Is energy storage the magic bullet needed to realise the potential of renewables? Paul Marsh reports from a panel debate in Edinburgh in October.
MATCHING supply and demand has always been a headache for electrical utilities. With the decentralisation of power generation made possible by renewables, the problem is beginning to prompt a fundamental reassessment of how the grid can best be managed. Similarly, on the user side, the intermittent nature of technologies like rooftop solar means that households and businesses export most of the energy they generate back to the grid – an obviously unsatisfactory situation.
The putative magic bullet of energy storage appeared to acquire a new sheen of credibility and seriousness earlier this year with the announcement of Tesla’s Powerwall home battery system, bringing the promise of grid independence within reach of a mass market. Around this time conversations at renewables events turned frequently, it seemed, to the topic of energy storage – often punctuated by nose-tapping allusions to the wisdom of investing in the sector (or so it appeared to me), if you had a spare fiver or two in your back pocket.
Networking group eCoConnect’s panel debate on the topic in October offered perspective on some of the obstacles that stand in the way of a commercial role for energy storage in the electricity supply equation.
The evening was opened by Professor Win Rampen, Chair of Energy Storage at the University of Edinburgh, who alluded to the diversity of technologies that can be used to store energy, from the spring mechanism in a watch to the pumped hydro installations that currently provide about 3 GW of storage capacity in the UK. As the grid evolves, this requirement will probably increase ten-fold, he said. Whatever technology aspires to meet this demand will have to be as good or better than pumped hydro. But whatever happens, “the grid’s going to have to change”, a revolution whose initial stirrings are discernible in places like the US, where consumers’ electricity bills have been lightened by the availability of solar.
As grid operators wonder where much of their income has gone, they still have to deal with the problem of trying to provide a stable grid that works even when the sun isn’t shining. “That’s a big problem for them… because most of their generation mix is nuclear and coal. Both like to run at one speed: On.”
“We’ve all been conditioned to think baseload is good. Baseload is terrible… that’s not what the grid needs going forward.” Baseload is basically intransigent generation, he said, and what we need is dispatch-able generation. The only thing that provides it at the moment in any reasonable quantity is Combined Cycle Gas Turbines (CCGT). “We’ve got to find technologies that store energy and give it back when we need it, and which can displace CCGT generation,” he said, expressing the personal view that we also need to get rid of coal and nuclear.
An era of flexible demand is coming, he said, and the internet of things is going to make this a much more sophisticated proposition. He foresaw a central role for “virtual storage”, where software will control appliances so that most energy is used when it is most available or affordable or green. “That’s going to take a certain amount of the role of storage,” he said, but it won’t do it all.
Getting into hot water
Energy storage has long been a familiar fixture in households in the shape of the hot water heater, now being replaced increasingly by combi-boilers and wood-chip heating systems – a disappointing trend for those who want storage, suggested panellist Andrew Bissell, CEO of Sunamp, a supplier of “heat batteries”. This is a piece of kit he said can store about four times as much heat as a hot water tank for the same space usage. It’s designed to work with a combi boiler and can also accept electrical input from a PV panel “when you want to avoid export”, or to accept heat from solar thermal installations.
Climate change and heat poverty were the problems he had in mind when he started his company, he said, and it is about to install 600 of the systems in social housing in Edinburgh – with government funding. Serial production of the product had just commenced that afternoon, he said.
He offered perspective on aspirations to deploy storage in the grid. “The regulatory framework we live in does not support most storage being on the grid,” he said, and this has been standing in the way of investment at the grid scale. For the various systems and elements within the grid at present, “it’s not at all clear who’s allowed to own it, who’s allowed to operate it, and who’s allowed to profit from it”. He saw the most accessible opportunities for energy storage being in the home.
Stuart Noble, Head of Wholesale Regulation at Scottish Power, attempted to offer the pumped storage perspective. If there was a clear vision of what was needed – say, 10 GW of storage capacity – would Scottish Power be likely to invest? There were so many uncertainties out there, and limitations in our ability to understand the true nature of the “stresses and strains on the system”, as things unfold, that – combined with the complexity of the business case around pumped storage – it would be “very difficult without some form of incentive to press the button on some of these investments”. There were clearly missing pieces from the puzzle at present.
