There has been quite a bit of news-flow about battery storage over the past couple of weeks. There was the announcement that Elon Musk had succeeded in his bid to delivery the world’s largest battery storage facility to South Australia within his 100-day deadline, while in the UK the new capacity market register was published against a backdrop of lower de-rating factors for short-duration batteries, and the All Party Parliamentary Group published a bullish report on the sector.
All of this can be taken as hugely positive for battery storage, and an indicator that the dreams of 100% renewables (outside Iceland!) might be achievable. But how much of this is real: can battery storage overcome the significant challenges of scale and capacity that I have described before?
Tesla’s Australian miracle
Back in September, Elon Musk promised the state of South Australia that he would deliver a 100 MW battery storage facility within 100 days otherwise it would be free. Last week it was announced that the plant began dispatching power to the grid on 30 November, officially the first day of the Australian summer, and just 63 days after the contract was signed (although according to Bloomberg, the facility was already half built before the official signing date, so the “bet” wasn’t really a gamble….)
Although the battery is likely to be used primarily to boost supply during periods of peak demand, Tesla claims it has the capacity to power 30,000 homes for up to an hour in the event of a severe blackout, which may be long enough to bridge the gap until conventional back-up resources come online. (At least for the c4% of households able to benefit.)
According to clean energy news site, RenewEconomy.com.au, of the 100 MW/ 129 MWh in the array, around 70 MW of capacity is contracted to the South Australian government to provide grid stability, with a duration of 10 minutes, while the remaining 30 MW will have three hours storage and be used as load shifting by Neoen for the adjacent Hornsdale wind farm.
The facility is part of a AU$ 550 million plan announced in March to make the South Australia independent of the nation’s power grid. The cost of the project have not been made public, however, Musk noted at one point that failing to deliver on time would cost his company US$ 50 million – this would relate to the 70 MW contracted to the government, which was the subject of the “bet”.
However, as one Energy Matters commentator suggests, this is may only refer to the ex-factory price, with the full cost being up to twice as much. The US$ 700 / kWh estimated by Bloomberg New Energy Finance, would give an overall cost of US$ 90 million for the whole 129 MWh plant.
Another article from RenewEconomy.com.au suggests that the facility is unlikely to make money for the developers. According to Dan Sturrock, investment director of the Australian Renewable Energy Agency, under current electricity market regulations, it is difficult to build a business case for grid-scale storage projects:
“You see a lot of (energy storage) projects, well designed as they might be, and optimised in many other ways, funding them can be difficult – the only bankable revenue stream is wholesale arbitrage. It’s very hard to build a business case around FCAS (“frequency control and ancillary services”) in the long-term, because it’s a very shallow market, and very hard to model … in fact impossible to model,” – Dan Sturrock
As an illustration, Sturrock used the proposed 30 MW/ 8 MWh battery storage project known as ESCRI that is being built around the Wattle Point wind farm in South Australia. The chart illustrates the funding gap after revenue from network, wholesale markets and FCAS. The installation benefits from a contracted network service to Electranet.
“Very crudely, although we might expect the cost of battery storage to come down by a third or even a half, if all we’re doing in… three to five years is maybe market arbitrage, that’s a very untapped opportunity,”– Sturrock
According to the article, the developers of the Lincoln Gap wind farm are not using any government funds to build their proposed 10MW/10MWh battery, but say it would likely not be economic yet as a stand-alone investment, but is only viable as part of a renewable-storage package.
Sturrock argues that the most useful quality of batteries is their fast response capability however, this is not currently being valued properly in the Australian market, making it difficult for battery projects to make money in the current market framework, including the Tesla scheme.
The 100% renewables dream, Australian-style
A number of recent reports have recently been talking up the potential for renewables in combination with battery storage to provide most if not all of Australia’s energy needs. Australia is associated with high levels of sunlight, which may make the prospect seem achievable, but detailed analysis suggests that these reports significantly understate the amount of storage that would be needed.
He calculates the storage requirements for Blakers’ two basic renewables scenarios – wind:solar ratios of 80:20 and 55:45, using 97 days of actual grid data (27 July to 2 November), and shows that at least 2.8 TWh storage would have been needed to support an Australia-wide renewables over this period.
This is six times more than the amount estimated in the Blakers’ study, and may well under-estimate the long-term storage requirements, this being a relatively short period of analysis.
“My generation estimates also implicitly assume that the grid data used to develop them cover a period of “average” wind and solar conditions. Designing storage to handle worst-case wind and solar conditions over a period of ten or twenty years would further increase storage requirements, probably substantially,”– Roger Andrews
This is important, because the proponents of high renewables penetration assume adequate storage capacity will exist to back up the resulting intermittency and frequency control issues, but failing to properly quantify the extent of that requirement will mean not only is the practicality of such a system over-stated, but also its costs will be significantly higher than expected.
