Energy Storage Europe 2019
24 March 2019
In part 1 of this report on Energy Storage Europe 2019 I reviewed Oliver Schmidt's paper in order to give an overview of the most competitive stationary storage technologies right now and in the future - but what about the size of the opportunity for stationary storage ?
In short are we looking at a billion dollar market or a hundred billion dollar market ?
A handy way to provide this overview is to dip into the excellent session at the end of the conference entitled 'The World of Energy Storage - International Markets' - where 'international' in this context means outside Germany.
Fortunately the first talk by Valeska Gottke of the BVES (Bundesverband Energiespeicher - the German Energy Storage association) covered the progress inside the German market so in principle it looks like we have all territories covered.
The BVES are one of the main supporting partners for this conference and their list of industry members, shown below, is hugely impressive. It is worth drawing attention to the differences between British and German preparations for the energy storage revolution and to note that the closest UK equivalent that I can find to the BVES, the UK Energy Storage Network 's last shown news update is dated June 2017 ... just at the point that the UK government has started aggressively funded the new Faraday Battery Institute (mainly eV batteries so far) it appears that the body supporting the commercial sector has just died - this cannot be good and hopefully will not turn out to be yet another case of 'developed in the UK but exploited abroad'. Anyway - on with Germany...
Valeska went on to analyse the growth of the German energy storage industry, showing that it has grown by roughly 10% year-on-year over the past 3 years and that all this growth has come in technologies outside of pumped hydro storage, i.e. primarily batteries - where the growth amounts to more than 15% year-on-year. Germany already has significant home energy storage - estimated at currently more than 125,000 units sized around 6.8 kWh. By 2020 it is expected that there will be 200,000 of these, and by 2035 more than 6 times as many. The very fast growth of this sector is down to the rapid adoption of household scale PV by many of those in the south of the country over the last 10 years.
Medium (<150kW) and Large (>150kW) scale storage has had a somewhat slower start, seemingly due to the remarkable regulatory inconsistency that still persists in the German market that means that storage is classified and charged as both a producer AND a consumer of electricity. Despite this classification trap by 2020 Medium scale (70kWh average) installations are expected to number 1,500 and Large scale (5MWh) 350. Both are expected to grow by a factor of 3 between 2020 and 2030 yielding a total countrywide Stationary energy storage capacity of 11 GWh by 2030.
To get an overall estimate of the scale of the German electricity system we can consult the Clean Energy Wire graphs here - as you can see below 600TWh electrical energy consumption sets a fair scale, before any increase due to future electric vehicle adoption is factored in.
The figure of 600TWh is worth bearing in mind as it sets a scale for Germany, as roughly the 6th largest electricity user in the world - we will return to these figures later on when we compare the US and UK with the situation in Germany.
The United Kingdom
Turning to the UK we come next to Philip Nicholson of the UK government who described the UK's flexibility and energy storage markets. The term flexibility in the title might suggest that this is simply an optional feature, a nice-to-have, like being able to chose between taking the bike or the bus to work. In the electricity supply field the term flexibility is all about how supply and demand can, and in fact must, be brought into balance all over the network, every second of the day - it is not so much an optional feature as actually the main point of what the grid is about.
In a deregulated energy market many different organisations can offer energy storage services that yield the necessary flexibility (sometimes called frequency firming or regulation) that the grid requires - these are conventionally broken down into primary, secondary and tertiary frequency firming as described in this excellent article. Whilst many other energy storage services (shown in the table in the previous article) exist these are typically handled entirely by the grid system operator and so currently are not tendered for via a market mechanism - that is not to say that will always be the case in the future, and with many more energy storage assets available it is more than likely that these markets will be established to allow these assets to participate.
Having described the somewhat mind-numbing array of the UK's National Grid energy balancing markets Philip then turned to the future of the UK's grid, as described in the National Grid's 2018 Future Energy Scenarios planning document. As shown above this models 4 different scenarios : Customer Evolution' (CE), 'Community Renewables' (CR), 'Steady Progression' (SR) and 'Two Degrees' (TD). The details of the different scenarios are somewhat complex but the basic idea is shown with the arrows on the two axes in the grid above :- the vertical axis represents the level of decentralisation that might be expected whilst the horizontal axis represents the speed of decarbonisation of the country's energy mix. What actually happens will depend on political incentivisation and the technology available at the time - though to be honest the second of these is much less important - if the technology doesn't exist today it isn't going to help us achieve any of these improvements - the precise balance of technologies employed though will depend on their relative cost at the time.