Seemingly giving the lie to such cool assessments was panellist Robin Duncan, Project Manager Energy Storage with AES, said to be the largest operator of battery energy storage systems in the world. His own firm has been working with systems operators, utilities and governments to introduce energy storage as a viable alternative to traditional methods of ancillary service provision and meeting capacity, most recently with a 10 MW installation set to go ahead in Northern Ireland, seemingly one of the first announcements to have been made of grid-scale deployment of battery technology.
A bugbear for households and businesses with their own renewables sources is the need to export most of what they generate to the grid – recent statistics put this level at about 80%. Bissell expounded on the various tweaks and adjustments his firm would make to such installations, which – used in conjunction with his own heat batteries – would allow people to hold onto much more of the energy they generate.
His system also includes the communications platform to make this energy available to the grid, seemingly anticipating potential future developments with despatchable demand. “But we don’t yet know what to connect it to,” he said, expressing bafflement as to “how to unlock what the Rocky Mountain [ Institute ] says today is 14 different value streams” from energy storage provision.
Noble offered some ideas on what these value streams might entail with the larger, grid-scale installations. Cruachan currently provides things like power assistance to the grid and ancillary services, on top of the basic megawatt-hours.
Enrique Troncoso, an engineering consultant with Systeng Engineering, which offers consultancy on energy storage, was also on the panel. There are many different ancillary services implicit in the functioning of the grid at present – you can’t apply the same business model across all of them.
Duncan said AES’s grid installation in Ireland was providing short duration storage as well as ancillary services.
Duncan agreed completely that “complexity is the no1 barrier to development” with the deployment of storage in the grid. He said AES had toured the country asking DNOs and TSOs the simple question: what is storage? Is it import or export? And how do we apply for a grid connection? There didn’t seem to be a clear answer.
Much of the debate seemed to centre on problems with the grid. For example, the amount of solar now present in the distribution network makes it difficult to run the transmission system. Noble suggested the solution was to have a large energy store at the transmission level – “you get the economies of scale, [ and ] it can be a cheaper solution than everyone doing it on their own at distribution level”.
Counting the cost
The cost of energy storage has seemed an area for optimism, with lithium-ion batteries like those used in Tesla’s Powerwall following a downward price-trajectory curve thanks to their existing mass market in electric vehicles – with Nissan Leaf having also recently announced an intention to offer its batteries for home energy storage.
Bissell offered numbers to put Sunamp’s offering into perspective. Solar PV is currently about £100 per kW-hour, if you go and buy the cells. As an appliance it’s more like £300 per kW-hour, and because Sunamp puts a 3kW heater in it, it’s bordering on £500 per kW-hour. The fact that it was “behind the meter” of the house or building made a big difference, he said. There you value energy differently than on the grid.” If you can displace kW-hours that you would buy from the grid with a self-heat solution, “that’s worth a lot more to you than it is out on the grid”, he said. Compared with electric batteries, Sunamp’s heat batteries are about a third to a half of the cost, he said.
Bissell and Rampen voiced the opinion at different points in the evening that discussion of grid connection issues was in danger of being swept away by what’s happening at the consumer side, in places like Hawaii where “grid uncoupling is happening at a tremendous rate” as Bissell put it. “You’ve got people paying 45 cents per kW-hour to buy electricity from the grid, getting penalties applied if they’re connected to the grid with PV and going ‘you know what? To hell with the grid, I’m off’”.
Duncan said the same was true in Florida and California, where there has been much talk amongst utilities policymakers about the “duck graph”, which illustrates the increasing severity of the problem that faces utilities at moments when a wild swing in net load occurs, such as when the sun gets low on the horizon on a typical weekday just as office buildings turn on their lights – a problem for which energy storage is being primed to tackle, with legislation now going through to mandate this in California.
Bissell said: “I think there is a risk that you guys have to get your planning horizons in line with the consumer, who’s not planning on 10 years and multi-million pound investments”.
Noble of Scottish Power felt this was a fair challenge, conceding it was possible that “by the time we get there, it’ll be too late – they might not need us”. On the other hand pumped storage is the most cost-competitive technology that’s at large scale [ deployment ], and can offer very good consumer benefits.
