In January, Eunomia published the first UK forecast for battery storage, which estimated that deployment could reach over 1.6GW by 2020. It generated a good deal of interest, with a range of organisations keen to understand how the forecast was produced. Now, six months on, there’s a new question – is the forecast still relevant?
Forecasting storage deployment for the UK certainly involved ‘sticking our necks out’ somewhat, but was something we felt was necessary to help inform investors and others interested in the potential of this technology. At the time there was very little else out there – and to a large extent that remains the case.
The forecast certainly involved a degree of art, but was backed with science. Working in the sector over the last two years provided valuable sources of information, but turning it into a forecast meant considering historical comparisons, drivers and indicators.
- Historic growth of solar PV: the experience of this market shows the potential for unit costs to decrease rapidly as deployment takes off – although solar PV’s rise was also largely assisted by subsidy schemes that are unlikely to be available for storage.
- Increasing need for ancillary services: National Grid’s System Operability Framework predicts that the requirement for Fast Frequency Response (FFR) services will increase by 30 – 40% in the next 5 years, and this will be largely serviced by batteries. This includes, for example, the 200 MW of extra capacity expected to be deployed as part of National Grid’s Enhanced Frequency Response (EFR) service, which is due to start in 2017.
- New planning applications for battery storage: a growing number of planning applications for wind and solar PV projects included the co-location of battery storage, or for batteries to be added to existing installations.
- Increasing non-commodity costs: Some of our commercial and industrial clients forecast significant increases in their peak electricity price as a result of non-commodity charges, providing a major incentive to go off-grid during peak times.
But did this all add up to a meaningful forecast? It’s necessary to look into these issues, and what has happened in the interim, in some more detail.
Happy EFR after
As expected, the EFR tender round stimulated significant interest, showing that developers now have an appetite to bring storage projects forward. Distribution Network Operator Companies (DNOs) have been inundated with connection applications for energy storage projects as developers work to submit EFR applications by the 17th June deadline. Anecdotal evidence suggests that UK Power Networks (UKPN) has received nearly 9 GW of applications for grid connection since the start of the year, and Western Power Distribution (WPD) a similarly huge number. Only a fraction of this volume will win EFR contracts, but it also seems likely that some ‘unsuccessful’ projects will still go ahead, with developers finding other ways to build-out so as to make a return on the capital they have already invested.
To meet its growing need for FFR services, National Grid has signed a four-year contract with Renewable Energy Systems (RES) to provide 20MW from its battery storage systems. This may signal a trend for further such contracts, let outside the EFR system.
Whilst these developments suggest that the market may grow as we forecast, in April, despite a recommendation by the National Infrastructure Commission, Ofgem ruled out (in the short-term) allowing DNOs to own and operate storage beyond their current licence restrictions. This would have been facilitated by giving storage its own asset class (i.e. so it was no longer classed as generation) and have removed the current restrictions to DNO ownership of such assets.
That said, the current cap of 2.5% of the DNO’s distribution business revenue which can be from generation should not constrain the number of distribution network-led storage projects in the short-term. Furthermore, the parallel 10 MW restriction on ownership of generation assets by DNOs applies at an individual installation level and therefore shouldn’t be a major barrier to the scale of batteries currently being considered in many cases.
However, DNOs are not currently incentivised to own storage assets. Although storage might help alleviate grid constraints and defer traditional reinforcement measures, DNOs appear to prefer to procure this service from third parties. SSE is currently tendering for services to support one of its ‘constraint management zones’; meanwhile, in South Wales WPD’s Project Sync is trialling a similar service to overcome issues associated with peak generation from solar PV during periods of low demand. Both include storage as an option to provide these services, alongside demand-side response (DSR). If this ‘toe in the water’ approach is successful, however, both may procure such services more widely across their networks. This could go some way towards the level of capacity we predicted being realised.
Although DNOs are generally technology agnostic, deferral of investment in traditional reinforcement represents a significant opportunity for battery storage in the short to medium-term. In the longer term, as DNOs potentially transition into Distribution System Operators (DSOs), and use flexibility from demand and generation to manage network flows. As a result, they are likely to procure ancillary services akin to those procured by National Grid, thus vastly increasing potential size of the market for battery storage.
Eunomia’s report highlighted the challenging economics for battery storage co-located with renewable generation, particularly for wholly new developments. We therefore forecast this to be the slowest growing sub-sector of the market. Only a small number of new applications have come forward over the past six months, suggesting that we were right to think that this approach is only viable in limited circumstances. Eunomia recently modelled a series of scenarios for the co-location of commercial battery storage with solar PV, and found that in most cases, the magnitude of risk usually still outweighs the level of return that can be realised. However, for existing solar PV or wind farms, some projects do appear to be attractive to specialist investors, usually those which also hold an interest in the design and construction of the project.
This study also highlighted the implications for developers when attempting to ‘stack’ revenues, as the ability to do this will be a key enabler to market growth. Little work had previously been done to clarify how the provision of ancillary services may interact with existing, or new, Power Purchase Agreements (PPAs) for renewable generation. Our research showed that very few aggregators or PPA providers appear to have a clear idea how these different contracts may impact upon each other. On the one hand, the addition of a battery (which is contracted to do FFR) to a solar PV installation reduces imbalance risk, but on the other, output will fluctuate in response to calls from National Grid to respond to changes in frequency. The research also suggests the impacts either way are not significant and will not constrain investment, but it remains a factor that requires consideration. As a minimum, developers will need to negotiate the provision of ancillary services into their PPA contract terms.
The most attractive investment proposition identified in Eunomia’s electricity storage report, and the sub-sector which we forecast to grow most quickly, was commercial and industrial (C&I) ‘behind the meter’ (BTM) applications, where batteries would allow major electricity users to avoid high peak prices, while stacking other revenues. Our recent work indicates that this remains the case, but significant uptake by businesses has not yet materialised. Possible reasons include:
- Limited awareness and understanding of storage among energy managers, and perceptions that the technology is ‘too risky’.
- Inability to demonstrate meeting a required payback threshold, which can be as short as 1 to 2 years.
- A reluctance to take on third party finance to facilitate investment.
Much-needed further work to fully understand the challenges and barriers faced by organisations that could benefit from C&I BTM storage is in prospect. Late last year, Ofgem set out a number of actions to help increase flexibility in the network, which included engaging with stakeholders to raise awareness of the opportunities and understand the concerns and needs of commercial and industrial customers. Unless such actions are taken quickly, this sub-sector may not meet our expectations.
So, across all three sub-sectors (DNO, co-location with renewables, and C&I BTM) is deployment keeping pace with Eunomia’s forecast? The EFR tender and other recent developments mean that it is likely that our prediction of over 200 MW of storage coming online by 2017 should be met. DNOs do not appear to have been deterred from opening up opportunities for storage in their networks, and could represent a further rapid growth area. Meanwhile, whilst the business case for co-locating battery storage with renewable generation remains as challenging, we are seeing projects coming online. The biggest surprise is that, despite the strong financial rationale, C&I BTM applications do not appear to be taking off. It may yet prove just to be a matter of time.
Six months on, the prospects for electricity storage in the UK remain buoyant. Provided that the costs of the technology continue to fall rapidly and the first fully commercial deployments demonstrate success, there is no reason to change the broad thrust of previous predictions: 1.6GW by 2020 remains a realistic – and desirable – ballpark figure.