customer refused and took the risk of floating on the spot market, and the generator in question punished them for this decision, withdrawing capacity in order to induce volatility. You can argue this was a reckless gamble, but it wasn’t necessarily a bad one – a week of volatility is ultimately easier to stomach than years of overpriced power.
This wasn’t the impetus for the RRO, however. Eight months later, that same generator, still seeking to extort South Australian customers, withdrew part of its capacity for the summer, leading to controlled blackouts in SA. This is what drove the creation of the RRO, the argument being that blackouts could’ve been avoided if customers had given in to the generator’s demands; that AEMO later took the generator to court for misrepresenting their availability seems to have been overlooked.
Many people recognized that this was actually a competition issue – if multiple generators had been competing to win customers, it’s unlikely that a large portion of load would’ve gone uncontracted. The RRO does nothing to resolve that problem and, in fact, makes it worse, as it reduces customer bargaining power. The MLO was stitched on to the RRO in an effort to counteract that effect, though who knows if it’s worked.
As the article points out, none of this has anything to do with actual system reliability, which is a bigger issue than the level of contracting over a short period.
Thanks Teffy, you’ve been part of an experience to which I was unaware.
The period I was referring to was the period immediately prior to the decision to close Northern early and offers by its owners to sell contracts from it. I gather you are referring to the challenging competitive situation after it had committed to close. I imagine that would have been very difficult time indeed for customers.
The general industry narrative, however, that was used to justify the construction of the RRO, referred to decisions of the customer being made prior to the decision to close northern.
In the end, this is all ancient history, based on peoples, subjective recollections of a series of negotiations. my point in bringing it and the neg up is just to show that the RRO was an accident of random history, rather than an outcome of any thoughtful design.
These short minimum don’t worry me as much as the many consecutive days of very low VRE output in Victoria and SA that occurs at least once per “winter” period per year. These events are often also followed or preceded by periods of poor VRE output. These periods are also during the time of year when demand is very high. They are also long enough that most BEVs will have been depleted even when used purely for road use as distinct from V2G.
If V2G becomes a large supply option, then combined with other storage such as residential and utility scale BESS, short periods of VRE won’t be a problem. The technology and economics are generally improving to assist in solving these short term VRE dips.
So according to Teffy- who seems to be mixing up several years – generators stop running for a couple of years to make money. Yeah, righto. Seems like a classic case of a customer expecting the market to stay static as they made a bad call. If you want prices to stay low, lock in a contract. If you don’t the plant that pushes down spot prices won’t run, and prices will rise. It’s not hard Teffy.
The project energy connect RIT-T modelling got much of its market benefits through it causing early closure of SA peaking plant, which lowers total market OPEX. This outcome seems to be confirming that modelling and should be no surprise.
Sadly Project Energy Connect will do nothing for SA when the wind isn’t blowing strongly and the sun isn’t at its peak.
There is a reason industry has fled SA, unreliable and expensive energy driven by the closure of fossil fuel plants.
Bravo Ben – abolish it.
Great, you’re celebrating the fact that the FCAS market will increase when coal generators close down. What about the small matter of the ENERGY that is lost when those coal generators are gone? Funny how little you talk about battery energy. Most battery-pushers, like yourself, always underplay energy and instead love to gloat about nominal capacity levels.
How does your battery future work when Eraring closes next year (which you’re obviously pushing)? Eraring’s daily energy is probably at least 50 GWh all day everyday (taking into account cycling and not running at full output all day). At the moment, the biggest battery in NSW is the Riverina and Darlington Point BESS with a whopping 0.3 GWh of energy! Incredible! It’s safe to assume that whopping 0.3 GWh would be a daily limit since batteries will generally run a single cycle in one day.
I know it’s really exciting that these batteries can enjoy some of that coal FCAS cake but what about the small issue of losing 50 GWh of energy and replacing it with technology that offers 0.6% of its energy.
“How much more comforting it is to know that we depend on millions of interconnected small units rather than a few large units.”
That’s incredible. You’re saying that managing millions of small independent units is way more comforting than several giant units. You should become an engineer! Lets totally ignore the worsening DER problems on the distribution networks such as voltage regulation, phase imbalance and congestion.
it’s 2024… how do we still have these coal power plants running!?
Because we can’t rely on distributed renewables as the transmission network is a weak point
There are pictures going around, but unless it is from a reliable source, then we still have to wait. If it is the case then we have had multiple events causing multiple failures with those 500kV towers. In both situations the towers were rated to handled higher wind speeds than those that generally occur around the events. Meaning we would have to either have a fault in the material, design, construction or maintenance of the towers, or we have had two previously extremely rare weather events within a year or so of each other.
I can only recall one incident with collapse of 500 kV towers, namely the collapse of the group of towers near Cressy in Jan 2020 which islanded SA. The 500 kV lines west of Moorabool are on double circuit towers. The 500 kV lines east of Moorabool, which tripped today, are on single circuit towers of completely different design.
There was the collapse of one double circuit 275 kV tower near Tailem Bend in Nov 2022 in a freak storm. A 220 kV tower near Poatina in TAS collapsed due to a landslip in Oct 2022. Back in Sep 2016 a number of towers on different 275 kV lines in different locations in SA’s Mid North collapsed in the storm that led to the System Black.
Thankfully this came out on a quiet day with not much else going on, I’ll just settle in and read/watch it.
The recovery to 50.12Hz would have been the initial fault clearing. The roll-over after that likely would be runbacks/control schemes for windfarms. LoyA units may have (mal)operated on ROCOF after the cascade of frequency.
There may have also been an attempted reclose on both the 500kV lines.
Thanks Liam, I appreciate the input and clarification!
When EMTT (emergency transformer tripping) scheme is armed then things could have definitely gone a lot worse!
On the mid afternoon of Tue. 13 Feb 2024 I saw both Sydenham T.S. to Moorabool T.S. lines down from several km
away.
The road which is over-crossed by the two lines was blocked by police; they showed me a phone photo of two
collapsed towers; but wouldn’t allow public in the area. The police said six towers were collapsed.
The photo that I saw showed the collapse of the section part way up where the tower bifurcates to make
space for the centre phase.
It appeared that three sequential towers and thus four spans of each 3 ph. lines were not visible
from a distance.
I saw evidence of a microburst in the area with many trees uprooted and others snapped or torn apart.
There were some wet patches of road in the area indicating very recent rain.
The 500 kV lines in that area were built between 22 kV / 12.7 kV distribution lines. There were no aerial crossings.
(There is an underground 22 kV undercrossing outside the MLTS fence in Anakie Rd. some km to the south west from the tower collapse area) I would be very surprised if there were any recent aerial crossings.
Also Heywood and Murraylink
The AusNet outage page has been really struggling to load since yesterday afternoon and still this morning. Pretty poor that they aren’t set up to handle spikes in web traffic.
What happened to A2? It seems to be offline again.
In reference to the frequency deviation, the event lasted ~10 minutes with frequency returning inside the 49.85 prior to 13:20 and then eventually recovering to the 49.9 allowing contingency response to ease. The most interesting part is the swing from nearly under to nearly over frequency back to an under-frequency event.
Thanks Mitch – I’ve corrected my error there (had misread the chart and said 18min, but I agree it was under 10 min – from approx 13:09:44 to 13:18).
I think the initial frequency dip was the initial fault and possible reclose, also possible that voltage harmonics on the signal your system was processing made the dip appear larger than it was.
With so much non-synchronous generation on the NEM nowadays, the rate of change of system frequency for events keeps evolving due to ever continuing lack of inertia imo.
My guess is that LYA holds the current Vic Trip to House Load contract with the NEM, and had all four Units selected active at the time of the event (to enhance the probability that the required minimum generators successfully Island to fullfill the contract obligations).
I believe the rapid ROCOF you have demonstrated above probably triggered the detection relays of all four Units, …..and unfortunately all four failed to Island effectively.
Hopefully the battery systems and EDG’s at the plant stood up to the subsequent Station Black.
My t2 bobs anyway.
The escalating loss of load from local distribution outages as the storm front increased in intensity on the outset of LOR3 load shedding would be the main reason for the low quantity of intentional load shed (300MW) and it brevity. Vic’s demand was on track to reach 8,000 MW before thousands lost power.
Opens the door for the trusty ole climate-driven unprecedented storms” gaslight option.
The Vic-NSW interconnector constraint equation “allowed” 6,574MW to flow Sth at yesterdays 14:25 DI.
More alarming is the huge volume of reported household phone/internet outages, coupled with loss of 4G internet at the same time, only hours after the event.
This issue alone needs to be looked at, NBN should be forced to comply with minimum battery/EDG supply performance guidelines. This would allow the public to check in on relatives before their land line is lost (like what happened to my 89 year old independent living mother last night.
I would be interested to see your insights into the performance of the large battery generation systems in Victoria for yesterdays event.
JDH Consulting are wind loading specialists and have been studying failure of high voltage transmission line towers in Australia, particularly in Victoria for about 25 years. Failures due to small but intense storms such as microbursts are common for a number of reasons, such the wind structure within the storms and the directional orientation of the line. The risk of failure along a line is also dependent upon its length. JDH has been in discussion with AusNet regarding a specific study of line risk for the Victorian network, similar to the extensive studies carried out for ElectraNet in South Australia, following the many failures that occurred there in September 2016.
Noted here with respect to ‘Some *initial* headline questions following the transmission damage, unit trips & loss of load, initiated on Tuesday 13th February 2024’, John
The article correctly highlights the age of the electrical transmission infrastructure in SE Australia, and the many failures from severe convective winds that are occurring. However, a factor that is often overlooked is the need to design for the particular characteristics of convective winds in comparison with those of larger synoptic scale wind events for which the older systems were designed, often based on methods developed overseas in much different wind climates. In layman’s terms, the convective wind events generate a single large gust that envelopes 3 or more spans, and results in the failure of several towers in a group as we saw on February 13. The result of this is that a convective wind event with a peak gust of say 140 km/h can generate wind loads on a span 50% higher than those from a large-scale synoptic wind event with the same maximum gust. It is essential that design methods for wind loading accounting for these characteristics are adopted for the new transmission lines under construction and planned for the energy transmission.
Another question wotth investigating is whether the wind and solar farms are actually negatively impacting the climate by producing more volatile systems.
factors in the tower collapse “still up in the air” I see what you did there 🙂
Was it just wind? Or was there a hail loading as well?
Even more questions come to mind
What was the response of industrial equipment to the frequency and voltage changes resulting from the event?
How did station auxiliary equipment respond to the disturbance, did this contribute to the trip of LYA?
How did Loy Yang B and basslink ride through the initial disturbance and trip of Loy Yang A (could there be lessons for inverter based and thermal generators)?
Also a partial answer to A2
At 4:27pm local time, AEMO issued notice 114641
RECLASSIFY CONTINGENCY
13/02/2024 03:27:46 PM
Reclassification of non-credible contingency event due to existence of widespread abnormal conditions – VIC region(s).
AEMO ELECTRICITY MARKET NOTICE.
AEMO has identified that a non-credible contingency event impacting the power system is more likely to occur and is considered reasonably possible because of the existence of widespread abnormal conditions namely severe weather in the VIC region.
45 minutes later, there was a trip of all four 500/220 kV transformers at the Hazelwood Terminal Station.
Market Notice 114643 “Significant power system event
– Victoria Region – Hazelwood transformers”.
A side note, during the LOR3 period I was unable to access my DNSP (Ausnet, offline altogether) or the AEMO (Gateway timeout) websites. The latter did delay my response as an Amber customer. Other parts of the state suffered large scale communications outages. I question whether our telecommunications and IT infrastructure is suitable for market participation or even advisory of outages during severe weather events. I also wonder about the effect of communications outages on virtual power plants or demand response.
Safety factors incorporated at the time these existing transmission towers were being designed were doubtless higher than those applied today. Nothing wrong with that since today’s material quality standards are superior and support lower safety factors. If heavier cables have been strung, that would test any ageing components.
The Sydney Harbour Bridge, West Gate Bridge, etc, should probably be reassessed against today’s operating circumstances with heavier trucks and more of them etc.
Who is auditing the condition of this ageing infrastructure?
This report is shocking.
– Automatic changeover for Y protection inoperative, configured such that both X and Y redundant protection are powered from the same DC supply.
– Switching sequence has risk of interrupting DC power to plant, removing redundancy from critical systems, and all protection, was carried out while plant is running.
– No recovery/rollback plan or procedure for loss of AC power to, or failure of, the battery charger while DC system is operating without a battery is mentioned. Restoring DC supply from one of the batteries would have improved the situation.
The accompanying video also mentioned that operators did not ask Calvale to isolate the generator as they feared a steam valve leaking by. However, reverse power would suggest the grid was not providing a mechanical load on the turbogenerator, so isolating it isn’t going to make the situation worse.
AEMO did invoke a constraint to procure 450MW of mainland raise reg – possibly to cover overspeed cutout at windfarms? I could see VIC bringing in a protected event category similar to SA – not sure how they would operate system more conservatively eg SA limiting flow on their one AC interconnector
Thanks HR10 … on Friday, the AEMO’s release of their preliminary report provided more details about the increase of mainland Raise Reg procured.
The tree’s in background look fine begs the question the quality and the integrity of the steel and design of the tower’s I Wonder where they where made and designed??.
With regard to the 2.3GW generation loss of LYA, given the 1GW of lost load downstream of the transmission failure, we should regard the low frequency event as a loss of 1.3GW.
Re Basslink/LYB/(and YPS) riding through the voltage dip and subsequent low frequency, I feel it is likely that LYA tripped due to the operation of its equipment that facilitates its Trip to House Load market ancillary service contract with AEMO?
I believe there is an electrical/magnetic component involved, as current is passed through the cables it essentially turns it into a “motor” which the geomagnetic storm twists the towers, generating a great amount of torque on the frames.
It reads similarly to how the University of Queensland accounted for the performance of their St Lucia BESS, where a significant source of revenue was the cap contracts they were able to forego as a result of being able to manage their own risk with the BESS.
It shows our energy system is becoming more and more fragile as we remove reliable coal fired power stations and replace them with distributed renewables that require significantly lengthy transmission networks and interconnectors with other regions that significantly expands the area of risk.
Network charges on our Electricity bills have been rising ever higher ,year on year to such an extent experts and many in the know have termed all the money being charged as ” Gold Plating the Network” suggesting …consumers are and have been paying too much money in their bills to justify the cost of keeping the grid and networks operating optimally.
NOW…this article suggests that major network infrastructure…Electricity Distribution Towers…at least in Victoria is averaging an age of 50 years+ and as such in some areas is reaching it’s expected life span!
SO… Why have consumers been paying higher and higher prices for Network Charges that are supposed to be needed to keep the Network/Grid in peak shape…and yet have major aspects of that grid / network reaching their life span ? Where has and is all that money for gold plating gone ?? And where and what has it been spent on ?
You CAN’T charge Network / Distribution costs that are supposed to keep things in peak condition , up to date and modernised …justifying the high charges being paid …YET here is an example of Network Infrastructure that HAS NOT been maintained and /or replaced since or before the Introduction of our Decimal Currency in 1966!!!
Where has the money gone ???!?
Brett, the investment has largely been in new lines and upgrading existing lines to connect renewables to the NEM.
So we have had billions spent and the network is now less reliable than it was before because we have much more distributed power generation that relies on flimsy line infrastructure.
It doesn’t show that the system is becoming more and more fragile at all. What bit of six of ten coal generators going offline don’t you understand.
Three years ago, a failure at one 400 MW coal generator in Queensland caused 2,400 MW of other generation to go offline and 400,000 customers to be disconnected.
In this case Victoria had sufficient reserves so that even though 2,960 MW of its coal generators were offline, within two hours of the incident all load shedding had been reversed and it was already exporting power.
The average transmission distance is falling as rooftop solar has near zero reliance on transmission and all of Northern and Western Victoria and even western Melbourne is closer to generation than it has ever been. More interconnectors allow diverse transmission paths so in fact reduce risk, particularly as the average load on interconnectors is declining
Try comparing like with like.
SA and Victoria both had massive blackouts across the state because they don’t have sufficient reliable generation and strong winds brought down the distribution network. Victoria was extremely close to having a system black event and things becoming significantly worse than they were. The risk of this happening again in SA, Vic and Tasmania is much higher as we rely on thousands of kilometres long transmission systems to get power from distributed renewables and don’t have sufficient synchronous generation located close to load centres.
On the other hand QLD had a catastrophic failure at an actual generation plant which of course affected the other units at the plant taking them offline. Meanwhile the majority of the state was unaffected. The Callide failure was not caused by a weather event and the risk of it happening again is extremely low.
Capacity revenue maybe from
“
The latest tender with Victoria and South Australia will target 600 MW of dispatchable capacity with 4-hour equivalent duration across the two states, in line with the Australian Energy Market Operator’s 2022 Integrated System Plan.
Bringing this new generation online will help cover the loss of capacity from the scheduled closure of South Australia’s Torrens Island B power station in 2026 and the 2028 retirement of Victoria’s Yallourn Power Station.”
If you go into document properties in the PDF you can see
Created 16/02/2024 15:30:38
So unsurprising that you only became aware of it today.
That is an insanely large amount of money to produce a system that is less reliable than the fossil fuel generation system we had in place up to 2010.
And does it also take into account the shorter lives of these renewables which means we would need to be spending a significant portion of that capital outlay again every 10-15 years, in addition to having to transport and dispose of the wind and solar components?
It isn’t a particularly large amount of money, replacing the coal we had in 2010 almost all of which will have to be replaced by 2035 would cost about $100-$120bn in today’s money not including new mines or wate supplies or coal rail tracks. Combined with the operating costs at today’s coal prices would mean that the plants have a breakeven of $120-180/MWh. Firmed wind and solar can comfortably make money at $100 without suffering 2GW trips.
Peter Farley, you’ve done it again!
“Firmed wind and solar can comfortably make money at $100 without suffering 2GW trips.”
That’s right, once we have 10,000 km of new transmission lines reaching the REZs, we can enjoy $100 energy! I’m sure that transmission lines will be both dirt cheap and timely. That maths definitely adds up.
And I’m sure that this high quality transmission line will not increase our chances for transmission line failures the next time there is a freak storm.
I also like the way you implicitly blame Loy Yang for what happened on the 13th 😉
Hi John, Great article.
I note elsewhere on WattClarity they mention: ” Sydenham (SYTS) No. 1 and 2 500 kilovolt (kV) lines tripped following failure of six 500 kV towers (three on each of the two 500 kV circuits).”
I would be interested to know how far apart these groups of towers were, and whether the collapse occurred within a few minutes of each other.
Peak winds in these microscale events are usually very small, temporally and spatially.
I would have thought two lines should be enough redundancy. Is it necessary to increase the separation between lines?
All solutions sound expensive!!
With regard to your highlight 2) of Dundonnell under Event2, the fist paragraph of Section 1 Overview of the report contains the sentence “Initial review indicates Dundonnell WF tripped as designed due to operation of the South West 500 kV special control scheme.” By my reading the reference to the special control scheme does not occur elsewhere in the report.
Relating to the loss of Stockland Hill, there appear to have been fire service notices issued about a fire in that vicinity.
There are five big variables which are difficult to quantify which will mean that the transition is different and almost certainly quicker and cheaper than the above model suggests.
1.How much of the current “offpeak” demand which keeps coal plants going at night will be shifted to daytime “solar soaker” rates
2. What is the optimum expenditure on existing hydro eg the upgrade of Clover could be just the start of hydro upgrades.
3. Will the perception of grid unreliability drive overinvestment in behind the meter generation and storage, which will supply 20%+ of total demand and clip peak demand by 2-5 GW
4. Will detailed economic analysis drive developers toward different configurations of wind and solar farms optimized for output in suboptimal conditions rather than just adding storage to conventional designs?
5. Will the TSO’s suddenly discover dynamic rating, reconductoring and storage as transmission to speed up transmission capacity increases
My understanding is that Stockyard Hill WF was asked to turn off to enable the fighting of a large grassfire in the area which was closer to the turbines itself.
Suggest the Yallourn units were responding to frequency via their primary frequency control.
Thanks for the comment. I don’t know the details of last week’s failures at Anakie and they have not been made public as far as I am aware. However, in previous events, several adjacent towers have failed as a group.
On your second point, the storms are not that small. A ‘footprint’ of peak winds in a ‘typical’ convective downdraft has rough dimensions of 10km along-wind x 3km crosswind. Those dimensions would be large enough to cover two parallel lines, unless they are separated by several kilometres.
Furthermore there may be several storm ‘cells’ – e.g. in SA on 28/9/16, there were at least three separate cells resulting in 22 towers failing or with severe damage.
The AEMO report suggests load shedding in Melbourne was to manage system constraints, mentioning thermal overload on a 220kV line from Geelong to Moorabool terminal station taking power from generation in Western Victoria to loads Melbourne. This situation complicates management of dispatch.
APD is possibly the participant referred to in section 7 of the preliminary report
“To manage power system security for credible contingency events, AEMO issued a direction to a participant to maintain power system security.”
since “a credible contingency event – the loss of the MLTS A1 500/220 kV transformer – would have resulted in the synchronous separation of South Australia and the APD smelter from the NEM”
@Peter Farley
1) None. With all the GIGAWATTS of capacity already in the NEM that your lobby promotes, you’re not making a dent in off-peak demand. Battery ENERGY is almost negligible when compared to coal plants.
2) Optimum expenditure? How about we use Snowy 2.0 as a proxy and go with $10 billion and a 10 year lead time.
3) Perception of grid unreliability? I guess what happened in 2022 was just a fantasy of mine. Clip peak demand by 2-5GW? You’re literally just making up numbers now.
4) LOL!!! All the best wind and solar regions are already taken. There’s nothing you can do about it, pal. That’s why your lobby wants the heavy 10,000km transmission line project subsidised.
5) “Suddenly discover”? That sounds like very sound reasoning to scrap all coal. Hopefully we’ll suddenly increase our transmission!
AEMO directed Mortlake to run from later on Tuesday afternoon. This would have been to manage the risk to Portland Smelter in the event of synchronous separation of the south west 500kV network from the rest of Victoria. As we saw in February 2020, Portland can’t be securely supplied from South Australia in those circumstances.
I was interested to read Paul Simshauser’s research about dynamic transmission constraints. The main thing I took from this, and I probably took the wrong thing, was that if the wind is blowing then more transmission on a line is possible because by definition the line will be getting wind cooling.
I guess the transmission constraints through the NEM don’t generally reflect dynamic max load as opposed to hot day static max load?
It’s interesting to me that because wind and solar receive an AGC set point developed from a wind / solar forecasting model, that any variation in weather conditions (sudden strong winds, calm spells, clouds) cause forecasting errors, which require FCAS to correct, which increases cost and risk.
I will certainly be interested in seeing how this changes in future iterations with transmission requirements modelled – as you mention, the results will likely be very different.
I would note that if you are designing a system like the one your model has chosen, where there are subregions that are substantially underserved with local generation and require near-permanent access to generation that is physically remote, that leads to critical transmission infrastructure that has to be substantially overbuilt, to allow for some transmission paths to be out of service for scheduled maintenance while still remaining resilient to events that trip some of the remaining paths.
To incorporate transmission I think the offshore region(s) will need a definite location as well.
It would also be interesting to start the VRE portfolio with existing/committed VRE projects as well rather than completely ‘from scratch’. This may not change much, of course.
I thought this was interesting on the LOR3 condition (from section 5 of the preliminary report):
There was enough generation in the Victoria region to meet demand, however, due to constraints on the network,
this generation was not able to supply the load. At 1420 hrs on 13 February 2024, an actual LOR3 condition was
declared and AEMO issued a direction under section 116 of the National Electricity Law (NEL) (deemed a clause
4.8.9 instruction under NER 4.8.9(a1)(2)) to load shed 300 MW of load in the KTS/TTS/ROTS metropolitan area to
maintain the system in a secure operating state.
There was some information on this in section 2.3 of AEMO’s preliminary report into Feb 13:
Prior to the event, AEMO was managing a sustained low frequency event including the loss of approximately
500 MW of generation from Stockyard Hill WF that occurred at approximately 1135 hrs and had increased the
minimum frequency control ancillary service (FCAS) regulation raise requirement. Constraint set
F-MAIN_RREG_0450 was invoked between 1200 hrs and 1355 hrs on 13 February 2024 to ensure mainland
regulation raise requirement was greater than 450 MW.
One thing to keep in mind for bids in the second half of the period at MPC is that from 1435, AEMO applied a manual override to set the price at MPC. That has to affect bidding behaviour.
Believed to be due to bushfire smoke now starting to cover the North West/West of Melbourne.
Tristan,
You discussed curtailment but did not take into account the period of Dunkelflaute.
Batteries will not provide power for periods of weeks or months. It requires pumped hydro. Even Snowy 2.0 will only provide 2GW and then only for a week.
Boston 2022* paper showed that during Winter 2010 there would have been significant blackouts, even with 10xSnowy2.0 for storage.
It also showed massive curtailment.
Other analyses I have seen like this start in July 2010 or 2011, but this paper shows peak plants (or massive batteries) could be on standby for 10-15 years.
Ref: https://iopscience.iop.org/article/10.1088/2515-7620/ac5677*Bongers discussion in Wattclarity:
https://wattclarity.com.au//articles/2022/06/seasons-of-low-wind-and-solar-output-will-influence-the-shape-of-the-nem/
The market price in SA and Vic are often very interesting. Yesterday in SA was once again interesting.
Both Queensland and NSW look like they’re heading towards very high levels of operational demand this evening, with levels at 3pm around where they were at 5pm yesterday (NEM times), and yesterday was a fairly high demand day.
The reasons for this are very clear in the market data – the output drops were Moorabool WF being constrained off by transmission limitations around Moorabool / Geelong, which preceded the 500 kV line trips (and worsened thereafter). The availability of the plant remained high until about 3pm, but in any case it was mostly constrained off from 12:40 NEM time onwards
From the market data, Berrybank 2’s ups and downs are clearly constraint driven (like Moorabool WF), although it was persistently below target output (by up to 10-15%) between 11:00 and 13:00.
On the other hand Berrybank 1’s availability dropped to low levels then to zero in the half hour following the main event and stayed there for all but a couple of DIs until about 20:00 when it recovered.
Interesting article — I’m a long-retired EE who worked in and around generating plants for 33y. Five of those years were in fossil and nuclear plants with 11 units. It never occurred to me to ask what the time interval was between the unit trip and the generator trip, nor do I recall ever seeing any records that would provide me with that information. Thanks for answering the question I never thought to ask. (USA)
Perhaps the Moorabool Wind Farm (and the Berrybank WFs) tripped on excessive wind speed at its location. This wind storm subsequently hit the 500kV towers causing their demise.
This will be interesting when subsidies for rooftop solar are removed the amount of systems will reduce over time as they start to fail creating more demand for the house hold that didn’t consume and for not being able to supply the network what it needs
AEMO Market Notice #115120 reports restoration of Moorabool – Sydenham No1 500 kV line at 1929 hrs 25/02/2024 on temporary towers.
Thanks Alex … now noted with “Temporary replacement towers up, for the Moorabool Sydenham No1 500 kV line (on Sun 25th Feb 2024)”, but noting the No2 line is still to be replaced with temporary towers.
Using AEMO’s own Operational Demand and Rooftop PV tables gets a value of 15,088 MW for the half-hour ending 15:00, close enough to your 15,100 MW Paul. David Osmond’s figure of 15,400 MW from OpenNEM seems very high, perhaps it’s based on 5 minute data and a different PV source (eg APVI).
I made some fairly detailed comments on a LinkedIn post by Geoff Eldridge of Global Power Energy that went into the basis of AEMO’s Underlying Demand figures and how they might differ from others’.
(https://www.linkedin.com/feed/update/urn:li:activity:7166574564234252288?commentUrn=urn%3Ali%3Acomment%3A%28activity%3A7166574564234252288%2C7166577230486204417%29&replyUrn=urn%3Ali%3Acomment%3A%28activity%3A7166574564234252288%2C7166610652499795968%29&dashCommentUrn=urn%3Ali%3Afsd_comment%3A%287166577230486204417%2Curn%3Ali%3Aactivity%3A7166574564234252288%29&dashReplyUrn=urn%3Ali%3Afsd_comment%3A%287166610652499795968%2Curn%3Ali%3Aactivity%3A7166574564234252288%29)
And that perhaps only the worst performing / highest cost to maintain unit at Eraring will switch off ‘some time’ after 18 Aug 2025, rather than all four being switched off on 18 Aug 2025, i.e. Eraring will stage the closure of it’s four units from August 2025, over perhaps even years.
Is there a tool where capacity factors for individual years or overall for all wind farms and individual wind farms that could produce a lust of the selected criteria?
Hi Peter, the closest thing might be our GSD reports:
https://wattclarity.com.au//deeper-insights/
AEMO Market Notice 115511 reports return to service today on temporary towers of Moorabool – Sydenham No2 500 kV line.
Thanks Alex, have noted here:
https://wattclarity.com.au//articles/2024/03/06march-temporarytowers-moorabool-sydenham-no2-500-kv-line/
Subscribe to updates please.
Thanks
Dan – Great graph.
So we need to double solar/wind in Apr and May, BUT the network curtailment will be an issue all months AND the economic curtailment will more than double in September to December. Or it’s not that simple?
Not that simple Dave, but I wish it was.. We talked a bit about the long-term future of solar and wind in our GenInsights21 report which Marcelle referenced on WattClarity:
https://wattclarity.com.au//articles/2022/02/what-does-the-future-hold-for-wind-and-solar-in-the-nem-seen-via-geninsights21/
Considering “operational demand” and “underlying demand, what is the data at “Annual generation capacity and peak demand – NEM” showing us? 2024 peak is looking low in this data which agrees with the analysis presented here. From that data, the peak demand was in 2010-2011, it is difficultto tell by looking so i downloaded the csv data to Excel. 2019-20 came close but fell short of 2010-11 by a considerable amount. Depending on what the overall peak is for 2024, I would say there is a downward trend in this data, flat at best.
There are other charts too. For example, “Annual electricity consumption – NEM” which paints a similar picture, and this one shows a significant downward trend from 2010.
These are all nice and pretty graphs. The real question should be “What is happening with the NEM to cause prices to rise at close to 100% over a period of flat or falling demand?”
There’s ample analysis published on drivers of NEM prices – a good place to start is AEMO’s Quarterly Energy Dynamics reports at https://www.aemo.com.au/energy-systems/major-publications/quarterly-energy-dynamics-qed
But a major factor in the long term increase in NEM prices (from the 1999-2012 period) is the increasing linkage and exposure of generation costs – specifically fossil fuels – to international energy prices. Another factor has been the retirement of ageing coal-fired capacity. Replacing old, fully depreciated capacity with new generation of any form is generally going to require higher prices to justify the new investment.
Assuming we get annual cycling equivalent to two full depth cycles, 120 half depth cycles and 200 at quarter depth mostly in the mornings, that means that simplisticly twenty-two GWh of storage would eliminate the problem.
Now let’s say 10 GW/18 GWh of batteries were available as well as 4.5 GW/400 GWh of new pumped hydro, we should be able to eliminate at least 75% of the curtailment. That would incidentally increase renewable share from 38.7% to 40.7% plus whatever rooftop solar curtailment is avoided.
Very interesting to see how the calcs would fall out if solar capacity was tripled and wind doubled
Did SA batteries respond to the sudden 348 MW ramp over 4 dispatch intervals? Because if there was a “possible” LOR3 event later that afternoon then a battery with a Capacity Investment Scheme Agreement would in the future have to have some of their capacity reserved for that possible LOR3 event.
Dan,
Thanks – another thought-provoking article. It was a difference world 7 years ago, pretty much halfway through it’s design life and 100MWh for $56-90m.
How much have the cost of grid scale batteries dropped since then?
Interesting to compare these to the Kurri Kurri peaking station. It will have taken 30 months IF it is completed by the new scheduled date, December this year. It will have cost $1billion but will produce the same energy capacity as all current 20 batteries in 3.5 hours. (2.4GWh)!!
Project construction time is highly dependent on project size, which should explain some of the spread in the FID-to-COD timeline between projects. Also, a lot of the projects with FID in 2021-22 were probably affected by supply chain issues which definitely led to significant delays. It would be helpful to include the project size in the table above
I think there were still some issue with supply chain well into 2023.
Didn’t the VBB suffer further delays due to the fire, or are you counting that in the operational mode?
Hi Craig, for the article I used my own definitions for each of the phases which I stated in the article.
The fire at VIC Big Battery was in Sep 2021 (which I believe happened during testing). In any case, by my definition it was fully operational in Nov 2021.
Thanks for the discussion Ben particulaly around the assessment process. Although a bond of a “few hundred thousand dollars” is not to be sneered at its tiny relative to capital cost of a project and, in my view, nowhere to provide confidence that projects winning contracts will actually be built. Building confidence that projects actually will be built is in a way the entire point of the scheme.
I also wonder whether greater storage depth really is more valuable. Would you rather have 1000 MW of 1 hour storage or 1 MW of 1000 hours storage?