As far as energy storage is concerned the FES scenarios show the following trends:-
As you can see the greatest adoption of energy storage occurs under the Community Renewables (CR - Pink) scenario, at 29GW in 2050, whilst the lowest adoption occurs under the Steady Progression (SR - Yellow) scenario, with 11.8GW in 2050. One thing to point out is that only two out of the four scenarios actually end up with the UK meeting its legal commitment (Climate Change Act 2008) to reduce greenhouse gas emissions by at least 80% from 1990 levels - these are the two scenarios on the right-hand side of the scenarios matrix (CR and TD). These scenarios should perhaps not be framed as 'aggressive' but rather as 'necessary to meet our own (UK set) legal obligations'. [ Aside - the 2.9GW value shown in the conference slide is, I believe, our current starting point, of which >95% comes from the UK's existing hydro-power energy storage.]
What CR (2050) = 29GW or SP (2050) = 11.8GW means in terms of actual energy storage capacity is not clear - for the sake of argument let us conservatively assume that energy storage durations average out at 1 hour (the typical duration of a Lithium-ion battery, though of course we know that VRFB's easily scale sublinearly costwise to 2,4,6 hours). Thus our take from this is that by 2050 we can expect between 11.8GWh and 29 GWh of energy storage to be required by the UK.
It is important to note that these figures specifically exclude energy storage by the UK electric vehicle fleet - those are counted separately as shown in the FES summary document:-
Thus the CR (2050) = 29 GW / 29GWh (from 50GW minus 20.6GW) is a genuine stationary energy storage value. The question of Vehicle-to-Grid (so called V2G) energy storage is an interesting and relevant one, but is a very large subject so I shall have to return to it on another occasion.
Having said that it is worth pointing out at this stage that the UK's National Grid does seem to be pinning rather a lot (40% to 55% of total energy storage) on V2G and being able to take advantage of the batteries on wheels that they assume the general public will be buying in future. It certainly would get them out of the situation of having to pay for someone else to provide the energy storage (between 35GW and 50GW total for the two scenarios that do meet our 2050 commitments) - the only problem is finding an eV owner who is prepared to invalidate their manufacturer's battery warranty by allowing the grid to use their battery whilst not driving anywhere.
The United States
Next up is the US - currently leading the worldwide charge towards energy storage deployment. This was covered nicely by Mark Higgins of Strategen, long time supporters of the Energy Storage series of conferences.
Mark described how, despite a federal government that is refusing to take any kind of lead in the country's energy transition, a range of tax incentives, strong state level government that has often mandated renewable energy adoption, and an advanced deregulated market for frequency/grid firming has allowed a number of US utilities to trial grid-scale energy storage - and generally they very much like what they have found.
California is clearly leading the charge amongst the US states, with a projected deployment of 1.7GWh of storage in 2019 and expected annual deployment of 4.4GWh in 2024. Whilst the total accumulated energy storage was not mentioned at just the current rates of adoption I estimate this as conservatively 12GWh by 2024.
It is not just California that has mandated renewable energy deployment (50% by 2025 and 60% by 2030), Hawaii has also previously mandated 100% renewables by 2045 and Nevada has recently joined the list requiring 50% renewables by 2030 . Arizona, with its incredible solar resources, second only to Nevada, recently rejected a mandate for 50% renewables by 2030 , however it may well meet this target purely as a result of economic pressures.
Mitigating the intermittency of renewables generation is a good reason to deploy energy storage it is by no means the only reason - and many other states, most prominently New York and New Jersey, are planning storage to replace peaker plants or offset grid strengthening in the forthcoming era of the electric vehicle.
Mark described how over the entire US Brattle group predicts that 50GW of energy storage could be installed if policies were enacted to allow energy storage installations to capture all the potential value streams that they are technically capable of:-
This prediction relates to the utterly memorable US Federal Energy Regulatory Commission's order no 841 which requires grid operators (there's at least ten Regional Transmission Operators) in the US to "come up with market rules for energy storage to participate in the wholesale energy, capacity and ancillary services markets that recognize the physical and operational characteristics of the resource." So next time someone asks you what FERC 841 is you can let them know it is not the latest drum 'n bass group.
So where does all this detail leave us - can we draw some general conclusions - and more importantly can we answer the question posed at the start of this article - 'exactly how big is the market for grid-scale stationary energy storage ?
