Solving the growing problem of wind curtailment and grid congestion requires a huge amount of new transmission, more flexibility and more storage. Whether bespoke solutions will play a role, however, depends on a raft of complicated policy choices.

Grid congestion and the curtailment of Scottish wind power is going to get worse before it gets better. Wind curtailment is costing UK consumers billions and the price of inaction will only rise. The best way to solve the problem is to dramatically ramp up construction of new transmission infrastructure, and use flexible assets such as batteries more effectively. Hydrogen offers a plausible bespoke use case for excess wind, but there is still much uncertainty over how that market will develop. Locational pricing might also help, but risks creating more problems than it solves.

Think-tank Carbon Tracker estimates that wind congestion has cost £1.5bn since 2021, and warns this figure could reach £3.5bn annually by 2030. Current costs arise mainly from transmission bottlenecks near the Scotland-England border – the infamous B6 boundary. Congestion forces the grid operator to switch off and compensate wind farms in Scotland, and pay generators in England – typically gas plants – to power up. Carbon Tracker expects Scottish wind capacity to grow five times faster than new transmission capacity between now and 2030. At that point, up to 20% of all Scottish wind production could be at risk of curtailment.

Able to build cable

The obvious and most urgent solution is to build more transmission infrastructure. The Electricity Networks commissioner Nick Winser published an independent report in August with recommendations for how to halve the time it takes to build new transmission lines from 14 years to seven.

This includes a Strategic Spatial Energy Plan, new short-term and long-term regional flexibility markets, and streamlining planning consent. Time is of the essence. Carbon Tracker’s modelling indicates that one year of delay in delivering adequate grid infrastructure could cost £3.6bn while delivering new projects one year early could save a similar figure.

There are already myriad plans to improve grid architecture: Recabling projects for existing lines is one. The use of ‘bootstraps’ – connecting an offshore generator through a subsea link to two points in the onshore transmission system – to reduce land points and infrastructure needs is another. Also, Ofgem has an Accelerated Strategic Transmission Investment (ASTI) framework, which aims to speed up the approvals process for building new infrastructure.

Wind turbines and electricity pylons on the Causeymire in Caithness.

If wind projects are threatened with curtailment or grid connection issues, a direct line power purchase agreement (PPA) with an individual commercial consumer is an option. This could be anything from a data centre to an industrial manufacturer. But for wind projects to agree to a direct line PPA, they would have to be very confident that the wholesale market will not deliver greater value.

Uncertainty over how quickly the UK will be able to deploy new grid infrastructure complicates the landscape for such bespoke solutions. If curtailment could be addressed in the next seven to 10 years, there is less incentive to investigate a bespoke business case.

At the same time, neither industrial manufacturing nor data centres are likely to locate to the Scottish highlands for slightly lower electricity prices. Industrial manufacturing facilities are exceptionally costly to build and have complex supply chains. Data centres need to be close to internet exchange points and urban hubs with multiple redundant fibre connections. This explains the plethora of data centres in Dublin and London, rather than Inverness.

Hydrogen hopes

There is understandable excitement around the possibility of hydrogen made with bountiful Scottish wind power.

In a June 2023 report, Carbon Tracker power analyst Lorenzo Sani said that, in an ideal market, wind congestion could provide hydrogen systems – electrolysers, storage and H2-turbines – with a business model strong enough to cover all their fixed costs.

His modelling suggests that 4GW of Scottish electrolysers running on electricity from wind would be able to satisfy almost half of the UK’s projected hydrogen demand by 2030. But that ideal market – along with the storage and transport network for hydrogen – does not yet exist.

The overarching issue that affects the outlook for all these options is that the UK has a nationwide electricity price. At present, there is no market mechanism that can properly incentivise more load above the B6 constraint.

Location, location, location

There are, however, demand response options that allow consumers in England and Scotland to adjust demand at times of congestion or constraint. In April, the ESO began a two-year trial of a new Local Constraint Market (LCM), which pays consumers in Scotland – residential and commercial – to use excess wind.

Demand flexibility is vital for a grid heavy with renewables. The LCM opens the door to aggregating smaller assets like households and electric vehicles, in addition to larger industrial operators.

There is also scope to increase demand response south of the B6 constraint. In August, Octopus Energy struck a partnership with UK Power Networks to provide free electricity during times of excess to consumers in areas of South East & East England.

Shifting the demand load

Octopus has also used the ESO’s Demand Flexibility Service to trial paying customers in England to reduce demand during peak times. Increasingly, third party firms are helping industrial and commercial consumers install tech solutions allowing them to adjust electricity use in response to shifting prices.

The more demand response in the system, the better. But this does not help get wind power from Scotland to English consumers. Paying Scottish consumers to increase demand is helpful. But there are only so many consumers, and the biggest – industrial producers, for instance – have limited flexibility in how much additional electricity they can consume. Production lines cannot run when staff go home at night.