Thanks David. I thought the same re bond. Indeed just the cost of preparing a bid would run to seven figures.
I think the modelling approach proposed sounds quite good for valuing the depth of storage to meet the needs of the NEM, certainly much much better than just dividing by 4 hours in the RCM, which I have heard (but struggle to believe) is also what is used in the much bigger capacity markets of the US.
Your concluding question may be even more rhetorical than you intended, consider this: Battery A can do both roles as the system requires it, but not vice versa.
Excellent insight from WattClarity as always.
One note is that the Vic-SA tender did not apply the four-hour minimum. Whilst present in the consultation draft, it was amended to 2 hours as the tender opened.
Would battery discharge by VBB be able to provide the increased import limits on VNI under SIPS when the Moorabool – Sydenham 500 kV lines were lost?
You refer to the V>>NIL_MLGT_MLGT_R2 and V>>NIL_MLGT_MLGT_R5 constraint equations. The names of the constraints suggest that they relate to the Moorabool – Geelong 220 kV circuits, which could overload during the 500 kV outage.
Further reading for the technically inclined. This is an open access paper.
Title: Modeling geomagnetic induced currents in Australian power networks
Journal: Space Weather Journal
Issue: 10.1002/2017SW001613
Publisher: AGU Publications
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2017SW001613
Paul, re: VBB SIPS, this is only armed for trip of a number of transmission lines, primarily any of SM-DD-MS 330kV lines
Could be related to the solar storm warning as electrical fires are a common side effect.
Great summary Ben. It seems the Federal Government were left with little choice but “to do something” to drive to 82% by 2030. Given this reality, the CIS is the best of the options on the table and as you say at least preserves much of the wholesale market functionality.
No-one actually knows what the safe level of inertia is, but in an inverter based system with distributed hierarchical control, the safe level may well be 1/10th or less of what traditional systems considered necessary. Inertia is actually an instantaneous but very small energy store. Inertia does nothing unless there is a frequency change. If there is a frequency change of say 0.3Hz, the same as occurred when LYA tripped, the energy available to stabilise the NSW grid is (1-(49.7/50^2)x I = 1.1% of the inertia.
When inertia is 20,000 MW.s, that is 239 MWs available. If the frequency change happened over 3 seconds the available power is about 80 MW. i.e. a 100 MW battery could provide similar fast frequency services as 20,000 MWs of physical inertia.
A key question is how to ensure all batteries don’t inject 80% of their available power at the same time otherwise there would be serious overshoot and potentially cascading instability. This is not a risk with physical inertia, but on the other hand batteries can arrest a far larger frequency excursion than physical inertia can.
In addition there is synthetic or virtual inertia on wind and solar farms which is also potentially significantly larger and more controllable than synchronous inertialarger
Great article! I see batteries are not providing intertia before an event however the FFR would provide a level of effective inertia. The question is how AEMO plans to consider this.
Ben Potter would be a trusted jourmnalist of the AEMO and other green protagonists. He would therefore be rewarded with exclusives.
Wouldn’t like to unnecessarily speculate but it is quite usual for pairs of adjacent units to share a certain amount of “common” plant and /or interconnections intended to provide reliability and the ability to maintain both units in service whilst minor maintenance activities are undertaken. Given it was simultaneous trip (and the unload seems to be vertical so either an electrical protection operation or turbine trip rather than combustion where the unloading rate would be slower) it looks like it is a secondary system of some sort. Recovery was good. Whenever these Toshiba units- or indeed most large units – trip from high load (normal metal temperatures) they need to cool for a bit before steam can be re-admitted (so called mismatch temperature). So it appears there was nothing really to “fix”. Just sort it out and get the units back on. It does happen, fortunately very rarely, but is the joy of working at the pointy end of a production operation. Something most commentators will never experience let alone understand.
It may be coincidence, but the increase in energy arbitrage revenue seems to roughly correspond to the introduction of 5 minute settlement in October 2021.
I’d guess it was caused by an incident in the DC distribution system.
Hi, I’m looking for information regarding the turbine availability (no. of turbine available for power production) provided by the Wind farm owners to AWEFS. As far as I have searched this information is not publicly available for research purposes. Am I right? If so, are there any proxy for the same?
Thank You.
Wednesday 10 April 2024
1/ The insurance will not pay,
2/ The part owner of C4 has a court order for a new enquiry
3/ The Australian Energy Regulator is taking legal action
4/ A industry standard Castell interlock System had not been fitted to the unit breaker to stop the isolation of the 220 VDC control supply when the C4 unit was in service.
5/ The original forensic enquiry by a structure engineer is on going
Uhh? Does anyone have a frequency plot for Tassie for event 1? A near 50% system load rejection is pretty huge. I assume the load was shaken off due to a severe voltage depression from switching onto a fault? Generation did not run back by 500MW so assume Basslink imports stopped. Dare I say a zone two distance protection operation from the opposite end so a slow clearance? What happened at 1.39pm? Generation went down so did this imply Basslink started to import again? Would be interesting to pull this one apart a bit more if someone has the operational data.
Archived copy of article!
https://archive.is/20240411103954/https://www.theaustralian.com.au/business/margin-call/secrecy-on-callide-c-disaster-continues-clough-boss-emerges-from-the-wilderness/news-story/4b105b44d91083580a76157ce607e5f2
SA the wind -eading state, burning diesel, not the first time lately.
Its all over Red Rover, check out Germany and GB to see our future, importing power and deindustrializing hand over fist.
So are we, although does anyone dare to keep a count of the energy-intensive firms that have folded?
And we can’t import from anyone, not until the Sun Cable is operating.
“so in the hypothetical future grid that’s heavy on wind and storage, such as we analysed in Appendix 27 within GenInsights21, you’d at least have had some forewarning to stock away what energy reserves you had available to be drawn down through this period”
Sadly the amount of storage you would need to cover even a whole day of low wind is astronomical in cost, and if you can’t recharge it because of low wind then each successive day it gets harder and harder to keep the lights on.
Great article Paul:
Based on modelling (Boston 2022* – Figure 1) the wind drought of June 2017 was small compared to 2006,2008 and particularly 2010.
There is now good hindcast ERA5 wind data for 40 to 50 years, so updating this paper would make sense. ERA6 will be available next year.
Climate change may also decrease wind speed so the future requirements could be tested against the various RCP models.
* Geoff Bongers mentioned this paper in: https://wattclarity.com.au//articles/2022/06/seasons-of-low-wind-and-solar-output-will-influence-the-shape-of-the-nem/ . It dies leave out figure 1
Nothing to see here I suggest. Well described by David Leitch “Seasonality and deep storage in a highly decarbonized NEM” published on your web site Sept 10 2020, and seasonally highly predictable. Have a look at some of the modelling work done by Geoff Bongers et al from Gamma Energy on the NEM. Basically says the same thing. You cannot sensibly afford enough storage to get through these doldrum periods. Perhaps you could get Messer’s Bowen, Albanese and Bandt to start reading some of this material. They might learn something!
Looking back month by month until October 2023 I haven’t seem any charging or discharging in the AEMO data for TB2BL1 or TB2BG1.
Thanks heaps Paul
April to June is a common time for slow moving highs to sit over SE Aus bringing very stable (and rather pleasant) weather. And it is astonishing how widespread such conditions are: from the Eyre Peninsula to north NSW and down to Tas. Also affects offshore wind around the 4 SE states. Geographic diversity of assets within this huge area doesn’t help.
You will however find solar was somewhat increased during this period compared to the week before, which is to be expected in this weather pattern. A silver lining to the non existant cloud, at least in daylight hours.
Thanks Ben
Here’s one dispatch interval in that time range that also had low solar:
https://wattclarity.com.au//articles/2024/04/13april-casestudy-low-vre/
(of course, close to sunset).
Paul
Fantastic article, Ben. Could I please query one detail though. If the interim unserved energy reliability standard is 0.006% and the old was 0.002%, then the interim would be less conservative than the old. But, I think it’s simply a typo, and the interim standard is 0.0006%.
Frustrating to hear that a Rule change can’t repeal this because it’s in the NEL. Hopefully some traction can be gained with customer representatives.
Well spotted Trevor and correct. An important zero has gone missing. We’ll get that fixed.
> In the wake of South Australia’s devastating 2016 grid outage, authorities moved to boost grid resilience in this way, building big batteries, more renewables and new interconnectors, while Australia’s energy market operator AEMO changed how it dealt with windfarms if grid issues occur.
None of these proposed solutions would have helped what took place that day. This is just another article pushing renewables at any cost.
> Big batteries: the maximum dispatch of big batteries in Victoria during the load-shedding period was 20 MW! The two hours of $16.6k spot prices had an average of 5 MW battery dispatch! That is quite literally nothing.
> More renewables: there are already plenty of renewables in Victoria. Stockyard (500 MW) tripped due to an unrelated issue a couple hours before the transmission towers fell. 1 GW of other wind generators tripped due to the transmission towers falling. How much more wind over-builds will be satisfactory for you?
> New interconnectors: an extremely expensive proposal with very long project lead times.
I also love the sophistication of your climate analysis. “We’ll have warmer air in the future which means we’ll have more moisture which’ll probably mean we’ll have more intense thunderstorms.” Top notch analysis!
Transmission towers just like the Baltimore Bridge are made up of ‘fracture critical’ truss members lacking redundancy-
https://edition.cnn.com/2024/03/26/us/baltimore-bridge-collapse-what-we-know
Now a tower lattice truss could well be designed to handle extreme wind all over it but maybe not a wind driven object seriously impacting a single member. Buckle that link and you introduce a chain reaction among load members bringing down that tower and similarly more interconnected ones with the resultant shock loading.
Some interesting analysis.
But worth emphasising that our existing hydro assets can dramatically reduce seasonal storage requirements.
A mostly wind & solar grid will have large deficits in winter (& large surpluses for Sept-Mar).
A lot of this winter deficit can be filled by existing hydro.
An average, NEM hydro discharges 1,500 GWh more in winter than summer. That immediately reduces seasonal storage requirements by that amount. But inflows are even more biased to winter, 6 TWh vs 2 TWh. So by simply restoring hydro discharge to be more seasonally similar to natural inflows, then we can reduce the winter deficit by 4 TWh.
During the years of the carbon tax, hydro discharged 6.5TWh more than inflows to take advantage of the temporary high prices, making it clear that our hydro assets are able to substantially shift generation from month to month & year to year to maximise profits.
Also worth mentioning that the Blowering & Hume reserviours at or near the bottom of the Snowy scheme largely decouple irrigation water requirements from Snowy hydro generation.
This is not a new phenomenon. There is a 220kV single circuit line in WA that has been screwed up and thrown on the ground by thunderstorms at least 3 times in 35 years that I know of. And this is the point. Microbursts are not dissimilar to tornados involving high velocity swirling winds. Towers are designed for dominant lateral and axial loads. Tortional loads from swirling winds tend to screw them off at any “skinny” point in the structure – a swirl does exactly this pushing wind loads on the conductors one way on one side of a tower and the other way on the other side. It would simply be aesthetically and economically prohibitive to attempt to design for these extremes. What it does challenge however is the validity of considering dual circuit towers as genuinely N-1 designs and indeed in this most recent Victorian case, putting single circuit lines immediately adjacent to each other in the one easement. Microbursts are usually very localised events touching down and following a bit of a path and then disappearing. Fully acknowledging the difficulty of getting line easements these days, but a few km of separation for critical network interconnections is a very good idea if you want to chase reliability.
This search may be useful.
https://news.google.com/search?q=AEMO%20Quarterly%20Energy%20Dynamics%20when%3A7d&hl=en-AU&gl=AU&ceid=AU%3Aen
Thanks Tim!
Hey Paul!
We see the future contracts are trending up significantly after the report came out, it’s quite contradictory that the report is indicating lower system price, but the market is reacting in another way. Any thoughts?
Hi Paul/team
Interested in your BoA preso, is this available to view (clicked on it and just the header slide opened)?
Fully understand if just for clients.
Many thanks,
Graeme
You are correct, Graeme. I’ll contact you directly.
Information on wind droughts has been available in the AEMO records, easily accessed at Aneroid Energy, since 2009.
See the work of Anton Lang and Paul Miskelly.
https://www.flickerpower.com/index.php/search/categories/renewables/20-2-four-icebergs-in-the-path-of-renewables-titanic
It has been well-known for over a decade that the wind and solar transition will not work until we have a radically new storage technology.
Why wouldn’t the turbines switch to liquids when the gas system gets tight as has always happened?
There are many challenges ahead. Have any of these predictions taken into account the steady rise in EV car charging during the night when the system will only be able to use gas for the expected “firming ” periods?
Fascinating article, thanks. As a retired engineer, now an outsider to the industry, I very much appreciate this free (to me) trusted communication.
I note the increasing scale of BTM energy, combined with talk of future industrial BTM generation for larger users. Does this tend to move supply from the green zone to “off the picture”? Maybe plot the estimated BTM generation as energy on the first chart, though that is already cluttered.
In the précis above, it is interesting to note the Abstract referring to “low-cost intermittent wind and solar resources”.
This “low-cost” claim is highly disputable. For instance, at the outset, the cost of a renewables based system must include the cost of all the required firming or back-up needed to make such a system work. In effect this amounts to building two independent generating systems – each capable of maintaining the system when the other isn’t.
Furthermore the renewables option must include the full cost of building and maintaining a vast web of new transmission lines needed for the express purpose of collecting the outputs from a huge number of generating units.
Note that these new lines collect electricity rather than distribute it, and they also inefficiently lay idle for much of the time.
Then there is the reality that short life renewables require perpetual rebuilding at unknown expense. Even the replacement of thousands upon thousands of worn out turbine blades presents as a huge expense item.
The gas based back-up generation system will produce a high level of emissions, perhaps far greater than anticipated.
However, the worst aspect of this discussion is the absence of any reference to the emissions free nuclear option.
A large, inflexible coal fleet in the presence of rapidly rising intermittent renewables can be
expected to confront a rising number of negative price events (Nelson et al., 2022)… prices fall to negative levels with solar plant capable of producing to the negative of ‘renewable certificate prices’, and rooftop solar exports being indifferent to wholesale clearing prices.”
The model seems to include an assumption that REC’s will be with us for ever, along with increasing negative price intervals and cost pressure on retail tariffs. Since VRE is often touted as the cheapest generation, why not do away with REC’s immediately, if not sooner? That would reduce market volatility and energy cost to the often poorest members of society customers.
Great article Dan
Just to be clear, your definitions of network curtailment factor and economic curtailment (offloading) factor use maximum capacity in the denominator, in simila fashion to capacity factor?
That’s absolutely fine, but every other presentation of these curtailment or offloading metrics that I’ve seen (eg AEMO QED and ISP, Energy Co REZ stuff, my own posts) uses available energy as the denominator – because this basis translates directly to the percentage reduction in potential output experienced. Numerically, using available energy in the denominator yields curtailment / offloading factors which are three to four times larger than the values in the Global-Roam results based on maximum capacity.
It would be worth highlighting this difference for readers, since most will be used to seeing the higher values derived as percentages of available energy.
Allan
Thanks Allan. Yes correct, this was done so that I could display the data as a 100% stacked area graph along with capacity factor (a metric that a wider audience would be familiar with), but I completely agree with your point and don’t want to confuse readers who might have thought that those numbers are artificially low – so have added in a note to make this clearer.
This is what happened at Callide C4. They tried to re-invent the wheel! It was ignorance. Just remember ignorance is just the lack of knowledge. Large 220 VDC Batteries are not designed to be connected together under loaded conditions because of the large fault currents produced when used for large critical loads in a power station. We are lucky it was only a coal fired power station
Indeed, the modification of the Callide C4 design, which included the removal of the Castell key interlocks, unfortunately led to a catastrophic failure. The attempt to improve the system’s redundancy and reliability inadvertently resulted in the complete destruction of the steam turbine, incurring significant costs.
The incident underscores the importance of thorough risk assessment and testing when implementing design changes, especially in critical infrastructure like power plants. It’s a stark reminder that while innovation and improvement are necessary, they must be approached with caution to ensure safety and reliability are not compromised.
The lessons learned from the Callide C4 incident will undoubtedly contribute to safer and more reliable design practices in the future. It’s a costly but valuable lesson on the potential risks of modifying established systems.
The Castell interlocking ensured that that the critical circuit breaker which needed 220 VDC to operate was locked open before doing any switching on the critical 220 VDC supplied from the unit 220 V battery, the unit should have been out of service. C3 must have been out of service when it’s 220 V battery was replaced
I don’t think we can assume that the shrinking quantity of bids in the 0-$300/MWh range necessarily increases reliability risks in the NEM. It would do so under the condition that “everything else remains equal” but this isn’t the case here.
There is a reducing quantity of generation priced in the market according to the variable cost of fuel in the price range 0-$300/MWh and this has occurred because we have added VRE priced at zero (or lower) and storage priced according to opportunity cost. These new resources tend to hold a good share of their bid quantities outside the normal price range and yet prices still tend to land within it. If anything, prices outside this normal range are much more likely to be below than above it.
More research is needed to understand bidding dynamics but outcomes suggest bid quantities are being shifted between bid prices above $300/MWh and below zero to achieve this outcome. If rebidding is occurring frequently (every five min for some resources) then even sudden and unexpected shocks might be corrected rapidly (ie in the next 5-min interval).
Thanks GregW
Apologies if what I wrote was not clear, but I was not focusing (just) on ‘reliability risk’ when writing this article.
The much broader pattern of prices in the NEM these days (which was happening anyway, but was also bumped along by 5MS) leads into different types of risk – including commercial risks for a range of different participant.
You mention storage in particular, but that’s only a small percentage of the volume of energy offered in bidding. There are also other reasons why bid prices have changed for a variety of different assets.
Paul
Looks like Bayswater and Mount Piper trying to punish Origin – ouch!
It seems that the answer would be some storage, perhaps 15-20% of windfarm peak capacity for two to three hours on the upstream side of the transmission constraint and a similar amount of storage near the loads. The load end storage limits peak demand on the transmission and the generator end limits both economic and transmission based curtailment
If I participate in the FCAS service, but happen to encounter a situation where I cannot provide the quantity as bid, will there be a fine? Is there any documentation on the penalty regulations? Thank you.
If I participate in the FCAS service, but happen to encounter a situation where I cannot provide the quantity as bid, will there be a fine? Is there any documentation on the penalty regulations? Thank you.
Hi Lu Ming, I’d suggest it is best to seek independent legal advice when it comes to regulations and penalties. But, by way of example, this article might also provide some food for thought: https://wattclarity.com.au//articles/2023/10/31oct-agl-fined-bayswater-loyyanga/
Thanks for sharing, Paul.
This statement is made in the concluding remarks section: “Detailed analysis of demand side reductions and the impact of electrification was not considered in this study.” Given the potential scale of household batteries and other demand side resources, I’d be keen to understand their impact on the modelling outcomes.
I believe the context here is gas network demand-side reductions (hence the mention of electrification, which will reduce domgas demand). It’s clear from this that heavy reliance on gas peakers is going to require either an increase in the storage capacity of the gas network, gas demand-side reductions, or more probably both in some measure.
To my mind, this should add some urgency to the gas-to-electricity household transition.
Wind was lower than often the case in the morning, but not the afternoon – when the longest peak prices were sustained. Only VP was substantially below. Eraring was not substantially below, it was half mast when 2/3rds mast on average.
Eraring might “normally” be at “2/3rds mast on average” Bruce but this was very far indeed from being an average situation. Prices were in five digit territory and in those circumstances Eraring would almost certainly have been running near full tilt if it had all four units available – but it had only two (again, not a “normal” state of affairs).
well I am still of the view that if we got on with it, it would be less of the problem. What ever the new supply source is I should have thought there was little doubt than unless demand is falling you can’t close old supply before you build new supply. The fact is that NSW has been ridiculously slow at getting new supply built given the age of the coal generation, the propensity of the coal supply to be throttled by flooding or competitive export parity pricing and the every increasing forced outage rate that requires $100s of m of maintenance capex every year.
It remains the case that getting new transmission built and more projects through the planning process is the only realistic and achievable answer. And it is realistic and achievable.
An excellent string of comments follow after Bruce Mountain’s article in The Australian.
One might say the cat is out of the bag.
Hi Bruce and Allan,
I think it’s worth taking note of the amount of transmission line outages at the time of the price spikes. Appears there were 3 x 330kV lines out. Wollar West – Wellington, Armidale – Tamworth and Collector – Yass. These outages were causing constraints on generation and interconnector flow too. I think they were all planned outages but some work may have taken longer to complete, hence the overlap.
Thanks Trevor.
I had managed to find time to post about one of the outages in the article ‘A quick look at the ‘N::N_CTYS_2’ constraint equation, with respect to Wednesday 8th May 2024’, but have not found the time to look at the others that were material.
Paul
Good points from both David and Trevor (and thanks Rob); those transmission constraints were definitely a very significant factor, and of course sufficient new supply – both energy and capacity (ie fully dispatchable MW) – should be built before existing stuff gets shut down. It’s also worth noting that there was plenty of undispatched coal-fired and gas-fired capacity available at the times in question, it was simply offered at very high prices.
I think Bruce’s main point (and David’s) about the growing gap between the rate at which the ISP assumes new stuff can get built versus the build rate we are actually seeing is the critical thing here.
For Trevor … as Allan notes, have a look at:
1) the spare (but expensive) capacity for coal units here, via AGL and EnergyAustralia.
2) the spare (but expensive) capacity for gas-fired units here … though remember that Uranquinty (bid cheaply), was constrained down.
Not quite sure I understand the logic here. The so-called network curtailment is almost irrelevant. It was never 100% of the wind farm capacity. You can only curtail something if it was capable of generating anyhow and none of the “Southern Tablelands were producing anything. If there was any wind they would look like (say) Bango Wind Farm #2. Touching a transient stability limit probably just implies the transmission network was getting full with Thermal and Hydro MW to keep the lights on and would no longer be N-1 fault tolerant for significantly more coming from this region.
Hi Andrew, the second chart is meant to show that the wind conditions for the wind farms in the Southern half of the state were still generally better than for those in the North-East – but NEMDE was constraining those wind farms down (to 0MW except for Bango Wind Farm 2) largely because of the N::N_CTYS_2 constraint equation.
As mentioned in the article, Green + Red = Availability and Grey = Unavailability.
While not being supportive of these problems, shortfalls of wind reliability would put in close focus the reliance of wind
ID fans are downstream and maintain a small negative pressure in the boiler, precipitators/baghouses to keep things inside. FD fans provide the bulk of the mass of hot (ish) combustion air to the furnace to ensure complete combustion occurs. Generally, controls to around 3% residual oxygen in flue gas stream to prevent CO formation. PA fans provide hot air to the mills to dry pulverized coal and transport the milled coal from the mills to the burners. No PA fans = no fuel.
Some may dispute ‘useful’, nevertheless The Australian also had commentary from Graham Lloyd (‘Powering the nation on wing and a prayer’) and Simon Benson (‘AEMO’s summer blackout predictions give Anthony Albanese another reason for an early poll’).
The latter is focused on its potential political implications – “highlights the desperation creeping into the government camp on energy reliability”.
Perhaps the more correct headline for the SMH and the Age might have been “Origin Energy throws NSW a Lifeline” ? So aside from all of the renewables running late – assuming 2 years is enough to get them back on track, who is building the 1500MW of OCGTs to keep the lights on during the cloudy, windless weeks in the future?
I would be curious to know how the aggregate wind yield compares to the aggregate energy consumed in restarting the turbines during these periods and what the ‘net’ might be (if I understand the operation of these machines correctly)?
This particularly long period(s) of low wind output is significantly increasing the wholesale prices in Victoria, SA and Tasmania. It has been compounded by lower hydro output in Tasmania due to below average rainfall over the hydro catchments for around six months. A further compounding factor is the lack of solar in Victoria, both utility and behind the meter. While relative solar output at this time of the year is lower, Victoria has disproportionally lower utility and rooftop solar compared to SA, NSW and QLD. Additional solar output during these low wind events would significantly lower prices during at least some of the days.
And whilst those wind turbines are not driven by wind they become 50kw motors to prevent annealing the blade bearings
assessing the wisdom of recent agreement re NSW and Origin requires an analysis of what the renewable energy contributions actually are, and how much projected input is likely to be realised, what other measures are available to the Gov to secure renewable energy ( including regulation and legislatory measures regarding investment, taxation etc) and how the grid itself can manage energy collection and distribution. I am keen to to see all that put together.
Given the prices prevailing I can’t see too much economic curtailment happening. Good coin to be made. Bottom line is the wind simply isn’t there. So how about someone doing a quick integration under the curve to work out how much wind overbuild, solar and storage (GWhr + MW)) is needed to keep the lights on for the last few weeks once all the nasty black and brown curves aren’t there….and then do a quick calculation of the costs. The number will be eyewatering. Then perhaps a visit to Canberra.
NSW has gone out to tender for 1 GW of LONG duration energy storage – in NSW the term long duration can be as short as 8 hours. What energy source will charge the 1 GW of long duration storage when wind droughts go for days or weeks and coincide with low solar output?? Its not just a case of the price to charge, its whether there is energy to charge. Sadly, the general public think batteries and pumped hydro are a like for like replacement for coal and gas generation. A position the general media is only too happy to promote.
It wasn’t just low on the Friday. It was low for the whole week, in fact, the lowest it has been in five and a half years.
I have now kept detailed daily and weekly records for wind generation for the last 295 weeks, more than five and a half years, and this most recent week ending yesterday, 26May was the lowest power generation across a whole week that there has ever been. Because that Nameplate increases regularly, then it stands to reason that delivered energy (in GWH) to the AEMO grid increases. So, the only comparison we have over time (taking that Nameplate increase into account, now an increase of 6108MW from my start point of 5301MW back in October 2018) is that all important indicator of Capacity Factor, (CF) where you actually can compare week to week.
That CF for the last week was just 10.33%, the lowest that figure has ever been for a full seven day week. The long term average CF for all wind generation here in Australia is just 30%, so when it is this low, then questions need to be asked. Only twice in those last 295 weeks has the weekly CF been below 20%, and the lowest prior to this week was 17%, and that was four years ago now.
Capacity Factor has been falling consistently every week now for more than 30 weeks, a loss of half a percentage point, and that’s pretty big considering that the full maths calculations are across 295 weeks. 30 weeks back, it was 30.41%, and now it is 29.90%, and the last time it was lower than 30% was also that four years ago. Perhaps the more worrying trend in all of this is the most recent 52 week CF percentage. As more efficient and larger turbines come on line, then you might think that they would tend to raise the CF figure. However, the exact opposite is happening, with that recent CF percentage for the last year now coming in at 28.7%, a full 1.2% LOWER than the overall long term percentage.
And hey, has nobody put together the fact that wind is low when those huge High Pressure weather systems hover over that area where two thirds of all Australian Nameplate for wind power is located, in SE South Australia and Central Western Victoria. Have a look at power generation for wind this last week, and then correlate it with the Synoptic weather maps. that looks a little like an ‘Uh Oh’ moment to me. Why place so much wind in that one area, and then watch as the bottom falls out of power generation with every ‘High’ that comes across.
Anton Lang.
And single digit wind output in NSW for over an hour since 15.00 hours.
If there is no wind it doesn’t matter how many wind farms there are, there is no wind.
I expect this will be short lived as a blocking high is forecast to sit over Australia from Thursday night so Friday and saturday should see the wind die down again.
Economiser leaks are lots of fun! By the time they show up as wet ash in the ash hoppers the high pressure feedwater has usually cut adjacent (finned) tubes. If you don’t spot it early and shut down then the cut tubes fail and the leak gets really big, really quickly seen as a big feed flow / steam flow split. Hopefully it is an accessible spot (economisers are usually jam packed, multi layer locations) or it will be cut your way in and weld your way out. Somewhere there will be a boiler Engineer about to have a very busy weekend sorting this out.
There is of course also some “transmission” component to the economic curtailment – specifically the limited interconnector capacity between regions. Eg if exports from SA to VIC are at the dynamic maximum capacity of the VIC-SA interconnector then SA will be economically islanded, resulting in apparent “economic curtailment” of generation that could other be dispatched to meet some out-of-region demand.
General scale predictions of wind are fine as can be seen from my comment on your article from Wednesday 29 May now being realised. Just this morning we have already seen 3GW of wind capacity reduction between 8am and 11am.
The reality is that wind and solar are unreliable. You can’t have a system where you just randomly start producing 3 Hazelwood equivalents of power and then 2 days later it all just dissapears.
We are on the precipice of collapse of the NEM because of how volatile the energy generation mix is so everyone is producing to ensure the lights stay on, which drives up wholesale costs which flow o to retail costs.
Given there is almost no storage in the market yet, I think you are grossly overstating the problem.
Halleluiah! We are now starting to get on the same page. The bottom line is you simply cannot expect to get base load energy generation out of inherently intermittent, essentially unpredictable, sources of energy without simply gross overbuilds of capacity and storage that will have to be paid for by someone. I suggest people go and read carefully the work done by Gamma Energy Technology that displays the results as exponential cost of decarbonisation for the NEM. (the reason any of this is being undertaken, remember people!). This is why simply comparing LCOE between technologies is simply silly. You need to look at the total system cost of energy supply to get the correct (least cost) mix of primary energy sources. Near zero carbon emissions simply becomes a constraint on that list of suitable technologies. Wind and solar certainly have their place and contribution, but the blind ideological belief that we can achieve (let alone afford) a 100% renewable power system is going to take us to a dark place.
Renewables are still cheaper when factoring in storage. The CSIRO report made this clear. In a 100% RE NEM, wind and solar are not the only energy sources. You are ignoring the not insignificant hydro component. There will be additional energy sources such as off shore wind (higher capacity factor than onshore), solar thermal (built in storage), possibly wave, geothermal and more.
Australia is not alone in experiencing lower wind output despite higher installed wind capacity.
From Reuters article 15 Feb 2024:
“Power generated by Texas wind farms dropped by 22% in January 2024 from the same month in 2023 as low wind speeds continue to stifle output across the main power system in Texas, the largest power market in the United States……
As wind power is the second largest source of electricity behind natural gas in Texas, the drop in wind output so far this year has forced utilities to sharply increase generation from fossil fuels to balance system needs.
Combined output from natural gas and coal was close to 50% greater in January 2024 than in January 2023, underscoring the enduring importance of fossil fuels within the ERCOT system despite the ongoing build-out of renewable generation capacity.”
Data from the Electric Reliability Council of Texas (ERCOT)
ERCOT –
What comes first:
1./ The removal of the legislated ban on a nuclear industry, or
2./ A Royal Commission where all the financial aspects of nuclear fuelled electricity generation can be put and thoroughly tested against those of a renewables based system.
Australia is becoming bogged down in a highly politicised faux discussion wherein truth is the first casualty while elsewhere in the world,evidence that supports the nuclear option is becoming undeniable.
Each passing day sees huge torrents of taxpayer dollars flowing fourth.
Is it really flowing in the best direction?
In the 18:30 interval wind generation in SA was 13 MW (out of a max capacity of 2,732 MW)
So a capacity factor of just 0.4%.
This is why LCOE should include interconnectors as well because all the batteries in the NEM wouldn’t have been enough to keep the lights on in SA if they didn’t have fossil fuel generation.
Actually it’s more an illustration of why LCOE analysis of any kind is a complete waste of time if you’re trying to understand whole-of-system economics. Semi-arbitrarily adding various system-level costs – real or imaginary – into technology-specific cost measures is a fool’s errand.
CSIRO’s GenCost reports have done the industry a disservice by pretending that this approach is meaningful or even feasible. Don’t take this comment to mean that CSIRO’s data on underlying cost parameters is necessarily wrong, just that they have gone down an unhelpful path in then trying to shoehorn other costs into LCOE calculations, in response to the clamour from people who don’t understand how to do a system-level analysis.
Conversion of open cycle gas plants to combined cycle should be part of the Capacity Mechanism as it enables 50% more electrical energy from the same amount of gas. As batteries are displace open cycle gas peaking plants for the morning and evening peaks, the new role for gas will be for high efficiency gas plants operating continously for several days at a time during wind droughts. Their lower marginal cost will enable them to cost-effectively charge peaking batteries at night and during solar hours. As an example, Mortlake Power Station was designed for upgrading to combined cycle. It is one of Australia’s newest gas power stations, and has its own gas pipeline to the Iona storage facility, so is exempt from the pipeline constraints described in the paper. Conversion of more gas generators to combined cycle could well overcome nearly all the gas capacity contraints outlined in the paper.
daily 500kV outages out West Vic for the next two weeks.
There was an unusual drop in Pelican Point output from 490 MW at 1600 to 81 MW at 1620 before resuming full output around 1800. Bolivar, Snapper Pt and Hallett all responded immediately to the change.
Yep, PP clearly had a very significant wobble (technical term) as its output was running up towards full capacity and that together with the 50 MW import constraint across Heywood, which is here every day for the next week and bit, drove the $9.5k/MWh spike.