Changes to de-rating factors reduce the value of UK battery projects
Meanwhile in the UK, new storage de-rating factors have been published by the government that will make short duration assets less valuable in the capacity market. Batteries that can only deliver for half an hour will be rated at around 20% of their maximum output, while assets that can discharge for four hours will receive the same de-rating factor as pumped hydro at 95%.
This is a reasonable adjustment, since there had been some concern that short-duration batteries were being over-valued in a market that is designed to ensure the availability of capacity when needed.
The Capacity Market is important for new energy projects since long-term (15-year) contracts are available for new-build projects. These long-term contracts form an important part of the investment case, providing a secure long-term revenue stream, against the shorter-term income available in the wholesale and ancillary services markets.
Around 4.8 GW of new battery storage has pre-qualified for the T-4 capacity auction, out of a total of 27 GW new build capacity. These storage developers will now need to evaluate whether to build longer duration batteries to secure higher value from any capacity contracts, or to drop out, given the overall prequalifying volume of generation is around 30 GW well above the government’s 50.5 GW target, suggesting that the clearing price may once again be on the low side.
“The changes to the derating factors for storage is significant, as it alters the business model for many projects depending on the relative power / energy ratios of the storage system. Inevitably there will be winners and losers. However, these changes were foreseeable. It is indicative of the fact that storage is becoming a mature technology,” – Georgina Penfold, CEO, Electricity Storage Network
Storage cheerleading by the All-Party Parliamentary Group
A new report by report by the All-Party Parliamentary Group on energy storage and the Renewable Energy Association group of MPs has predicted the UK could have 12 GW of battery storage by 2021,up from 60 MW in 2016, with the right government policies in place, claiming this would have a positive impact on UK energy security and “empower consumers”. The report claims that storage can be “a simple, fast, and cost-effective means of balancing fluctuations between electricity supply and demand”.
The report contains three different scenarios for the future of battery storage:
The “high-deployment” scenario predicts 12 GW of additional battery energy storage by 2021, based on more projects being co-located at solar and onshore wind sites, storage connected to EV charging points, behind the meter applications at smaller scale, and larger, grid connected projects;
The “medium deployment” scenario sees 8 GW of battery storage with a lack of clarity regarding installing storage at RO and FiT accredited sites a key barrier to deployment, EV storage being predominantly deployed at rapid charge points, lack of half-hourly metering inhibiting behind the meter storage, and lack of DNO participation limiting grid-level projects;
The “low deployment” scenario, forecasts just 1.7 GW of storage with minimal regulatory support: the lack of clarity regarding co-locating with wind and solar sites means that in certain circumstances renewable projects risk losing their renewable support by installing storage, regulatory barriers mean that EV charge points are not rapidly deployed, concerns about consumers going “off the grid” lead to specific barriers being introduced for battery storage, and changes to grid charging and lack of capacity market and ancillary services reform inhibits grid-level deployment.
The report wishes to position the UK as a market leader in battery storage. It is worth noting that it is possible to become a market leader in a technology that may only have a niche application in the long term, and that this might even be desirable if meaningful economic benefits can be realised over this period.
However, the report makes no attempt whatsoever to present a balanced view of the case for battery storage. There is no analysis similar to that carried out by Roger Andrews above to determine whether batteries can make a meaningful contribution to the electricity system, and whether other technologies might not be more suitable to the challenges presented by increasing renewables penetration.
Some of the report’s recommendations are sensible – regularising the regulatory framework for storage to ensure a level playing field would be helpful regardless of the extent of the role batteries might play, but overall, it’s hard to take the report seriously.
Despite the seemingly positive news-flow, it’s difficult to be optimistic about the longer-term future of short-duration lithium-ion battery storage. Beyond the smoke-and-mirrors, at some point policymakers and consumers will realise that this technology can only meet a very narrow need.
Batteries lack the essential characteristics to support a high-renewables electricity system, so without significantly over-building either generation or storage capacity, it’s hard to see how they could play a major role.
The thousands of UK consumers without power this week due to heavy (for us!) snowfalls, would find any domestic battery system completely inadequate for dealing with the current supply disruption, and grid operators will soon discover that even the largest battery installation in the world will barely scratch the surface of the system needs as a whole.
Policymakers should take care not to develop overly-generous incentives for battery storage, which would put further pressure on affordability for already hard-pressed consumers.