Energy storage is increasingly big business in the UK and alongside intermittent renewables, will play a key role in getting towards a zero carbon electricity grid.

Business may be getting bigger, but for the time being it is also getting tougher. More capacity reduces the amount of revenue that short-term facilities can provide, thus squeezing margins.

The focus in recent years in terms of storage capacity growth has been on batteries, although there are a number of different technologies at play.

Great Britain had around 800 MW of battery storage in early 2020 and that has risen to around 4.4 GW today. During 2023 alone, operators added 1.5 GW of capacity and there is currently more than 4 GW in construction.

The government has set the aim of reaching 30 GW of energy storage by 2030.

Driving this is the goal of a decarbonised electricity grid by 2035 and the huge rise in intermittent renewable generation. Installed renewable generating capacity in the UK increased by a factor of seven between 2009 and 2023.

At the end of 2023, renewables accounted for well over half of total installed generating capacity. This ratio will continue to rise.

At the same time, we have seen a significant reduction in dispatchable thermal generating capacity, especially coal stations. A key consequence of this is that the Electricity System Operator (ESO) requires far more fast-response flexibility of other kinds to maintain grid stability – operating frequency – within acceptable tolerances when renewable generation drops unexpectedly.

Well-located energy storage can also help to manage network constraints at both transmission and distribution levels.

Battery emergence

Battery storage has boomed in the last few years as costs have fallen. The International Energy Agency (IEA) assesses costs are down by 90% in less than 15 years.

Lithium-Ion is the dominant battery technology at present. One factor driving down costs has been increased investments and interest in EVs.

Geology and project development capability are key determinants of natural gas storage. Battery storage, though, is locationally flexible and typically modularised.

Constraints on battery installations are around planning permission and a suitable grid connection.

However, the commercial optimisation of battery storage revenues is relatively complex. This is particularly apparent in the need to combine and balance multiple sources of income.

The average duration of existing grid-scale battery storage facilities in the UK remains around 1.3 hours. Battery designs are typically for 1-2 hours of storage.

Sources of storage revenue

Transmission-connected energy storage can access a variety of revenue streams. These are principally the wholesale market, the GB balancing mechanism (BM) operated by the Electricity System Operator (ESO) and a number of ESO-operated mechanisms for procuring fast-response flexibility.

A recent Modo Energy assessment shows that inter-temporal price arbitrage in the wholesale electricity market generally remains the single most important source of revenue.

However, this revenue is now significantly less than it was in 2022, for example. That year, when Russia invaded Ukraine and gas prices spiked, the UK saw high and volatile wholesale energy prices. That exceptional period was a highly profitable time for battery storage operators. However, that was the exception and market conditions have now returned closer to pre-crisis norms.

Modo has also shown how revenues – both total incomes and the mix of revenue sources in the stack – can change significantly from one month to the next. For example, the fall in dynamic containment revenues (one of the frequency response services procured competitively by the ESO) between October 2023 and January 2024 is very stark. Frequency response revenues then recovered quite well in March-April 2024, but these fluctuations serve to highlight a relatively high-risk investment proposition

One important observation is that little storage revenue can be “banked” in advance of construction.

Just in time

For understandable reasons, the ESO now purchases most ancillary services – including frequency response – much closer to real time than it often did in the past. From an energy storage point of view, this means revenues can fluctuate and are also vulnerable to increasing competitive pressure.

BM revenues have generally been a modest percentage of the total. Market commentary suggests the ESO has often been unable to accept as many battery storage offers as operators might have expected. IT system constraints have driven this problem, although these are now being addressed.

Accordingly, BM revenues have been increasing and could well be a greater source of income in the future.

Some grid-scale battery storage is distribution connected. The Distribution System Operators (DSOs) also have a need for flexibility services. This comes as decentralised energy activities – such as embedded solar, EVs and heat pumps – continue to grow and require a more sophisticated approach to managing regional and local electricity grids.

The DSOs are now procuring such services on a significant scale. Battery storage facilities are important participants in that process.

Derated

In addition, battery storage can and does bid into the GB capacity mechanism (CM). However, nameplate storage capacity is heavily derated, based on an assessment of its expected contribution at times of peak system stress.

These derating factors have also been falling over time as more battery storage comes on stream.

In the latest T-4 capacity auction for delivery year 2027-28, for example, derating factors of only 8% and 15% for one-hour and two-hour facilities were applied. A one-hour store with a 50 MW nameplate was thus remunerated as if it had only 4 MW of firm capacity.

This substantially reduces the amount of revenue available to storage operators from CM contracts. In other words, the longer-term revenues that storage operators can lock in via the CM is relatively minor.

Squeeze on

Overall, the recent profitability of battery storage in GB has been severely squeezed. This is a result of more benign wholesale market conditions and increasingly competitive ancillary service markets, as a consequence of the rapid growth in installed storage capacity.