In your 15 March 2024 article “Need for speed: How long has each battery project in the NEM taken to deliver?” you recorded Tailem Bend Battery as being partial operational from 27 October 2023. To date I have not seen any activity in the AEMO data for the battery.
Hi Alex, I checked again and from what I can see in the AEMO data there was non-zero SCADA output for the generation-side DUID for that battery at 13:25 and again 16:30 on the 27th October 2023. This could have been very early testing or a SCADA anomaly as it doesn’t look like there was any other activity until last Tuesday the 3rd of June 2024 – where it looks like it output 42MW at around 18:00
CS ENERGY CALLIDE C4 https://www.csenergy.com.au/ArticleDocuments/251/CS-OHS-53%20-%20High%20Voltage%20Electrical%20Isolation%20and%20Access%2008%2023.pdf.aspx
4.1.1.4 is the way the CS Energy has ensured that they will not switch off the 220vdc again with unit in service
What has CS Energy changed with their 220v batteries and what happened to the castell interlocking key systems that ensure that the UCB is open and the unit is out of service. Where are the critical HV isolation instruction sheets. Who turned off the 220vdc? The C4 unit operator involved would have PSD poor guy. What has CS Energy done to him? He was left in the dark! CS-OHS-53 is available on Google. Snowy hydro HVSI is a good information source
Thanks for covering the Dunkelflaute Paul.
It is now common to see a loss of 6GW of generating capacity across the NEM in the space of 24 hours due to blocking high pressure systems.
This is going to become more and more of an issue as fossil fuel generation is wound down and can’t be relied upon to meet demand, especially in times of drought and/or night.
The rains are late and storages are closing in on the ‘prudent storage level’. Storage levels are the lowest at this time of year since at least 2018. Reference to the Tasmanian Economic Regulator’s website – link hopefully included. [https://www.economicregulator.tas.gov.au/about-us/energy-security-monitor-and-assessor/tasmanian-energy-security-monthly-dashboard]
Next day’s bid data shows that Yabulu (Townsville gas turbine) was directed to remain online from 21:00 to 22:00
Something very interesting technically going on here. Sounds like one are of Tassie is electrically very weak – not many synchronous machines running and a line fault (lots of reactive needed to clear it – Basslink won’t provide that) and it all got the wobbles. MW swings you don’t want. Sounds like some power system stabilisers not doing what they are supposed to (or not enough of them in service).
Hydrogen plant in SA ready by the end of next year! LOL…Seriously, AEMO? A new type of power plant is going to be up and running from scratch in 18 mths? has it even reached FID yet? Even the SA government’s own website says “early 2026” which is I suspect still very optimistic.
So let me get this right. We have wind at about 10% capacity factor for 12 hours, no solar (its night) and the nasty coal and gas have probably delivered 250GWhr overnight. So, let’s just assume we ended the 18th with full batteries – each with about 250MW of inverters and 2500MWhr of storage and we have 100 of them – the lights might still be on when the sun comes up…..and then we need to not only need to supply the higher daytime load but also have enough spare to recharge the batteries to do it all again. And this is before we try to “electrify everything” and as a minimum shift an increasing proportion of liquid transport fuel energy into electric energy stored in EV’s ready for the daily commute. Even if the 127,000MW of wind proposed (shown in the AEMO forecasts) gets built and it behaves like this we have spent a truckload of money and we still have a problem. I must be missing something again.
You’re right Paul that 30202 MW is a record for NEM-wide AM demand, at least in the last 15 years. But not far above the previous record from last year of 29996 MW on 21/6/2023 at 7:45. Andrew makes an interesting point in his comment about “how will this be done in future?” My view is that the “electrification of everything” will be part of the solution rather than part of the problem. If I’ve got an EV in the driveway with 50kWh in the tank, wouldn’t I be ready to sell some of that for of $0.70/kWh (=this morning’s spot price of $700/MWh)? Governments should be legislating a new FiT to motivate people to plug in their cars like we all plugged in our rooftop PV. They should not be spending a truckload of money to duplicate infrastructure that people will end up buying anyway (at least, a version with four wheels attached). Forget 100 grid-scale batteries, and imagine a couple of million cars.
We were extremely lucky that this occurred during daylight hours when solar production was able to offset the complete lack of wind in the NEM.
Having said that, these blocking highs do tend to occur on a somewhat regular basis (another one forecast for Saturday/Sunday) which is why the claim of just building more and dispersed renewable capacity is not a reasonable answer.
Jo Nova has some good analysis here: https://www.joannenova.com.au/2024/06/20-billion-in-wind-power-across-australia-can-only-guarantee-as-much-power-as-two-diesel-generators/
How did a Jo Nova reader find their way here?
Oh she linked directly to this blog
Hi Jo! Hope Shell/BP/Chevron/etc is paying you well 🙂
Related is that gas deliverability from Iona storage is itself dependent on storage volumes once the threshold limit is reached. Deliverability doesn’t simply stop when it’s empty but begins to decline well beforehand.
Plus the ability to refill storage is in all cases much slower than the ability to empty them. So a single day of low gas demand and storage filling cannot offset a day of high demand and storage withdrawal, the refilling process is considerably slower.
They’re seasonal storages, they’re not capable of refilling rapidly so once depleted it’s a slow process to refill.
I have come to WattCkarity of recent times – I WILL be back, thank you!
You ARE on the Right side of Oz National energy grid history – please keep it up for the benefit of the technical understanding Oz citizens though now a little past it!
The coverup continues, the contact American technical writer is being extra diligent and taking all necessary precautions!
“The safety of our people and plant is our priority,” Mr Busine said. (The rebuild accident rate is unknown)?
Why is the forensic engineers report secret?
The critical 220 VDC Battery was disconnect during a unauthorised switching by persons unknown apart from a flash cartoon video
The C4 unit operator was disconnected from his controls, and left in the dark.
Unit C4 has been impacted by the availability of overseas based specialists needed to perform critical work, is more contact technical writer coverup.
The Contract Technical Writer gained most of his experience working for Boeing so he is use to big wreckage.
The cost of this debacle is top secret.
“Makes me shake my head” It makes me shout at the computer!!
It is easy to get 30 or 50 years of wind data at 100m from ERA5 from open-meteo say at Sheparton:
https://archive-api.open-meteo.com/v1/archive?latitude=-36.38&longitude=145.39&start_date=1980-04-01&end_date=2024-06-21&hourly=wind_speed_100m&wind_speed_unit=kn&models=era5
They even supply a few climate model data as well for future forecasts!
Of course what further complicates the use of historic wind data is climate change. Meaning that the probability of low wind periods could be changing over time. Of course 30 plus years of data is still going to be much more meaningful than two or three years.
A further complicating factor is that as more wind farms are installed, more wind energy is removed from the system meaning an overall reduction in generating capacity.
The law of conservation of energy makes it pretty clear we can’t just use the wind to spin blades and not have any negative impact on wind patterns.
The 66 MW Barron Gorge Hydro Power station is just coming back on line since Cyclone Jasper in December 2023 BUT due to a very long and good Wet Season in the highest most reliable rainfall region in Queensland (if not Australia), the 88 MW run-of-river Kareeya Hydro Station in Far North Queensland has operated almost 24/7 since Cyclone Jasper ie its basically operated as a baseload rather than peakload power station for over 6 months and continuing to do so as the Koombooloomba Dam that supplies Kareeya Power Station is at 90% capacity. Its over 60 years old so its printing clean $ for its owner – the Queensland Government.
It would have been very helpful if some or all of the curtailed power could have flowed into Victoria or SA.
Hi folks,
I’ve got a much more basic question – how are interconnector import and export limits calculated by AEMO? How much relates to interconnector physical limits / design and how much relates to NEM demand / supply?
And why are the import and export limits often different to each other? For example, on the VIC-NSW interconnector right now, the export limit (VIC–>NSW) is 183 MW and the import limit (NSW–>VIC) is 18 MW.
Never did get a response to your question Andrew?
Going out in time we can also look forward to the combination of low wind and old wind farms that need. 1. replacing and or 2. lots of maintaining coupled to a financial depression. But I am sure Hydro will not be in a once in a 10,000 year drought at the same time.
Heading needs an update: “Seven years since the ‘Wind Drought’ in 2007”
The C4 220VDC was isolated with the unit in service, a written high voltage isolation was not used, the insurance has not made a payment to CSEnergy,
The RTS insurance will be self insurance by Qld Gov, their is a problem with staff turnover, there is a problem with operator training, maybe the CS Energy mess should be given to Snowy Hydro as surely they would not turn off the 220VDC. They provided a picture of the C4 220V battery, is that small battery able to supply all the critical services for the C4 unit for one hour with battery charger not available. Does the employment of a extra 40 Engineers help, remember the C4 unit operator did not know that that the 220VDC had been isolated
Most years in the non-summer months, we typically have two periods of low wind output combined with low solar output which last around 5 days. However this year has been different as it has been low wind output combined with low rainfall on top of a drought in Tasmania, meaning much less hydro output from there. Conditions from a year like this are one of the most challenging to solve in a future 100% RE grid as the typical projected storage and up scaled RE won’t solve it.
There are several factors that will assist, but resolve the problem. Victoria needs much more utility scale and rooftop solar in the mix of RE generation. It will require a significant over build of solar due to the poor performance during the winter months, but this appears to be one of the cheapest options.
A doubling of capacity of the QNI flows into NSW is potentially a good solution, but not cheap and complex on the NSW side of the connection. The addition of BESS at multiple points would be required to increase the carrying capacity of the existing links in addition to upgrades of substations. This of course assume QLD continues to build out VRE sufficient to have excess.
Additional wind and solar in Northern NSW is another option, but of course this is where there is great resistance to this, particularly wind.
None of that fixes the problem ofhe wind not blowing across the NEM at pwriods when solar isn’t producing.
The fact that we have over 11GW of wind energy capacity and it was producing just 88MW shows oversupply of capacity doesn’t fix the fundamental flaw of renewables meaning you can’t guarantee output.
So a $37.50/MWh bounty then – doesn’t seem too bad, really.
I would suggest your sketched trendline overstates the case, since it is assuming the 5-day period of very low wind extends out to 9 days. A more reasonable comparison to see “what would have been” if the large burst of wind generation at the end of month hadn’t occurred would be to extend it at the rate of the overall May average?
Thanks for sharing this report. it outlines a common scenario in many industries where risk competency relates mainly to personal risk ( PPE and site conditions ) and not to the hidden risk of unmanaged process risk. In my many years of visiting many mine sites/generator sites / and general industrial areas, I have always been conscious of both risks. However, it is not the case that these two areas are taken seriously enough. Hopefully, this report will be taken up with all industries waking up to the combined scenarios stated.
We are seeing more anomalies like this due to the weakening magnetic field, shift of the magnetic poles and being near the peak of the 11 year solar cycle.
The final paragraph should say “150 GWh”, not “150 MWh”.
Nice to see WA included, very helpful
Multiple market notices for unplanned outages in Northern/Central NSW yesterday:
– MN117313 at 05:41: unplanned outage of Lismore – Coffs 330 kV at 0511, influencing constraints on Directlink
And then after load shedding (from 18:17):
– MN117323 at 19:41: unplanned outage of all 3 cables of Directlink at 18:45 (- but it looks like flow was restored by the next DI)
– MN117325 at 20:01: unplanned outage of Eraring – Newcastle 330 kV at 19:40
Speculation: some correlated transmission failures around Lismore, with load shedding in that sort of area and limited impact on overall NSW prices or supply/demand balance.
An article on RenewEconomy “Load shedding in NSW holiday spot after multiple transmission line trips” adds some more details.
Macarthur Wind Farm has produced nothing since 6 June, which at 420 MW is one of our largest wind farms. Surrounding wind farms have produced well when the wind has blown, and other wind farms connected to the same 500 kV power line haven’t had the same issues.
I’m curious how the logistics of firing up for the first time in 5 years works. Do they have to start by hiring in engineering and operational staff on temporary seasonal contracts?
It is interesting that the wind drought is caused by a succession of high pressure zones which in turn minimised cloud cover, so while wind output was down across the NEM by 10% probably 13-14% adjusting for capacity changes, solar output was up 20% so combined wind and solar output was actually up 5%. The big problem is that hydro output was down 22%, 30% in Tasmania and 40% in NSW so total renewable generation was down 0.1% for the quarter
Running a CC gas plant continuously and using hydro as the swing ends up using much less gas than running gas peakers
So CS Energy’s process safety management systems were not where they needed to be and the critical 220 VDC supply to unit C4 was turned off by maintenance staff without the operators that where in charge of the power plant operation knowing. So the insurance payout is lost and the cost is many $100M! It’s time that management was transferred to Snowy Hydro as they are the only public authority with the ability of managing power plant without wrecking it by ignorance(ignorance is only the lack of knowledge).
The event this morning comes a day after planning approval was granted to the Limestone Coast Energy Park battery. Will having that storage online allow the 50MW constraint on the Heywood interconnecter to be relaxed?
Hi Tristan, Thanks for the question. It does indeed seem like it could support the grid in that way, being quite large – 500 MW / 1500 MWh. Perhaps a SIPS (or similar) contract might be entered into. We’ll have to keep an eye out. Noting the anticipated timeline for operation is in 2026.
Really sterling article Dan.
Do you have any view on how the marginal benefits of energy connect compare with where we would be had the original $95M been spent on Riverlink in 1997?
Thanks Mark. Many would argue that the situation would have been easier, cheaper, quicker, etc. if things were started two decades earlier.. I don’t disagree, but hindsight is 20/20.
A few what-ifs and other questions worth thinking about:
– How things would have panned out had Riverlink been in place before the renewable development supercycle (and the system black event)?
– Would it have actually cost $95m?
– Were the expected benefits the same back then, as they are now? (and how easy was it to predict these when the original Riverlink didn’t go ahead)
– Are all the benefits measurable/model-able?
Sorry that should be “Project Energy Connect” in my comment!
No, or at least I think it is still too early to call it. The trailing 30 day levels are still below that of last year. The other factor in this wind drought weather pattern, a lack of rain particularly over Tasmania, is still resulting in Tas Hydro using fossil gas for generation and importing significantly from Victoria. Even though the key catchments in Tasmania seem to have good rain in the last couple of weeks, it would appear the catchments are still a bit dry and the dam levels low for Tas Hydro to want to return to normal operations. While this situation continues, it will result in higher prices in the southern regions of the NEM.
Here’s hoping we continue to see good wind levels and rainfall over southern Australia.
A major difference in projects cost blow out is whether its govt or private funded. Your statement of solar and wind farms not running over budget is because the estimate is real and there’s little room to keep throwing money in to fund the project. Compare that to govt funded projects where the real price is rarely disclosed, as politicians don’t want to admit the real cost as it will be hard to justify and won’t win votes. Hence cost blow out from original budgets.
Interestingly Snowy timeframe is near a decade, $12Bn in costs + $6Bn in federal buyout of state ownership + Humelink sums to approx $20Bn. This equates to ~$10Bn/GW…. kind of makes the argument against Nuclear null and void.
I’m certain that the masses do not understand the magnitude of the energy storage problem. Pipe dreams like Snowy Two will continue to crop up until the people understand the magnitude of the problem and grasp that even when complete this project will not provide a sensible backup to the wind and solar power grid.
As a simple example. A fuel tank in a typical car may hold 60 litres of petrol. Petrol will have an energy density in the range of 9.3 kWHr/litre. So that tank will hold 558 kWHr of energy.
Lifting one litre of water 680m will provide approximately 6800 Joules of potential energy, which is 1.9 WHr.
So to get the same energy storage as per the petrol in your tank, (ignoring all losses), would require lifting nearly 294 thousand litres of water up the 680 metres of Snowy Two.
Petrol really is an amazing dense fuel. If you want to replace petrol with electric propulsion, how many cars will be allowed to recharge each week on Snowy Two? And remember, you have to RECHARGE the Snowy Two scheme between discharge events.
I’m certain that the politicians who are pushing this scheme either won’t know the above or won’t listen to you if you tried to teach them. If the truth is known, the bubbles bursts.
It doesn’t seem unreasonable at all that 60 litres of petrol can raise 294 tonnes of mass 680m when used efficiently – after all, you don’t expect that driving your 2 tonne car up from Penrith to Hazelbrook will use much of your tank, even when it’s burning that fuel very inefficiently in an internal combustion engine (where the bulk of the energy stored in the fuel is going out the tailpipe and radiator as heat).
If you’re talking about replacing petrol with electric propulsion, you need to take that efficiency difference into account. The petrol vehicle is going to get about 700km out of that tank, so about 800Wh/km, whereas similar sized electric vehicles use about 160Wh/km. So drive the EV as far as the petrol vehicle goes on one tank requires less – only about the energy stored in 58,000 L of raised through the 680m head.
Also 294,000 L sounds like a large amount, but it’s almost exactly what the existing Tumut 3 station can pump uphill *each second* (the downhill generating flow is considerably higher). The Tantangara top reservoir in the Snowy 2.0 scheme holds 800,000 times that quantity when full.
Too bad it doesn’t last. The blocking high expected to impact for the next 5+ days will see a significant reduction in windfarm output.
As usual when it comes to real life, it’s not the peak performance but the minimum output amounts that truly matter as peaks are often short lived.
More batteries will help a little, but those 4 to 5 day low wind output events are going to be the largest challenge for SE NEM to transition to 100% RE.
The quick answer about highest 7-day average is yes, by a long way. Midnight to midnight, 19-25 July averaged over 5,950 MW, 600-odd MW up on the previous high in July 2023.
Someone raised a point about minimums. Those have increased as well – the Q2 2017 calm period saw a minimum 7-day average of 420 MW. This Q2 the minimum 7-day average was 1,320 MW.
Can someone outline how semi-sched and batteries are treated in this calculation please? For very little variation in Market Demand, the IRPM seems to be almost precisely out of phase with maximum generation. Not unsurprising, but if this IRPM is calculated assuming charged batteries and predicted wind then once you get down around 10% you really are only a hiccup away from a problem – particularly in a region unable to be supported via a transmission constrained market. Gets even more precarious if some of your assumed generation is coal that has been decommitted for a few days and is now cold.
I did miss poor Peter Hannam’s mention of emissions in the body of his piece about power price changes, so have updated my LinkedIn post accordingly, but I reckon my broader point still stands – it is super weird that such a significant change of direction – one that seems partly systemic beyond week-level fluctuations in wind output – isn’t getting more coverage!
The rich people of Elizabeth don’t seem concerned.
Blocking high causing a significant downturn in wind production.
Callide C3 is apparently offline as well for another month.
Businesses in general are at a 25yr high for bankruptcy/insolvency. Nothing to see here folks.
“Large-scale government-initiated projects unlock many smaller privately-funded projects. In the case of transmission, this can be in terms of actual project approvals – but in a more holistic sense, many of these mega-projects eventually play a role in enhancing investor appetite and ensuring long-term confidence in reliability.” Whoa, Dan. And your evidence is ….?
Bowen said it would be fine.
I wonder if he hears the mob forming out the front yet, pitchforks and torches to the front.
Is the battery in SA flat yet? It’s supply proportion of the local grid has fallen from 6% to 1.5%. Is it being kept partially charged to allow for a black start?
Due to the extension leads from Vic being out or seriously limited, SA is operating like an island. And if you’re exposed to the going rates. OUCH.
Glad Bowen has plans for the rest of us that don’t look like an extension of SA’s current woes. /s
Hi Ian, you might have already seen it, but if not we posted a follow up looking at bids by ‘fuel type’ on the 31st, looking back at the evening of the 30th. Your observation on the battery storage sector that evening seems to be supported by our initial review.
https://wattclarity.com.au//articles/2024/07/an-initial-pass-over-energy-bids-during-price-volatility-of-30th-july-2024/
So CS Energy’s process safety management systems were not where they needed to be and the critical 220 VDC supply to unit C4 was turned off by maintenance staff without the operators that where in charge of the power plant operation knowing. So the insurance payout is lost and the cost is many $100M! It’s time that management was transferred to Snowy Hydro as they are the only public authority with the ability of managing power plant without wrecking it by ignorance(ignorance is only the lack of knowledge).
So CS Energy’s process safety management systems were not where they needed to be and the critical 220 VDC supply to unit C4 was turned off by maintenance staff without the operators that where in charge of the power plant operation knowing. So the insurance payout is lost and the cost is many $100M! It’s time that management was transferred to Snowy Hydro as they are the only public authority with the ability of managing power plant without wrecking it by ignorance(ignorance is only the lack of knowledge).
At the beginning of the first volatility period 17:10 wind output was 124 MW.
This is despite SA having 2GW of installed wind generation capacity. So a capacity factor of just 4%
CS Energy spent $38m of taxpayers’ money hiring lawyers over the past two years, including to fight the release of the landmark report into the company’s woes. The investigative report into the 2021 Callide C explosion compiled by forensic engineer Dr Sean Brady, cost taxpayers a further $10.4m.
Richard Baum.
CS Energy spent $38m of taxpayers’ money hiring lawyers over the past two years, including to fight the release of the landmark report into the company’s woes. The investigative report into the 2021 Callide C explosion compiled by forensic engineer Dr Sean Brady, cost taxpayers a further $10.4m.
Hello, and thank you for this information.
Regarding the black line marked “total volume produced” on the “Trended Aggregate Volume…” graph, is that the total energy that was actually delivered after bidding?
Thank you.
Giles is a hack. He always blames fossil fuel generation for every outage or price spike despite them actually being caused by the lack of renewable energy generation.
Glad to see you shining a light on his lies but sadly his true believers will likely never see your rebuttal.
It’s fascinating how people will see what they want to see in any given piece of data or analysis.
Take the dispatch interval highlighted in Paul’s bid chart, ending 18:00 on 30 July. Putting aside how much supply capacity was renewable or not, there were 37,281 MW offered to the market and only 31,144 MW needed to be dispatched. However the proximate reason that the spot price went over $15,000/MWh in every region was that of those 37,281 MW offered, 6,273 MW (17%) were priced above $10,000/MWh.
Maybe the technical and competitive dynamics leading to this bidding behaviour would be more worthwhile subjects for commentary than whether Giles Parkinson is a hack or the level of renewable generation at the time?
It would seem the lack of competition from Victoria (or NSW via EnergyConnect) contributed to the high pricing. Of course more battery operators combined with additional storage capacity would have assisted as well.
Not sure if I’m missing something here but in the CS Energy response to Brady’s report it appeared that the key interlocks have now been removed to allow 2 batteries to be paralleled on the unit 4 dc switchboard. Given this was done quickly it presumably did not include upgrading/replacing the unit and station dc switchboards to allow double the fault current. Does this mean these dc systems are now in a much less safe state than their original design? It would seem to me that changing over from unit to station dc supplies with the unit online would never have been allowed for in the original design. This is unit protection after all. Just wait for the next outage. Did the dc electrical design documentation allow for dc switching with unit online? If not was a dc trip while replacing battery chargers with unit in service in 2021 an entirely foreseeable risk? And Brady states that the dc system relies on the charger, with the battery only providing a backup to the charger. I’m having trouble with this when dc oil pumps must always be available in an emergency. If a unit trip removes the ac then the dc oil pumps are the only thing preventing a catastrophe. It’s a long time since Callide C was designed, but maybe someone can reassure me that CS Energy’s “additional redundancy in the battery charger system since Unit C4” hasn’t exposed it’s people to far higher risk?
I thought it was the downramp of solar @ 5~6PM in South Australia that initiates the price volatility. Maybe I have to hate fossil fuels more in order to properly understand.
Expect those low wind conditions to continue until Thursday with a big blocking covering most of the NEM.
The saving grace is that temperatures are not forecast to be as low.
Fantastic article as always Allan.
I’m curious about how you estimated the state of charge of batteries that don’t report them.
I’ve heard that the 5-minute Initial_MW or other 5m data for batteries doesn’t accurately represent the average charge/discharge amount over the 5 minute period.
You mentioned calibration against the Torrens Island battery. Was that to determine a relationship between the 5-minute Initial_MW data & the actual 5 minute average, or something else? If so, and if you’re happy to share, I’m curious what you found out.
Thanks David. The charge estimation is based on:
– taking the mean of SCADA (InitialMW) values at start and end of each 5 minute interval as average interval dispatch, to yield energy consumed or sent out in each interval (yes the reality is more complex if Regulation FCAS is being provided by a battery but this ‘trapezoidal’ approximation is not too bad)
– applying a conversion loss rate to both charge into and discharge from storage
– assuming a constant rate of ‘parasitic’ losses (auxiliaries etc) from stored energy
Playing around with the last two rates yielded a calibration against the published TIB storage levels which then seemed to yield reasonable results for the other batteries after picking a start-of-day charge level for each that left them with minimal energy in storage at the time their availability fell to zero in the evening.
It’s only a rough estimation that wouldn’t necessarily work over an extended period but I’m happy it’s consistent with the behaviour of the batteries over the day in question.
Allan
In Victoria Newport Power Station contributed nothing through the morning and evening price spikes, while Mortlake was well below full output. Are gas pressures low through inadequate withdrawl capacity from the Iona storage?
The state and federal governments have been forced to re-enter the business of electricity generation, storage and transmission and they are making a mess of it. Look what happened when CS Energy’s Callide C4 steam turbine was destroyed by ignorance (they turned off the critical 220vdc supply) and problems will continue until training,examination and advancement of critical personnel is controlled by a central authority. God help us if nuclear power stations are required.
Critical personnel at CS Energy lacked experience and knowledge of process safety.
Callide C4 management allowed unauthorised untrained maintenance staff to carry out the critical switching that destroyed the C4 turbine.
Their claim for an insurance payout is lost.
What is the total cost including lost income, damage, legal , enquiry and repairs?
The coverup continues, what is the history of the critical 220VDC batteries, what happened to the critical key interlocks that have been used for over 100 years to avoid what happened at Callide C4.
220V batteries should not be paralleled because of possible large fault currents, switching has be done offload, that is why key interlocks use to be used and only authorised trained operators should carry out these critical switching proceedures with the unit out of service.
This makes me wonder if the gas generators aren’t doing a similar rough-and-ready estimate of their battery competition’s State-Of-Charge level. It seems like the optimal strategy for them might be to provoke the batteries into emptying before the market loosens, allowing them more pricing power with less competition remaining with available capacity.
You’d hope the battery bidding would assign a certain amount of variance to future market outcomes, not just take the pre-dispatch forecast as the 100% expected price (ie you only have to assign a 4% probability for a future $15,000 price to have a better expected value than a $500 price available now).
You can just imagine where this would go had Eraring closed as scheduled – its energy overwhelmingly would have moved to OCGTs resulting in higher capacity factors. People need to understand that gas is not the same as coal – coal stockpiles are easy – managing instantaneous demand flows in a gas system is difficult and will cost money. Pipelines, compression and localised storage. LNG storage starts to look sensible. You need a very big battery to store the energy equivalent of a 50,000 tonne tank of LNG. Otherwise you end up in liquids. And the logistics and costs get eyewatering.
We still appear to be in conditions with blocking high pressure systems that are resulting in lower than average wind conditions that are typical for this time of year combined with low rain fall in Tasmania. At least we are starting to see some short periods of very good wind conditions, but regions like SA lack the storage capacity to fully take advantage of these short periods. Lack of transmission capacity is further compounding this issue.
Hi there, sorry if this is a dumb question but why do gas storages on AEMO’s GBB dashboard continue to empty when there is a net surplus of supply over demand? For example, right now (Tue 8 August 14:35) there is supply of 5,721 TJ and demand of 5,580 TJ giving a net supply of 132 TJ. Yet Iona, Moomba, Roma and Silver Springs storages are all emptying with an aggregate rate of -76 TJ. Shouldn’t these 2 numbers be the same unless gas is being lost somewhere in the system?
Hi Andrew, I probably can’t give you a firm answer but, initially, two reasons come to mind. Yes, there is a thing they call UAFG – unaccounted-for gas – that represents fugitive losses along the network ~3 – 5% ish. Secondly, The AEMO note on Demand in the GBB (fur the current day) suggests that some calculations are performed for VIC and some demand for gas powered generation data might be excluded due to the methodology.
Great information. it is interesting the minimum to the maximum, better than 25% which confirms that solar produces even on days of cloudy weather.
Pretty much a standard frequency dip for 500MW loss with plenty of coal/gas/hydro units operating.
Inter-connectors not analyzed
The dispatchable power may not have been available where it was needed
You have to repeat your analysis State by State.
How close did we come to LOR1 and LOR2 warnings?
Availability of water for hydro plants may also be a factor. If the long term forecast is for dry weather that will impact on really dispatchable MW
Same for gas
Thanks for your thoughts Matt. In general some of those factors could impact regional reserve levels and prices of course. But in these specific cases of multi-region spikes:
– there weren’t binding constraints on the relevant interconnectors (which is why prices spiked jointly).
– there were no instances of actual LOR1 let alone LOR2 declarations.
– if you’re contending that being “close … to LOR1 warnings” would be sufficient explanation for prices near the MPC then I strongly disagree.
– hydro plants had enough water to offer the availability that they bid into the market. Are you seriously suggesting that the opportunity cost of undispatched water was a 5 digit offer price?
– for gas, see water.
Of course you’re welcome to run and present your own analysis of the individual region cases including interconnectors, I’d be genuinely interested in your results and conclusions.
Cheers
Allan
Would be interesting to unpackage these events to see who / what was the marginal cost generator. This is the problem with energy only markets. Batteries have no hope of paying their way without a very large spread on their energy arbitrage. Gas turbines have to cover their start up costs (loss of life of very expensive hot gas path components) over a relatively short generating window the duration of which the generator needs to predict before committing to run. This will only get worse as coal plants are pushed out of the market. From their min gen position, they can ramp to max relatively cheaply. Once they have to be decommitted it takes a long period of predicted generating time to amortise a restart. This is where the gentailers have inherent advantage.
Thanks for your comment Andrew. I’m not sure that the identity of the marginal generator is all that important here – there were signficant volumes offered at 4 and 5 digit prices from all fuel types other than solar and wind, and by more than one participant. I think that’s the key observation; which specific generators happened to be marginal in the high price intervals doesn’t really add any information, since there were clearly a number of participants offering at very similar prices. Allan
Tarong North NEM registration has Reg Cap 443 MW, Max Cap 480 MW.
I don’t understand why we are getting this continual rhetoric of coal plants being “inflexible”
Why do coal plants even need to be flexible when they have contract volume secured for months/years in advance.
A lot of Renewable advocates really get upset by the advantages low-cost scheduled generators have.
Maybe they should strive/mandated to be just that, low-cost scheduled generators instead of this semi-scheduled nonsense of doing half the job.
There have been a few papers published by Griffith University regarding Australian energy, most interestingly, P. Simshauser, 09/2021, “Rooftop solar PV and the peak load problem in the NEM’s Queensland region”. It is definitely worth a read in my opinion.
This one, Freddy?
https://www.sciencedirect.com/science/article/abs/pii/S0140988322001748
yes, that’s the one
is there anyone actually taking full advantage of the negative prices during the day? or are we relying entirely on batteries being built?
at this rate it looks like the price is going to go negative near daily for the next 6 months
Energy storage is probably the only way it could be taken advantage of. Unless there is another way to short the NEM?
Buy electricity from the NEM during the day for cheap to charge batteries etc, and then sell it back to the NEM when it’s more expensive.
Great read thanks Dan.
On umbrella/rain independence – whilst I accept the theory, it doesn’t matched my lived experience. 🙂
Mother Nature trying to average out the wind conditions by reminding us of the power of the wind.
Interestingly wind generation in Victoria in Q3 has continued to perform below average for this time of the year, while SA is close to average. Tasmanian dams and catchments need more rain, but I’m sure they could get by without floods.
It is time for a Royal Commission into what happened!
Was the critical 220VDC supply modified from original and who planned the process that allowed it to be turned off, with the C4 unit in service.
The C4 unit control room operator did not know that the 220VDC was switched off.
Why the need for a coverup, the fancy SC employed by CS Energy, commissioned the so called enquiry to avoid public disclosure of what really happened.
CS Energy should be given to Snowy Hydro
What is the total cost including increase cost of supply to the grid of CS Energy failure?
For those not following too closely, note that it’s a different unit than Callide C4 (the one that blew up)!
Callide C3 is the same as C4 its 220VDC switching happened with the unit out of service.
C4 was wrecked by ignorance (which is a lack of knowledge)
Casstell Key interlocking has been used for over 100 years to avoid what happened to ensure that the critical 220VDC battery was always available to stop the turbine motoring.
Beleaguered state-owned CS Energy spent $38m of taxpayers’ money hiring lawyers over the past two years, including to fight the release of the landmark report into the company’s woes. The investigative report into the 2021 Callide C explosion compiled by forensic engineer Dr Sean Brady, cost taxpayers a further $10.4m.
The 1,500 reactors’ll explode in case of an unprevented long blackout/war/chaos https://GlobalBlackoutPrevention.wordpress.com
CD-ROM for the win!
