Twice in the past 7 days we think we’ve seen some form of gremlins invade AEMO’s dispatch process:
- To inject what looks to be ‘phantom’ demand into a particular region;
- Which drives NEMDE to dispatch more megawatts of supply than is actually needed;
- Which leads to frequency spiking higher than it should be.
A question has come up in the course of investigating these two events (more on those specifically in other articles in the coming days) … how often has this happened in the past, and has the incidence been increasing recently?
- Well, really two questions.
- That is, of course, on top of
- the ‘why does it happen?’ question that we are also pondering
- plus some others as well
Somewhat helpfully, in recent years we’ve taken to collating articles we’ve written about ‘Operational Glitches’ here, but readers should be clear that there’s a mixed bag of causes here, not just the subject de jour.
Purely from these pre-written articles, I’ve started to tabulate this list in reverse chronological order of similar events:
| Date
(in reverse chronological order) |
What happened?
with ‘Market Demand’ |
What was the effect on frequency?
if already analysed in articles |
|---|---|---|
| Tuesday 11th November 2025 |
We’ve only collated two articles about what happened on 11th November here (with respect to the operational glitch) at this time … but more to come. |
|
|
Looks like the maximum ‘error’ in ‘Market Demand’ was of the order of 1,130MW in Victoria in the morning. |
As noted in this article, the frequency rose up to boundary of the NOFB, and may have exceeded (we’ve not checked the details). |
|
| Thursday 6th November 2025 |
We’ve collated 5 articles about what happened on 6th November here (with respect to the operational glitch) at this time. |
|
|
As noted in this article, it looks like the maximum glitch size was ~1,812MW in Victoria around lunchtime. |
As noted in this article, the frequency peaked at 50.229Hz at 12:45:21.2 (NEM time) … well outside of the NOFB. |
|
| Wednesday 19th February 2025 |
We collated 3 articles pertaining to Wednesday 19th February 2025 here |
|
|
There were two sizeable drops in ‘Market Demand’: |
As seen in this first article, the frequency prior to the 1st drop in Market Demand had climbed to around the level of the NOFB. |
|
| Friday 31st January 2025 |
We’ve collated 3 articles about what happened on 31st January here (with respect to the operational glitch) at this time. |
|
|
As noted in this article, it looks like the maximum glitch size was ~950MW in NSW in the late afternoon.
|
As noted in this article, the frequency was restrained to only 50.1Hz (within the NOFB). We explored what helped to restrain the frequency here. |
|
| Monday 9th December 2024 |
We’ve collated 2 articles about what happened on 9th December here (with respect to the operational glitch) at this time. |
|
|
As noted in this article, it suggests that the size of the glitch might have been as large as ~3,100MW in terms of the Victorian demand. |
Also illustrated in this article, the frequency did not appear to increase markedly outside the PFR deadband. … so we wonder if this was a different type of glitch?
|
|
| 24th January 2024 to 7th March 2024 |
A quick scan also reveals Allan’s article ‘Missing megawatts and phantom generation’ that covers a period of longer than a month where there was an operational glitch. But this was perhaps different? |
|
|
Nothing to add |
We did not look |
|
We’ll return to this list (in a subsequent article) to expand the list of events, based on more longitudinal analysis (by a method to be explained later).
Leave a comment