FES 2023 and its significant flaws

The Electricity System Operator for Great Britain (ESO) publishes its Future Energy Scenarios (FES) every year. The latest version, FES2023, uses various scenarios to model supply and demand up to 2050.

All scenarios have similar features including:

• Most generation capacity to 2050 will be in wind turbines whose output is extremely variable in nature
• The virtual elimination of gas, which is currently the primary controllable source used to counter this variability
• A massive increase in interconnector capacity, leading to reliance on other countries being willing to maintain supplies to the UK when they also may be experiencing power shortages
• The use of storage to bridge gaps in generation.

However, the duration and frequency of wind lulls in the UK mean that the reliance on wind as the principal energy source, without some form of controllable back up, will lead to an increase the incidence of blackouts.

Generating Capacity

Table ES1 in FES2023 gives details of the way in which demand will be met in each scenario. Taking “Leading the Way” as an example. It predicts that demand will increase from 289TWh in 2022 to 507TWh in 2050, or 1.76 times.

To meet this increased demand, generating capacity must be increased as shown in ES1.

 

* Nuclear includes small modular reactors in 2050

** Gas is CCS Gas and Biomass is CCS Biomass in 2050

Whereas the grid currently relies significantly on gas and coal, which contribute over 50% of generating capacity, by 2050 the majority dependence will be on offshore and onshore wind, which represent over 80% of generating capacity.

Back Up

Table ES1 also details proposals for non-generation.

 

*** DSR = Demand Side Reduction - which is as big a component as generation by nuclear in 2050

This main back up by 2050 will be interconnectors, mainly from mainland Europe, and storage. Interconnectors are reliant on power being made available from suppliers. Storage, including pumped hydro, is reliant on power being available for charging. Demand side reduction is reliant on being able to restrict supply without significant effect on industrial, commercial or domestic activities.

Wind Lulls

Much was made earlier this year of the day when wind contributed more to demand on the National Grid than all other fuel sources combined. It featured in newspaper articles and on the BBC News. Wind contributed 51.5% to demand on 22 March as shown below. It can also be seen that gas and imports (interconnectors) are ramped up to meet peak demand.

 

However, in the same month, on 5 March, wind only contributed 6.4%. So is the high or low figure more typical? The table below shows periods where wind has contributed less than 10% of demand in 2023 to date. These have been called wind lulls.

 

The table confirms that low wind speeds are not unusual, in fact they are frequent and often extended. This can also be seen in the graph for the whole of March 2023 shown below.

 

As can be seen, even in a “good month” for wind, as March 2023 was, there is massive variation in the contribution of wind to grid demand. Gas (CCGT) is the major contributor overall and, most significantly, is called upon to meet peak demand every day, even when wind is making a major contribution. Variation in demand is currently about 10GW between the minimum and maximum each day and this will increase as maximum demand increases as predicted.

Extended periods of little wind

Looking at the table again, the total number of hours when wind contributed less than 10% of demand is 617.5 hours, or 25.7 days and the longest single period is 55.5 hours, or 2.3 days. This pattern is similar to that observed over recent years despite a substantial increase in installed wind turbine capacity over that period. Assuming weather patterns remain the same in the future, it can be assumed that wind lulls will not significantly decrease as the capacity of operational wind turbines is increased.

FES2023 acknowledges the problem in pages 211–218 covering “Dunkelflaute”. The document says “‘Dunkelflaute’, refers to a period of cold weather with low light and little to no wind across Northern Europe. Our analysis shows that a range of low carbon flexible technologies, from both the supply and demand sides, are required to maintain security of supply in periods of cold, dark, and still weather in 2050, in the absence of unabated gas generation.”

FES 2023 then goes on to model a period from 19 February to 3 March 1985 which it refers to as a period of “wind drought” and projects this to 2050 illustrating that peak demand can still be met by renewables with a very small baseline contribution from nuclear and CCS Biomass plus demand side reduction.

1985 was a time when there were very few, if any, wind turbines attached to the Grid. It is therefore unclear as to why ESO have used this period as the basis for their modelling when a more recent and representative period could have been chosen.

For example, taking June 2023 as the most recent example of ‘Dunkelflaute’ and simply multiplying the actual data from that month by the increases in demand and capacity given in Table ES1 of FES2023 gives the following graph.

 

It can be seen that wind does not contribute much to demand over extended periods and there is a significant reliance on imports (interconnectors) when demand peaks. But as ‘Dunkelflaute’, refers to a period of little to no wind across Northern Europe, are these imports from Europe going to be available in a period of ‘Dunkelfalute’?

This leaves the other sources of power, principally storage, to fill the gap. The data from which the above graph is derived shows that demand in June 2050 would be 17263GWh and supply would be 13405GWh, leaving a deficit of 3858GWh. This would have to be met mainly by storage. But over such long periods of low wind, how could that storage be topped up and would there be sufficient to cope with the lack of wind power?

FES2023 indicates periods of DSR, Demand Side Reduction, to help. In other words, periods where use of electricity would be curtailed.

The Solution

It is suggested that the National Grid should look to more reliable and predictable sources of renewable energy rather than rely on wind as its principal supply of power to the Grid. Taking the predictions for June 2050 as an example indicates that the pursuit of such a policy will result in extended periods when demand cannot be met and hence there will be prolonged blackouts.