Whitehall ‘predictions’, meet real-world data

Regular readers will recall my ongoing correspondence with the Department for Energy Security and Net Zero over their (ahem) optimistic take on the costs of future renewables.

This has focused on offshore wind, which is supposed to form the backbone of the future energy system (although it should be noted that DESNZ’s figures for onshore wind and solar are equally dubious), and in particular on the figures for windfarms commissioning in 2025, for which we can hope to get real-world data against which to test their assumptions.

As windfarm projects speed up and slow down, the identity of the ones that will actually commission in 2025 changes. At this point, it looks as if Moray West may be the only one that switches on in practice, although Dogger Bank A might also squeeze in before the end of the year.

Firstly, let’s look at what we know already about DESNZ’s assumptions.

On capex, DESNZ claim that capital costs will fall to £1.5m/megawatt is already falsified, as against a typical cost for recent windfarms of around £3m/megawatt. We can be sure of this because we know that Moray West has spent more than this amount just putting in its foundations. Dogger Bank A doesn’t look to be very different.

On opex, DESNZ claims that lifetime average costs will be around £90,000/megawatt/year, which is way below the £200,000 that you might expect from recent windfarms, but for more definitive proof we will have to wait two or three years for some data.

However, on output, there is something new, because although it is yet to be fully commissioned, many of Moray West’s turbines have been installed, and it is now starting to produce output data.

Recall that DESNZ are saying that the lifetime average for 2025 windfarms will be an absolutely astonishing 61% of capacity. Recent windfarms have been opening at around 50% and then declining from there as they age. The lifetime averages will probably be well under 40%.

How is DESNZ’s prediction playing out? According to Elexon data, since the start of December, Moray West has had an operational capacity of 436 MW, around half of the planned total. (I’m slightly suspicious of these figures, because there was no increase in capacity between the December and January editions of the dataset – if it’s actually higher, then the capacity factor figures I present are commensurately overstated).

Then, using Elexon’s Settlement Data, we can get a feel for how the windfarm is performing. The graph below shows a 21-day moving average of Moray West’s daily capacity factor, which suggests that it has been delivering an average of about 65% over the last six weeks. The Renewable Energy Foundation’s data suggest that it has lost around 10% of output to grid constraints, so arguably it could be delivering 70%.

What might that mean for its annual average? Broadly speaking, the annual average might be around 25% lower than the average for Quarter 4. So if we scale down the figures we have, we will get to something just above 50%. This is above the first year performance of Moray West’s sister windfarm, Moray East, which managed 44%, and its other neighbour, Beatrice, which managed 48%.

So it looks quite good, but we are much more interested in the lifetime average. We know from the history that windfarm output declines over time, and that bigger turbines decline more quickly than smaller ones. Moreover, Moray West’s turbines are huge, at over 14MW each. If we assume that the decline is linear and the lifespan is 25 years, then the capacity factor in year 12 or 13 will be similar to the lifetime average. The graph below suggests that something just short of 40% is a likely figure.

The Moray West estimates presented here are extremely crude of course, but I think they are good enough to show that the windfarm’s output is nothing like high enough to justify a claim that its lifetime average will be 61%. In other words, our working assumption – that DESNZ’s figures are grossly overstated – appears to be confirmed. Once again, it seems that policy in Westminster is grounded more in fantasy than in the real world.