An audience member posed a question about anomalous events that challenge the grid’s ability to cope. “Every 10 years the UK gets a very big anticyclone that sits there for about 2 weeks,” he said. It’s cold, it’s dark, and there is no wind and very little solar. How do you cope with that? At the moment the best answer is open cycle gas turbines. But if there’s enough of them to make up for your 33 GW of installed capacity then there’s going to be enough of them the rest of the time when you don’t have an anticyclone. Duncan said he still couldn’t see a way past the need for thermal generation in a case like this. Bissell felt that increasing interconnection of the electricity infrastructure was key. Noble said: “I think pumped storage and its discharge times are more suited to the problem you’re talking about, and it’s a natural demand curve”.
The government’s recent decision to slash the feed-in tariff for renewables installations has greatly increased the interest in energy storage amongst rural, community and industrial consumers and developers, in the range 50kW to 5MW. Is energy storage ready to provide a solution here? Bissell said his firm was working towards having large scale heat stores based on Sunamp’s technology, which is modular and scalable, allowing lots of these small cells to be aggregated – the “server farm” approach. However, he suggested they also ought to do the R&D on big cells, to see which is better. He said they were talking to Scottish Enterprise about an R&D grant but the application was currently blocked because of a perception that there was no market for it at present. He appealed to anyone with evidence of a market need here to contact him.
Duncan said the technology was available to support these FiT-level projects. However, he added: “At the moment battery technology bolt-out is relatively expensive, I don’t know if the economics would work to implement that”. Bu it would all depend on the individual constraints, the amount of storage needed, how long it was necessary to store the energy – all of which would dictate the size of battery bank needed.
The debate seemed to mostly focus on storage as it might be used with conventional generation, but as Rampen pointed out, the future looks like it might be very different. An imminent step change in the cost of solar was shaping up as a massively disruptive influence. The seasonal and diurnal variation in loading “is going to have such a massive effect, that the current grid isn’t going to know what to do”.
Who should lead, in the uptake of storage? Bissell said: “I do genuinely believe that the things that can be adopted easily by consumers can ultimately be regulators for the grid as well.” This was the ethos behind Sunamp’s storage offering for consumers, he said – the fact that it has all the telematics in it so that, one day, when the aggregators are there, “they can be assets that can be used in frequency response or balancing services”, he said. This would mean that if, say, there was more wind energy being generated than we could deal with, we could absorb the energy. It would provide “the right amount” of feedback to the grid.
Moderator Robert Hokim highlighted an approach being taken in Germany, where a generator is paying households to put battery banks in their basements – providing them with a discount on their energy bill, in other words.
Bissell assumed storage would become ubiquitous, driven by a Bill Gates style figure perhaps, and might happen quickly, as the adoption curves in the consumer world can be incredibly rapid. The recent slashing of Feed-in Tariffs didn’t help but would likely be a temporary blip, given the downward cost trajectory of things like solar PV around the world. “And then it will make sense to generate electricity at home, store some of it as electricity and some of it as heat, and give some away to the grid”.
Which storage technology will predominate in the future? “We’re probably going to see a basket of technologies”, said Troncoso, who didn’t think a single technology was likely to be able to address all our storage requirements, in terms of size and response time. Hydrogen power will play a role in many of these markets, though it is currently still at the demo phase.
Troncoso also felt the potential of energy storage went beyond the power sector, and could facilitate a cross-sector integration across areas like heat and transport.
The panel seemed unanimous that targets weren’t the answer to driving the uptake of storage. Targets have been used elsewhere, such as in California, mandating 1.3 GW of storage, procured by three utilities. Duncan felt uptake in the UK was “happening naturally”, making a target redundant. “Targets may or may not be the solution”, said Troncoso, adding that what was more important was clear signals from government. “There’s a broad range of technologies there, which are nearly ready”, he said, but unless investors get a clear signal, it can’t happen.
Bissell suggested an incentive of some kind would be a great help, as it has been for solar PV – “maybe a few thousand Euros per installation”. This kind of thing provides a leadership sign from the political establishment – ‘we want this’.
He also felt it was vital to have transparent, fair and accessible markets for all of the ancillary services used in the management of the grid. Wary of the bilateral relationships that have sprung up between storage providers and utilities – “very cosy little deals, negotiated one on one”. These were necessary in the beginning, “but in the end we need a kind of open pool where everyone can bid what they’ve got”.