For example, Gresham House Energy Storage Fund is a leading GB market player with an operational portfolio of 840 MW, as of September 2023. It has a market share of around 20% and a current market capitalisation of almost £300 million.

The fund’s share price approximately halved between the start of January and late April 2024. Its recent trading update referred to “the most challenging operating environment since the company’s inception in 2018”.

At present, battery storage in GB is arguably a victim of its own success. Rapid market entry on a large scale has been one important factor behind falling unit revenues.

Like natural gas storage in GB over the last 20 years or so, it is plausible to think that energy storage will become a cyclical commercial proposition. The logical commercial reaction of existing storage operators could be to slow the pace of development until the market comes back into more of a balance. However, there are still many aspiring new market entrants.A recent Renewable UK report identified a project pipeline of more than 95 GW. Of this, over 30 GW has already consented.

Longer duration

Short-duration battery storage is a good complement to intermittent solar generation. It bridges the gap of perhaps a few hours between times of significant solar generation and peak evening demands for electricity.

Batteries and intermittent wind generation are less well suited to each other. This is because there may be several consecutive days of low wind conditions.

Batteries are just one type of energy storage, though. There is around 2.8 GW – or 32 GWh – of pumped hydro storage capacity at four sites in Scotland and north Wales. This is expanding.

Plans are under way for a second 600 MW facility at Cruachan, in Scotland and Drax is also investing to expand the existing plant by 40 MW. A recent report by Scottish Renewables identified six pumped hydro projects under development in Scotland, including Cruachan 2. If they all go ahead, they would add 4.9 GW to the UK’s existing capacity.

The electricity grid requires longer duration energy storage (LDES) in the future. This is an area of active government policy development and financial support for emerging technologies.

A recent report for the government defines LDES as facilities that can discharge continuously at maximum power output for at least six hours. There are multiple competing LDES technologies and it is not yet clear which will emerge as longer-term winners.

Although the future mix of energy storage technologies in GB remains unclear, it falls to the ESO to present a range of plausible scenarios to 2050.

Innovation and change

The technologies considered in the National Grid’s 2023 Future Energy Scenarios were batteries, pumped hydro, liquid air (LAES) and compressed air storage (CAES). They present four scenarios to 2050, of which the first three are consistent with the goal of net zero by 2050.

One active developer of LAES is HighView Power. This company has a 50 MW/300 MWh project in construction at Carrington, near Manchester. On a smaller scale, Scotland’s Innovatium has a live demonstrator project in operation at Aggregate Industries’ cement plant in Staffordshire.

A prominent UK example of compressed air energy storage development is Gaelectric’s Northern Ireland project, which European Union funding has supported. Gravity storage developers such as Gravitricity and Energy Vault are exploring project opportunities at former coal mines. In 2021, the former developed a small demonstrator project near Edinburgh.

Finally, the grid may also benefit from other forms of energy storage in the future.

For example, SSE Thermal is working on possible hydrogen storage at Aldbrough, in East Yorkshire. This links up with plans to develop a hydrogen-fired power station at Keadby, on Humberside. If these plans come to fruition, they would provide a flexible source of clean, dispatchable power for the GB grid.

A look to the future

The National Grid’s expectations for energy storage are dramatic. Total storage capacity increases from 5.3 GW in 2022 to a range of 13-29 GW by 2030, depending on the scenario. It then reaches as much as 22-52 GW by 2050.

While battery storage continues to dominate installed capacity, other technologies will provide the bulk of total energy storage volume.

The ESO projects that LAES and CAES would develop substantially, but pumped hydro remains the single most important storage technology by energy volume in all four scenarios. Given British topography, this sounds like good news for investment in Scotland and the Scottish economy.

Energy storage in GB is currently a merchant business proposition par excellence. Projects can only secure relatively little revenue in advance. Much depends on the volatility in the wholesale power market and revenue from grid services, for which there has been increasing competition.

Limitations in the current business model are apparent. The storage margins and investment returns are much reduced at a time when the medium to long-term needs of the grid require substantial investment in expanding capacity.

The key question for debate is therefore this: is the current storage business model still fit for purpose? Or do we need to find an alternative that allows investors and financiers to ‘bank’ at least some of their revenues before making a final investment decision (FID)?

The provision of flexibility was certainly one of the issues highlighted when government launched the ongoing Review of Electricity Market Arrangements (REMA) in 2022.

If intervention is required to reinforce future investment incentives, then some form of cap-floor regulation (as applied to electricity interconnectors) would be one obvious candidate.

Interestingly, the government has already proposed a mechanism of this kind to support the development of hydrogen storage.

Related Content

#TheGrid: Storing up problems for the future