I think we are unlikely to see many, if any, new utility scale solar farms announced that don’t have a matching battery of roughly 4 hours at the rated output of the solar system. Even with all the storage announced there is still going to be an excess of solar production on the weekends at this time of the year. What is more important is to look at the levels of curtailment on week days and how that compares to the capacity of the existing and announced storage.
So what does that do to LCOE (seeing that is the metric everyone loves to use) given it is the same amount of energy produced now servicing a lot more capital? From an owner’s perspective it is probably cheaper to simply forgo the revenue and work on a slightly lower capacity factor when making the FID. If the marvelous capacity investment scheme is underwriting things…..who cares? Certainly not the investor…..
An acid test of the CIS will be whether it underwrites any utility PV or focuses on wind/batteries/PHES for now. If the former, we’ll know it is flawed. There’s no need for more utility PV until a few more coal plants have exited and even then it’s likely rooftop PV will have already filled the gap since it’s competing against retail prices not wholesale.
Thanks Kieran
In the recently released GenInsights Quarterly Update for 2024 Q2 we pose a question about whether the Capacity Investment Scheme should only support VRE additions if they are fully integrated hybrid units.
Unfortunately, that’s not my understanding of the way that it’s structured at present…
A most interesting generating frenzy. Paul as Tasmania’s wedge tailed Eagles are an endangered species how about a regular story on their deaths at our glorious windfarms.
You should check on the 350,000 MWh storage of Snowy2 on page 47 of ESOO 2024. It seems the whole water resource used to turn this storage into 2,200 MW over 160 hrs is not fully recyclable and Tumut3 will be impacted
Firming capacity is classified by AEMO as shallow (less than 4 hours energy storage), medium (4 to 12 hours) and deep (over 12 hours). The Capacity Investment Scheme is primarily enabling shallow energy storage.
The government owned deep storage projects of Snowy 2.0 Pumped Hydro (2,200 MW) and Borumba Pumped Hydro (2,000MW) are unlikely to make their indicated full operational dates of December 2028 and September 2031. A Final Investment Decision is yet to be made on Borumba PH yet AEMO has classified it as an anticipated project for the ISP.
Its not just IF we have enough firming capacity its also IF there is sufficient medium and deep storage and what IF medium and deep storage is not operational by 2030 when the Capacity Investment Scheme has enabled another 23,000 MW of Variable Renewable Energy.
A question:
Last week the Victorian Government published its Cheaper, Cleaner, Renewable: Our Plan for Victoria’s Electricity Future which included “nation-leading storage targets of at least 2.6 gigawatts (GW) by 2030 and 6.3GW by 2035. Curious to know what are the medium and deep energy storage projects that will be operational in Victoria by 2028 ie by the time the 1,450 MW Yallourn power station closes?
The NSW government has only underwritten 6TWh pa of Eraring (if Origin opts in) which is around 40% of last year’s generation. What does that mean?
Why does the ESOO state that “These systems [rooftop solar] do not tend to contribute to lowering the scale of peak demands during extreme hot conditions in the summer..”?
Geoff Eldridge wrote on here in February about the new NEM peak of 38.6GW on 22/02/2024, of which ~8GW was coming from rooftop solar. Finding this fact hard to square with the ESOO statement.
I think you’ll find that both are talking about different measures of demand.
1) Geoff is talking about ‘Underlying Demand’ or ‘Native Demand’ (not quite the same as each other, but similar)
2) The AEMO most cares about ‘Grid Demand’. Because rooftop PV has already shifted the time of peak Grid Demand to evening periods after sunset (in summer and winter), adding more rooftop PV has no additional benefit of reducing peak Grid Demand further.
To understand more read this explainer:
https://wattclarity.com.au//articles/2018/04/an-explainer-about-electricity-demand-take-1/
“The Callide Unit C4 incident was the result of the simultaneous failure of key electrical equipment and system back-ups in a complex series of events that could not have been anticipated, with some of the contributing factors being traced back to the original design of the power station,” Mr Busine said.
What is the total cost? $50M on coverup, maybe $500M on rebuild because the Insurance has refused a payment all because CS Energy allowed maintenance staff to switch off the critical 220V battery with the C4 Turbine in service.
The C4 operator in the central control room did not know that the 220V battery had been isolated.
It has been an incredibly rough five days in Tas. Never seen or experienced anything like it. The greater Launceston / central north has been decimated. Saturday night saw at least one transformer explode in Launceston. Sunday night has been the worst by a long shot. We live beside a main feeder in the country which managed to ride through many dips in voltage as the Hadspen substation was managing other feeders from the distribution connection point that were going to ground. In the end, despite an apparent re-close attempt, the main feeder that passes us finally failed. Just to finish things off, 45 mins later, lightning struck the dead feeder just near us, generating a current rush to ground via the surge protection on the transformer causing the HV fuses to drop. Current then found its way into our house via the in ground phone line (yes, land lines are still a thing!), then back to ground via our network and local earth stake! We are the sole customer on this trannie and don’t expect power to be restored for a week. As of 0230 this morning, TasNetworks call taker said they now have 42 000 customers powerless. Would be interesting to see how much our frequency fluctuated down here given all the chaos that has ensued tonight.
Thanks for this Linton, fascinating!
You’re probably already aware, but Wattclarity had an article on the SAM back in 2017 when we had possibly the worst wind drought of the last decade. https://wattclarity.com.au//articles/2017/07/the-southern-annular-mode-sam-contributes-to-recent-wind-drought/
Thanks David! Another great resource to be reminded of – a positive SAM in that case. The state of the SAM is definitely one to monitor.
While these are one off events, the trend is definitely there. There are unlikely to be anymore large utility scale solar farms that don’t have a very large battery with them. This of course is due to the large amounts of “excess” solar in the grid, the lower cost of batteries and the change in regulations that now allows hybrid systems.
The smart money is, or should be, building MEM-attached [battery] storage that store power when prices are negative, and kick in during 4 to 9pm when we all pay the time-of-day premium tariff.
I wonder if anyone is doing this?
The time to build solar farms was back in the 2010s or earlier.
According to Prof. Simshauser’s modelling, rooftop PV is basically forcing the shutdown of coal plants, otherwise the NEM will become unstable, (or evening tariffs are going to get very, very high to compensate!)
I really think there needs to be a weighted standardised measure of temperatures that better reflect their impact on electricity demand. I know all too often I look back at the average temperatures for a month and find they don’t really reflect how hot or cold some of the periods were during the month.
For example a single hot day doesn’t place create as much demand for cooling, compared to five days in a row of the same or even slightly lower temperature.
Typo in the date above (it’s the 5th not the 9th)
Likely something to do with that low frequency noted in your earlier posts Paul.
A strange outage is due to Scada and Tasmania. Presumable direction?
Thanks Linton for this article. Importantly, the reduction in gate closure has allowed improved forecasts to be used by NEMDE. AEMO’s own figures quote an 11-14% improvement in MAE to their ASEFS/AWEFS forecasts and I know a significant part of this improvement is due to being able to use later site measurements as inputs to the forecast.
At ETSI we see an even greater improvement when moving from a 60-second to 10-second gate closure. The result of these accuracy improvements is downward pressure on regulation FCAS utilisation, evidenced by AEMO not having increased regulation procurement volumes despite the large increase in VRE capacity (admittedly PFR has helped frequency control around the PFR Control Band).
What happened on AEMO’s page under Renewable penetration for Sept 5th Thursday ? Somehow black and brown coal wasn’t recorded at all and so the % was over 90 % which is false.
“The Callide Unit C4 incident was the result of the simultaneous failure of key electrical equipment and system back-ups in a complex series of events that could not have been anticipated, with some of the contributing factors being traced back to the original design of the power station,” Mr Busine said.
What is the total cost $50M on coverup, maybe $500M on rebuild because the Insurance has refused a payment all because CS Energy allowed maintenance staff to switch off the critical 220V battery with the C4 Turbine in service.
The C4 operator in the central control room did not know that the 220 battery had been isolated.
Readers need to remember that ‘Installed Capacity’ can be a significant overstatement of aggregate windfarm capacity physically able to generate, because for each new windfarm connected, its nameplate Maximum Capacity arrives in AEMO’s systems in one hit whereas actual physical capacity can take up to 12 months to fully commission.
So ‘Installed Capacity’ can be well ahead of actual capacity on the ground, especially when larger new windfarms are being connected and commissioned.
As a consequence, Global-Roam’s calculated fleet “capacity factor” can be materially understated because it is comparing apples and pears – generated output from actually commissioned capacity versus ‘Installed Capacity’ not all of which is actually commissioned and able to generate.
The impact on fleet capacity factor, compared with using an apples and apples approach (eg by excluding newer partially-commissioned farms from the calculation) can be of the order of 2-3 percentage points – eg 30% vs 33%, which is fairly material.
So three years to procure a cooling tower rebuild, new; turbine, alternator, transformer, switchgear and associated electrical auxiliaries.
So I would call this 30% of a new unit, but like only 5~15% timeline of the full brownfield power station extension eg. the original building of C3C4, with the whole development approvals, environmental approvals etc…
So how does this factor in the “little bit optimistic” view that building a mature technology nuclear PS on a existing Brownfield site as “eh, I say we can do in a short 10 yrs eh??”.
Can we have a fact based deep dive into the topic of timeline of major infrastructure planning in the NEM please?
“The Callide Unit C4 incident was the result of the simultaneous failure of key electrical equipment and system back-ups in a complex series of events that could not have been anticipated, with some of the contributing factors being traced back to the original design of the power station,” Mr Busine said.
This is classic technical writer cover-up “gobbledygook”!
$50M is the cost of the legal advice and inquiry which was meant to be secret by claiming legal privilege after maintenance staff isolated the critical 220V battery supply with the C4 Turbine in service so the insurance was voided!
The C4 unit operator “who most likely now has PDSD” in the principal control room did not know that the 220V battery had been isolated, so the C4 Alternator motored to complete destruction!
If you take a look at the participant UIGF, it would appear that they are using their actual output rather than a forecast of their actual capability given the current/forecast irradiance and cloud cover. More importantly, they appear to be shading this/biasing it lower (possibly only around gate closure?), presumably to optimise their causer-pays factor – critically at the end of an interval where they are subject to the 85MW limit/have a target for that level and a semi-dispatch cap flag, their self-forecast will be somewhat below 85MW, the following interval they will receive a target less than the constraint limit, they are no longer subject to a semi-dispatch cap, and with the flexible nature of targets for semi-scheduled plant they can operate up to their fuel availability… This particular situation highlights that perhaps a more robust approach is required to impose a maximum output on semi-scheduled plant. Arguably biasing one’s self-forecast on the low side to enable provision of “raise” PFR (or just raise in general) is not necessarily bad for the system, and it wasn’t done with the intention of gaming the constraint – but I look pretty dimly on a self-*forecast* of *unconstrained* fuel availability that appears to be based largely on actual output and not actual fuel supply ie irradiance and cloud cover in this case…
Great update. My observation is that the green band reduction must be constrained by the number of baseload units available. As more coal units are retired, less volume can be bid $0-300.
This exposes the truth in the idea that ‘cheap’ renewables displace the most expensive generators in the bid stack ie merit order effect.
What’s actually happening is the mid-merit baseload (coal) is being forced out.
The evolution of this trend means the end of baseload generation and much higher system costs.
Clearly more government intervention is required – either to force more floor price generation into the market, or force more baseload.
I’m voting for baseload.
CS Energy Ltd’s annual report 2024 is not available today Monday 23rd September 2024, it might have a total cost of the Callide C4 outage caused by the maintenance staff’s isolation of the critical 220VDC supply with the C4 unit in service?
What is the total cost of wrecking the Callide C4 steam turbine in 2021, cause? maintenance staff isolated the critical 220VDC supply. maybe the annual report 2024 for CS Energy Ltd.will supply some information.
What a thought-provoking article! 🌟 It really highlights the struggles many of us face with “smart meters” and how misleading the data can be. It’s frustrating to see our energy-saving efforts not reflected in our bills! 😩 I love how you pointed out the importance of having accurate data, especially for those of us with rooftop solar panels. 🌞 It’s essential for making informed decisions and maximizing our sustainability efforts. Let’s hope energy companies take note and start improving their systems! ⚡️ Keep spreading awareness! 🙌
It would be very interesting to know what the CO2 emissions reductions were throughout this process.
Here’s some higher resolution frequency charts – https://www.dropbox.com/scl/fo/m3kpuizmi8a95hqxp2v2v/ALs_te-T8vXXSyVH8lKqrIE?rlkey=2e168f7ybwbv7lasncsvoxopl&dl=0 – free to use how you like. I’m more curious about what happened at around 07:29:50 where there’s sudden rise in frequency.
2nd Oct 2024. What is the total vast cost of switching off the critical 220VDC battery with unit 4C in service? The C4 Operators in control of unit 4C had not been informed!
This is how you wreck a base load power station which is at the end of its designed life.
This is worth a closer look at the rate of roll off of both load and frequency and then a look at satellite imagery for cloud cover. One that people didn’t see coming I suspect. Has already happened a couple of times her in WA where unforecast cloud banks rolled in from the ocean and cruelled rooftop solar production. Chewed up all system spinning reserve and then some.
Looks like there might be a minimum load condition at about 13:30 AEDT today Wednesday 9 October? Prices on the AEMO NEM dashboard are showing as being ‘under revision’.
Great summary Dan, as always, thanks for posting.
CS Energy’s Ltd Callide C4 Management allowed maintenance staff to isolate the critical 220V DC Battery supply with the C4 unit in Service without informing the unit C4 operator who was in control. He was left with his control panel switched off.
Process Safety proceedure had been removed by CS Energy’s management’s lack of knowledge or understanding
A critical insurance payment has been lost as it is impossible to insure against a lack of knowledge or understanding and behaviour that shows a lack of good sense of judgement.
Technical Writer gobbledygook has been used to try and hide what happened
CS Energy’s Ltd Callide C4 Management allowed maintenance staff to isolate the critical 220V DC Battery supply with the C4 unit in Service without informing the unit C4 operator who was in control. He was left with his control panel switched off.
Process Safety proceedure had been removed by CS Energy’s management’s lack of knowledge or understanding
A critical insurance payment has been lost as it is impossible to insure against a lack of knowledge or understanding and behaviour that shows a lack of good sense of judgement.
Technical Writer gobbledygook has been used to try and hide what happened
CS Energy’s Ltd Callide C4 Management allowed maintenance staff to isolate the critical 220V DC Battery supply with the C4 unit in Service without informing the unit C4 operator who was in control. He was left with his control panel switched off.
Process Safety proceedure had been removed by CS Energy’s management’s lack of knowledge or understanding
A critical insurance payment has been lost as it is impossible to insure against a lack of knowledge or understanding and behaviour that shows a lack of good sense of judgement.
Technical Writer gobbledygook has been used to try and hide what happened
Its not just the stock of towers. Next year is the midway point of the Solar Cycle which is traditionally the most active ie potential for transformer damage. This would appear to be an issue for Tassie and maybe Victoria. As an energy consumer advocate I’d be interested to know the status of transformer stocks and if engineers have plans to increase stocks when mainland states become dependent on Marinus Link/Battery of the Nation to keep the lights on. Maybe the supply chain lessons of COVID have already been forgotten.
Boo hoo and? We still paying highest power prices in the country..what a joke
Very interesting to me that the wind farm (199 MW) and solar farm (54MW) cannot synchronise with the diesel generator.
Yes, it seems unfortunate that they can’t at least use the windfarm and BESS alongside the diesel generator, which would seem helpful in the evening peak.
The locals keep to SA time.
You can see why.
Any angular momentum would be greatly appreciated!
Clearly a voltage control issue. Run out of fault level. Would love to see what the transient stability model said! Damn physics again!
* Feb 2024, 120 km/h winds near Anakie, VIC, 6 towers, 530,000 homes, thunderstorm
* October 2024, Menindee, south of Broken Hill, to Tibooburra near the Queensland border, 7 towers, extensive blackouts
Q1 Not sure inertia is particularly related to minimum demand – coal units won’t decommit
Q2 Around 7GW NEM operational demand, see 2024 ESOO Fig 44
Appreciate the detailed analysis
With respect to comment about Prominent Hill – both Prominent Hill and Carrapateena Mines are supplied from a separate 275 kV transmission line ex Davenport, and not the same 275 kV line that normally supplies the load to Olympic Dam (also ex Davenport) – ie there are 2 275 kV lines / 2 completely separate systems, so I’d assume both Carrapateena and Prominent Hill mines are still connected and operating as normal.
A good summary of battery opportunities in the NEM but I’m curious as to why you think there is an immature contract market. OTC trades have existed since before the NEM started in 1998 and the ASX started trading futures in 2001. I can appreciate that getting contracts of sufficient tenor is a challenge as most customers don’t contract past 3 years whilst project owners seek contracts as long as 20 years. Surely, this is a mismatch of the price and volume risk that a customer faces as compared to that of the project owner more than any fundamental problem with contract markets per se?
Some reports say the backup generators are “diesel”, others say they are “gas turbines”, does anyone know?
Might be helpful if people read the “Lessons Learned” document prior to pounding keyboards. Appears the project was all about studying the stabilising potential of a grid forming BESS in a fringe of grid radial system with relatively high renewable energy (inverter connected) inputs – hence ARENAs involvement as a research project. It was not about keeping the lights on. This is entirely consistent with what would otherwise be a somewhat odd combination of a 50MW inverter and 50MWhr battery (ie one hour of storage at rating). Indeed, I would be surprised if without major re-engineering of protection systems in the town it would be possible to run on the grid forming battery and renewable energy sources because of a very low faut level. Once a synchronous generator (or generators) were running it is possible provided the dynamic response of the power electronics does not end up pushing the generator off. Hence why things are being taken very gently. Sounds like sensible and prudent engineering practice to me. Not a great situation for anybody, but it appears the battery was never designed to keep the lights on, despite what the media people might have written. The best it might do is accept the risk of fault level issues whilst the GTs ran up and synchronised following a line trip – say 10 minutes.
The major transmission lines *never* have the towers replaced throughout the their life because it’s so hard to get the lines out. The job to replace towers and restoring lines overnight (as required by AEMO) has to be done with temp towers. It is an enormous undertaking.
The second part of this statement is a carbon-fibre, kevlar reinforced bow to draw. If only it were that simple.
“This equipment will ……….. allow renewable energy to be transferred between states which will help provide cost of living relief by delivering access to cheaper clean energy.”
Have a look at the solar forecast last Monday night 😉
Whoops, delayed again 🙄
Dont forget the initial event can often create harmonics on the system in the region that the frequency analyser is sampling at. Can create initial false high/low values.
Brilliant article, thank you Warwick.
Thanks for this Warwick, I wonder what happens to to liquidity in the futures/swaps market as coal exits, surely it is difficult for renewable energy owners to contract with retailers because they in most cases already have PPAs, and besides cannot be sure enough of volume and price realised to take that side of the trade.
Yes, I live on the Eastern most edge of Adelaide, and that last time slice would be when the sun hit my panels, after the rain departed.
The by-now-familiar constraints on transmission from southern NSW towards the load centres north seem to be playing their part in this volatility as well.
A few different constraint equations evident yesterday, such as these:
https://wattclarity.com.au//articles/2024/11/07nov-3dis-manyconstraints/
… you are referring to the ‘N-MNYS_5_WG_CLOSE’ constraint set.
Does a slight drop in frequency affect synchronise clocks and is the effect cumulative?
Terrific explanation of the events, and a nice outline the linear program behind NEMDE by zooming in on “simple” constraints. Perhaps not the points you intend to make, but it does serve as a reminder for
1) how it is becoming more common (or is already the baseline case) for it to be inaccurate to analyse electricity markets with whole-of-market bid-stack / merit-order framework. Assuming away binding constraints is assuming away reality.
2) that the system-wide impact and consequently economic value of injections at different locations (and by different technologies) at the same time within a NEM region can significantly differ. (I’ll spare everyone a rehash of the outcomes and incentives that follow from zonal market designs — it was probably all best understood and explained 20+ years ago! https://scholar.harvard.edu/whogan/files/nezn0227.pdf )
Hi Paul, I am a retired AEMO System Operations Control Room Operator. The statement that it was a “Small Frequency Wobble” is grossly inaccurate. The system Demand increases by approximately 2400-2600MW in the 30mins to 1400hrs. The Frequency in the previous 30mins to 1330hrs was struggling to get to 50hz. I have not looked at what Turbo-Generator plant was dispatched but it seems to be insufficient for the rate of change in demand. In other words the system does not have enough inertia. If the largest generator was to trip when the frequency was sc low, the remaining plant with active governors probably would not be able to pick up the loss of generation because it could not even cope with the increase in demand. I suggest that the system was “Insecure.”
Thanks Ian,
Next step in analysis here:
https://wattclarity.com.au//articles/2024/11/07nov-mainlandfrequency-part2/
I think you and Ian are both correct. Extended cloud band playing havock with the system load, and a lack of system generation inertia. And for my 2 bobs, the almost constant mid daily state to state tie line constraints to near 0 MW transfers cant be helping either.
So are you saying that if the constraint equation wasn’t in place and Victoria which at the time had negative prices and was only running its coal plants at 35% capacity was exporting to NSW instead of Tasmania, prices in NSW and Queensland could have fallen 80-90%
Can you explain why the Vic market occasionally has negative price spikes? For example at 2:10pm today the price was -$396/MWh.
Re the openining quotation: It may have been Pierre de Fermat on the wonderful proof of his “last theorem” that he’d found but didn’t have space to write in the margin of his notes – ultimately the proof took Andrew Wiles hundreds of pages of extraordinarly abstract concepts that very few humans are capable of understanding. Or perhaps Eugene Galois who scribbled out everything he’d discovered about group theory the night before he died in a duel.
Guardian reporting a boiler tube leak unit 3, and trip triggered when isolating that unit. Perhaps a mistake in a switching sequence?
Phase shifting transformers, allowing a ‘dial a power transfer’ control (assuming Heywood and VNI are functioning). I guess daytime negative prices and solar curtailment in SA are about to be far less common.
Do deviations from 50 hertz down affect mains powered clocks and is the effect cumulative?
did something similar happen QLD?
Thanks KC … might help:
https://wattclarity.com.au//articles/2024/11/13nov-more-about-nsw-demand-drop/
Strange
The NEM generation graph did not show any corresponding drop
https://explore.openelectricity.org.au/energy/nsw1/?range=1d&interval=30m&view=discrete-time&group=Detailed
How quickly did the load return? Pretty normal for big hits on the transmission system (voltage depression) to “shake off” load (things like voltage held contactors etc releasing and require a restart). These days can also shake off generation in the form of small rooftop PV inverters that fall over on low voltage – then restart after a few minutes. This will get worse as system strength (particularly fault level) declines.
They won’t be doing anything they don’t have to in the run down to decommissioning. These outages will be driven by statutory inspection stuff for pressure vessels and piping including testing of protection systems and the like. Can probably get away with an extension of time for the last unit if nothing untoward found at the last inspection if the thing is being decommissioned. Probably planning it nicely to decommission plant all due for maintenance as any good asset manager will do. Just don’t change your mind and ask them to run longer at the last minute (even if the lights are going to go out)!
Those network constraints also limit the contribution of some NSW generation as well – Uranquinty gas generation, wind in the south-west, Snowy scheme hydro can all be affected depending on how the constraints bind.
Yep, all trying to squeeze into a smaller sized pipe to Sydney. Hence the comment at the bottom of the article about potential further reductions on already reduced VRE
Yep and with more power trying to flow around the Bannaby – Mt Piper – Wollar – Bayswater 500kV route to get into Sydney from the north it’s possible the Bayswater-Liddell bottleneck could limit contributions from an even wider set of NSW generators. I don’t see how AEMO would let the Avon – Macarthur line 17 outage proceed if it was having any material effect on reliability.
Well, since that earlier comment the line 17 outage does appear to have been cancelled!
Between the relevant STPASA runs a number of NSW transmission outages affecting lines in both the north and south of the state were removed or deferred from AEMO’s schedule. Looks like that played a significant part in reducing the amount of projected load shedding.
Hi Paul.
I’ve had this tagged to read for a couple of weeks and have finally got around to it – very sensible commentary from Origin/Frank (as we’ve come to expect). As Origin have commercial benefit from both traditional and future generation tech they can speak truth to power without it seeming that they’re just talking their book.
A question on your Section B – did anything ever turn up in the media outside AFR and the Aus? How is this call to arms or wake up call not being picked up by a broader network? He’s the CEO of an ASX20 company speaking at a premier non-political think-tank event and still seems like it’s hard to get cut-through, doesn’t do much for one’s confidence that we can help the discourse mature!
Nice one as always. Thanks.
Dave
Hi Dan – Thursday is looking really tight, with a ~16h LOR2. Curiously, about 5h of that LOR2 is sitting at exactly 0 MW reserve under the latest STPASA. Do you or the WC team have an explanation for what’s providing that balancing – it looks almost to be a market loadshedding mechanism that acts before forced loadshedding?
G’day Tom (and Peter who has also commented below). We’ve fielded the same question from someone else – and a member of our team also observed that behaviour.
The STPASA process description says: “The linear programming formulation attempts to optimise reserve sharing between regions along with other constraints which are described later in this document to maximise the amount of reserve that is available to the system as a whole.”
We are not 100% sure, but another member of our team suggests that it could be that the model judges that there is benefit to leaving one region fairly ‘thin’ on reserves if the excess can be of benefit to a neighbouring region.
Excellent question Tom!
And now followed by a notice 121083 to the Waratah BESS (which is still commissioning): 96 MWh to be held in reserve. I was wondering what we’d see today from the commissioning BESS, and it seems that Waratah at least will be in action (to be joined by the Capital BESS?).
You reap what you sow… Good luck. I’m insulated from this crap
My wife works for Transport for NSW. Office staff have been told to adjust aircon temps up to 24-26deg and are not allowed any plugged in electronic devices at their desks etc.
So I’m at a commonwealth agency
It’s known for us as the “Government Energy Action Response” and in their words “is designed to manage and mitigate energy demand during critical periods”. I do know the idea was borrowed off the NSW government.
Participating agencies sign up and are notified by DCCEEW when to activate the plan.
For us, it involves raising the air con temp, turning off unused office equipment and all the external facing blinds are closed
`The best response would be by the NSW Planning Department to stop approving apartment and office towers and by the Immigration Department to reduce the number of immigrants who fill these towers. 2 towers with 750 dwelling units generate a peak load of 5.2 MW
It would be interesting to see an overlay
of proposed you transmission developments to see what extent these pinch boys in the network would have changed the outcome. What role did wind help yesterday
Hi Dan AEMO has stated it is going to look deeper into the change of use in gas in the future for the 2026 plan. The UQ Gas and Energy transition research centre is doing some great work on this matter. I attended their annual presentations on research work yesterday. I am confident the change can be managed however it comes with a cost for gas storage and low utilisation of gas production. There is option feasibility and optimisation studies required. I think this area is the remaining missing issue. Through in biomethane/biofuels.
Thanks Ian – I was also at that event last Thursday afternoon so sorry I missed you.
Agree with your thoughts. If we go down that path, I think there is still some work to do to communicate to the wider public that the cameo/insurance role of gas is worth having and paying for.
What do the 3 [Zone:…] terms represent in the Predispatch-RHS? Is a zone an aggregated value of generators in a region or is a zone expanded to a series of individual generator/load terms? All the generators/loads named in the Dispatch-RHS are in the locality of Yass/Canberra/Tumut or in SW NSW.
Yes you may very well be on to something there! There are a number of other constraints with similar constructions. Although, in this case, the magnitudes still seem a bit different.
The difference between dispatch and predispatch here was only about 200MW so in retrospect I think there’s more in play here than just the one constraint. Negative residue management playing a part too.
Nice work.
Thanks Peter!
The ‘priced out’ description works well.
I understand why inverters will be curtailed, but GFG? Curtailed implies it is priced in but can’t run due to network constraints.
Thanks Ben! And yes, gas fired gen was also curtailed. We see it more often for the solar and wind because there is generally more transmission network between them and the load center, so more flow pathways to be managed with network constraints. But all gens are connected to the network at some point so there’s potential for certain network conditions or settings to impact all types of gens, at times.
Nice work Paul. It’s also notable that despite the work done to provide guidance on how participants should classify outages in MTPASA, some of them routinely seem to ignore that guidance.
Without naming individual generators, there are cases we’ve seen of various units extending planned outages (sometimes for significant numbers of days and/or multiple times) but continue to label the additional outage period as a ‘Basic Planned Outage’, when the correct classification would obviously be ‘Extended Planned Outage’.
Other NEM participants and spectators could perhaps infer these generators are:
– unaware of AEMO’s guidance
– aware but unable to interpret or implement it,
– aware but choosing to ignore it, or
– trying to avoid scrutiny of maintenance slippages.
It is intriguing that a report of this nature is not given the same status and visibility as other similar reports.
Minor correction: Aidan is with the CIS – centre for independent studies – not the IPA.
Thank you for drawing attention to this issue. I think it is a matter of great concern, not just now, but into the future. There are many very large VRE projects under construction and announced for Southern NSW, plus EnergyConnect plus a likely VNI-West. While HumeLink has been announced, the capacity of it just doesn’t seem any thing like sufficient to meet the needs of the future, let alone what additional announcements of VRE are likely for Southern NSW.
I am surprised that we have not had any announcements for a tender for one or more batteries to provide virtual transmission capacity on the existing transmission and plans for virtual transmission for HumeLink.
check out the NSW demand against the VIC demand, might be an error
Great insights Linton!
Always important to look ahead.
This might be improved by including EnergyConnect and observing if NSW would have been helped or hindered by SA demand / generation profile at that time with Eraring gone?
EnergyConnect won’t assist with the transmission bottlenecks between southern NSW (Wagga / Yass / Canberra) and greater Sydney. Humelink will ease those limitations but construction of that project hasn’t commenced yet.
It looks like there might be some issues with VNI flows since the Initial MWs were following exactly the same as Target MWs from 13:35 to 14:40.
Perhaps, AEMO has lost visibility of actual VNI flow reading, and NEMDE only could calculate combined demand of NSW and VIC but not individually.
With Initial being the same as Target, NEMDE saw/used the flow into VIC increasing = NSW demand decreasing and VIC demand increasing. The correct Initial value seemed to have come back at 14:45 with the NSW and VIC demand corrected.
Great article Linton,
I suspect batteries might prove the key to keeping the lights on in 2027. About 3GW/10GWh committed or anticipated for NSW, primarly 500MW/2GWh Liddell, 850GW/1.7GWh Waratah, 415MW/1.7GWh Orana & 275MW/2.2GWh Richmond Valley.
Are batteries, including ones like recently added Waratah super battery, able to assist with power supply during unplanned outages?
The class action covers a period before Callide C4 was wrecked by CS Energy Ltd “management failure” that turned off the critical 220V battery with the unit in service and the alternator motored to complete destruction.
The damage insurance of the alternator was voided by ignorance and the RTS insurance is self insurance.
Total cost maybe $500,000,000 The coverup continues!
What is the rationale behind the highest bid, which then positively affects all other bids? I have never seen anything like this in business.
All the above assumes an engaged and sophisticated energy consumer. This is unrealistic given the generally poor levels of energy literacy of most residential and small business consumers. The consultation overview states it is seeking insights from “electricity consumers (industrial, large and small-scale)” – not sure what their definition of “small-scale” is….. if they truly wanted insights from the customer classes that consume the bulk of grid supplied electricity why didn’t they just write “residential and small business” electricity consumers? Just another consultation to add to the myriad that electricity consumers and their advocates are already expected to participate in for free. No wonder Australia’s electricity system has failed electricity consumers…. because it has poor representation from electricity consumers at both a state and national level. So tired of the worn out phrase of we put consumers first. Nothing could be further from the truth.
Just on constraints in that forecast, there is a second relevant constraint set due to a current outage on Liddell-Tomago line 82 (set N-LDTM_82). This isn’t due to finish until 16:00 on Monday so overlaps with the Avon-Macarthur line and Kemps Creek transformer outage set (N-X_AVMA_KCTX) highlighted in the article. The joint effect of these outages would be significant limitations on flows into Sydney from both the north (QNI) and south (NSW southern generators and Victorian imports). There is no way known that AEMO will permit the Avon-Macarthur / Kemps Creek outages to begin on Monday. So the forecast currently looks worse than it will really end up being (unless there are generator failures between now and Monday).
Thanks Dan.
It might be interesting to plot these projects on a map also showing known areas of congestion
Looking at that chart, there’s no denying that the peaks and general grid utilisation these days is slightly lower, or possibly the same, as when compared to the grid back in its heyday, <=2010.
And that's with an extra 4 million people in Australia since 2010, presumably all using electricity.
Not sure you can infer much from the chart in terms of utilisation of the grid other than peak utilisation. It’s trending monthly peak demand.
One would need to look at average demand to deduce much in terms of utilisation.
