Tight supply/demand balance forecast for SA, VIC and TAS on Thursday and Friday


I’ve been offline a couple days and have tuned back in this afternoon to forecasts of extreme temperatures across several spots later this week.  I see Adelaide’s already hot, with the forecast for the heat to remain and build through Thursday and Friday:

2019-01-17-weatherAdelaide

As noted on the image, consecutive days of hot weather have more effect on electricity demand – partly because of physical properties (like thermal mass in buildings), but also because of the psychological factors (i.e. “screw it, I’m turning the air-con on because I can’t take it any more”).  Hence we’d expect to see demand in South Australia highest on Friday, if these predictions turn out to be correct.

In Melbourne I see that the forecast is for the heat to build to Friday (from a lower start today) with a cool change coming through sometime Friday night:

2019-01-17-weatherMelbourne

Given that we have access to the popular NEM-Watch v10 entry-level dashboard to the NEM, it’s easy for me to have a quick look at what might be seen from this window into the NEM – with the following snapshot taken at 18:00 NEM time (so 18:30 in Adelaide DST and 19:00 in Melbourne DST):

2018-01-17-at-18-00-NEMwatch-forecast30000MWtomorrow

I’ve annotated with a couple points:

1)  We see demand currently in South Australia up in the yellow zone

2)  We see tomorrow that the NEM-wide Scheduled Demand is only forecast to peak up at 30,034MW in the 17:00 trading period (this is about 5,500MW below the all-time record and really not too noteworthy for a summer’s day).  I see that the forecast is for Scheduled Demand in Victoria to rise tomorrow to about 8,600MW at 16:30 (almost 2,000MW below all-time max) and in South Australia to 2,800MW later at 18:30 (about 600MW below the all-time max)

3)  However I also know that there are supply-side considerations as well:

3a)  I remember that since last summer Hazelwood has closed, and Northern also went in South Australia a while before that – hence the portfolio of dispatchable generation has reduced.

3b)  Unfortunately, we see that the current wind harvest from all wind farms across the NEM is quite low today (ranging between 200MW and 400MW through much of today), and I wonder what the forecast is for the next couple days.

3c)  As an aside (because they cannot be known in advance), I also have seen that the Australia Institute has been flagging unit trips at thermal stations in more recent times, and wonder whether the forced outage rate and availability figures are significantly different than that seen in past years (remembering my prior “clapped out second hand car” analogy in speaking to concerns about reducing availabilities, and making a mental note to look into this further when time permits).

4)  Hence it’s not really too much of a surprise then to see that the current predispatch forecast is for prices to spike in SA, VIC and also TAS up above $10,000/MWh (all 3 regions for 16:30 and 17:00 trading periods, NEM time).

5)  I’ve highlighted the Market Notices AEMO has issued indicating the least severity “Lack of Reserve” notices (i.e. LOR1 level) for both Thursday and Friday in Victoria.

Opening up a pre-prepared template trend from the ez2view Time Series Data Viewer (which clients can access themselves here) I see that the peak in demand in Victoria tomorrow is forecast to reach the sum of available generation capacity (meaning that Victoria will be heavily dependent on imports from neighbouring regions (which we presume will be NSW, given SA will also be feeling its own heat).

2018-01-17-ez2view-TSDV-forecastVIC

As a quick last look at one of the factors mentioned above, I’ve used the “Forecast Convergence” widget within ez2view to look at how AEMO’s successive forecasts for aggregate wind output across Victoria have been varying – with particular interest in looking forward for the coming couple of days, and at times of peak Scheduled Demand:

2018-01-17-at-18-45-ez2view-ForecastConvergence-VICintermittent

As a later addition, I have included the same chart but looking at aggregate wind in South Australia tomorrow – and, unfortunately we also see more recent AEMO forecasts (to 20:30 on Wednesday 17th) are for there to be not much contribution at all from wind at the time of peak Scheduled Demand, which (at 18:30 as above) will be when there will be not much small-scale solar PV injection as well:

2018-01-17-at-20-30-ez2view-ForecastConvergence-WindinSA

 

We’ll watch with interest tomorrow and Friday…

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.


*