Matthew Lockwood, IGov Team, 9th December 2016
A few days ago I heard a presentation by Paul Massara of North Star Solar, a new solar PV + battery home energy system start-up. One of his points was that scale manufacture of lithium ion batteries means that electrical storage is getting cheaper and cheaper, and PV + battery packages are now cost effective in the UK with the right financing package. Certainly such systems seem to be taking off in places like Australia, and are now required in new installations in Germany. These are home system batteries, but it seems very likely that in the near future they will also be joined in many homes by electric vehicle batteries.
So what are the implications for the way that electricity markets work if batteries really do become cheap and ubiquitous? My aim here is to do a quick thought experiment about what could happen, definitely not a prediction of what will happen, by when, and so on.
Start first on the demand side, with that rooftop solar PV + battery combo. In the UK, solar output peaks in the middle of the day. In a lot of households, demand is low at this point, and if we assume solar PV is also cheap and households have a lot of it, generation is likely to exceed demand for much of the middle of the day, allowing batteries to charge up. Switching to batteries at the evening peak then reduces net peak demand on the grid (in my thought experiment households have smart meters and home systems which receive some kind of scarcity price signal and automatically switch to the cheapest source– this of course may not happen in reality). If it has been really sunny there may also be enough left over to cut down the morning peak as well.
Of course, demand in businesses and offices during the day is much higher, but if these have solar PV as well, their net grid demand will also be lower, on sunny days. Because demand and supply are more matched here, storage may be less useful.
The overall effect will be to flatten the total demand curve (industrial demand is already pretty flat); it would also become smoother at the granular time level.
Second, on the supply side cheap storage could also be used to regulate wind output. This is already being trialled in different forms, for example in Scotland and in Germany. At the limit, wind, and indeed all intermittent renewables could, with sufficient battery capacity, become baseload.
At the extreme, the resulting electricity wholesale market would have a flat demand curve met with what we currently call baseload generation. There would be no price variability, as (automated) storage would have arbitraged all that variability away.
If the UK were a lot sunnier throughout the year and we had more roof space (somewhere like the south-west USA?) then the market might end up looking a lot like this. Because PV output in the UK is very low in winter, the effects are going to be much more seasonal here, and in cities especially roof space is limited. So the effects will be less extreme, at least if or until power-to-gas technology makes inter-seasonal storage cost-effective. However, even here there is likely to be a significant flattening of demand and supply curves.
There are some potentially interesting implications here. One is that demand side management, in the literal sense of shifting demand away from peak, is going to be useful mainly only in the winter. Another is that a lot of the current and anticipated challenges for the transmission system operator (and for distribution system operators) arising from intermittency may retreat again. On the other hand, transmission networks in particular are likely to face the dilemmas that are already happening in parts of Australia, where solar PV + storage is leading some households to off the grid, meaning that network charges have to go up for those still on the grid, providing a greater incentive to the latter group also to go off grid, i.e. the ‘network charging death spiral’. In places like the UK, things are unlikely to get that extreme, but in the summer especially, transmission networks are likely to be much less used, raising questions about whether our current approach to funding them will work.
A third possible implication is that investing in large scale electricity generation will become much lower risk, because prices and revenues will become much easier to predict accurately, and there will be no fuel costs to predict, so that the cost of capital for the sector should be pretty low. Negative pricing and wind cannibalisation would become a thing of the past. Not only would the balancing services market shrink back, but intra-day and day ahead trading would also play a smaller role.
There will doubtless be many other effects, but the overall point is that cheap, ubiquitous electrical energy storage will lead to a very different world and may change the focus of many of today’s energy policy debates. For example, efforts at making demand side response work may become less important. However, such a thought experiment also makes clear the need for new approaches to governance and institutions in energy, approaches that can engage with the rapid nature of change and the possibilities that new technologies offer us.