When I see terms such as “extreme weather” and “catastrophic fire danger” used to describe a day which perhaps MAY be the hottest in just 5 years, the only thing that gets hotter than the projected 45 degrees, is my collar out of a sense of frustration over the increasing trend to invoke such hyperbole when discussing weather.
@Freddy: Australia’s de-industrialisation has probably outpaced the increase in consumer electricity demand.
Just to clarify,
a/ large scale wind and solar being below forecast doesn’t directly affect market demand – they are part of the fleet that supplies that market demand (except for non-scheduled plant which is a small fraction of wind/solar).
b/ but large scale solar availability being below forecast could well correlate with small scale rooftop PV also being below forecast – this does directly affect market demand, as explained in Paul’s post.
As an (irrelevant to the topic) side bar, the pronounced saw tooth profile of the AEMO small scale solar is interesting. Any idea what gives rise to this? Given these are <100KW and the sun rises continuously, one might expect a smooth profile.
It’s just an artefact. The solar data is a half-hourly estimate. When superimposed on generation data with 5 minute granularity, the sawtooth pattern results.
One more key takeaway Paul – if we add rooftop PV estimates to operational demand to get underlying demand (ie a measure of total power consumption from all sources), then yesterday’s half hourly values peaked at about 10,600 MW, higher than market demand in Jan 2009 when there was no rooftop PV to speak of, and almost certainly higher than peak underlying demand in Jan 2014 when there might have been a hundred-odd MW of rooftop PV output, if that, at the time of peak market demand.
Since yesterday was not the culmination of a multi-day heatwave here in Victoria, this does point to a material rise in underlying peak consumption – not surprising given population growth etc.
Good pickup, thanks Allan!
Dumb question, why isn’t the Waubra wind farm (WAUBRAWF) included in the analysis?
Not a dumb question at all … Waubra is Non-Scheduled and, as such:
1) Does not bid;
2) So is not shown in that filtered list of units in ez2view.
We could look at a trend of output for the day, but would not be able to see more than that.
Very helpful and well written summary thank you James.
Suggest this data by time of day (instead of month) might be interesting too.
Can someone point out which column in ‘Figure 4.9’ represents ‘Liquid Fuel’ as appears in the AEMO dashboard for South Aust? FWIW, another name for ‘Liquid Fuel’ would be?
Great graphs Linton. I wonder if you could shade on a NSW map the areas containing the constrained generators at the time of peak congestion?
The links in “network curtailment or economic curtailment” go to 404.. 🙂
At the time of writing, Vales Point is still operating – does this mean they have secured financing? Has some flexibility been provided?
The methods and assumptions document published alongside the report (section 10) notes that “share of fuel type analysis included only coal, gas, battery storage, hydro, wind and solar. Other fuel types such as liquid and biomass were excluded.”
What is the current price trend of Queensland’s physical and financial contract prices?
MTGELWF is connected to the grid at a sub-transmission level, on the Powercor distribution network.
It has a permanent 100MW export limit, imposed by Powercor as a condition on the connection agreement. This was lifted from 66MW sometime in 2023.
Two additional important points on this:
– the constraint is invoked because of a pre-existing planned outage of Marulan-Yass 330kV line (Transgrid Line 5), scheduled to run over 11-19 Jan. This is the same outage that led to some very high prices in NSW and Qld in early November 2024.
– when the NRM constraint clamps NSW -> Vic flows to zero, it’s true that this forces down generation in southern NSW. However this doesn’t reduce net supply to meet demand in NSW, because any higher generation in southern NSW would simply flow south into Victoria – there is a hard restriction on flows from southern NSW towards Sydney because of the Line 5 outage.
A very informative article Oliver,
I really enjoyed reading all three.
No doubt I’ll read again.
However, I have one query
“Players will bid in their contracted level of output at SRMC.”
I can’t see any evidence of that in MMS Bid-data.
CFPSs tend to bid around $-900/MWh up to 80% of available capacity. which is $950 BELOW their SRMC.
My understanding is they do this to secure volume to dispatch to fill their (mostly base load and peak) contract swap positions.
During the day when the spot price is below their SRMC, they have a bit of fun where they shift volume into higher price bands to intentionally run “short” to boost day-time revenue.
(I also imagine they realise some benefit of consuming LRCs as well)
Then during peak/high wholesale prices they outbid the other thermal plants and run “long” with their 20% or so marginal capacity to boost peak revenue.
Certainly seems to be a big incentive for them to invest in ramping capacbilities.
It’s Win-Win happy days for them.
My view is that Contract positions along with Gentailer control of the few powerful scheduled assets seems to be the biggest driver of bidding behaviour.
The underlying cause looks like a sudden drop in northern NSW large solar output, particularly at New England Solar Farm – a “system strength” constraint out of nowhere dropped its output target from ~370 MW to zero at 1400. The N::N_MNYS_2 constraint also played a role because the only NSW generator that could ramp quickly enough to cover the drop – Tumut 3 – sits on the wrong side of that constraint. I half suspect a rogue input to the system strength constraint could be the real issue.
Some further digging shows the system strength constraint came on because of a forced outage / trip of a transformer at Muswellbrook in northern NSW that took some QNI-related lines out of service – that’s the fundamental villain of the piece.
Thanks Allan, have noted more in the article ‘About the ‘N_NEWENSF1+2_100-INV’ constraint equation that violates at 14:00 on Wednesday 15th January 2025 – Part 4’ here.
Hi Allan & Paul, thanks for the commentary. I’ve seen this happen before when TUMUT3 ramping upwards causes the NSW interconnector to flip 180 degrees (once was a couple days ago, once on 21/11/24). As touched on by you Allan, I think because TUMUT3 as a large positive term on the LHS of N::N_MNYS_2, N::N_2_5_2 (among others), when it ramps up, NEMDE seems to choose to flip the interconnector (force it’s target flow negative), as opposed to constraining other generators on the LHS? It seems unusual to me but intuitively, given the congestion around Marulan-Yass, energy from TUMUT3 has nowehere else to go but southwards. To make things even more confusing, the price setter for 14:05 shows TUMUT3 price setting with their -$1000 bid… but for an increase in -139MW. What is the price setter file implying here? that to meet a marginal increase in demand, the marginal cost arises from the cost of TUMUT3 turning off? Thanks guys
Update, Jan 2025: AEMO has decided to increment the maximum requirement volumes for the VF Raise (R1) service from 425 MW to 500 MW from 1:00pm (AEST) Monday 20 January 2025.
Unless you want to vacuum up LRCs to get your Green Stamp, I really cant see any value in Renewable PPAs
which for buyers have anti-hedging characteristics.
Looking forward to your analysis on the benefits Gentailers get, two-shifting their scheduled assets and the effect on the value of independent retailer contract positions.
Batteries anyone?
QLD was too slow to install batteries, even just to provide grid stability services, let alone bulk storage to help with evening peaks. Today QLD really needs more batteries than they have or have been announced.
It will also be interesting to see how much the lack of transmission through Wagga Wagga makes this situation worse.
Actually, at the time of peak demand*, rooftop PV contributed nearly 2,800 MW of output Paul!
* That’s underlying demand of course, which hit something like 12,335 MW. Due to the high penetration of rooftop PV it’s inevitable that it won’t contribute much at the time of peak market demand – because market demand will only peak when rooftop PV is nearly done for the day.
Hi Paul, in reporting the peak loading, you need to factor in the administrative load curtailment in effect (which has an associated economic loss) in order to obtain fair comparison of events. Over-loading the grid is one thing, having to shut down the economy is quite another cost!
I must be out of touch as I have never seen QLD demand (10,897MW) exceed NSW demand (~8,175MW) by 2,722MW.
It would be interestung to see a monthly peak demand comparison between these 2 NEM regions over the past few years.
Thanks for this. Many had been wondering if PEC1 had actually started operation, rather than just doing some testing. It would appear that PEC1 combined with the transmission restrictions around Wagga Wagga is going to make it very difficult, if not impossible, for SA to increase their percentage of RE in their grid, until PEC2 comes online. Even with PEC2, progress in SA’s percentage RE may not occur until transmission problems around Wagga Wagga are resolved.
So the TL;DR is that imports to SA will be higher at least sometimes when more expensive and SA large solar (marginal producers) will be curtailed more? A followup on this when there is more data will be interesting.
I thought PEC included phase-shifting transformers to allow the energy transferred on the link to be controlled? Is that only going to be actively used once PEC2 is commissioned?
Looks like a data error – I imagine the overfrequency was *prior* to the drop, as AEMO was dispatching generation well in excess of the actual demand.
Monday evening demand also looks a little suspect – just not as blatant!
I sort of agree but it’s not a data error. A 0.12 hz frequency swing on a 26GW system is a bit under 900 MW based on 4% droop and a bit of vagaries in hydraulic governor systems.
It is a data error – go plot the frequency vs demand. Over-frequency sitting at edge of PFR band from 16:35 when NSW demand starts going crazy, breaks through PFR band around 16:55… and then starts dropping back to normal over course of 17:15 -> 17:20. Ie AEMO had a data/forecasting error such that they were over-procuring generation earlier in the afternoon, and the -1000mw demand drop was taking it back in line with actual demand – over that dispatch interval the surplus generation situation reverts to normal, with frequency dropping back to ~nominal.
Thanks to both of you, I’ve invested some time on Monday to post ‘Last Friday’s large and (mysterious) apparent drop in NSW ‘Market Demand’ appears some data glitch?’.
Thoughts?
Pretty much spot on HR10 – between DI 15:55 and DI 17:15 something screwed up AEMO’s calculation of NSW actual demand from metered generator outputs and interconnector flow levels. You can verify this from the relevant MMS tables. Between 15:55 and 16:30 the calculation was too low, then up to 17:15 it was too high. The problem got fixed for the 17:20 DI which explains the sudden “demand drop” for that interval. These errors correlate reasonably well with frequency being on the low side of the PFR band between 16:05 and 16:30, then on the high side until 17:20.
Problems with the EBA negotiations
Thanks for the tip, Darryl … have posted ‘About the Industrial Action at Gladstone Power Station (potentially) taking all units offline … and LOR forecast for QLD, VIC and SA’ as a result.
Industrial action…
Thanks for the tip, Fred … have posted ‘About the Industrial Action at Gladstone Power Station (potentially) taking all units offline … and LOR forecast for QLD, VIC and SA’ as a result.
That’s the highest Victorian 5-minute TOTALDEMAND recorded for any non-working day – more than 300 MW above the previous high on 28 Jan 2018. So it’s consistent with the underlying picture of demand growth revealed by the Q4 record Victorian demand we saw on 16 Dec 2024.
Thanks for the tip, Allan
A roughly 4pm there was a facebook comment on meuqld facebook page saying that they had come to an agreement. I suspect up until the agreement they were running the units at min generation in goodwill?
I think we wait for the investigation. Vestas quote a withstand of around 180km/h for this model of turbine. Either there was a specific issue with this tower (the failure seems to be at the top lift joint of the tower) or it will be a very localised wind event beyond its design capability – not unusual in thunderstorms. The photos clearly show the proximity to other turbines that survived – though perhaps not without damage. We have seen the enormous power of swirling winds on transmission towers in recent history and this is likely no different.
Well said Andy!
looks like it was at a bolted joint in the tower. One photo I saw you could see the flange on the fallen tower if you zoomed in.
Thanks Roger … you must have better eyesight than me (and or saw better quality photos) to see a flange. What I saw was just what’s above in the article.
The top approximately 10 metres or do of the supporting tower appears in the ABC’s photos. It seems, as commented above, that the bolted joint failed. This structural failure makes me wonder whether it could be a common fault across a specific design, or perhaps a construction fault due to a single contractor’s work.
Presumably we will soon know the results of the investigation.
Unless lightning struck a blade whilst rotating & caused a massive imbalance causing a failure of either 1) the boss that the nacelle bolts to OR 2) a tower segment weld. Many questions – was it feathered & locked (braked) at the time or still operating ? The pic of the top of the tower whilst ‘clean’, still not detailed enough to see of 1 or 2 above. Nothing in the wreckage to indicate either (on the photos I’ve seen) – what looks like a nacelle bolt pattern I think is the root end of a blade. But, if the winds were only at the speeds mentioned, unless a blade did indeed have an utterly catastrophic failure, that’s a worry that the tower has dropped the nacelle. That 1 & 2 are back running seems to point to it not being a fleet tower issue and maybe a blade has shattered and smacked the tower. The root cause analysis will be a good read.
The coverup continues, CS Energy wrecked Callide C4 turbine by switching off the critical 220vdc supply battery with the unit in service, the unit operator was left in the dark and the turbine was destroyed. What happened to the interlocks that have used for over 50 years with critical 220vdc batteries
AER should be paying the $500,000,000. lost because they allowed this to happen by not having procedures in place to control such process procedures.
https://www.abc.net.au/news/2025-02-05/victoria-berrybank-wind-turbine-collapse/104895810. Is the tower, blade would have splintered at the crease like the others in picture. Colour is the right for the inside of a Vestas tower.
Guys, I realise you do not cover Western Australia to any extent however your assumption re rooftop PV are spot on. I suggest you start using the BOM satellite cloud maps and the answer will become obvious. We see enormous rates of change in operational demand on hot cloudless days as the sun sets (gets worse as we move past the summer equinox – remember we do not have daylight saving in WA) and the evening peak arrives. If you get a hot overcast (hence humid) day in Perth then the operational demand comes up in the middle of the day and keeps growing. These become big energy days for conventional generation and as we back out of coal, we end up in more gas than can be delivered so hence into liquids. Batteries are there but if there was no spare solar / wind around to charge them it simply implies more fossil primary energy. Train wreck arriving soon!
Willogoleche resumed output early afternoon on Friday 7th Feb.
Portland smelter has two potlines so when this happens you’ll see roughly half then no load (like you see in the charts). You can change power usage a little but there’s a bit of a window to operate in to be make aluminium and not just heat. The Dec 16 outage looks pretty long for an aluminium smelter, probably close to the max they would allow. Based on the load dropped then returned it was one potline for a bit, then the next pot line for a bit with some overlap – so roughly 1 hour per pot line.
Alcoa Portland Aluminium does do RERT – not sure if AEMO activated it then though?
Thanks Michaela, that looks right. Seems one of the potlines draws about 300 MW and the other a shade under 200 MW. The load profile on the 16th fits with the smaller one being tripped for one hour, then the second larger one going off roughly half an hour later, also for a one hour period. As you point out, they probably wouldn’t want a line off supply for more than an hour as they get more temperamental to restart the cooler they get. Allan
PS RERT definitely wasn’t called on any of these dates.
The first point to consider is that making comparisons with 2008 is fraught with difficulties. For starters, this article overlooks the massive reduction in Victoria’s energy intensive manufacturing base. Brand names like Holeproof, Dunlop, Hard Yakka, Fletcher Jones were all Victorian manufacturers in 2008. Ditto intensive energy users like Ford, Holden, and Toyota have subsequently packed up shop. Another big load that has disappeared is the Port Henry smelter at 360 MW. A additional chunky +40MW load is the Altona oil refinery that has closed since 2008. Further, there are all of the upstream suppliers, component makers, and contractors that have subsequently disappeared. The list of loss of manufacturing business goes on and on. Green activists look at the reduction in energy usage and jump with joy. I look at the reduction in energy usage caused by loss of manufacturing and see an economic calamity unfolding in slow motion.
The second point to consider is how the network is having to do double duty boomerang parenting for solar PV owners when a bit of cloud comes across on a hot summer’s day. Customers are quite happy to install a solar PV system, hold up the middle finger to networks by completely undermining revenue when the sun is shining. But then, like wayward children, expect to be able to return to the network in force on a hot summer’s afternoon, when clouds come across or the sun goes down, and demand there is network capacity spare for them. It is the equivalent of an _infrequent_ customer expecting Qantas keeps seat 9C free on every 737 in their fleet for them, on the off chance that they want to book a flight sometime in the next few months. In the real business world, this practice just doesn’t happen. Yet politicians and regulators are encouraging this behaviour. Having idle network for just a few hours every few years is inefficient CAPEX in the extreme. Perhaps it is time for a new rule that specifies a domestic battery of a certain size is installed with every solar PV system.
Making comparisons with 2008, 2023 or any other year since NEM start isn’t “fraught” at all. It’s just data. If you want to get into reasons for changes, let alone your own personal views on their merits, that’s your prerogative but it wasn’t the point of the post.
But on your last point, why wouldn’t we also make a rule to require purchasers of (say) air conditioners or electric vehicles buy a generator of a certain size?
Then I have missed the point. There was a “remarkable” conclusion made that population had grown by 30% but peak demand hadn’t. Yet, the analysis overlooked “the why”. Surely a big part of the “why” is the large base load and large industrial consumers that had closed in Victoria.
By contrast, the population in Queensland has also grown by about 30% over the same period. Yet its system peak demand has grown by about 4GW. Obviously, a big factor is the increase in base load due to LNG and coal mining, the upstream suppliers in machinery and manufacturing, and the population growth who assume comforts like AC are a mandatory in the home.
As for the point about air conditioning and electric vehicles, on households without solar PV systems, they are generally paying their fair share. Particularly AC units that have long daily run times. (AC that only come on for a few hours per year are a burden.)
From a grid perspective, EV on TOU or load control are one of the best customers have around as they suck up electrons in the middle of the night to improve network utilisation when it would be otherwise idle.
While the wholesale electricity market is an important part, 45% of the electricity bill comes from TUOS and DUOS expenses. Declining network utilisation (which is a function of short and spikey peak demand) drives additional network CAPEX that everyone must pay for.
Looks like a record today over 3700MW!!
I suspect you’re looking at another measure (perhaps ‘Operational Demand’), George?
But, regardless of the measure, as noted in ‘‘Market Demand’ soars in South Australia on Wednesday evening 12th February 2025’ a very high end result!
Did pelican point lose a GT? No LOR2 was forecast?
If the frequency is wandering it is clear indication that there is not enough governors active on the system. For turbogenerators to have effective governors the machine must be above minimum load and below maximum load. From my experience as a AEMO Load Dispatcher (retired), AEMO needa to apply some constraints to get better regulation of frequency.
My (uncalibrated) measurements has them dropping ever so slightly under for a very brief time.
https://www.dropbox.com/scl/fi/rmz78wijwdn4g0d7otgk7/Screenshot-2025-02-13-at-9.03.56-pm.png?rlkey=p8hbrcwwpf08ytxq47ogawxp8&dl=0
Roughly 10 seconds?
Governors respond in less than 6 sec and adjust steam flow to compensate for change in steam pressure (rise or fall) to keep turbine speed at 50Hz. Fuel flow is determined by Steam Pressure and Flow rate and is slower than the Governor response and should not change until the governor demands more average steam flow which takes more than 60 sec. So the problem is that System frequency is on a slow trend up/down which is causing the fuel flow to oscillate in response to the slow change in frequency. This will be found to be caused by largish slow load changes which were not traditional in load caused by consumer demand (diversity) but changes in large solar and wind generation caused by cloud patterns and wind variability. As solar and wind farms in common localities get larger the problem will get worse. Have you ever watched the output of your rooftop solar panel for a few minutes? You will notice the power level change by up to 30% for very little change in solar brightness. So imagine a 1000MW solar farm with some wispy high cloud passing by, the change in output can be around 300MW. This kind of change system wide on a diversified basis will become more noticeable as large solar farms in a common locality become more prevalent (eg. on the Darling Downs and New England areas)
Absolutely, Ian McG.
Trends to more variable (wandering) grid performance is a serious matter.
Meanwhile, records are being set and AEMO’s messages are about the past; mothing done to reduce “wandering” to historical levels.
This will end in tears.
Also, extremely windy… at same time as being clear skies
Wind farms just like solar farms will see continual degredation throughout their life.
https://www.ref.org.uk/publications/280-analysis-of-wind-farm-performance-in-uk-and-denmark
Some of this is driven by general wear and tear but also by the addition of new wind farms which results in reduction of wind energy in the area.
Great piece, thanks Allan.
Agree wholeheartedly with Aaron.
Not great to have CVPs setting price, but also darn hard to solve a network this complicated with a hub and spoke model.
This is an excellent article. Consumers used to ‘simple’ energy tariffs are sheltered from the reality of a highly volatile underlying energy market. This can only be done by retailer hedging or self-hedging (production). When the methods of self-hedging (coal fired power plants) disappear, then some other source or ‘supply’ for the financial market will need to step in OR retailers will need to pass on this increased risk (cost) to consumers.
Is this an opportunity for long duration storage (and a revenue stream for Snowy)??
Yes, long duration storage definitely has a roll to play but so has many other asset classes.
Adding my appreciation to Aaron’s above – thanks Allan, this is a great piece. I genuinely had no idea that the shadow price associated with constraint violations could flow through to dispatch prices in this way.
One question: the parallel flow paths of V-SA (over Heywood and PEC1) seem unusually likely to drive this sort of situation where a constraint might have a lower-bound greater than its upper-bound, and would necessarily then be violated. It seems reasonable that this could happen even under relatively normal system conditions and cause violation-set price spikes out of nowhere.
Does this actually happen anywhere on the NEM? Or is AEMO good enough at their constraint formulation process to make these sorts of infeasible constraint combinations very unlikely?
Great info. I’d love to see an update on these charts with time of day for peak demand included.
Knowing what time of day the peak demand has occurred, from before rooftop solar was measurable to now, would give us a good idea of the real impact of rooftop solar on peak demand.
The problem actually started about 1pm. Note the slope on the operational demand curve starts to get very steep. A quick look at the BOM satellite images shows a nice bank of cloud pushing up the gulf then sweeping over Adelaide. The joys of uncontrolled, weather dependant generation.
What was the problem exactly, Andy?
It would certainly be interesting to perform another study on this since we now know the impact of distributed PV means yesrly grid demand increases are closer to 0%.
It is also interesting that the 2015 study mentions that 20 to 50% of revenue is earned in the top 20 days of the year, and now we know most of that is in the 5 hours between 4pm and 9pm.
Thanks Paul, as always.
Where does the number finally selected come from, do you know? Is there a financial model somewhere that tries to predict how often we’ll bat the MPC and how much revenue will acrrue during those outlier events and this is meant to be assumed that generators can bank on that equivalent to a capacity payment?
When I write that, it seems convoluted…but I’m not sure I’m wrong (definitely not sure I’m right either).
Cheers,
Dave
Is past wind performance a reliable indicator of future performance given the IPCC has said Australia will face slowing winds in the decades to come and the type of wind turbine chosen to be installed today is based on past wind characteristics?
I suspect that forecasting wind and solar separately is not the right approach. For example although wind output in May June 2024 was down 24%, all renewables were only down 13%.
Wind stilling is a genuine issue, if it is evenly lower speeds it is not a big problem, taller towers and larger rotors solve the problem. If it is extended periods of overcast twith low wind, more like a Northern European Dunkelflautte then significantly more generation and storage is required.
Shouldn’t your 24% load factor at 18mw be 37,843 GWh / LGC, not 48,893?
Best wishes
Julian Barton
Yes, there was a mistake in the text but not in the chart – it was 38,893 LGCs in 2024 (not 48,893 LGC) … corrected now.
Up, up and away … Market Price Cap to increase by $2,800/MWh (or 16%)
Hi Paul, so what does that mean for consumers?
Also, big picture, what is your opinion as to whether the price of electricity, paid by the consumer, will be more or less under a 100% renewables scenario?
Cheers
Great article. One minor point:
The frequency diagram seems to have ‘low’ and ‘high’ back to front assuming the fulcrum and the horizontal bar are rigidly coupled together.
Supply up or demand down = frequency rise (not fall)
Supply down or demand up = frequency down (not rise)
Hi Robert, thanks for reading and commenting! I also took that view, initially. The reason I say initially is that upon reflection (in the course of writing the article) I realised the diagram represents the balance as see-saw. Adding supply (supply up), makes that side heavier on the see-saw, and that tips the frequency pointer to the high side.
Great Post Lindon, really informative of the facts.
I know it is difficult, but perhaps some times Global-Roam could stretch itself into the implications???
Thanks Boyd. Appreciate your encouragement in that direction. There’s more to come from Watt Clarity in this space.
RUBBISH ARTICLE.
“On one hand, proposals like the Federal government directly intervening to build more state-owned generation (including nuclear power stations) does not bode well for a more competitive market. ”
Really ? It certainly would. Its a new competitor capable of filling and offering hedging commodities and contracts.
Having a very low SRMC there is no real benefit in nuke plants two shifting to chase extreme pricing to fatten contract positions and rape the market.
Warren/Allan this is in conflict to many of your articles about hedging and contract positions.
With nukes I see a less volatile market, forcing cooperation between intermittent sources to produce competitive hedging services.
A less volatile wholesale price. Ah yes, Flow Power, with a Retail arm like Amber that profits from volatility by offering retail tariffs based on a function of the wholesale price, under the guise of their consumers producing a less volatile load profile than the spot market, and profiting from the difference.
Aha even more let’s check the references, there they all are. The Super Power Institute and Climate 200 crew with Simon HAC/Malcolm Turnball.
It will be interesting to see what impact this has on the congestion and counter flow between Vic and NSW.
The Solar farms are probably being limited (down) due to Spinning Reserve requirements, and/or Fault Level requirements, that the system needs to maintain reliability, for such a case as expected reduced solar output due to weather conditions.
Sorry guys, can’t get my head around the graphs. The small solar is obviously some sort of derived number from perceived total load and generation and imports. Bass link is not clearly shown here. The abrupt drop in small solar (to zero) corresponds to your fall in “demand”. Faults (big ones) will shake off small PV due to voltage depression but not all of it. Will also shake off load (voltage held contactors releasing and also inverters on VSDs stopping but the load reduction does not seem to correspond to the timing of these reported faults. Wind and hydro had been moving around prior. What was Bass link doing? ie are these market pricing driven load changes or real time responses to system load changes caused by disturbances? I note there did not seem to be a lot of Hydro on at the time so I’d be interested to see what the system fault level was at the time. Big influence on clearance time to faults and voltage nadir during a fault (and hence load shake off). Sorry, not enough info here to pull this apart.
Hi Andy, thanks for the thoughts. Not enough time for more detail but worth noting that the NEMwatch snapshot was a real time view and, as such:
1) Shows the most recent estimates for rooftop PV … which, by virtue of the slower cadence of the data, is always behind the 5-minute updates for ‘InitialMW’ (SCADA snapshots) of the bigger generators.
2) A trend for Basslink Target Flows and limits is accessible in NEMwatch, just was not shown on that image.
When time permits in future, might post an ‘after the fact’ analysis.
Do you have to have an Australian energy retail license to participate in Very Fast FCAS (FFR)? I have invested in a community battery being built in Goulburn and I am wandering if, as shareholders, can participate in FFR with only a small and/or large BESS.
Hi Emilio, you have to be a registered ‘participant’, and there are different categories of participant to facilitate different involvement in the markets (including the VF FCAS markets). More information can be found on the market operator’s (AEMO’s) website https://www.aemo.com.au/learn/market-participants/electricity-market-participants
Thanks to the tireless work of Energex we had power restored two days earlier than first predicted. We are so grateful for there massive effort after cyclone Alfred.
One would imagine that these rules will quickly incentivise wind to add or at least contract with storage or even syncons to smooth their output. It may even make sense to add capacitor storage as 10-20% of expected output for 5-10 minutes would be enough to virtually eliminate risk over the next bidding period
You can really see the limitations on fault tracking caused by the lack of smart meters in QLD. The lack of data from smart meters means that Energex is only making guesses on where many of the outages are, and the related impact. Simplistically this can be seen by the many overlapping areas, but also by the fault zones not following street patterns. I looked up where I used to live to look at the outage impact and it was very clear to me that this was the case.
Thanks Dan; some interesting observations here.
On your comment about volatility vs spreads, I’ve also noticed a lot of confusion on the topic – everything from questions about ‘typical daily volatility’ to revenue trajectory assumptions extrapolated from maybe 1 or 2 data points (e.g. the whole-of-station LYA trip).
I’m not sure if this is standard terminology, but I’ve started using ‘variability’ and ‘volatility’ as consistent terminology to separate periodic/forecastable outcomes (e.g. daily or seasonal cyclic trends) from unpredictable scarcity events. This has helped a bit in explaining some of these concepts to clients, such as why we might expect the incoming wave of shallow BESS to tamp down variability and short-duration volatility events but maybe do less to offset longer-duration volatility.
Thanks Tom. I think that logic and terminology is appropriate.
VRE tech needs to start to deliver profiles with hedging value for retailers
or consumers will continue to provide with higher charges.
We’re already seeing CFPS falling “short” of their contracts with failures,
chasing high profits two-shifting their contract positions.
When are VRE/batteries going to start offering caps or even a new evening peak style hedge commodity?
They are still not prepared to take the risk.
oh well. I guess my retailer will be paying inflated prices for base and caps and I’ll be paying.
Kinda makes state run baseload nuke plants with a very low SRMC, who’d would not be interested in two-shifting their hedge position,
quite appealing.
The LGC concept is fundamentally flawed once renewable energy sources become a large enough % of the market for them to end up being curtailed. If a business enters into an agreement with a counterparty building a wind farm and said windfarm ends up being curtailed for whatever reason (for example if lots of wind farms happen to have lots of wind available on a given day and all end up being partially curtailed) then the total number of LGCs produced over the year may well fall short of what was contracted. Now you can argue the generator should have to procure, but this will be a commercial nightmare. Renewable energy proponents only want risk free (take or pay) market position with an offtaker to keep their bankers happy. The only guys able to do this are the big energy retailers who have balance sheet capacity to underwrite this sort of risk. The cliff in 2022 was the end of the RET bubble – stuff signed up to before COB 2020 all came on line and that was that. This is the problem with intermittent generation – if it is not there then industry is in a hole looking for energy and when there is a glut it may well get curtailed – so industry doesn’t get the LGCs it wanted at the end of the year. Smart thing might be to just get rid of semi scheduled generators and make it such that the generators do whatever they want or need to do to become scheduled generators. (Effectively this is already happening with so called solar hybrid projects.) They then contract accordingly. Trouble is I suspect the resultant LRMC numbers will be ugly so the projects still won’t work.
Has there been any movement on this topic? Timing do not seem to have changed much (?)
Does anyone else thing that ~10s for dissemination of data is a bit outrageous? Should be milliseconds!
Are we talking 13GW or GWh
I would imagine that it is discharge capacity.
It is going to be a very complex market to model in terms of estimating how much energy a particular battery might have left in storage, and how much capacity there is in the transmission network connecting these BESS.
I don’t believe there will be that much more competition in the evening peak market by 2027 and certainly not 2025. Fossil gas and hydro are still going to set the price during much of the evening peak, even while fossil gas use drops significantly. A major problem is the transmission networks, not just between states/regions, but within states/regions. We will increasingly see cases where there is capacity and energy available, but the transmission network is operating below its rated capacity, so these batteries can only have limited impact.
SA hasn’t seen as much benefit as should have from the additional BESS in the last 12 months due to a lack of competition.
As per Craig’s comment, perhaps a region by region breakdown of control would be more relevant in considering pricing?
Have they considered a possibility of being able to feed into the grid from under-utitilised home battery systems?
Say I install a 20kW home battery while my kids are still living at home and staying up to 1am playing their PCs and running aircon, etc, but after they’re moved out, all that battery capacity would simply not be needed. Is there any way to tie that back into grid export?
Including Callide C3, that appears to show that the entire coal fleet of Australian generators had an equivalent forced outage rate, EFOR, of ~5% for the period in question.
Not a disaster, then?
On this occasion, when the clinker detached, it resulted in a pressure spike within the furnace. Our initial assessment is that repairs are required to cladding and lagging on the upper section of the boiler.
Pressures spike?
Is this technical writer gibberish for tube leak?
Was it a furnace loss of ignition?
It would appear that the trippimg of the 500kV line caused the trippimg/off-loading of an aluminium potline smelter at Portland which in turn caused a drop in system demand and thus a rise in frequency. When did the smelter return to service?
Hi Robert, correct – looks like it was off-loading rather than a trip at the smelter. From this morning’s data it looks like the smelter began its return to normal loading just after 08:45. Noting that time is before the line was officially returned to service (as per the market notice, 09:48).
The link to the report appears to be broken. I came here after watching a recent Youtube video by Zoe Hilton then going on a search for the report she refers to. No luck so far.
On this occasion, when the clinker detached, it resulted in the release of steam from the ash conveyor water system. The released steam extinguished the flames of all four fuel pulverisers (mills), prompting an automatic unit trip, in accordance with the unit’s safety design.
Is this more technical writer ignorance, and another CS Energy coverup?
Was it caused by oil torch ignition by the unit operator after the loss of the PF mills?
Good article – although I question some of the conclusions drawn around the role gas CAN play or rather the assumptions used to draw these conclusions. What would be interesting to know to qualify some of the conclusions drawn in this article are the following:
– How much additional gas could be brought into the east coast gas system by increasing gas exploration? Given Victoria’s moratorium on gas supply – one would assume a decent amount without considering the additional gas production potential of QLD and NSW that could be acheived from further exploration. How far would this additional gas, combined with a gas reservation policy (which in my view is desperately needed and stupidly never implemented) go to meeting the gap?