To compare the three countries (US, UK and Germany) and the state of California I have plotted the installed Stationary energy storage capacity (in GWh) against the yearly total consumption of electrical energy (in TerraWatt hours, TWh. 1TWh = 1,000GWh ) for each of these cases. This needs to be done on a logarithmic scale in order to capture the large range of different electrical grid sizes.
I show only the two most extreme UK FES scenarios - the Community Renewables (CR) in red and the Steady Progession (SP) in yellow, though as previously mentioned the second of these would not meet the legal requirement of the 2008 Climate change act to reduce carbon emissions by 80% by 2050. All estimations, eg where only a GW power figure and not a GWh Energy figure are reported, have been made as conservative as possible so that the overall conclusion can also be guaranteed to be a conservative one.
The dark grey dashed line is a guide to the eye that shows a trend of proportional growth (i.e. a doubling in storage for each doubling in energy consumption - we expect small country's grids to behave in an essentially similar fashion as large ones so we should expect to see trends with this slope.) This line helpfully goes through the 10GWh storage capacity point at an annual consumption of 500TWh. A quick sanity check on this shows the storage capacity is 1/50,000th of the annual consumption - or about 0.7% of the daily energy consumption.
This may seem rather small but remember that this is just the installed capacity and it says nothing about how many times you might use such battery capacity each day. If the battery were used only once per day then the amount of energy provided by it might seem trifling, but if we sweat the battery asset and use it over many different timescales for grid balancing the energy provided to, and supplied by, it each day might be 2-5% of the total energy supplied - a valuable contribution and important grid balancing asset.
Whilst the level of the line arbitrarily coincides with this 10GWh/500TWh point it also makes some sense in that it shows Germany, with its strong interconnections with other country's grids and energy storage infrastructure as slightly below the average, and the UK (once it gets going) as an island without those advantages, somewhat above it.
The actual penetration may be somewhat above this dark grey trendline - the lighter dashed line shows the trend at double the storage integration - i.e. 20GWh for every 500TWh of consumption. The UK CR scenario is still expected to go significantly above this level of integration for the Community Renewables (CR) scenario - following in the footsteps of California, which despite a smaller overall electrical grid is likely to remain ahead of the UK in terms of the total amount of energy storage deployed. The US, taken as a whole might appear to lag behind these trends, even with FERC 841's 50GW storage dividend - however this conservatively only assumes 1 hour storage i.e. 50GWh - if 50GW translates into 100GWh storage then it rises straight up onto the original trendline.
The two trend lines form a useful band that we can very conservatively expect stationary storage integration levels to reach. What does this mean when applied for the potential worldwide stationary energy storage market ?
Worldwide energy consumption is currently 25,000 TWh - so using this as basis the theoretical worldwide demand for stationary energy storage would be between 500GWh and 1,000GWh. As can be seen from the detailed UK Future energy storage scenarios the greater integration of renewables and electric mobility will only increase the electrical energy demand going forward - add this to the growth in energy usage in China, India and other developing countries and the International Energy Agency estimates that by 2040 global electrical demand will have increased to between 36,000 and 40,000 TWh, This would increase storage to between 720GWh and 1,600GWh by that date.
How do these predictions compare with other existing predictions for Stationary Energy Storage, and if they agree, how much might this market be worth ?
Their estimate is a total installed capacity of 468GWh by 2027 and c. 100GWh installed per annum at that point - this would yield over 1,000GWh by 2030 and something close to our upper estimate of 1,600GWh by 2040 once retirements of early projects are considered. Our figures, derived purely from individual country's planning estimates, agree well with those of Navigant.
Allowing for an average life of 15 years per battery asset (ok, ok, we know that VRFB's will last much longer than Lithium-ion) in 2030-2040 we might expect battery replacements to conservatively require 100GWh replacement capacity per year at that point, even if the energy storage fleet does not expand past 1,600 GWh.
Using the assumed Navigant deployment cost of $0.5/Wh we can conclude that in 2030-2040 the global stationary energy storage market would be worth at least $50 Billion per year.
How much of this might be captured by VRFB's is discussed in the previous article.
This article only conveys the personal opinion of the author. Whilst every effort is made to ensure the content is accurate, we cannot guarantee the accuracy of the data shown. This article does not constitute professional, financial or investment advice and must not be used as a basis for making investment decisions.
Site content is not authorised by the FCA and you are not safeguarded by the Investor Protection measures of the Financial Services and Markets Act 2000. See our full disclaimer