The combination of wind congestion and demand response may make periods of cheaper power increasingly common in Scotland. “But it’s very difficult actually to forecast how this is going to work, which means it’s hard for this to underpin a business case,” says Susie Elks, a senior policy advisor at thinktank E3G.

Nodal pricing quagmire

There is one solution – simple in theory, immensely fraught in practice – that could provide a price signal strong enough to draw commercial consumers north of the B6. This is the thorny issue of locational pricing, a topic recently explored by E-FWD.

There is already a locational signal for where generation assets are built: a network charge to connect based on distance from consumption and many other parameters. Scottish generators already pay more to connect because they are further away from demand centres in England.

“The problem is that it’s hard to predict how this charge will evolve in the future, and it’s unlikely to be large enough to influence a big investment decision on its own,” says Lorenzo Sani, a power analyst at Carbon Tracker.

Using network charges to influence location is fiddling at the edges of the problem. Locational Marginal Pricing (LMP) would provide a far stronger, clearer signal. Consumers in Scotland would pay less for electricity because the marginal cost of generation there is lower, and be incentivised to consume more.

Generators in Scotland would be paid less, disincentivising additional generation capacity above the B6 constraint. If there is more wind generated in Scotland than can be used, the marginal price would fall to close to zero.

“The argument is that it would drive smarter behaviour from consumers,” says Tom Edwards, senior modelling consultant at Cornwall Insight. “And if the wholesale price in Scotland is close to flat or even negative, that’s when you might see data centres and hydrogen producers and all sorts of other flexible demand.”

But locational pricing is controversial. Designing and implementing a locational pricing system would take years, and the uncertainty could threaten business investment. There would almost certainly have to be some form of grandfathering to protect projects built before a certain date, as well as a way of insulating consumers. Just having locational pricing on the table of options creates challenges for business, which is why many industries are lobbying against it.

For Edwards, a better solution would be to implement locational pricing in 10 years’ time after building out all the necessary grid infrastructure to make it work.

“I think LMP is a good system that works where you’ve got a relatively stable system, transparency about what’s happening on the network and clarity about future generation plans,” he says. “But it’s probably not what we want to do when we’ve got to build 30 GW of offshore wind in the next eight years.”

Feast for flexibility

Regardless of whether LMP is implemented, better use of battery storage should help tackle curtailment in the near-term. And there is a clear business opportunity for flexibility providers operating in a grid facing serious congestion issues.

National Grid ESO operates the Balancing Mechanism (BM) – a continuously open online auction to balance supply and demand. Carbon Tracker’s Sani notes there is potential for flexibility providers to benefit from a double payment – receiving payments from the BM to increase demand during congestion, and then selling that electricity during periods of high demand.

His modelling indicates that – in an ideal market system – storage providers in Scotland would be able to cover between 50% to more than 100% of their annual fixed costs from payments due to wind congestion.

But even in the existing (sub-optimal) system, the BM is not operating as well as it could. It operates on a legacy IT system designed decades ago for large-scale generators. Control room processes are partly manual. This contributes to what is called the “skip rate” problem.

National Grid ESO struggles to accommodate small generators like batteries, which means that investors are seeing their assets skipped over in the BM dispatch process. Edwards said control centre staff might opt to dispatch a single more expensive but more familiar source of 20 MW, rather than having to select 20 individual 1 MW batteries.

“What we’ve seen this year is much more storage than ever before being available to be used in the balancing mechanism, and it’s just not getting used the way any of the battery business models anticipated,” says Wendel Hortop, markets lead at Modo Energy

He notes the ESO is preparing to release the latest software update for the BM – a new Open Balancing Platform – which will hopefully improve how batteries are dispatched.

Time to upgrade

There are other issues that touch on IT, data and modelling capacity. The market has only started to build out storage in Scotland in the last year or two, based on business models that expect the assets will be used in the BM and – potentially – be able to take advantage of negative pricing during periods of excess.

“The big challenge is that this isn’t really happening in practice,” says Hortop. In addition to the skip rate issue, the ESO is also hampered by an inability to see how much power a battery has beyond the next 15 minutes.

There are two and four hour batteries available to soak up congestion, but they cannot realise full asset value because the grid operation only has certainty they will be available for a quarter of an hour. Modo Energy estimates there is 389 MW of battery storage capacity operating in Scotland, which will rise six-fold to 2.9 GW by 2026. For the grid operator, improving data visibility is a clear priority.

Ultimately, the best option of reducing curtailment is to build out more transmission. Batteries can certainly help, especially if IT system upgrades allow the ESO to take full advantage of their flexibility. The outlook for other solutions like hydrogen production and private wires to industry are more opaque. So much depends on market design and policy choices where there is little clarity and more than a little controversy.

Related Content

Domestic car makers pay price of EV competition

Wham, bam, thank you CBAM