– Taking into account the above, how much additional renewable energy would be needed to then fill the gap over the next 15-20 years until nuclear is available? The statement above that the Coalistion’s plan assumes that renewable energy industry will collapse in “a few years” seems overdone – what is this based on? If it’s simple the frontier economics report, well we know that is just a model and is probably wrong just like the ISP is probably wrong.
– My gut feel here, it is critical to continue to accelerate renewable energy development, gas development and start nuclear development to meet our future energy needs – the politicisation of energy is preventing thisand I think its risky to draw a conclusion that gas supply won’t be enough and nuclear will never be built. This is self fulfilling rhetoric that doesn’t focus on solutions to the problem, and adds to the polarisation of the debate. This article, without saying it seems to imply that the only solution is full scale renewable energy deployment – which while I agree this is part of the solution, discounting other potential parts of the solution (a.k.a gas and nuclear) is risky rhetoric, particularly given the massive uncertainties we face over the next 50 years in not just energy markets but international markets and politics.
But as the research also says, “operators often make strategic decisions to scale back asset management and capital investment in ageing coal plants, opting against major upgrades due to their limited remaining operational life and the shortened window to capture a payback on upgrade investments”.
If the planned closure is pushed back, won’t that likewise delay the onset of this ramp-down in maintenance, thus deterioration will be similarly delayed?
A very similar thing happened before in 2021.
https://www.entsoe.eu/news/2021/08/20/outage-of-french-spanish-interconnection-on-24-july-2021-update/
Great to see some knowledgeable analysis of the event, its so difficult to decipher the media reports! Please keep us updated if more information becomes available. I’m fascinated to also know how their blackstart process was supposed to respond and why its taken so long to restore power.
hip shaking and shoulder rolling .. cool man.. tre cool.. 🙂
There is so much going on in that video. I had to watch it multiple times.
It really shows there are far more regular constraints than I was aware of. I wasn’t aware of problems in NE Tasmania.
It also shows how large and wide spread the transmission problems are in NSW. It will be interesting to see if some of these new batteries assist with this issue. Given I haven’t heard of much in the way of grid forming inverter batteries for NSW, I suspect the impact of the batteries will be some what limited when it comes to transmission problems.
It has been reported that they had 5 blackstart units at hydros across the country, but three of these were on outages or breakdowns. If true, that implies their contingency grid management wasn’t the best.
Here is the source of my comment and other related issues
https://x.com/energybants/status/1917206847389389216
I guess you are just waitng for the official report before making any statements. If you have nothing to say best to say nothing.
I’ve always liked Peri-Peri.
Apparently, many voters who did not like the election results, turned off their TVs
Those pesky Negawatts !!
This is just another example where electricity markets ignore financial opportunities for players to make bids out of what policy makers imagine to be a physical forecast. It isn’t entirely unusual to find rules that effectively bet market participants will ignore good financial decisions and where, when they don’t, things go awry rather than self-correct. This isn’t a good design strategy. A well designed market is one where players operating in their self-interest coincidentally operate to maximise social benefit (thank you, Adam Smith).
Another example where this ‘mistake’ occurs is the day ahead market in the US. One of the primary selling points of a day ahead market is that it forces market participants to submit bids that best approximate their output and demand in each trading interval of the following day. This provides an incentive for participants to spend the goldilocks amount of money on forecasting because they have to pay or be paid for quantity differences in real-time at real-time prices.
However, it turns out that wind farm owners are better off deliberately bidding below their mean output forecast day ahead because of the financial effect of the curve in the supply stack (they lose less money if they overbid than if they underbid). Their actions cause the day ahead schedule to represent an optimal financial position and to diverge from a physical schedule, which upsets the market/system operator and policy makers no end who were hoping to kill two birds with one stone (bid).
great article
thank you Dan
Great article, thanks Dan. My feeling is that the balance is a little bit off here. I think the MPC should be lower, but the APC should be higher.
Main reason is that I would like to see policies encourage more consumers to participate in the spot market. I view the high MPC as a deterrent for consumers exposing themselves to the wholesale price.
There have been several instances over the past few years where the MPC has been hit due to local transmission constraints rather than genuine supply scarcity. Whereas the triggering of the APC would be more linked to genuine supply scarcity or energy resource issues.
Hi Dan – I think you’ve covered a lot of ground here and that is to be commended. However, doing so can tend to trigger some conclusions and convictions in people that are shaky when you consider them in more detail.
Here is just one example. We need to understand how economists differ on the matter of market power to better understand the problem. US economists tend to use SRMC and variable cost interchangeably and any differences are considered an abuse of market power (they tend to ignore the presence and operation of hydro generation for some reason). Efficient pricing should account for opportunity cost but this is often misunderstood.
Market power in the US is considered such a scourge that even relatively recent attempts to allow generators to offer according to opportunity cost of fuel in ISO-NE have failed. Similar attempts to allow small generators to vary from variable (fuel) cost (“small fish swim free”) in Ercot also failed, which is why they adopted an Operating Reserve Demand Curve to ramp up prices when generation becomes scarce. This takes places despite the fact that consumers and society is better off if scarce fuel stores are priced according to opportunity cost.
Also, the incidence of very high spot prices (ie above $10K/MWh) needs to be considered in the context of the event that caused scarcity to occur. They are often due to sudden and unexpected events and this is where rebidding rules and trading behaviour needs to be well understood. I’ve published articles on Watt Clarity about this issue and which rules need to be reviewed.
Thanks Greg — very useful points, especially the context around how different markets handle market power. Appreciate you sharing the examples; it’s given me a bit more to think about for future analysis.
We covered it on The Energy 🙂 https://theenergy.co/article/renewables-setting-nem-wholesale-prices-more-often
CS Energy was refused a payout of Insurance in May 2021 because of ignorance when the critical 220V C3 battery was isolated without the unit operator being informed! How will EMG help?
Maybe a Royal Commision would!
“We’ve not delved further into why intervention pricing was so high in the 2017-2019 years in this article, and whether it relates to the directions for system strength gaps in SA from late 2017 onwards”
Yes initially synchronous unit requirements in SA were delivered via intervention (with associated intervention pricing). There was a rule change at the end of 2019 which clarified applicability of intervention (ruling it out for these circumstances), subsequently AEMO started directing the units instead (with no intervention pricing applied) – see https://www.aemc.gov.au/rule-changes/application-regional-reference-node-test-reliability-and-emergency-reserve-trader
Thanks HR10 for the confirmation and background!
If the NEM at the time was about 21GW then yes, the bump is about right for the loss of 300MW odd. Interesting the relatively slow recovery. Must be about the deadband setting. I thought the magic batteries were bidding in for small signal raise????
How does a regular user add that all-time max wind generation line to this chart (trend) in ez2view?
I blinked and I missed it.
Perhaps check out “veracity” vs “voracity” although perhaps the winds were “voracious”.
Good to see this result, and a safe prediction is that the continuing commissioning of a number of large wind farms means that new peaks are likely over the next few months.
oops and thanks … although perhaps ‘voracity’ might also apply in terms of any games that self-forecasters can play!
nice topic
Farewell Liddell – thanks for the memories! For sliding doors, safety glass, and mirrors, visit us at sliding door Services.
Wouldn’t it be simpler to require semi scheduled generators to become scheduled generators by having (say) 30 minutes of storage behind their network connection so they can actually comply with their offers to the market? I say 30 minutes because this is probably enough time for a gas turbine or hydro machine (the most likely things needed for a big block of additional energy) to be dispatched to fill the hole when they bid down (or out). Surely just putting the risks and costs of non-performance where they belong. If they want to contract with a third party to provide the service to them great – but no double dipping.
It has dropped back down to around minimum load with prices dropping below $0/MWh!!
Old coal fired units need to operate as constant designed base load units,to run them up and down will wreck them as they approach the end of designed life.
The coverup continues?
Now it’s 50% Sev.en Global Investments and 50% Queensland Government.
Who is paying for the repairs to Callide C power station as there has been no insurance payout for the loss of Callide C4 turbine when the critical 220 DC supply was isolated with turbine in service and the turbine motored to destruction?
VP6 was clearly taken off “hot” ready for restart. If you bring the unit loads down really low before desynchronising the throttling losses chill the incoming steam to the turbine which is hard on the machine in terms of stress management on thick walled components. Better off tripping from a higher load as this will not cause through wall temperature differentials that Toshiba machines don’t like but still sufficiently cool to be a sensible target temperature for restart.
MP1 clearly tripped over its shoelaces whilst unloading. MW came off pretty quickly (but not dead vertical) so probably some form of protective trip.
It is disappointing to see this review referring to the ISP’s Step Change scenario as if it is an expected outcome.
Friday 13th looks like the beginning of a wind drought. Just bad luck?
Not sure what data you’re looking at, Scott … might want to can see in this live Dashboard view?
As expected, all Bi Directional Units are now empty (no generation bids), but demand is still near peak, making more price spikes.
I have successfully installed a 2kVa ‘run of the river’ micro hydro system for an individual off grid (at that time) resident downstream of the Tinaroo dam in NQ. Unfortunately the 1980’s flood washed everything (including the house) away. The principle is very sound and could be repeated. Where can I find an organisation to discuss this idea.
John Pietila
0421 240 241
Brisbane
Is one of Wivenhoe’s pumps out ?
I have just been exposed to WattClarity and I can see I now have a new best friend in the Australian Energy Space – great articles, awesome raw and analysed data and a treasure trove of information. Thank you.
Thanks for the kind words, Barry
Certainly expensive in Victoria, after a cool cloudless day, those with home batteries and wholesale pricing must be making a mint.
In terms of generation, hydro is at 5.4GW batteries are at 1.3GW, wind is just above 900MW. However brown coal appears to be lacking. Unit 3 at YWPS is out of service and Unit 4 shut down mid afternoon. Unit 1 at Loy Yang A is also out of service, while unit 4 is operating below normal capacity since mid afternoon. Both units at Loy Yang B have also dropped in output coincident with the high prices, which they also did yesterday. Cannot think of a physical reason to be down by 30% in mild conditions, perhaps there is a financial or FCAS reason?
The AEMO has a report on this incident, root cause was one phase of a circuit breaker failing to open while taking a generator out of service. This tripped YPS No.1 220kV busbar and as a side effect, YWPS unit 2 lost auxiliary power supply. Some screws holding the trip coil in position were found to be loose on the phase of the circuit breaker which failed to trip. The faulty circuit breaker was an Ausnet asset.
https://aemo.com.au/-/media/files/electricity/nem/market_notices_and_events/power_system_incident_reports/2024/trip-of-yallourn-220-kv-no-1-busbar.pdf?la=en
Victoria as of writing is sitting at the market price cap, the other states are not far behind. It appears the shortfall in capacity is in Victoria and New South Wales, with 1100 and 550MW respectively of demand supplied from other states.
Does anyone have thoughts on why the bidding for VIC was so often 9c short of the market cap?
Dear Paul,
Thank you for writing this thought-provoking article. However, I’m struggling to reconcile the idea that BESS traders and their auto-bidders are simply making poor decisions and consistently losing money. The paper you linked mentions that the provision of contingency FCAS costs $10/MW/hr, which raises a couple of questions for me regarding their methodology:
a) Could the BESS be holding other underlying financial contracts? For instance, is it possible they are strategically withholding capacity as part of a broader hedge against wholesale price movements? Do we have any visibility into the terms or pricing of such contracts?
b) How is BESS Capex impacted by charging and discharging patterns? Is this accounted for in their analysis?
It would be helpful to really understand the bidding behaviour of BESS so DR participants can identify and exploit rare opportunities when they arise.
I’d appreciate your thoughts on these points.
Liam. Loy Yang likely operating fully hedged or “short” with a low-cost contractor(s) filling their residual contract margin.
Contracts are complex and obfuscate apparent operations. Wholesale market price often gives very little indication of direct commercial outcome.
The Spot price is more representative of bidding outcomes based on underlying contracts and contractual obligations.
Hi T C,
Thanks for your comment. I agree, it’s unlikely to be a case of poor decision-making. My article was intended to draw attention to the consequences of whatever is driving this behaviour.
a) Protecting a 3–4-hour hedge seems like a plausible explanation. Unfortunately, the NEM is not ‘perfectly competitive’ in the economic sense, as details of such arrangement are not, to my knowledge, made available to all market participants. I’m also not clear on how modern tolling arrangements might result in this behaviour.
b) Battery replacement costs are certainly a likely factor. That said, with battery prices continuing to fall, most analysts would argue that it is best to maximise revenue in the short term rather than conserve capacity.
Greater transparency around out-of-market contracts would help all participants understand how these deals influence market outcomes.
Kind regards,
Paul M
I agree with the conclusion that the story was largely a beat-up. Your chart shows that average daily Vic gas generation so far this year is below last year.
As to whether the peak would have been higher with more gas~fired capacity? Even at the extreme prices of last week there was some capacity undispatched, because it was only offered at even higher prices.
You are right.
It was another beatup. There was a severe wind drought for three days and poor solar as well but gas peaked at about 75% capacity and averaged 26% for the week. On most days where gas generation is required, 2 GW/5 GWh of storage would displace it entirely. Further as wind and solar capacity increase coal and hydro utilisation will decline during normal operating conditions, leaving even more capacity to fill the gaps in wind and solar
I can see why there is an argument for geographic diversity of supply in provision of these services (assuming this better ensures power system security than geographic concentration).
But I struggle with the concept that ‘technology diversity’ is required. These contingency services, particularly very fast frequency response, are very inexpensively provided by BESS (in pretty much the same way that provision of inertia used to be effectively costless in a fully synchronous system) – I don’t see any good argument at all for forcing inclusion of higher cost FFR technologies into the supply stack.
The article seems to be arguing that we won’t have enough of these services at some future date if economic competition drives DR providers out of the FFR markets, but given the level of investment in BESS, that seems a very long bow to draw.
Hi Allan, thanks for the comment.
Just to clarify, the article isn’t arguing that we’re heading for a shortage of FFR supply. The current low prices reflect an abundance of BESS-based supply, and projections from Modo Energy suggest this supply will continue to grow rapidly in the NEM.
The case for technology diversity is more about risk management. Each technology brings its own physical response characteristics and commercial behaviours. With the rise of algorithmic bidding and increasingly strategic dispatch from BESS, those commercial behaviours may become a system risk in their own right.
My argument is simply that it would be prudent to maintain some technology diversity within the FFR, and broader FCAS, stack. That diversity could be at risk if prices continue to clear at very low levels, potentially leading to over-reliance on a single technology class with tightly correlated incentives and response profiles.
It is interesting that overvoltage was suspected as the cause.
What are your views on the fact that there is increased solar weather activity recently (for the last few months) that could have induced voltage in power lines?
A similar event occurred in August 2003 in Northern USA and Canada, and this was also during a period of increased solar weather activity, although, according to Google’s AI, it was ruled out as being the cause.
In the higher latitudes, solar flares can alter the earth’s magnetic field which in turn can induce DC currents in transmission lines. These DC currents can cause a transmission line to trip out of service due to such currents appearing to the protection systems as earth fault currents. This has happened in Canada on very rare occassions. In Spain no transmission lines tripped until after the power system was in a system crash spiral.
In Spain the lack of power sources with voltage control incorporating stabilising circuits meant that steady state voltage oscillations
were undamped and grew to such an extent that over-voltage casued the tripping of 2 large solar PV farms that then led to the collapse.
The large amount of solar PV stations with Grid Following inverter systems that do not control system volatge but only control reactive power output according to a fixed power factor was clearly a problem.
Lots a batteries with Grid Forming Converters and voltage stabilising circuits should fix the problem.
David’s model is a copper plate model. As such, it could prove what is impossible; but not what is possible.
For example, if a battery capacity of X GWh in a copper plate model was insufficient to get through a particular low wind night, then it would be impossible for any real world placement of X GWh of batteries to service that night. But the fact that X GWh IS sufficient in a copper plate model to get through some night tells you close to nothing about the possibility of finding a suitable configuration to get through that night. David’s not an idiot, he knows this, but he still repeatedly misrepresents his model as being relevant to real world feasibility … it isn’t.
Used Co-pilot to translate and summarise in 1 page – could ask for a full translation if you want one:
Here is a **1-page summary** of the **non-confidential report from the Committee of Analysis on the April 28th Electricity Crisis** (“Informe no confidencial Comité de análisis 28A”):
—
### **Summary: April 28, 2025 Iberian Electricity Crisis Report**
On April 28, 2025, at 12:33:30, the Iberian Peninsula experienced a **complete voltage collapse**, disconnecting Spain and Portugal from the European electricity grid. This unprecedented event triggered a formal electricity crisis declaration under EU Regulation 2019/941. The Spanish government established a multidisciplinary **Committee (28-A)** to investigate the causes, impacts, and propose corrective measures.
#### **Key Findings**
– **Multifactorial Origin**: The blackout was not due to a single failure but a **cascade of events** involving voltage instability, system oscillations, and inadequate reactive power control.
– **Five Phases Identified**:
1. **Pre-event voltage instability** (9:00–12:00)
2. **System oscillations** (12:00–12:30)
3. **Generation losses due to overvoltage** (12:32–12:33:18)
4. **System collapse to zero voltage** (12:33:18–12:33:30)
5. **Supply restoration** (12:33:30 on April 28 to 14:36 on April 29)
– **Root Causes**:
– **Insufficient dynamic voltage control** due to limited synchronous generation and inadequate reactive power absorption.
– **Unusual oscillations** (notably at 0.6 Hz) possibly triggered by specific photovoltaic installations.
– **Overvoltage-induced disconnections** of renewable plants, leading to a self-reinforcing collapse.
– **Delayed or inadequate response** from some generators and infrastructure operators.
– **Cybersecurity**: No evidence of cyberattacks was found. However, **vulnerabilities and poor configurations** were identified in several digital systems.
#### **Restoration Efforts**
– Power was restored to **99.95% of demand by 7:00 AM on April 29**, using a strategy of **islanded grid segments** reconnected progressively.
– Support from **France and Morocco** was critical in re-energizing the grid.
#### **Recommendations**
– **Technical**: Accelerate deployment of voltage control technologies (e.g., synchronous compensators, FACTS), enforce compliance with voltage regulation standards, and improve grid observability.
– **Regulatory**: Update operational procedures, strengthen oversight of shared evacuation infrastructures, and enhance coordination with European partners.
– **Cybersecurity**: Implement NIS2 and CER directives, improve authentication, logging, and network segmentation, and expand coverage to smaller operators.
– **Market Design**: Reassess the impact of quarter-hourly market intervals and interzonal gate closure times on system stability.
—
Would you like a translated version of this summary in Spanish or a more detailed breakdown of any specific section?
Great article Linton. Appears that the count of forced outages may have stayed at the same level over the last 5 years but the EFOR % number has more than doubled over the past 5 to 10 years. Thats suggests that frequency of the events is steady while the durations of the outages are much longer.
Hi Ian, yes your thinking aligns with a feature Paul identified in 2023 when looking a the trend of unavailability. The trend was increasing and the recent extended forced outages were contributing to that.
( Part A of this: https://wattclarity.com.au//articles/2023/03/06march-geninsights-q4-coal-units/ )
Thanks Linton, great analysis as always.
Does the Δ300 metric risk excluding the ten Queensland 350MW Hitachi units if they’re operating at partial load? Gladstone runs 280 MW units, so is entirely excluded from this analysis.
Does a Δ200 metric risk including planned shuts or things like clinker deloads?
Hi Alex, you’re right. The measure would miss unit trips when not running higher than 3ooMW. And smaller unit, yes, wouldn’t make the cut. A Δ200 metric could be some thing we explore in the future!
Further to Linton’s reply Alex, we chose the Δ300 metric in order to make the trips large enough to feature more clearly in what Linton’s going to analyse in a follow-on article … frequency changes.
What I found interesting was all the asynchronous plant was supplying the grid at fixed power factor. They were using the synchronous plant on to do the dynamic VARs control but the grid wasn’t dispatching them to do it. – Their static equipment had all already been taken out in an unsuccessful attempt to lower the voltage. The report authors don’t seem to understand capability curves of generators, but that may be an auto-translate issue of word choice.
From my thinking, for consideration on this topic of frequency and voltage…
– Traditional turbines have more trouble turning when there is more load on them, making frequency drop.
– Inverters, theoretically, can continue to pump out 50Hz even when past max load, the voltage just cannot keep up.
– Flywheels that are not generating will follow frequency regardless of voltage, including traditional turbines at minimum (steam) pressure.
So in theory, couldn’t a grid that is mostly/completely inverters and flywheels have a grid collapse while the frequency is still within bounds?
And the graphs above showing that the frequency drops less on big generation loss seems to back up this theory.
Hi Hughman,
While I can’t answer your question, I did note that AEMO released recently (10 June) the draft Power System Risk Review report
https://aemo.com.au/energy-systems/electricity/national-electricity-market-nem/system-operations/general-power-system-risk-review
which goes into quite some detail on these and related aspects including the NEM’s needs and opportunities for the future. Submissions open until 5pm, July 1 2025!
Not so great on the 26th June. Coal to the rescue. How can we operate an electricity system that is so intermittent and volatile?
Great and yet the next day there was no wind.
Thanks Paul
It’s really quite amazing how widespread Aus weather conditions are-high or low wind speeds seem to occur simultaneously across half a continent; some of these farms are over 2000km from each other!
I never realised this until the advent of wind generation. Presumably applies to the nearby coastal waters also.
Another lull in wind production this afternoon down to 400MW again
Various lack of reserve forecasts issued and cancelled for the first few days of next week. Hard to keep track of the situation from the market notices alone. I assume associated with forecast uncertainty provided by the east coast low.
High voltage and cascade failure, reads like too much reactive power. It will be intetesting to find out why this occured, as it depends on other components (Statcom, SVC, reactors, capacitor banks, HVDC converters …) as well as the generation sources. Transmission and distribution lines themselves can act as reactive sources, it may be there is insuffienct capacity to sink reactive power under the configuration in use that day.
At the risk of repeating some comments I made on LinkedIn (https://www.linkedin.com/posts/carl-daley-0308956_thu-26th-june-2025-event-activity-7344178338963312641-c23j?utm_source=share&utm_medium=member_desktop&rcm=ACoAABn7upMBG6-yj4xnUebiiT_FACaSSRNFnqo), this is an interesting piece but unfortunately it conflates market outcomes with “system vulnerability”, or high prices with “critical limitations in energy storage technology and renewable integration”.
The physical power system was under no major stress: there were no LOR conditions and reserve margins were around 20%. There was over 3,000 MW of available but undispatched (ie spare) gas and hydro capacity through the high priced periods in the NEM excluding Queensland (where prices were not so elevated).
So we didn’t get the extreme prices due to the power system running short of capacity, “wind collapsing” or for other hyperbolic reasons. They arose from a more involved and opaque set of factors involving the ownership of peaking generation in the NEM, the market contracting strategies of a wide range of participants, which physical capacity happened to be offline (and yes low wind is part of this), and how this all then played out through bidding behaviour.
Good article. Wind production certainly collapsed going from 6600MW earlier in the day to just 400MW when the demand was peaking.
Unreliable renewables will continue to cause disruption, price spikes and the closure of industry.
This article exposes the vulnerability of the NEM to load shedding (or just as damaging for consumers high wholesale prices) when coal is retired and we have insufficient gas powered generation and medium and long duration energy storage. The Capacity Investment Scheme’s continued support of mainly 2 to 4 hour batteries needs a radical rethink before its too late. Wind droughts are real and could go for weeks. The wind drought in this article lasted for a day. Even with the NEM’s current coal and gas generation capacity, what would happen if there was a 5 day wind drought combined with low solar and moderate demand in 2 or 3 states of the NEM?
Jennifer, on what basis are you saying there was insufficient gas capacity? There was more than enough bid available, its owners just wanted more than $17,400/MWh to turn it on. Same for a big slab of hydro.
Hi Allan My comment states “exposes the vulnerability of the NEM to load shedding (or just as damaging for consumers high wholesale prices) WHEN COAL IS RETIRED and we have insufficient gas powered generation and medium and long duration energy storage”. My comment implies we got through the event last week because the NEM still has coal generation. But the Capacity Investment Scheme is incentivising 23 GW of more variable renewable energy by 2030 backed up by 9 GW of “dispatchable” clean energy. The article shows how 2 to 4 hour batteries can’t plug the shortfalls “as storage capacity became exhausted, contributing to price escalation”. The CIS needs to radically rethink its definition of dispatchable as coal is retiring and new gas generation is not included in the CIS.
Fair enough Jennifer – of course if we don’t have enough dispatchable capacity in future then reliability will suffer and prices will be high. And of course a range of dispatchable capacity types, not just short duration storage, will be required. No-one who understands power systems would disagree.
I just don’t see the connection to an event where prices went high for reasons other than insufficient dispatchable capacity – of which we had plenty on the day in question.
There may some confusion around this section
“Congestion, connection delays and security constraints. Because revenue is divided by output, declines in revenue due to such curtailments is worn by the investor. Again, this is necessary to encourage the investor to locate and operate in a way that minimises these risks.”
The documentation specifically builds ‘curtailment & congestion’ into the revenue calculation ie the project is allowed to claim MWh lost to curtailment.
It’s included in the calculation of the contract MWh –
Nominal MWh = sent out MWh + curtailed MWh, if the curtailment is from AEMO.
That says to me these CIS projects are protected against curtailment from anything to do with the network, which includes curtailment to protect the grid against minimum demand. This will naturally be the biggest problem for solar projects.
This fits well with the premise of the CIS – to massively overbuild wind and solar.
The fact that the wholesale price in Victoria was significantly higher in 2025 than 2024 means there is a management problem. 2024 was one of the worst years for wind conditions on record, so to have higher prices is really saying something went wrong.
“Several planned transmission outages”
I think it is time that the exact timing of these outages needs to be made more flexible. The cost of delaying these works by one or two days until wind conditions were better, might have been small compared to the additional electricity costs to consumers. Some one needs to run the numbers.
Is the limitation with VNI, or with transmission around Wagga Wagga? I was under the impression that it was the later. It is simply the reporting system that makes it appear as though it is VNI.
It seems all renewables were operating in constant power factor mode (general practice as per FERC and ENTSO-E in US and Europe is to control to unity power factor according one AI site) They did not participate in dynamic voltage control mode and hence did not do much to help with the overvoltages. The full non confidential Spanish gov report recommends that market rules be changed to allow renewables to be remunerated for taking part in dynamic voltage control. One wonders if the blackout would have been avoided if the renewables had been in this mode.
Don’t we want to know the frequency of events per numbers of units? If the number of units declines so should the number of events. Removing the closed units won’t account for that underlying trend.
Hi Steve,
There’s a multitude of stats that we could calculate. In the end, we aimed to understand whether or not we are seeing fewer ‘events’, compared to the past. Some drivers of fewer events are closures (but we eliminated that in the second chart), potentially longer durations spent unavailable, and potentially more time spent operating at low loadings (below 300MW).
In the next article (that Paul referred to) [ https://wattclarity.com.au//articles/2025/06/frequency-change-associated-with-coal-generation-events/ ] the aim was to see how these events impact system frequency.
Thanks Dan, plenty to talk about, and interested to see others’ interpretation.
The question now is: how many 100s of kms of transmission lines do we have which will not survive wind speeds under global warming conditions?
So how much are the total costs for the system and how much are consumers paying for system stability in all types of FCAS and FPP payments
Really good article series this one, thankyou.
Is work on the Tungkillo-Tailem Bend line the reason for a wierd AEMO dispatch interval? At 5:25pm EST the AEMO dashboard had Vic price $193.82, SA price -$19.97, NSW price $138.48. Both the SA and NSW inter-connectors had strong counter-price energy flows. From Vic to SA it was 620MW on the AC inter-connectors, which was at its limit, 15MW Vic to SA on the DC connector (at its limit), and from Vic to NSW it was 1088MW, again at its limit. Being close to 5pm SA time, work on the transmission line may have been finished for the week and the line re-energised, but as this capacity was made available to NEMDE it couldn’t find a stable solution.
Triple-C alert… *danger*, fossil fuels ahead.
I’m definitely going to have a look through the report, but i have noticed the problem nobody seems to address so far is the diminishing operational demand as distributed PV keeps increasing, and now the government is pushing everyone into getting solar batteries. I’m certainly not complaining about that, but surely that is going to affect the grid electricity price and ROI of any new large-scale grid-attached generation projects?
Buried the lede, CSIRO says “We won’t release the modelling for general consumption.”
Take that hoi polloi!
Kate Summers speaks a technical language suitable only for power engineers. However she speaks in such a wonderful and passionate fashion that even amateurs like myself cannot help but be entertained by. It’s no disrespect to her fellow panellists to note that her star shone more brightly.
Also of great interest to me was the session on REZs which was of course the session on transmission.
Suppose at Eraring we have an accident similar to what happened at Callide C4 in 2021. Do we really have the renewables and storage to replace this coal plant at full capacity? What would happen during a dark, cold winter week as just experienced end of July?
3 Aug 2025
Premier NSW energy guzzler in winter: coal plants running at 90 % capacity during critical evening periods
https://crudeoilpeak.info/premier-nsw-energy-guzzler-in-winter-coal-plants-running-at-90-capacity-during-critical-evening-periods
My website is on oil. But I got interested in power supplies because everyone assumes petrol and diesel cars can be easily replaced by EVs. (The real problem are actually our diesel trucks!)
Excellent non-biased discussion. Thanks Paul
“This distributed PV output offset underlying demand growth across the daytime hours and led to an
overall reduction in operational demand (Figure 6). NEM-wide operational6 demand averaged 21,760 MW, a
reduction of 151 MW (-0.7%) from Q2 2024.”
Looks like the trend continues.
It begs the question why would major grid generation projects of any type, green or otherwise, be attractive for investors when the opportunity for sales keeps reducing?
Perhaps it is simply the replacement of the fossil fired plant creating this need?
Perhaps the government also backstops dispatch targets?
Just on Waratah’s maximum discharge to date Paul, where you mention that it doesn’t include any Raise FCAS enablement volume – the battery isn’t yet registered to deliver FCAS, I guess that will come later in its commissioning programme.
Isn’t the question not whether the NEM can operate on 82% renewables by 2030 but whether ALL consumers, particularly the largest consumer group SMEs (now called Business Mass Market by AEMO), can afford their power bills from a NEM with 82% renewables. After all, if energy consumers can’t afford their power bills then the plan has failed the National Electricity Objective. SMEs don’t get government bailouts like large energy users (eg steelworks and smelters) and can’t even access hardship programs. Over 5 million Aussies depend on a small business (under 20 employees) to put food on the table and a roof over their heads. This is not just an engineering problem.
Re. KKGT – there appears scant info on their gas supply – contracted &/or spot, TJ/d?
Nothing on Snowy website, etc – they talk up the lowering of costs…?
The fact we have had an average rainfall winter, the BoM are forecasting a drier period ahead and storages are still recovering from lows last winter … may indicate some prudency in water storage management is required.
file:///C:/Users/TimASTLEY/Downloads/Storage%20(17).pdf
https://www.economicregulator.tas.gov.au/Documents/25%202485%20Tasmanian%20Energy%20Security%20Monthly%20Dashboard%20-%20August%202025.pdf
Sorry – a link to my C drive won’t work – try the bottom of this page – https://www.hydro.com.au/water
Hi Paul, you have probably got this a hundred times already, but in case it was missed; in the headline you have Fri 13th instead of Wed 13th. Spooky.
oops!
It shouldn’t be too difficult because the amount of grid electricity generated and sold keeps falling every year, so the 82% target becomes relatively easier to achieve as time progresses.
But don’t ask how the rising cost of grid maintenance gets paid…
So it looks like the Basslink trip is the cause of the demand drop. Basslink was carrying ~330 MW southward at the time; there’s a Frequency Control Special Protection Scheme (FCSPS) set up in Tas to manage the risk of Basslink tripping. In this event the FCSPS immediately dumps an amount of Tasmanian load sufficient to offset the loss of imports.
The 292 MW scheduled demand drop is about the size of lost Basslink imports, indicating that it’s likely the result of operation of the FCSPS after Basslink tripped from 330 MW.
The annual report suggests that the STATE Government paid lOrigin between approximately. $230 Million and $3OO million which are availability payments and consistent with reports on State Government financial support.
Not unexpected. A very mild day, limited cloud cover with a bit of a warm breeze about.
Meanwhile demand is through the roof tonight as the chill sets in and solar/wind has died off.
Thank you (both) for publicizing this event, which is potentially (pun intended !) Impactful on Tasmanian community activities.
Dead silence emanating from public media and Government.
Strengthens the case for a new cable link to mainland Australia?
I am still waiting for a public explanation (by a responsible authority) explaining intemperate cable overload protection adjustments, possibly in response to desire for increased revenue, directly related to previous BassLink cable failure.
