Unplanned outage at Loy Yang A1

Reader Beware … just because you’re reading more about coal unit outages here in recent days, it does not necessarily mean that these outages are happening more frequently!

———————-

We’re having parallel conversations with several of our clients (each who have a licence to ez2view) with respect to using the software to alert on, and then interpret, coal unit outages … which are topical for obvious reasons:

1)  so it suits our purposes to share this type of article here (rather than duplicate information in parallel one-on-one conversations);

2)  as we have also recently for:

(a)  This trip and (brief) unplanned outage at Loy Yang A4 on Wednesday 14th August 2024 … not noted in that article was that the unit was back online a couple hours later.

(b)  About the ‘Wobbles in recent days, at Loy Yang A’; and

(c)  Yesterday evening the trip and (brief) unplanned outage at Tarong North on Monday 19th August 2024 … not noted in that article is that the unit was back online soon after.

(A)  Broader aspects of coal unit performance

All of these events feed into a broader conversation about coal unit performance.  On that note:

1)  Next week (time permitting) we’ll post some insights generated from our recent reviews of 2024 Q1 and 2024 Q2.

2)  Also next week we’ll be curious to see what the AEMO reports in the 2024 ESOO.

(B)  Outage today at Loy Yang A1 (current visibility)

So adding to the list above, Loy Yang A1 came offline this afternoon, as captured in this ‘Notifications’ alert from a display copy of ez2view we have running:

Different from the other two cases we documented recently (links above), we note that this unit has come offline from a lower load position – which makes us wonder if this outage was planned in advance?

… unit trips from close to full load are highly likely to be unplanned outages

As is our standard process to quickly understand if the outage was planned or unplanned, we flip to the ‘Generator Outages’ widget in ez2view to the pre-filtered view of MT PASA DUID Availability data for all 44 x coal units remaining in the NEM:

As was the case for Tarong North yesterday, we see that this outage has appeared ‘out of the blue (green)’ … from which we are confident to infer that it’s not an outage that was planned in advance (despite the unit coming off from low load).

Out of curiosity, I’ve clicked through to the ‘Unit Dashboard’ widget in ez2view to see a time-series of the output profile of the unit (looking back 2 days) and, up until 04:00 this morning, how the unit had been bidding:

I’ve also selected the 04:00 dispatch interval in the chart to see (in the table) the most recent bidding information … which talked about commercial factors in the market (i.e. and did not hint at any technical issues in that particular rebid reason).

(C)  Impact on System Frequency

In the past two examples (trip from close to full load) we saw frequency drop outside of the NOFB, which will presumably have triggered Contingency Raise FCAS  response to augment the Primary Frequency Response and the Regulation FCAS to help the frequency recover.

A quick reference to our live and high speed dashboard shows not the same level of dip in the past 2 hours (which is what we would expect, with a disturbance considerably smaller in size).

(D)  If we wanted to delve further…

For those readers following on with their own copies of ez2view … then I might suggest some of the following:

1)  We hope you’ve already configured your own companion notification alert to email operations@YourCompanyDomain when any [Coal Unit] Switches ON

(a)  i.e. the inverse of what was triggered above.

(b)  that will let you know if Loy Yang A1 makes it back online this evening.

2)  We’re running out of daylight hours (i.e. fitting in with the cadence of the MT PASA DUID Availability data updates), but the last update today (i.e. 18:00 today) might contain updated details on outage plans for Loy Yang A1…

(a)  i.e. if it turns out there’s some technical issue meaning that the generator can’t bring the unit straight back on;

(b)  Given how late in the day it is, however, the generator may not have time to make that assessment to publish the data to AEMO and hence into the data update for 18:00 …

(c)  So more likely that any update won’t be seen until the update for 09:00 tomorrow.

3)  Also tomorrow (after the ‘Next Day Public’ data drop after 04:00) we’d be able to drill into the rebid reasons (e.g. in the ‘Bids Table’ within the ‘Bids & offers’ widget in ez2view) to see what AGL has nominated as the reason for the outage there.

4)  If the outage was planned in advance, the ‘Forecast Convergence’ widget might be useful for reviewing successive updates of the MT PASA DUID Availability data set to see how outage plans have shifted around.

(a)  for instance, in January 2024 we posted ‘Hot & Wet Weather pushes back expected return to service dates for Callide C3 and Callide C4’ utilising this widget to show changes for Callide C3 and Callide C4.

(b)  however as there’s no outage currently pencilled in via MT PASA DUID Availability for this unit, this approach is not really relevant at present for Loy Yang A1 … other than to show that there’s currently no outage planned for the unit until April 2025:

… remembering that this widget allows the user to ‘look up a vertical’ to ‘see that other dimension of time’, we see that the 2-week long planned outage has recently been shifted back a couple weeks in the schedule, for reasons unknown (though not concerning, as this type of shuffling is quite common).