Transitional challenges require greatly enhanced analytical capability (exemplified in MT PASA Rule Change proposals, for instance)

Yesterday evening, I spoke (with co-author Jonathon Dyson for his Greenview team) in Sydney at an event organised by the Australian Institute of Energy about the Generator Report Card.  The crowd was good, and the questions were great (with some spilling onto social media, into email – and it’s too early for the carrier pigeons to have arrived).  Thanks to those who attended.

On that note, we’re looking forward to a similar Q&A session with AEC members (and others welcomed as well) in Sydney next Wednesday evening, 21st August.  I expect that the questions will be different, because the audience is different again.

However the story today is about something else I was able to do yesterday – afforded by the opportunity of being in Sydney yesterday, and with submissions closing today at the AEMC.  I was invited to meet with the relevant people in the AEMC to speak about our views on two rule change proposals submitted by ERM Power relating to the AEMO’s MT PASA process (in part due to prior comments at the end of this note here on 7th August), with suggestions both for:
1)  Improved transparency of data flowing out of it along with an improved methodology; and
2)  An extended data range included in the process (to line up better with the specifications of the Retailer Reliability Obligation, amongst other thing).

In talking through the proposed changes (and the challenges we’re already experiencing with the energy transition) with the AEMC, the parallels between what ERM Power had proposed and what we had summed up in Theme 14 within Part 2 of our Generator Report Card were unavoidable:



We summed up this theme by noting that the “Transitional Challenges require greatly enhanced analytical capability” (and talked through a few examples of how we have seen this present during the AIE session).

We did not have the specifics of what ERM proposed in mind when we wrote the report card (released on 31st May – nearly 3 months ago) – however we do believe that the general thrust of what ERM is asking for is aligned with the general sense of what we were stating under this theme:

It’s worth noting that what ERM has proposed was through two different rule change proposals – but the AEMC have (thankfully) chosen to address them together.  My shorthand (with limited time) of what the ERM are proposing is as follows:

ERC0270 Rule Change Request:

MT PASA ”Transparency and Accuracy Amendments Rule Change Request ”

ERC0271 Rule Change Request:

MT PASA ”12 Month Extension of Duration Rule Change Request”

This first part of the  ERM submission is here.

If we’re talking about “parts” of a rule change request, then this one would seem to be able to be considered in 2 parts (making 3 parts in total, including ERC0271).

Part 1 =  Firstly, to extend data transparency into the MT PASA time horizon:

1a)  It was late in 2010 (a long time ago now!) when real-time generation data began being published for all Scheduled and Semi-Scheduled (and some Non-Scheduled) generators – a great boost to transparency.

1b)  the proposal asks that DUID level available data also be published with each weekly update (or ideally every 3 hours, we hope!) with the MT PASA Region Availability data sets.  Ideally this should be done for all DUIDs for which real-time data is also available – not just Scheduled Generators.

(i)  This would include Semi-Scheduled plant
(ii)  It ideally should include significant Non-Scheduled plant (using the same logic as used to determine their need to be visible in real time) ; and
(iii)  It should also include the DUIDs that would be created for the new “Demand Response Service Provider” category that might be implemented (discussed further here – notwithstanding the concerns raised previously here about mooted change)

1c)   there’s also a suggestion to require publication of the forced outage rates assumed in the probabilistic modelling.  Having invested effort through the Generator Report Card process in trying to interpret forced outage rates purely from the public data published in the MMS, this would be a welcome addition.

1d)  finally, there’s also suggestions about what specific measure of demand should be published in the data set (see this prior article for discussion about the confusingly large number of measurements)

Part 2 = the second part relates to the methodology used in the MT PASA process.  Recently this moved from a deterministic process to a probabilistic process (resulting in the publication of metrics like LOLP (Loss of Load Probability) and USE (Unserved Energy) such as noted here for the coming summer) – there are a couple components of this

2a)  modelling the outcome of a 90%POE assumed demand trace , and including this in the combination of 50%POE and 10%POE

2b)  ERM suggests that the AEMO include consideration of the probable contribution of generation projects that are committed and under construction, but not yet registered (so not included in the modelling methodology currently).

This second part of the  ERM submission is here.

Part 3 = this component of the proposal asks that the time-range covered by the MT PASA is extended from a 2 year time horizon out to a 3 year time horizon.

Amongst other benefits, this will make the MT PASA updates:

3a)  line up better with obligations for wholesale market customers that could well be triggered under the Retailer Reliability Obligation as soon as this month of August (somewhere in the next 2 weeks) when the 2019 ESOO is published; and

3b)  It will also line up better with the notice of closure obligations

… though it would seem that these types of requirements would essentially imply the need for MT PASA to be extended somewhat beyond 3 years, perhaps?


To the extent that we have been able to think this through, it all seems a fairly reasonable request, and likely to deliver good value to wholesale participants, but also to the growing number of retail customers signing up to renewable PPAs stitched to the output of particular renewable projects, also (the anticipated maintenance schedule for which should also be visible in MT PASA with the same level of accuracy required for Scheduled generators).

It is understandable that the AEMO (in particular) might be daunted by yet one more change to the information management processes on top of the huge job that is 5 minute settlement.  Seems to me that this means our conversation should focus on “what do we have to do to enable this to happen”?

Submissions to the AEMC are due today – further details from the AEMC are here (perhaps the AEMC will take these thoughts here into account).

About the Author

Paul McArdle
One of three founders of Global-Roam back in 2000, Paul has been CEO of the company since that time. As an author on WattClarity, Paul's focus has been to help make the electricity market more understandable.

Leave a comment

Your email address will not be published.