Thank you, Dan, for sharing such a comprehensive and insightful perspective on the unique energy challenges facing Timor-Leste. The contrast between reliance on fossil fuels and the urgent need for reliable, sustainable energy access really highlights the complex balancing act many emerging economies face during the transition. Stories like these remind us how critical tailored solutions are—both for improving quality of life and supporting long-term economic resilience.
This situation will continue while the political imperative of (uncontrolled) renewables at any cost exists. I think that every grid scale solar (and possibly wind) installation should be required to have a battery system capable of producing and absorbing power to limit the rate of change of power generation so that other generators have a chance to react.
There were 4.8 million people in Victoria in 2002 and 6.5 million in 2019. Have you normalised the demand for population?
As Australia’s push for net zero is looking increasingly unlikely (and too expensive), the prospect of extended use of coal should go way beyond trying to 2shift large coal fired units that were never designed for this purpose – progress to much smaller capacity units that can be more suited to allow dominance of solar power when available. These should not be high tech supercritical units, but basic design – with the cheaper fuel costs, super high efficiency should not be an issue (it never was in the coal fired units on which we still rely). What politicians will have the courage to go down this track – not the current mob! Best regards
Its interesting to note that Victoria’s aggregate electricity demand decreased from 52,976 GWh/year in 2002 to 46,885 GWh/year in 2019 in spite of a population growth from 4.8 to 6.5 million.In that period solar/wind generation increased from zero to 17% of total generation.At the same time average annual wholesale electricity prices increased from $35.44/MwH to $124.88/MwH which appears consistent with the increased volatility as shown in the graph above.
Could this article be any worse written.
Bring back the last 2 weeks of authors, pls.
Would be very interesting to have a look at a longer period – before and after. There appears to be a frequency oscillation here with an approximate 15 minute return period prior to the one at 11.45am taking off. The exceedance occurred at 11:55 but it all started 10 minutes before that.
These sorts of excursions were so rare a few years ago, and now it’s just another week.
I feel sorry for all those who have to monitor frequency excursions and evaluate Unit response to prove compliance nowadays. Just about a full time job!
I am interested to know how much of a negative it is for Australia that we are working to such a failing plan. Not necessarily that high levels of wind and solar are unworkable, but that we cannot implement at anything like the touted rate. We don’t understand when coal plants can close, we don’t understand when we will need more gas fired peaker plants etc etc. A plan that is plausible would surely keep costs down.
I’m Ok with accepting predictions that Eraring’s closure is likely to be delayed. I’d prefer if they came with some detail of how much NSW is falling short of replacing it, particularly when the comment is ” … or perhaps in 2029 “. And what is the potential uptime and reliability of plant four years of more more after the originally planned closure? I think there is an emerging “real world” gap presupposing that it can indefinitely continue to ramp up and down and be relied upon.
Agree with you on several counts, Paul. Whose job is it to provide ‘some detail of how much NSW is falling short of replacing it’?
The Elephant in the room is: Does the system have sufficient Inertia for reliable operation? If the large heavy generators are shutdown and replaced by Solar, Wind and Batteries which contribute little or no inertia because the inverters cannot remain working during severe frequency excursions, then the inertia required to contain frequency decline will not be there. Low inertia results in steeper frequency decline during loss of generation because there is not enough governors on the system to restore frequency or even to control the frequency at 50hz. The result, as we are seeing even at this early stage, is a wandering frequency and an inability to identify what equipment actually caused the frequency excursion.
Just say it – 80 year lifespan Nuclear plant is the only safe and sure way to power our future!
“..4 different analysts quoted in voicing their concerns about the unrealistic notion that Eraring can close in 2027 … or perhaps even in 2029”
Bit of an exaggeration that 4 analysts are voicing concerns specifically about Eraring, given only Tom Allen is directly quoted in the article on this. From the article:
– Jarden are quoted as doing modelling of Eraring staying open beyond 2027 with no normative statements that this is what should happen.
– Neither Matt Rennie or Greg Elkins mention Eraring at all in the their quotes, just the 2030 and 2035 targets.
Hi Paul
Is it possible to have an article that summarises minimum synchronous generator requirements across each of the NEM regions as they are current set, and how they are planned to reduce – assuming that is actually the case. Feel free to remind me I’m attempting to simplify a complex set of rules
Thanks for the idea Paul, it’s certainly something we want to cover so I’ve added this to my to-do list — though I can’t say exactly when I’ll be able to get to it.
Dan
Great work Paul and Linton in cleverly using the available public data to unpack the likely causes of these events. Very interesting and I think important towards understanding frequency resilience. Publishing this sort of investigation may plant the seed towards institutional reporting which can access more data.
A quick look back at combined availability levels for the affected wind generators shows that this constraint could bind about 10% of the time. The energy potentially curtailed might be something like 2.5% of available output. Not insignificant.
In practice the figures will be something lower because of other constraints and price-based offloading.
Great article. Brings some balance to the discussion. Too many are focused on the output of “aging and unreliable” coal generation and ignoring the limitations and transparency around VRE output… which is absolutely necessary when navigating the impact of extended wind droughts.
Nice work. I like to think it should be possible for market 4s data to feed and automatic viewer for frequency events to filter the larger causers and performers. A significant factor not being considered though in the frequency observation space is whether there actually is any controller steering frequency back to 50Hz because the mainland is actually at 49.980 or 50.020 Hz six times more often than the midpoint of 50Hz so whilst less >or< NOFB events are occurring the consolidation is starting to resemble continual slow variations between the slightly wider range than the AEMO PFRR PFCB but is actually rarely settling suggesting alsupply and demand are never well balanced. In my recollection of days pre NEM NSW frequency was much steadier.
The 4-second causer pays data for 11th September does not appear to have yet been uploaded (as at 15th Sep) to the AEMO webpage. Is there an issue that you know of that could be causing this delay?
Great remarks Paul, and really supporting the view that 2033-35 is where the rubber really hits the road with coal retirements. Not only is that where capacity is forecast to fall off a cliff, but those retirements are also technical retirements as much (more so?) as economic retirements: Bayswater, Vales Point, Gladstone (if still active?), Callide B (if the extension sticks), and Tarong will all be climbing the far end of the bathtub curve around that time and there’s very little public information to be had on what the options look like for delaying that.
I think it’s underappreciated how different the technical EOL aspect is from the discretionary, policy-driven pre-EOL economic retirement discussion that has captured attention, and I also wonder how well QLD and NSW are doing at recognising that a retirement wave hitting both states near-simultaneously may limit the ability to rely on imports to solve local capacity problems.
I hope you’re not verballing Oliver with that “90% of the coal fleet disappearing” comment Paul 😉
That’s a strawman argument. Anyone who understands the very different availability characteristics of wind and thermal generation would never expect the output profiles of the two to be comparable.
Thanks for the comment, Allan – I did check each of the articles written about this week’s 2025 QCES with the relevant parties.
With respect to the differences between coal and wind, I’m sure Oliver understands the difference:
1) His presentation on the second day was focused on explaining the size and depth of storage required to meet firming requirements (with a focus on the QLD region)
2) So (my sense is that) Oliver’s comment was a way to help the audience understand the size of what will be required in a VRE + Firming world without coal.
Thanks for sharing.
The pages dedicated to clarifications and fragmented state of the ‘GFM’ environment goes to show how important it is that International Standards be set up as soon as practical. It will be a hot mess in years to come dealing with the variety of vintages, manufacturers and capabilities.
Hmmmmmm.
When politics overrides physics.
subscribe to updates
From my reading of the report, they say the models say they should work well – excluding “niggles” like close-in fault current behaviour. As those in the power industry know, there is a massive difference between models and actual proven near 100% reliable performance under myriad fault behaviours – many which weren’t covered in contingency planning. The non-credible events – with Callide being a good example.
Just how much importance is to be placed on preventing blackouts? Would you stake lives on the risk that they won’t fail to do their job? I wouldn’t.
This table about yesterday’s, today’s and tomorrow’s problems is a very good planning tool. “out of scope” is routinely used to shovel away the real problem
https://www.theguardian.com/environment/2025/oct/01/gladstone-queensland-biggest-coal-fired-power-station-could-close-six-years-early
Alison Reeve from the Grattan Institute just told ABC News that power mainly goes to alu smelters which need reliable supply 14/7
When you look at the last 7 days “baseload” was 550 MW. But Gladstone also provided up to 1,230 MW during peak demand
This is just the usual posturing we are seeing from industry claiming they will shut down and then having the Government throw money at them to stay open.
Very blatant that they choose to make this announcement with the new QLD Energy Plan to be released in less than a fortnight.
The title should of course have been 5th October
Suggestion for an article:
To which extent were recent NSW market interventions caused by calculations as described in:
26/09/2025
“A minimum level of synchronous generation must be in service to ensure the stable operation and protection of the New South Wales transmission network. Transgrid and AEMO have jointly agreed that this requirement is met when the equivalent of six large synchronous generators are in service. Under these conditions, the network is considered to be in a satisfactory state. To be considered secure, the network must still have the equivalent of six large synchronous units in service after any credible contingency— typically the loss of the generator with the highest contribution”
https://www.aemo.com.au/-/media/files/electricity/nem/security_and_reliability/congestion-information/2024/transfer-limit-advice—system-strength-nsw.pd
Suppose Eraring were to shut down, would that number be (12-4)=4?
What other criteria are there for the minimum of rotational, synchronous inertia in the grid? How about imports?
The title should of course have been 5th October
Suggestion for an article:
To which extent were recent NSW market interventions caused by calculations as described in:
26/09/2025
“A minimum level of synchronous generation must be in service to ensure the stable operation and protection of the New South Wales transmission network. Transgrid and AEMO have jointly agreed that this requirement is met when the equivalent of six large synchronous generators are in service. Under these conditions, the network is considered to be in a satisfactory state. To be considered secure, the network must still have the equivalent of six large synchronous units in service after any credible contingency— typically the loss of the generator with the highest contribution”
https://www.aemo.com.au/-/media/files/electricity/nem/security_and_reliability/congestion-information/2024/transfer-limit-advice—system-strength-nsw.pdf
Suppose Eraring were to shut down, would that number be (12-4)/2=4?
What other criteria are there for the minimum of rotational, synchronous inertia in the grid? How about imports?
Good thoughts, but it may be even knottier. Rio needs Gladstone PTI committed to close GPS while Powerlink need a GPS retirement date before 2030 to justify Gladstone PTI. So in a way, GPS is forced to announce retirement before that retirement is feasible if they want any chance of the retirement becoming feasible later.
I think the early closure of Gladstone has been speculated for many years. The author doesn’t even mention the subsidy the station receives via CS Energy. Nor do I personally think system strength was the major reason for the Eraring extension. And although the firming for Boyne Island is not yet known I suspect QLD plans to build a new gas plant and that coal from other plants will provide the firming while it’s being built. That’s not a solution I favour but what do I know.
Its getting hard to ignore the rollout of batteries even for the most conservative of “experts”, but what is clear is we are still in the high growth phase so forecasting where we will end up is little more than a guess, and then you have to reestimate what power and energy looks like in such a system.
An excellent article. Thanks. Interesting how one state government actions can change plans somewhere else on the eastern seaboard.
The inbuilt conflict between government’s ultimate responsibility to maintain power supplies and the commercial decisions of companies and investors on the timing of closing aging power plants was, in the case of NSW, created by the very government having privatized what was and should have remained public infrastructure in the 1st place. With this privatization, the State government wanted to rid itself of the difficult transition away from coal. This opportunistic approach has backfired.
Great article. The Author could include some other points or solutions. Most coal power plants in Queensland are Government owned so I don’t think this will be a systemic/repeat issue. Other solutions could include: (1) other Queensland coal power plants can have their life extended, or alternatively (2) another coal plant could close instead to improve the economics/operation of Gladstone.
It appears that all projects have been reported on the basis if their peak energy sent out (nameplate) capacity, not reduced by capacity factors. This leads to a loads served estimate that is wildly overstated – something in the order of 3 times overstated. The number of hypothetical houses served is thus overstated by the same factors.
Surely, after more than a decade of optimism, reality should be applied to all Ministerial statements and other public guidance papers.
IMHO, the promised 6.6 GW of new power gen will actually deliver only 2 or 2.5 GW, with lower reliability than the Australian Eastern States Grid provides currently.
I note sync condensers in the mix. They obviously have some doubts about grid forming inverters being able to completely meet grid strength needs. I wonder what their black start plan is?
Is VP6 working at only half capacity?
Thanks for the great coverage Dan – always a pleasure to read your summaries.
Cheers,
Dave
Clouds.
A great summary of NEMDEV 2025.
Thanks Dan
Very interesting. The casino that is NEM. Know when to hold them, know when to fold them, know when to run!
I wonder if there is a way of calculating how efficient the fleet was at capturing the high prices with the benefit of hindsight? Then you could track this statistic over time to see if there is a trend.
Suggest that recommendations from the Nelson review will be binned unless Bowen gets it legislated before the next election.
Looks like someone is either unfamiliar with the dead time of the wind systems, or they’re using a square wave to join the dots rather than the triangular wave used in the charts above.
Statement on Callide Power Station availability – Updated
27 Oct 2025
Updated at 5.15pm
https://www.csenergy.com.au/news/statement-on-callide-power-station-availability
Frequency plot consistent with the above – near vertical drop implying an electrical fault / trip. Would be interesting to understand the amount of load and solar shaken off with the voltage depression.
It seems Vales Point VP6 is offline since 28 Oct 1 am. At least it’s when reporting ended
https://explore.openelectricity.org.au/facility/au/NEM/VP/?range=3d&interval=30m
So coal generation on 28 Oct was only 4,780 MW at 21:00 instead of 5,336 MW a day earlier
Is this because Vales Point VP6 is off-line since 28 Oct 1 am?
In NSW last night Oct 30/Oct31 we had CONTINUOUS hydropower between 551 MW and 1,379 MW
VP6 back 15:30 31 Oct. Tell us what happened with Vales Point
I had done this post:
7/10/2022
Sale of Vales Point: How will Czech brown coal baron Pavel Tykac with
business registered in Liechtenstein impact on NSW coal and energy markets?
http://crudeoilpeak.info/sale-of-vales-point-how-will-czech-brown-coal-baron-pavel-tykac-with-business-registered-in-liechtenstein-impact-on-nsw-coal-and-energy-markets
A terrific article. Thanks. The science/maths of “emergent” effects is little understood and could lead to counterintuitive behaviours. But very, very interesting.
Here’s the raw hz data from Melbourne. 49.84hz is the lowest I saw
https://www.dropbox.com/scl/fi/x1t4c6wku2fw6dt21c2m0/hz-data.csv?rlkey=vz5w7ppo8m8umf1md09jrelxw&dl=0 (note it also has noise but
Interesting. Thanks. Isn’t about time we stopped approving large solar farms unless they have at least 30min battery backup to hold up output until scheduled generation can ramp up?
I was at the All Energy conference on the first day last week and there was no discussion about All Energy at all it was just about renewables and batteries.
They need to include all forms of energy in their discussions.
The number of batteries on show were enormous.
What an absolutely absurd policy by Labor. They created this mess by promoting renewables despite common knowledge of the duck curve.
Retailers will increase their shoulder and evening tariffs to compensate which is when moat people have unavoidable energy usage.
Free aircon but more expensive dinners and hot water.
When are all these battery farms getting recharged? Have we got this much solar power abundance. Why not shut the coal and gas fired power stations down now and save us from future hailstorms and cat.5 cyclones,rising sea levels and 50 degree temperatures and bushfires from hell
If the frequency signal was measuring fine, the error could have been in the AGC delivered set point telling generators to increase output, or in the controller’s understanding of actual supply.
Another error source may have been the automated wind and solar forecasting system? Could that have predicted less wind and solar about to occur?
If the frequency measurement was ok, could the error have gotten into the demand forecasting which in control systems is a feed-forward input to the controller?
Fun times.
Good series of articles Dan, all seem to be pointing in the direction of why local control of key electrical variables is necessary and showing the weaknesses of centralized control. Controlling a global system means if something goes wrong – it goes wrong everywhere.
If the idea is successful and 15 GWh/day is moved from evening-night demand to soak up curtailed wind and solar even if it pushes up the current midday price by $30/MWh to replace coal and gas at $130/MWh that is a saving of $1.5m/day in energy costs for the retailers.
Further for customers it will make switching from gas and immersion hot water heating to heatpumps more attractive, which will reduce household energy costs even more. One heatpump hot water service can replace almost two gas heaters and two immersion heaters with no net increase in electricity demand.
The alternative is to invest about $3bn in batteries to store the curtailed wind and solar and add that to the RAB
The reality is most people’s demand profile is morning breakfast and then afternoon/evening for dinner, dishwashing and showers.
Ultimately most people can’t shift those demands so what will happen is people will use their airconditioners in the middle of the day and maybe do a load of washing but then get slammed with higher pricing for the morning and evening when the majority of their demand is.
Hi David,
Quick suggested correction:
“that’s an additional 8.5 TWh/year additional demand”
This is out by a factor of 10, I’m assuming you forgot to multiply the annual consumption by the efficiency loss of batteries (around 10%?). It should be closer to 0.85 TWh/year of additional demand (still a lot, but a little less concerning!)
Hope it is not reminiscent of the 10 day outage in 2008, when an Antnov delivered a new 40MVA Aux Transformer for Xmas 2008, some months later.
Not often a Wilson transformer fails – especially not 2 out of 3. I wonder if they were wrongly specified (eg can’t handle the harmonics), or damaged during testing? Be interesting to get more “inside knowledge”.
Oh this is an interesting failure mode. It highlights there’s a risk when a large amount of data entry is done by a single provider a single update/patch/issue could heavily impact a lot of units. Might require some thought around ensuring diversity.
Hey Dan, nice paper
Did you look separately at daily max demand / demand at time of minimum reserve? And bucket by max demand?
Anecdotally AEMO was consistently under-forecasting QLD evening peak demand through Q1 2024 – you might hope there was a human in the loop to nudge the forecast in the right direction, it took their model several weeks at least. Constantly had a forecast lower than prior day’s outturn, despite similar or hotter weather…
It’s much harder to unpick now given the impact of rooftop solar, but you did used to see demands outperforming AEMO PD for some time before the forecast changed to something more realistic (see Allan’s forecast progression chart https://wattclarity.com.au//articles/2017/02/initial-analysis-sa-load-shedding-wed-8-feb-2017/ )
Thank you and I agree on all points.
I have made an attempt to create the visualisation you’ve described and it confirms this thinking. Will post as an article in the coming days/weeks (when I find time!)
See here: https://wattclarity.com.au//articles/2025/12/the-long-term-trend-of-forecast-errors-for-daily-peak-demand-in-qld-over-the-past-seven-summers/
Walgrove setting prices despite plenty of other available generators and very little volatility in the FCAS markets. Keen to get our detailed WattClarity treatment on this blip! Interestingly the same $14k got triggered again on the 21st morning.
Renew Economy wrote on 4th August 2025:
“The original timeline for the project was for it to be delivered in full by May, although that timeline was allowed to slip after the decision by the NSW to underwrite a two-year extension to the Eraring coal generator, that is now expected to close in 2027, although it may stay open longer.”
https://reneweconomy.com.au/australias-most-powerful-battery-is-now-officially-operating-as-the-grids-biggest-shock-absorber/
More on the slow-down is here:
https://reneweconomy.com.au/australias-most-powerful-battery-charged-up-quickly-and-helped-keep-lights-on-but-progress-has-slowed/
The original GHD phasing of 2023:
https://majorprojects.planningportal.nsw.gov.au/prweb/PRRestService/mp/01/getContent?AttachRef=SSI-48492458%2120230220T015054.349%20GMT
Very interesting dynamics on the constraints.
N>>NIL_85_86_S definitely hit max at 12:10, but I think N::N_CNLT_2 is an interesting feature.
Based on dispatch constraints…
N::N_CNLT_2 was at capacity the entire time, but around 12:15 it suddenly had the RHS drop by 300MW.
https://nemview.hgn.id.au/graph/dispatch_constraint/N::N_CNLT_2
I would guess this had a ripple effect combined with…
I also notice the NSW Wind was mostly unconstrained up to 12:15, and constrained down about 60% for the rest of the day.
Quite the interesting day.
Thanks, but of course an easy record to break every year given MPC is indexed 😉
Don’t forget that not so long ago we would get frequent intra half hour swings mpc to mfp. 5ms solved that.
Yes Ben both points are definitely true.
Interestingly, since the start of 5MS there have been relatively fewer occasions where a region has both hit the MPC and fallen deep into negative territory (well beyond “negative LGC” levels) in the same day. With the MPC making a step change up from July 1st — a 16% increase for FY25/26 — the gap between the market floor and cap has now widened beyond CPI too.
It seems to have been the Eraring battery ERB01
https://nemview.hgn.id.au/graph/dispatch_scada/NSW1/es2
Hi, I noticed that previously both the output and input DUIDs had a MAXAVAIL value. This makes sense because the charging capacity might be offered differently to the discharging capacity.
But the bi-directional DUIDs only have a single MAXAVAIL.
Is the MAXAVAIL considered symmetrical n for the bi-directional DUIDs ie same MW for charge and discharge?
Hi Ben, MAXAVAIL is a parameter submitted in the bid and bidirectional units submit a 2-sided bid for energy. The DIRECTION field in the bid distinguishes the ‘side’. It will be gen for production or load for consumption thereby allowing for a different maxavail on each side.
It looks to me like the lack of battery capacity close to Sydney contributed to the delay in pushing the price back down. There was only a 250MW increase in battery output at the time of the price spike on Wednesday, which was insufficient to cover the increase in demand in the Sydney region. Of course the lack of transmission capacity within NSW was setting this situation to play out, then dramatically compounded it when the demand spike occurred. Clearly relying on only one large BESS close to Sydney is not looking like an optimal solution. Even at full capacity the Waratah Super Battery would not have fully handled this sudden change in supply.
I think the implementation of the Negative Residue Management (NRM) constraints needs to be reviewed. The current process is producing unnecessary supply/demand shocks on the grid, which often occur at times when the grid is under strain. It would make far more sense to stage the introduction the NRM constraints over several intervals to avoid the shocks to the grid.
On NRM constraints, 1) they do cut in progressively, IIRC they don’t try to cut back counterprice flow by more than 100 MW per dispatch interval; 2) they don’t reduce supply available to the high price region, since they target reduced exports out of that region.
You write that “The rapid movement of clouds can cut rooftop PV output with little notice — and because this generation is largely invisible to the market operator, it adds another layer of uncertainty to their forecasting models.” but a reasonable proxy for that data is available via the large and growing network of home weather stations that measure and report insolation.
For example have a look at the Ecowitt map at https://www.ecowitt.net/home/map drill down to my weather station at https://www.ecowitt.net/home/index?id=168614 and you can clearly see the history and realtime data for insolation. Aggregate this up by postcode with a multiplier for estimated rooftop panel capacity and you are in business!
The web site will even provide the data in tabular format and you can export it. To automate this would be a very small project that could yield a relatively reliable forecast model.
Nice writeup Dan
I reckon you’ll find that the sudden curtailment of wind and solar shown in your charts will coincide with Snowy rebidding a lot of its Tumut generation capacity down to MFP in order to dispatch against the high spot price. By 11:15 those generators were running at over 1,600 MW, having been at 0 MW until 10:45. They sit lower in the firing line of the 16_8_39 constraint than some of the southern NSW wind and solar generators, which would have been first cabs off the curtailment rank.
Thanks for the article. It was interesting to read the following statement, “Gladstone has already been operating at minimum stable load (Pmin) almost daily for the past year. In effect, the plant is already mothballed.”
There is nothing unusual about a unit at GPS running at “Pmin”. Historically this may have been after evening peak demand. It also was dependent on which units on line were contractually supplying Boyne Smelters. The other units would at, or be close to, Pmin. The change is that this Pmin period has shifted and extended each day.
Regarding reliability, it should be noted that reliability of a power station depends on the resources available, including reasonable continuous funding for maintenance, asset refurbishment/replacement and associated periodic shutdowns. Since privatisation, GPSGOS (Gladstone Operating Services) has had a funding model based on reliability, not MWe (export). This funding was fixed and eroded dependent on having the contractually agreed number of units on line. This can be a challenging environment when operating a coal fired power station that has been operating since the 70’s. Particularly when you don’t have control over the quality of coal you need to burn.
Paul
Notwithstanding the solar forecast issues, isn’t the immediate problem that primary frequency response and regulation FCAS are not providing sufficient response to avoid these frequency sags, whatever their cause?
It seems arguable that AEMO ought to be enabling a lot more regulation volume, at least during solar ramp periods. This would have the physical benefit of keeping frequency closer to target – which after all is the key objective here – but also a market benefit in that it would raise the cost of regulation FCAS being incurred by the under-performers via frequency performance payments, so might prompt them to forecast better (or at least more conservatively)?
It could take a very long time to get any changes to the semi-scheduled generation category and rules. As far as I know AEMO could vary the volumes of FCAS enabled with very little lead time provided there is a reasonable justification for it. Which there certainly seems to be here.
Allan
Thanks Allan
My understanding is that the AEMO is exploring this as a possible first step … because (as you say) it would be a relatively simple/quick approach.
I’m wondering if doing that would be enough, as the Semi-Scheduled capacity continues to scale, and the mix of Scheduled capacity changes (perhaps with aggregate MW installed declining from current levels). Various people have looped me into different conversations where various options are being considered – at some point might have time to write about some of these.
Yep, sounds like a situation where multiple things might have be done, let’s do the quick obvious and easy to reverse ones first!
Never saw it coming..
The cost of coal is irrelevant. Burning it increases the cost to our future. Enter that of our children and grandchildren are such a large amount that will dwarf any current costs.
It would help the average non electricity espért to understand this article if you spelled out what the many acronyms stand for and what they mean , in layman’s terms please ?
What the data shows, so far, is circa 10% reduction in peak output compared to the average for the previous week. Exactly in line with the fire involving a fraction over ten percent of the area covered by panels.
Monitoring for a longer period will give us a better picture but suggesting that there is no long term damage seems premature at least
Thanks for the prod, Richard … note now that ‘Wellington North Solar Farm back to 94 ELAV compared to 104 in total’.
I just looked at the infographic. Coal gone but gas generation to increase?
Here’s one social media post of possible interest, with following thread, and a promise of more to come.
https://x.com/FootnotesGuy/status/1998653261516255727?s=20
Elav may not be such a good measure of availability of solar plant. ELAV reflects the number of inverters in operation – which may, or may not, reflect the maximum output available if damage to an installation affects strings only.
A week has gone by and peak output is still sitting 10% below its recent maximum. I can find no update from Lightsource BP
Dear Allan,
Regarding multiple generator contingencies, how do you suppose this will be implemented? Would a second layer of runway pricing be implemented from within the tripping scheme? My brain hurts trying to think about what a fair scenario would be.
Also, curious how the cost recovery for DPV tripping would be handled https://www.aemo.com.au/initiatives/major-programs/nem-distributed-energy-resources-der-program/managing-distributed-energy-resources-in-operations/der-behaviour-during-disturbances
Thanks for a great article as usual
These coal stations are getting old with nothing to replace them.
Based on NSW SCADA tonight (so may not be trustworthy), I think there could be 3 of 9 operational coal turbines in derating after 7pm tonight.
A NSW LOR2 of only 80MW remaining tomorrow came an hour later.
What happened to small scale solar? Could prices be elevated due to a monitoring / reporting problem with that data?
Worth noting against this post for future reference (though don’t take as official values)…
NSW market demand maxed around 16:00 AEST at 13,000 of the NEM wide 30,000.
NSW estimated underlying demand maxed around 14:00 AEST at 16,500 of the NEM wide 39,000.
That would annoyingly trigger fault recorders for those who have to monitor GPS compliance.
A quick look on the Zoom Earth cloud cover shows increasing cloud cover over Woolloongong, Sydney and Newcastle from 11:40am on into the afternoon. Could that be the cause? I would say a good chance.
What a year! Thanks for the wrap.
Thank you Paul for keeping us all updated over the holiday period, especially since Christmas Day was the big MSL day.
However, the focus is now on Victoria and SA for 1st January. Waiting to see what battery in Victoria receives directions if AEMO deem it necessary for MSL management.
“Whilst this notice speaks to ‘forecast MSL1’, the lowest level of System Load forecast here is below MSL2 level, as well”. Paul think your pickup here may not be the first time an MSL2 has been classified as a MSL1.
I believe the run to late 2022 was related to speculation the Albanese/Bowen team would use LGC demand mandates as tool to deliver their decarbonisation ambitions. Once the expanded CIS was announced in Dec 2022 that story began to fall apart, and the introduction of the REGO scheme has made it apparent the government has little desire for a meaningful certificate price in future.
The shape of AEMO’s estimated actual PV curve for SA looks pretty jagged even at half-hourly level. Fast changing clouds likelt part of the story.
2018 fall had a lot to do with a CER change of position on the morality of deferred surrender. I covered it at the time here https://www.energycouncil.com.au/analysis/traders-taxes-and-renewable-energy-certificates/
A classic example of one or more participants being able to exercise pricing power through the event. Have a look at which participant/s had online capacity at batteries or synchronised gas units – ie not fast-start GTs etc – offered at MPC!
I understand ‘small solar’ is basically an AEMO estimate of RTPV but what is ‘mediun solar’ and why is it invisible?
Thanks for the question, Bob
Should probably add a 3rd article in the series here:
Part 1 = ‘The opacity of distributed, small-scale solar PV output’; then
Part 2 = ‘The (ongoing!) opacity of distributed, small-scale solar PV output’
Basically:
Large Solar = includes any unit for which AEMO publishes SCADA readings on a 5-minute basis (e.g. InitialMW). So typically Semi-Scheduled and a few larger (Registered but) Non-Scheduled units
Small Solar = AEMO estimate of rooftop PV, which uses a sample-and-extrapolation method, with the extrapolation being governed by aggregate MW capacity of units registered in the SRES part of the MRET … so units no larger than 100kW.
So Medium Solar = everything in between … which, readers will understand, is a growing scale for which there is no visibility.
While not a direct explanation for batteries on the 10th, I note there were minimum load concerns from 11am (NEM Time). Demand was not as low as forecast, possibly due to shading of solar in Melbourne by bushfire smoke.
“132659
MINIMUM SYSTEM LOAD
10/01/2026 04:17:29 AM
Forecast Minimum System Load (MSL1) condition in the VIC region on 10/01/2026
AEMO ELECTRICITY MARKET NOTICE
AEMO has detected that there is an elevated risk of insufficient demand to maintain a secure operating state in the VIC region on 10/01/2026.
Minimum System Load (MSL) conditions are forecast when demand is less than the relevant MSL threshold. The advisory MSL thresholds are:
– MSL1: 2116 MW
– MSL2: 1616 MW
– MSL3: 735 MW
The demand is forecast to be below the MSL1 threshold for the following period:
From 1100 hrs to 1430 hrs 10/01/2026. Minimum demand is forecast to be 1783 MW at 1230 hrs.”
A little background on the VBB System Integrity Protection Scheme (SIPS):
The scheme reserves 250 MW of the VBB’s 300 MW capacity between ?November & ?March each year to provide AEMO the ability to run transmission between Snowy/NSW and Melbourne (the Murray-Dederang and Dederang-South Morang 330kV lines) at higher capacity levels. These are parallel circuit pairs and normally have to be run below a level that would not rapidly overload the remaining circuit if one line in a pair trips. The SIPS scheme allows AEMO to run the lines to higher ratings by triggering the VBB to rapidly discharge at 250 MW if there is a line trip: this extra supply near Melbourne reduces flow on the remaining line to a level that allows sufficient time for market redispatch if the tripped line remains out of service.
The VBB discharge events discussed in Paul’s article show the scheme operating when different lines on the Murray-Dederang-South Morang corridor tripped, presumably due to bushfire smoke or lightning.
Note that the market bids are “housekeeping” to reflect what the battery has physically done in response to a line trip: the SIPS scheme operates outside the market and doesn’t rely on the 5 minute bidding and dispatch cycle.
There was:
132687 RESERVE NOTICE 10/01/2026 06:39:59 PM
Actual LOR1
11 out of 12 coal units were operating. Despite plenty of solar power, they started to ramp up already in the early afternoon to reach around 7 GW for the whole evening
Of interest in this period, I think I saw ‘games being played’.
Based on the past days, NSW large solar drops to practically zero at NEW 19:15.
But I watched the P5 forecast through the hour up to this time, and there was bidding of way higher than that to the end of the P5 hour. Like bidding that 500MW will be produced at 19:45. Obviously, the very last bid of an interval was much closer to real output to avoid penalty, which means after 19:15 was zero.
But how can a forecast be trustworthy when there is such large amounts of false data in it?
I wonder what took up the slack between the forecast and final?
So youre speculating that the transformer will come online then?
Because that’s not testing, testing would have already been done offline. Soaking for a second time and then taking load is proving.
Noting that, on Thursday morning 15th January 2026 we posted ‘No, it looks like the entry for Waratah BESS was a Data Entry Error?’
