In between times these past couple weeks, we* have been trying to find spare hours to complete the focused analysis for 2024 Q1 for the latest iteration of our GenInsights Quarterly Update series.
>> * These analytical reports are a collaborative effort between us here at Global-Roam Pty ltd (publishers of WattClarity) and the team at Greenview Strategic Consulting.
I’m currently working through the trended quarterly statistics for ‘Aggregate Raw Off-Target’ for all Semi-Scheduled units as part of a longer-term series of analysis – some of which we’ve shared under the ‘Trend of AggROT (across Semi-Scheduled units)’ series of articles.
Through 2024 Q1 there were a number of interesting days:
1) Even before the quarter had ended, we’d published ‘Aggregate Dispatch Error across all Semi-Scheduled units (NEM-wide) throughout Tuesday 13th February 2024’
2) In this article we take a quick top-down look at Thursday 22nd February 2024.
3) There were others, as well …
Remembering back to Thursday 22nd February 2024.
Over the day and since we have published a number of articles about that day by a range of analysts – where we have remembered to do so, these articles are all collated here.
Some key points about this day:
1) It was a hot summer day, particularly in the southern regions
2) This heat drove ‘Underlying Demand’ to a new all-time record … for the NEM, and for Victoria.
3) Bushfires in VIC and TAS affected a range of electricity infrastructure, including:
(a) some particular power stations, like Stockyard Hill Wind Farm and Waubra Wind Farm; but also
(b) rooftop PV yield, and the predictability of that; along with
(c) some transmission capability.
With those articles setting the scene more broadly, let’s look particularly at this metric for Semi-Scheduled units.
Aggregate Dispatch Error, across the Day
Remember that Aggregate Dispatch Error is a step towards calculating Aggregate Raw Off-Target for any group of units, but is not quite the same. It’s just easier for me to pull the data out to present the following analysis.
Here’s a graphical representation of calculated Aggregate Dispatch Error across all Semi-Scheduled units that were registered at that time:
Remember that:
1) To click on the image for a larger view; and
2) That in terms of the number sign:
(a) Positive numbers represent collective under-performance across the collection of units against their Targets, whilst
(b) Negative numbers represent collective over-performance across the collection of units.
Some quick observations
1) The numbers line up into essentially three time periods:
(a) From midnight through until the sun comes up, the numbers are all relatively small, and mostly positive
(b) Through sunshine hours the numbers on this day are almost always positive, are larger, and with some very large incidents.
(c) From sunset the scale of the numbers drop off, and we see from ~22:00 in particular they are mostly negative.
2) In terms of large instances, we’ve chosen to highlight the 4 instances where Aggregate Dispatch Error was greater than 500MW, as follows:
(a) At 14:15 NEM time we calculate Aggregate Dispatch Error to be +653MW (i.e. collective under-performance)
PS1 we have specifically drilled into this dispatch interval in the subsequent ‘Case Study of 14:15 on Thursday 22nd February 2024 (large collective under-performance across all Semi-Scheduled units)’ article.
(b) At 15:00 we calculate Aggregate Dispatch Error to be +550MW (i.e. collective under-performance)
(c) At 15:50 we calculate Aggregate Dispatch Error to be +547MW (i.e. collective under-performance)
(d) At 17:05 we calculate Aggregate Dispatch Error to be +609MW (i.e. collective under-performance)
3) Incidentally, we see a brief period of volatility for VIC, TAS and SA for 15:50, 15:25 and 15:30:
(a) During which period the price for one of more of those regions was well above $500/MWh;
(b) Coincidentally, the Aggregate Dispatch Error during these dispatch intervals dropped … not sure if related or not?
We’ll leave this here, for now, but possibly drill in further later …
Leave a comment