Kafka on the Offshore
Magical Realism or a Cunning Plan ?
UPDATED: having found an error in my numbers - on the % increase of wind (overall a 2x from today not a 4.5x as per the original form of this article - the 4.5x is the planned increase in the offshore wind capaicty) - herewith a corrected version. apologies.
The British government has announced a “bold” “ambitious” plan to “boost Britain’s energy security ... Bold new commitments to supercharge clean energy and accelerate deployment, which could see 95% of Great Britain’s electricity set to be low carbon by 2030”
2030 you say? Thats less than 8 years away… Looking back 8 years, 2014 was the year that the Tesla model S was launched in the UK. I don’t recall 2014 being that different to 2022 in terms of the whole industrial-energy system, but OK, let’s leave that thought hanging there for a moment.
So what is proposed? If we cut out the word-salad and ignore terms like “ambition” and “where possible”, “we will look to”, we can focus in on the numbers.
Offshore wind: up from 11 G to 50GW by 2030
Onshore wind: recognition that ths will be almost impossible to increase
Solar: …to increase the UK’s current 14GW of solar capacity which could grow up to 5 times by 2035
Nuclear: up to 24GW by 2050 to come from this safe, clean, and reliable source of power. This would represent up to around 25% of our projected electricity demand
Whilst there is also text concerning heat-pumps, oil and gas production, and hydrogen, I will focus here on the electricity generation part only and look at installed capacity and actual generation.
So to resume:
Wind to go from 24 GW to 50 GW - all the growth being offshore (2x increase overall, but 4.5x for the offshore)
Solar to go from 14 GW to 70 GW - 5x increase, bizzarely assuming most will be rooftop.
Nuclear to go from 7 GW to 24 GW - albeit by 2050
Elecricity demand roughly doubles by 2050
Again, just for reference, the current 10.4 GW of offshore wind comes from 2,297 turbines. Whilst turbines are getting bigger, we can estimate that this would need to increase to roughly 10,000 to achieve the 50GW target.
Installed Capacity is of course a poor benchmark as it does not relfect the firmness or utility of the generation. But as a headline, the UK has just over 105 GW of installed capacity currently, for a demand that rarely, if ever, exceeds 43 GW. This is a big over-build already, far more than the traditional Reserve Margin of 15%-20% above peak demand. This is reflected in low capacity-factors which can be back-calculated from the ratio of theoretical TWh and the actual TWh produced over a year. Solar comes in at 11%, wind at 30% and natural gas at 39%.
Solar and wind have low capacity-factors because of their inherent irregular availability
Natural gas is low because it is used to plug the gaps when wind and solar are not available - that is, it’s low utilization is an outcome of choices made to prioritize wind and solar in the supply side.
Nuclear has a capacity factor that often approaches 85% because it cannot easily load-follow and is used as base-load.
So, if we already have installed capacity of over 2x peak demand, why do we need to do a 2x in wind, 5x solar and a 3x on nuclear? All that because we can’t plan on ensuring reliable and cheap gas?
Assuming that new nuclear won’t be available by 2030, how much wind and solar would we need to be at 95% renewables?
To answer this, we can’t simply look an annual averages, because as the proportion of renewables increases so there are periods where generation will exceed demand and -in the absence of storage at orders of magnitude above anything available today - these will lead to curtailment (I’ll come back to this later).
So, there will be days where renewables generate more than needed and there will be days where renewables generate less than needed - at which point the “legacy” solutions will need to be available - otherwise the lights go out.
By looking at data from Gridwatch which is in 5-minute intervals for the whole of 2021 (4 months shown above for illustration) we can see that with the current configuration renewables managed to provide up to 80% of the UK demand in the very best 5 minute intervals (averaging about 60% over a 24 hour period - which is what the headlines report).
In the 2021 data there was no overall curtailment above 100% of demand. In fact, for the year, wind and solar contributed 22.8% of the electricity. However, if you increase “installed capacity” of wind and solar in the spreadsheet and calculate the theoretical output (pro-rata to the 2021 actual data), there are time steps where generation starts to exceed 100% of demand. Demand “caps” the generation and there is now a bucket of “excess” power. By the same token, there are still many time-steps where legacy generation has to provide. As the over-build increases, so there is more “excess” and less “legacy”. By summing all the 5-minute intervals for each over the year, we can see how this becomes asymptotic.
By the time we have 8x the current solar and wind (160 GW wind and 112 GW of solar) we would be at 89% - so not far off the target of 95% - but 8x is significantly above the government targets.
Plotted differently, we can see that it is not until we have a 3x over-build (relative to 2021 infrastructure) that we start to see modest amounts of excess energy - although there are significant reductions in the requirement of legacy generation (and thus emissions) in these early steps.
A crucial point is that these increasingly rare requirements of legacy generation are typically on cold winter nights - so are essential for society. It is also worth noting that there is increasing cost in the final steps for limited gains - and this should be considered when planning a “net zero” outcome - how good is good enough?
The Boris plan calls for 2x wind and 5x solar - so this would sit between the 2x and 3x bars on the above chart. At this point the system would be 54% renewables with almost no TWh of excess generation. Not quite the 95%.
If the build out is greater and/or there is a wedge of nuclear - we could maybe get close to the 5x scenario.
Which is where (I assume) that hydrogen comes into the discussion. If (and it is a huge “if”) the excess generation could be converted into hydrogen for use as storage, then it would (theoretically) be possible to further reduce the need for legacy fuels. At this point its getting well beyond the scope of a spreadsheet - but the 73 GWh of excess at a 5x over-build with a 40% roundtrip efficiency (electricity-hydrogen-electricity) could bump the “renewables” share up to 90%.
Which maybe explains why nuclear will be developed 3x to 24 GW
So far it kind of looks more like planning than theatre. However, this is a very theoretical analysis just playing with numbers in a spreadsheet.
Specificlly, major concern with all of this is the costs.
Firstly there is the cost of building-out of 5x wind and solar (as well as upgrading/replacing exisiting wind and solar)
Secondly, using irregular excess wind and solar to make hydrogen will render ths horrendously expensive - economic calcualtions of green-hydrogen rely on 100% utilization.
Thirdly, as the utilization rates for legacy generation drop, so they will need capacity payments to remain on standby.
Finally, a bigger wedge of base-load nuclear will be being built in parallel, but as huge over-build of renewables depresses prices for long periods, nuclear will require fixed pricing or be sub-economic.
In addition, I have assume constant demand - something that seems very unlikley as we “electrify everything” - in a case of higher demand, clearly the percentages of renewables will drop relative to a flat-demand scenario.
95% green electricty by 2030? - whilst conceivable using a spreadsheet, I remain of the opinion that it is mostly magical thinking.