Late to the party but here’s my frequency monitor data
https://cloudisland.nz/@xssfox/115903623773077534
https://www.dropbox.com/scl/fi/lbusy94nzder4il9gw1y4/hz-data-2026-01-16-18_39_40.csv?rlkey=4agdgqthw4afto4kf9fndlj3a&dl=0
Nice one Michaela! Captured the two steps well, and more clearly showing the first unit impacted frequency just on or moments before 16:55. Suggests, also, some latency in the timestamps of the MW readings.
The obvious question here is whether it was merely coincidental that immediately after the Callide C trips, aggregate off-target level for semi-scheduled solar leapt to nearly double the level it had hit in any previous interval on the day?
Or did the frequency disruption due to Callide somehow trigger some solar farms into underperforming over the next 5 minutes? – I can’t quickly think of a mechanism, but it does seem strange.
You should have pictures of C4 that blue up a couple of years ago
Given the evacuation of staff, the ‘control system fault’ may have been caused by a problem with DC systems and CS were concerned about a repeat of the events of May 2021 when C4 turbine disintegrated.
I noticed that in the https://nemoutages.com.au/mtpasa, QBESS will be out of action until 2029/01/13 which is a fairly decent time away without market participation given that its only been fully operating at full capacity around Jan-Feb 2023.
To clarify my earlier comment: I’m basing this observation on publicly available MTPASA data at the time of writing. As far as the eye can see with the data available to the public, QBESS shows as unavailable until 2029/01/13. I acknowledge there may be additional information not reflected in public datasets.
More articles being collated here, Jovi
https://wattclarity.com.au/market-operations/major-outages/2025-10-02-queanbeyan-bess-outage/
now described as an internal IT network overload caused by human error. Does that relate to the text above at all?
RE : wind farms dont like the heat… Not sure that is explained correctly.
Are you sure it’s not the air density that is lower during hot conditions that leads to reduced output from the wind turbines – I’d be surprised if their components care about being hot…
As a aviation pilot , hot weather doesnt bother the engine but the lower air density can double our runway requirements….
very cool blog!! Excellent .. Amazing ..I am happy to find a lot of useful information here in the post, thanks for sharing
Air density
Fan forced cooling in nacelle for gearbox oil & electrical components (or feathering of blades to reduce output)
Reduced electrical efficiency at high temps
Air density
Fan forced cooling in nacelle for gearbox oil & electrical components (or feathering of blades to reduce output)
Reduced electrical efficiency at high temps
Would really like to see a similar modelling for the hunter transmission project and northern inflows and wether this will eleviate flows into sydney. Looking at closures of coal in this region epecially with recent news of extention of closure dates.
Accumulated aggregate state of charge is a good metric.
I expect that when EnergyConnect is fully commissioned the SA volatility will flow into all the other states. Will be interesting to try and track that.
Ben,
Been said before – the state that can’t(SA) will be used to put the finishing touches to those that (barely) still can.
And very high power prices
Widespread acknowledgement that increased renewable electricity generation is driving down wholesale power costs. No mention of politicians agreeing with this. Hopefully the current quarter will continue to reflect the increased storage capacity, solar and wind generation. And if so, what effect will continuing wholesale power costs at lower levels have on retail power costs. Just noting that jumps in retail power costs due to removal of state and federal subsidies have contributed to recent increases in cost of living.
The customer focussed and missing part or this analysis is “what has happened to the average retail household electricity costs and bills?”
Has the average household electricity bill increased or decreased in the quarter? . What are the reasons for the changes?
I was surprised by this, but as a fairly new battery owner and Amber customer benefited. Looking at Open Electricity trying to work out what might caused it, significant exports from NSW to VIC seemed to occur around this time while VIC started curtailing Wind and Solar. (I know there are a lot of other factors I can’t see and am not aware of that may have contributed.)
Hi, can you explain why a bunch of wind sites appeared to be either dispatched down, or if the wind disappeared?
For those in control of such a large response, it is in your interest to go to an effort to voluntarily become part of the scheduling process as, by directly influencing price, you will moderate these spikes rather than having to react after they are published, or, when restoring, you can avoid provoking a new spike.
This “analysis” is embarrassing to see on Watt Clarity. You guys are better than this.
The article is fine until the author seemingly decided it was too difficult to work out the impact on coal generators and instead decided to assume that every coal generator in NSW would run at ~50% load when prices are at their peak. If the headline wasn’t based on this final assumption, it wouldn’t be such an issue.
I’d love to see an analysis diving into the dynamics of greater battery capacity on the grid. Including higher daytime pricing, reduced emission intensity (less gas for peaking) and coal retirement projections. It seems that batteries won’t have a huge impact on coal (compared with their impact on other generators). There won’t be enough of them to justify bidding lower than coal to stop coal from matching demand (minus wind) during the night.
Date: February 2026
From: A Retired Power Plant Operator (20+ Years Experience)
Subject: Formal Record of Technical Negligence and the Path to Professional Safety
1. EXECUTIVE SUMMARY
As a former Marine Engineer and 20-year power plant operator, I am providing this record to ensure the 2021 Callide C4 explosion is never repeated. The incident was not a random technical fluke; it was a consequence of bypassing engineered safety barriers for production goals. To restore safety and insurability to the Queensland energy grid, we must move away from “in-house” company training to a Statutory National Standard for plant operators.
2. THE TECHNICAL “SMOKING GUN”
The 2021 disaster resulted from two specific, documented deviations from core engineering safety principles:
The Interlock Bypass: Maintenance protocols bypassed physical Castell Key Interlocks. These “trapped key” systems are designed to make it physically impossible to isolate the 220V DC protection supply while the turbine is online. By bypassing these, the turbine’s “emergency brain” was killed while it was generating power.
The Lack of HV Isolation: Management opted for “live” maintenance on critical protection circuits to avoid a planned outage. Under standard engineering practice, High Voltage (HV) Isolation is the only safe method for this work. Without it, the unit was left with zero redundancy, leading directly to its kinetic disintegration.
3. THE ECONOMIC COST OF NEGLIGENCE
The “white-wash” narrative was dismantled by the following financial and legal facts:
Total Financial Impact: Taxpayers face a bill exceeding $1 Billion for repairs, rebuilds, and grid stabilisation.
Insurance Crisis: Following the failure, CS Energy became virtually uninsurable; premiums surged by $40 Million annually.
Legal Penalty: In February 2025, the Federal Court issued a record $9 Million fine against the operator for breaching National Electricity Rules.
Will the role and actions of the Federal Minister be scrutinised at the same time?
Reneweconomy-tier analysis from WattClarity with this article. Sad.
Thanks Bill. I’ll go back to Reneweconomy. By the way other than the podcast I have no professional relationship there. I contribute to put something back after a fairly successful career in in investment banking research. I have a huge amount of time and respect for Wattclarity and the quality of its work and I’d hate for it to be diluted with my “reneweconomy” tier. Very happy to have my points debated, never happy but will always accept where I am wrong and actually over a very long period of time have mostly kept towards the front of the analysis curve. If you want to be early you will be wrong. Perhaps batteries wont have any impact on evening coal volumes any more than solar had an effect on daytime ones. I would be open to the point that charging the batteries has an impact on midday volumes but equally there will likely always be plenty of low cost solar. Still I abandon the watt-clarity field. It’s not that I mind criticism Ive had a fair bit of it in my career and from very very senior people, but I really don’t want to upset the quality diet that Wattclarity has. regards
Interesting, Oliver.
I’ve questions about the attribution to the structure of bid prices following the legacy format (because your chart of ‘average variance’ starts escalating at 2016 … which is coincidentally when we started noting the changed pattern for Q2:
https://wattclarity.com.au/areas-of-deeper-analysis/review-of-price-outcomes/review-of-q2-price-outcomes/
More later on that.
But I also found Gabriel’s comment of interest as well:
https://www.linkedin.com/posts/laeragabriel_cutting-through-the-noise-activity-7427641570604093440-QPbw/
Why have Hydro and Gas units have largely refused to bid at the floor?
Seriously guys – are you suggesting that all electricity consumers should engage in this process in order to get their electricity bills under control.?
Why not just go off-grid and be done with it? (And do it at one’s own expense?)
And why not reflect on the incredibly generous taxpayer cash splash that drives this behaviour.
Then there are all those consumers whose residential circumstance preclude using panels and batteries but are being obliged to help fund them anyway.
And if this circumstance is not fair and just as it stands, pray tell what taxpayer subsidies will flow when all existing batteries and panels (and wind turbines) reach end-of-life status?
Let’s be honest – most people just want reliable and cheap electricity whenever they need it!
And an electricity grid based on nuclear power is the only honest way to deliver that!
A really insightful analysis.
A few things which would have been great to see in addition:
1. Energy inflow breakdowns to battery, load and grid export on Sankey diagram.
2. 12% energy lost from battery?! Much higher than I would have thought (and what is generally touted by industry). Is this the efficiency reality, or are there other factors contributing to “battery losses”?
3. Hot water being the largest energy consumer is surprising too. Resistive system? Not much distance travelled in EV? AC used infrequently or otherwise efficient building envelope (or moderate climate)? Some discussion on this would be illuminating.
4. Any import earnings (i.e. getting paid to consume)? Or are the negative wholesale price instances needed to overcome the various variable network/market charges just not there yet?
5. In Fig 9, another useful comparison would be to add the upfront capital expenditure to the solar + battery curve in a new data series and adjust all $ numbers for inflation. Could include another series with un-subsidised capital cost as well (if known). These additional series would show the overall payback period (eventually) and should include the costs of the EV & charger (if counting the fuel savings in the BAU data), the 2nd battery, and any other infrastructure, efficiency or fuel switching changes that have been made. It feels like the true payback period might be a while off yet, but that’s not accounting for any non-monetary value gained from the whole exercise.
Can we see any trend in battery charging costs either increasing or decreasing?
I ask because higher charging costs will translate into higher selling price and associated higher wholesale prices eg will batteries eliminate negative price periods?
Hi Ben — I’m not sure I agree that it’s valid to think of discharge prices as ‘charging cost + a margin’ in a cost-plus sense; both charge and discharge outcomes are mainly set by system conditions and volatility, not by what the battery paid earlier.
In the animation in the article, we can see that charging prices sit in a much tighter band than discharging prices, which strongly suggests that the relationship between the two isn’t linear and the spread is often dominated by what happens on the discharge side.
It’s difficult to understand why AEMO would resort to unrealistic – even impossible – capacity factors as inputs to their planning.
My view is that AEMO was entirely correct to provide pure capacity factors. What is actually achieved is of course different. In ITKs LRMC modelling we adjust for capture ratios as we call them which allow for both economic and physical curtailment.
And although we are talking wind in solar it’s the same thing only more so. And we have what used to be called an inverter load factor but now I think ACDC ratio is the more common term.
Within reason the underlying pure wind capacity factors will remain constant but economic and physical curtailment and availability may well change.
Good read.
Wind traces are generally produced using one of the major climate reanalysis suits (ERA5, MERRA2, BARRA-2). I’m not sure which one AEMO uses but many studies use ERA5, which significantly overstates annual capacity factors. The Australian BOM’s BARRA-2 is much better. We published an analysis comparing actual wind farm output versus modelled output using the reanalyses above and found that all the reanalyses can produce good results at times, but the difference between modelled and actual is highly variable. Modellers should be aware of the gaps between actual and modelled.
Validation of BARRA2 and comparison with MERRA-2 and ERA5 using historical wind power generation
https://doi.org/10.1071/ES24028
G’day Paul, I was wondering if you knew where I could find a more detailed table (in this format, ideally), including ‘Native Demand’ and ‘Regional Demand’? I know you have this one from 2018 https://wattclarity.com/articles/2018/04/an-explainer-about-electricity-demand-take-1/, but that is missing ‘Underlying Demand’ and ‘Market Demand’.
The failure at Callide C4, which resulted in a $500 million loss, was attributed to a combination of technical failures and systemic organizational shortcomings, with the primary responsibility resting on CS Energy, the plant operator. Forensic engineer Dr. Sean Brady’s report, released in February 2025, concluded that CS Energy failed to implement effective process safety practices, including the absence of formal risk assessments for critical switching procedures, which led to the catastrophic failure of the Unit C4 generator on 25 May 2021. The report specifically highlighted that the decision to carry out a switching sequence with the unit online—despite known faults in the 220VDC switchboards and the absence of redundancy—was a direct result of inadequate management oversight and a culture that did not prioritize process safety over cost savings.
The failure was triggered by a cascade of events: a voltage collapse in the new battery charger, the loss of the AC supply due to a false arc flap protection trip, and the failure of the automatic changeover switch, which had been disabled since January 2021 due to a low-consequence classification.
CS Energy’s processes were found to be deficient, with no formal requirement for risk assessment of operational changes, and a lack of a dedicated process safety team from April 2019 to July 2020.
The company’s shareholder mandate, which emphasized cost reduction and debt management over safety, created competing pressures that undermined safety protocols.
The insurance company refused to pay the $500 million claim, citing the failure to follow proper procedures, which further underscores the organizational negligence.
As a result, Callide Power Trading, a subsidiary of CS Energy, was fined $9 million in February 2025 for breaching the National Electricity Rules, with the court emphasizing the need for a deterrent penalty to prevent future failures.
The responsibility for the ignorance that led to the failure lies with CS Energy’s leadership and governance structure, which failed to enforce safety protocols, despite repeated warnings and known risks.
How is the money actually raised to pay Basslink? Is it simply extra TUOS charges in Tasmanian and Victorian consumers bills? On Ruth Forrest’s website she said the money comes from the IRSRs (https://ruthforrest.com.au/uncategorised/hydros-challenges-are-part-of-a-system-wide-pattern-affecting-every-other-major-public-entity-in-the-state/). And I have also heard that the money doesn’t come from the IRSR but from the proceeds of the IRSR auction…
Any information, especially with AEMO or NER sources, would be really appreciated.
Hopefully my Woodside shares will do a little better!
So are the Retailers buying peak hedges elsewhere, or just found other easier ways to hedge, swaptions etc.?
Visibility of trade on the OTC exchanges (e.g. ToD Markets) would be a valuable addition to NEMReview.
Could the problem be that the power electronics or cabling is undersized with insufficient heat removal capacity and the thermal overload alarms come up? The plants may have also been built down to a price point and the cooling was downsized as it was regarded as just a nice to have.
There are two effects. Air density at 40C is 10% less than at standard testing temperature of 15C so energy in the airstream is correspondingly 10% lower and all machines,cooling sytems and electronics are less efficient, particularly cooling systems as the cooling effect of a radiator is dependent on the Delta T and the mass flow which are both lower.
Ummmm – can I suggest you find some representative sites to say what the wind is doing rather than the temperature. Dare I suggest the temperatures became very high when the wind dropped and you will note the error declined rapidly with temperature decreased….dare I say when a breeze came in and the windies actually did something. Simply highlights the vagaries of trying to forecast the output of these things. We see this characteristic all the time over here in the west in summer – on hot days the easterly winds die and nothing happens until the sea breeze (south westerly) arrives. On really hot ones there is no sea breeze so we get nothing until the easterlies start up again in the late evening / night.
At midday, every state except Queensland has demand outstripping supply.
Wind generation is sitting at 1GW from a nameplate capacity of 13.5GW and dropping fast.
It appears FPP calculations were impacted as RCR, Usage and FPP settlement volumes were not published from 11:10 to 12:40
Particularly striking (and I think underdiscussed?) is that transport fuel security & 1-in-N-year electricity system resource adequacy look to have the same least-worst solution: giant tanks storing a few dozen PJ of diesel to hedge domestic transport needs against foreign conflict *and simultaneously* hedge dual-fuel GTs against gas network constraints through long periods of high residual demand. That’s two rare and uncorrelated events, so there’s no reason you can’t use the same assets to do both simultaneously while deploying renewables and EVs as fast as possible to reduce fuel demand wherever possible.
Hi, what does ‘electricity’ fuel group represent?
Hi Ben, those are demand response units (wholesale demand response).
I should also add, there are more scheduled and semi-scheduled units in NSW. We only charted the ones that rebid.
“…the coal trains are (I believe) all electrified…”
All coal trains in the Hunter Valley run on diesel, whether to the Port of Newcastle, or to the various coal fired power stations (Vales Point, Eraring and Bayswater).
Thanks for helping correct that one, John.
great article
the big thing im wondering about is what impact battery will have on FIT’s because as you noted the wholesale price of energy is down to $0 in the middle of the day
even a blind man can see this as easy arbitrage but my question is how much battery can be installed and slurping up that cheap $0 excess solar before the cheap excess solar runs out?
if you have say 10gw of batteries slurping up cheap daytime solar but only 5gw solar being output are we going to see daytime prices smooth out to nighttime prices?
will FIT’s start to go back up again and a restart of the solar/wind boom where batteries want more cheap renewables installed so they can keep the easy arbitrage going?
Thanks — my sense is that the dynamic might play out a bit more slowly than some anticipate, given there’s already an enormous amount of solar capacity when you combine rooftop and large-scale. On top of that, there’s 3+ TWh a year of curtailed solar (effectively more $0 daytime energy):
https://wattclarity.com.au/articles/2026/02/keeping-up-with-the-curtailment-2025-beneath-the-headline-numbers/
Less about “what happened at Macarthur Wind Farm” than “what happened in the operations area(s) meant to be overseeing its market dispatch?”. Its bids appear to be made on a set-and-forget basis, and no-one noticed that AEMO’s AWEFS availability forecasts somehow got stuck at ~80 MW for nearly 18 hours, well above what it could actually produce.
excellent article. Not sure how I missed it earlier.
Sorry, but storage is not critical and coal closures are not accelerating.
The ‘market’ is broken because of the ongoing interventions ie subsidies and targets, and more of the same is not the answer.
The duration of a battery is just its inverter size. A 2hr battery can operate as an 8hr by simply discharging at 0.25 of its max power, but it can ramp up to cash in on price spikes. For example, on 5/6 April the NSW price was ~$75 all night, so Eraring discharged at half power from 5pm to midnight, so it was effectively a 7hr battery. I don’t think we’ll get many 8hr batteries because their ROI will be slower. Having a smaller inverter is cheaper, but it limits your ability to cash in during price spikes.
In a cooperative system it would make sense to have 8-16hr batteries. But we have a competitive system where they need to take advantage of negative prices to charge (typically 4hrs per day) and peak prices to discharge (typically 1-2hrs per day).
Price volatility will be with us for some time as whenever batteries get on top of things, that will trigger another coal plant to retire, which will set off another round of volatility.
“Pumped hydro highlights the issue. It provides long duration storage, synchronous system services and asset lives measured in decades, but much of its value is not captured in project revenues.”
Hmmm. I wonder what other low-emission generation technologies might be favoured by the called-for “market design changes that deliver long dated revenue support aligned with asset lives”.
Seems to me to be insufficient governor action. Add up all the non-conformance in solar and wind generation and compare to steam and hydro generation units off Target. I reckon the system is too light on inertia. We are headed for a serious system event (reference Spanish system collapse)
550+MW of over performance? Call it what you like, but I call it Nonconformance. 550 is a huge matter, particularly if all the governor controlled units are near minimum load. This is serious territory that cannot be spoken about lightly. I hope the Under-frequency load shedding system will save some sections of the system on the way down.
Not quite. The answer to your observation was in the 10th April write up. It talks to a 17 day combined outage. Not surprising – maybe a bit of scope creep since April. Lots of common plant in these sorts of stations and it talks to switchboard replacements etc which is never quick or easy. Also talks to a control system replacement so common plant also has to be done.
Great chart Paul
Although PEC-1 operation has seen clear increases in the maximum export and import limits for V-SA, a similar chart showing average limits applying in dispatch would show reductions in both directions, particularly on imports: flows from South Australia into Victoria.
This counter-intuitive outcome results from PEC-1 inclusion causing V-SA to appear in a number of regularly binding constraints that it wasn’t previously subject to – for example V^^V_NIL_KGTS. The reasons for this are outlined in detail in the article you linked above.
Quite frequently these limits have forced counterprice flows out of Victoria into SA when the “natural” flow direction would have been opposite, one reason why average flows have trended more strongly into South Australia since the advent of PEC-1.
Allan
ER01 off line, BW02 off line
I would be curious to understand if there are any requirements from Solar OEMS to provide AEMO access to there API for data on behind the meter solar and Grid data, the amount of curtailment occuring ect. understand and modelling this is imporant and would make the Soalr inverter OEMs more accountable for clear and concise data. If its not it should be in the pipeline or being written into the standards otherwise made mandatory to gain CEC approval and rebates. Making the chinese OEMS accounatbe and provide clear metrics, when normally that stuff is hidden.
And the free lunch is coming with 3 hours free power in the middle of the day.
Let’s not worry about the price, suggest worry about where the energy is going to come from on days like this when all of the nasty black stuff isn’t there anymore. It won’t happen every day but these events are far more common at this time of year than the green lobby would like you to believe – and the wind drought goes right across the bottom of the continent. And the quantity of energy needed to plug the hole is far more than Snowy#2 and a few batteries will provide. Who is building the GT’s and the gas (liquid??) infrastructure needed to fuel them? Business model will be challenging.
@Andy
Snowy 2 will have the energy for days like this, just not the power (batteries by 2030 vice versa). IMO the NEM needs a capacity system for gas so it gets built even if it barely gets used, and I’m not sure if the ESEM will perform this function. I don’t think gas fuel will be in short supply at this time of year (Iona currently at 102%) given batteries are taking away its intraday shaping role but we’ll see what happens..
Love that your alert needs to distinguish the non-rustbucket coal units from YWPS and GPS.
Using my experience with Domestic Battery traded on the Electricity Market through my supplier, the trading software is not informed on the severity of the lack of solar for a General Rain Event. My 13kW solar panel could only manage around 1 kW in the middle of the day, so the solar could not satisfy my domestic usage, let alone recover the battery sales from the previous evening. I was forced to intervene and charge the battery in the “Off Peak (1100-1600hrs) period to have secure supply during the next 2 nights and ensure the battery did not reach a discharge level that could damage the battery.
It seems to me that the major battery operators should take notice of the forecast size of rain events coming across the country and not be so aggressive in trading capacity the night before. That way they can remain competitive for unforeseen System Events and trading the next evening if the rain event is prolonged for a couple of days.
My experience is weather forecasts are useless beyond 4 days ahead. Can you do an analysis that looks at RMS error in estimate vs days ahead. I suspect you’ll see 3 days ahead is where it gets fairly reliable.
For a grid without coal, 48hrs accuracy is really all you need. Running 2GW of gas for 48hrs is a lot better than 10GW for 10hrs.
Good points Mark. We also need to keep in mind that weather-driven uncertainty in demand likely exhibits relatively rapid growth beyond the same 3-4 day horizon too. With an energy-constrained supply side it’s going to be those first 3-4 days in the projection that are most important for reliability over the succeeding days anyway.
What will be interesting is whether the current spot market mechanism will be enough to bring the gas turbines online early. The batteries should foresee the VRE shortage and withold capacity in hopes of cashing in on the price spike. That should increase the spot price, but will it bring gas turbines in early enough? 48hrs @ $400/MWh is cheaper than 1hr @ $23,200/MWh
Is your NMSE normalised to nameplate rating? I.e. a 100MW unit with 0.2 NMSE has an average 20MW error in its prediction. If that’s so, then anything over 0.3 is bordering unusable for estimating shortfalls. If you’re estimating 50MW, but on average it’s either 80MW or 20MW, then it’s not a very useful forecast. CLRKCWF1 being at 0.4 after just 2 days is a real problem.
I would have liked to have seen the 0-2 day numbers, even though it would be crowded.
Most of them “stay in their lane”. MEWF1 is an outlier as it starts in the upper band of poor predictions, but after 6 days is in the lower band of better predictions. Nearby KABANWF1 does the same thing. Both are low on your first “figure of thirds”, so they’re wind farms that spend 60% of their time below 40% capacity. If they’re below 40% most of the time, then predicting the output within 33% after 6 days is easy. WAMBOWF1 also fits this, but MCINTYR1 does not, even though it’s also a “consistently low” unit.
Excellent point about overestimates being more problematic than underestimates. An underestimate results in a little gas be used when it wasn’t needed. An overestimate could result in load shedding. Maybe it’s better to plot both average underestimate and average overestimate.
Could you pick 3 wind farms with high, low, and typical predictability. Make a plot with y-axis showing error in estimate from -60% to 60% nameplate, and x-axis showing days ahead forecast. Average overestimate (estimate – actual > 0) will of course curve upwards (because we’re only averaging positive numbers) and average underestimate (estimate – actual < 0) curves downwards in the negative axis. Plotting every 6-day forecast would get too garbled, so ideally a colour density plot of how often the error in the estimate is at that point. More red area below the x-azis in the density plot would indicate systematic underestimate.
Hi Mark – Thanks fore the comments. The nmse is the mse normalised with the mean square of the actuals. Would love to dig into this further, along the lines you suggest… when time permits!
It would be great if you could do an article looking at the SA yields over the past 2 weeks. There were a few points where total output was below 10MW, this would be a record low in the context of the total installed capacity.
Should all home batteries be called “price responsive”? Most are essentially off-grid with grid backup and don’t respond to price signals at all. They’re effectively removed from the grid, except for some solar exports in the afternoon and some imports during cloudy weather. Only VPP/Amber and other peak-bonus plans (Globird Zero Hero) are price responsive.
Hi Mark, the two halves of the slide are different datasets:
So no one was calling all home batteries price-responsive — the 2.8 GW is just the price-responsive subset of behind-the-meter batteries. Apologies if that wasn’t clear.
Ah ok. The numbers vaguely lined up (420k installs × 5kW export limit = 2.1GW, plus a bit extra for 3-phase systems). What’s the stat for total home batteries? Most recently reported numbers seem to be just CHBP only.
It seems like AEMO must be factoring the whole inverter size into that 2.8GW number. I’d use export limit rather than inverter size. Seems wrong for AEMO to claim there’s 2.8GW on tap during a high price event when most of that will be blocked by export limits. Obviously it gets complicated with dynamic limits. Also they will likely be self-consuming some amount, but that’s pretty much a given that they will always self-consume regardless of price.
Is there a stat for number of UPRR batteries?
Hi Mark, good catch — off the top of my head I don’t have a total home battery number beyond what DCCEEW are reporting, and I’m not sure if AEMO publishes a UPRR battery count (vs just capacity).
Better data about behind-the-meter storage is needed, in my opinion.
To clarify, the ISP doesn’t produce generation mixes based on modelling outputs. The ISP starts with demand and generation mixes decided up front – all inputs.
Then the ISP fits transmission to that.
Excellent article, Oliver!
It was a “complex chain of factors” and “unforeseen circumstances” that could not have anticipated that the 220 VDC was isolated!
They had removed or never installed the mechcanical interlocks that had been used for many years to stop just that from happening in most power stations!
Simple, but super insightful chart, Paul, thanks so much. It emphasises how challenging it is to optimise batteries through periodic dunkelflaute. So easy to get wrong, yet I suspect will set our future critical reliability limits.
I wonder if 2-3 day-out market forecasting systems are as good as they could be? They are obviously crucial to the process.
Well thought out, and indicating risk and outcome considerations often left out of corporate and government analysis. And absolutely not sensibly considered very often in political dialogue bereft of long term considerations e.g. claims like “net zero by 2050 is a proven failure” being made in 2025. In the context of this article I specifically worry about “…Gas-powered generation is the reserve capacity the system leans on in precisely the conditions that produce unserved energy, yet gas supply and transport limits are too often left at the edge of the model. ” If I look one and then two decades out, I think there is much less gas being used in homes, small businesses and probably a number of large users. Removal of uneconomic retail gas networks is already occurring at a small scale while states are banning new connections and/or requiring higher up-front payments. Some are also subsidising the move off retail gas with a general intention of ending it. How far do these trends go before the economic rationale of maintaining long distance pipelines for gas becomes undermined noting the growing age of much of this infrastructure. Where is the new gas burning backstop generation going to be built and what supporting sources of gas will exist by then. Anywhere south of the Qld border seems doubtful. Has anyone really modelled the possible futures of gas transmission and distribution over the next 10 to 40 years? If we only need gas for backstop generation ie. rarely but with high volume for days, is it at viable cost even with large subsidy for the pipelines and storage as well as the generators?
SMH: https://www.smh.com.au/politics/federal/business-use-and-energy-hungry-ai-tipped-to-drive-doubling-of-power-grid-demand-20260624-p609rb.html
ABC: https://www.abc.net.au/news/2026-06-25/batteries-boom-wind-stalls-aemo-isp-plan/106837876
I wonder what would happen if Fed & State policy conflicted, how would that get resolved in terms of AEMO modelling the ODP ‘while meeting government policy’? Eg. Fed Net Zero 2050, Qld extends life of Kogan’s Creek to 2052.
Would be good for AEMO to produce a counterfactual of ODP with only national carbon targets but no other gov policy (eg. No offshore wind in Vic, no state targets).
Should be no real surprise. The WEM has seen about 3 weeks of low renewable output- a bit longer than usual for this time of the year but increasingly the norm. the couple of great days were cloud cover and no wind. Every scrap of remaining coal ran base load to make the energy stored in the batteries. Just as well the carbon filters weren’t on. For SA however it does beg the question what is going to keep the lights on when TIPS B and Osborne aren’t there. Where is the market signal to put the capital up for OCGTs to deal with these sorts of events? Whilst your topic here was which was worse, 2019 or 2026 this is still less than a 1 in 10 event and happens pretty much annually to some extent. But you won’t make money out of a power station that runs only a few days a year without a capacity market.
Hi Paul,
To assess the recent wind drought, I selected a capacity-weighted mean location across current South Australian wind farms and extracted ECMWF IFS 100m wind speeds at that point. After applying a 4-day running mean, the recent period came in at just under 3 m/s — and looking back through the 9 km resolution model record, you’d need to go all the way to 2017 to find anything comparable.
The ERA5 reanalysis, while coarser at 31 km resolution, extends the record back much further. There’s a 5-day lag in availability, but it should be valuable for putting this event in historical context — how rare is a wind drought like this, and how often has South Australia seen something similar?
*Using the Capacity-weighted mean location is a huge simplification, but it’s quick and practical.
I will provide the similar ERA5 plot when the recent event values are available.
The ECMWF IFS plot can be seen at:
https://www.linkedin.com/posts/david-duncalf-3a76738a_to-assess-the-recent-wind-drought-i-selected-share-7476607188489117696-_HyW/?utm_source=share&utm_medium=member_desktop&rcm=ACoAABL2tW8BD-z-K9gs-qD03PmZXz81JZm5FlM
This is really interesting. The three BESSs are using grid forming technology with a contract with Transgrid to provide system strength in that region. And they were directed to disconnect due to system strength.
…..especially when BESS is touted as the darling ‘System Strength King’
With Amber electricity and price spiked huge this evening. Was curious to find out why.
Whilst BESS growth does impact the intraday price trough, by far the larger driver is that LGCs (which VRE offer their energy at roughly the negative of) have dropped from somewhere around $50 to well below $10
Thanks HR10 … your comment was one of the factors prompting the article ‘Spot LGC prices down the gurgler…’.
Absolutely a factor in raising prices, LGCs falling from ~$50 to under $10 does make VRE bid less aggressively negative.
But that gets you shallower negative prices, not positive ones, crossing zero needs an actual buyer bidding positive, which is what the charge-bid data shows batteries doing in QLD and VIC.
Both matter, but I’d push back on “by far” for the positive crossing specifically.
For QLD vs VIC, I’d argue latitude is the biggest factor. QLD seasonal demand tracks quite closely with solar output – highest demand in summer, low demand in winter. Lowest is late May and early Sep, so not quite a perfect match to seasonal solar output, but certainly an enviable position to be in. To a large degree, rooftop solar is meeting all their demand growth and seasonal variations. VIC is opposite with highest demand in winter (and growing due to degasification [is that a word?]) and low-ish demand in summer (barring the occasional heatwave). Summer and winter demand are about equal in SA and NSW, and that is where fuel mix and interconnector dynamics are the main differences.
Agree that latitude is one of the factors. A big part of it is also that VIC and SA have highly correlated wind patterns (and a relatively high amount of wind capacity). VNI limitations too obviously, amount of ‘baseload’ coal in the fuel mix, etc. are other reasons why we commonly see a north vs south divide.
Very interesting thank you,
what is your opinion of the eventual flattening of the 6pm afternoon peak, due to BTM battery systems, and what effect that will have on retail prices?
Thanks Freddy. The impact of behind-the-meter storage is hard to quantify but the reported numbers coming out of the Cheaper Home Batteries scheme point to the fact that it will/is going to have an impact. Some of the AEMO numbers suggest installations of behind-the-meter storage aren’t uniform across regions (i.e. QLD installing at a much faster rate). I covered this a little bit during my presentation to the EUAA back in May:
https://wattclarity.com.au/articles/2026/06/reflections-on-supply-and-demand-across-two-very-different-industry-conferences-last-week/
Re: retail prices — some impact, but obviously retail prices are made up of a lot more than just the wholesale component.