by Fereidoon Sioshansi, PhD, Menlo Energy
Falling costs, improved performance and mass scale will make energy storage hard to beat.
To say that energy storage is a booming business may be an understatement. In 2010, there were seven large battery storage systems with 59 MW of capacity in the US. In 2015, there were 49 systems with 351 MW of storage capacity. By the end of 2018, there were 125 systems with 869 MW of capacity.
Part of the reason is a 61% drop in the cost of storage, from US$2153/kWh in 2015 to $834/kWh in 2017, according to the Energy Information Administration (EIA). Now there is talk of prices in the sub-$150/kWh, and lower, as the technology improves and is scaled up.
When Tesla built its 100 MW battery in 100 days in South Australia, it made headlines. The latest issue of the US Energy Storage Monitor report, published by Wood Mackenzie and the US Energy Storage Association expects the US energy storage market to expand to annual deployments of 7,3 GW in 2025.
Figure 1: The difference in net loads with and without storage.
The more important reason is that the value of storage is on the rise. A new study has concluded that – unsurprisingly – the economic value of energy storage increases as variable renewable energy generation in the electricity mix increases. The study, published in the journal Applied Energy, was a collaboration among researchers from the Massachusetts Institute of Technology (MIT) and Princeton University’s Andlinger Centre for Energy and the Environment (ACEE), supported by General Electric (GE).
According to Dharik Mallapragada, one of the lead authors, “Battery storage helps make better use of electricity system assets, including wind and solar farms, natural gas power plants, and transmission lines, and can defer or eliminate unnecessary investment in these capital-intensive assets.”
As documented in numerous other studies, the MIT report notes that this capacity deferral or substitution value of batteries for generation or transmission capacity is the primary source of their value. Other sources of value include the ability to provide operating reserves to the grid operator, avoiding fuel cost and wear and tear incurred when cycling gas-fired peaking plants on and off, as well as shifting energy from low to high value periods.
The report’s key conclusions are:
- The economic value of storage rises as the percentage of variable renewable energy generation rises.
- The value, however, begins to decline as storage penetration increases due to competition between various storage resources for the same set of grid services.
- As the amount of storage increases, the majority of its economic value is tied to its ability to displace the need for investing in both renewable and natural gas-based energy generation and the need for more transmission capacity.
- Finally, at current costs, a relatively small amount of storage is cost-effective in grids with a high percentage of variable renewable generation, say above 50 or 60%.
The value of storage, like everything else, is subject to diminishing marginal returns. “As more storage is added, the value of additional storage steadily falls,” said Jesse Jenkins, one of the MIT researchers. “That creates a race between the declining cost of batteries and their declining value, and our paper demonstrates that the cost of batteries must continue to fall if storage is to play a major role in electricity systems.”
The authors also note the need for reforming electricity market rules to enable energy storage developers not only to participate but to better monetize the value of substituting generation and transmission capacity.
According to Mallapragada, “Depending on their … design and market rules, capacity markets may or may not adequately compensate storage for providing energy during peak load periods.” The good news is that storage costs are indeed falling, and grid operators are changing the rules for storage to play a more active role in markets with high renewable penetration. In late July 2020, for example, CAISO (the California Independent System Operator) agreed to allow owners of hybrid power plants – for example, solar paired with batteries – to bid and operate the two sources in tandem.
This was in response to the surging trend of pairing storage with solar projects. It is also aimed at filling the 3300 MW gap at peak hours as natural gas-fired power plants retire in California. CAISO is also easing the way for hybrid projects, for example, allowing solar plants with storage to ramp up and batteries to ramp down in response to signals in real-time.
As reported in CA Current, University of California Berkeley’s Prof. Severin Borenstein, who sits on CAISO’s board, pointed out that hybrid projects are ideally suited to respond to real-time conditions, both better and faster than peaking plants.
Greg Cook, CAISO’s executive director for market policy noted strong interest from developers in adding more storage to existing solar and wind plants, which is much faster and less expensive than installing batteries in separate locations and because existing renewable plants already have the necessary hardware and interconnections to the grid.
According to Elizabeth McCarthy of CA Current, the change was long overdue. There are currently over 63 GW of proposed hybrid projects – with energy storage – in the CAISO interconnection queue. More than 68 GW are battery projects. “There is more storage in the queue than renewables,” Bob Emmert, CAISO’s senior manager for interconnection resources, said.
The utility-scale batteries favoured by hybrid plants, however, are not the only game in town; there is also growing interest in distributed storage provided by EVs and/or distributed batteries. In late July, CA Current reported that three community choice aggregators (CCAs) in San Francisco have formed a partnership with Sunrun, the biggest US solar PV installer, to develop 20 MW of new behind-the-meter (BTM) solar plus battery backup to keep electricity flowing during emergency power shutoffs.
The three CCAs, East Bay Community Energy, Peninsula Clean Energy and Silicon Valley Clean Energy, are offering lower bills and financial incentives to some 6000 participating customers including a $1000 sign-up bonus for offering demand flexibility – which Sunrun will aggregate towards meeting California’s resource adequacy obligation, a highly coveted product in CAISO market, especially given what happened in mid-August 2020.
One of the selling points of the battery plus storage scheme is to protect customers in case of future emergency power shutoffs during the fire season. In 2019, Pacific Gas & Electric Co (PG&E) cut off power to some 2-million people at the height of the devastating brush fires, some for several days. Sunrun is not alone in trying to monetize the rising value of storage. Stem, another San Francisco based company, is expanding its offerings by managing some 345 MWh of behind-the-meter storage assets in 25 commercial and municipal customers while providing local controllable capacity to the utility, Southern California Edison Co. (SCE). Stem claims to manage over 790 MWh of flexible resources at over 1000 sites in nine states and three countries, more than any of its competitors.
113 GW of storage in hybrid plants waiting in the queue
As battery prices fall and wind and solar generation rise, developers are increasingly combining wind and solar projects with on-site batteries, creating “hybrid” power plants. A report recently released by the Lawrence Berkeley National Lab (LBL) provides a summary of the existing and planned hybrid and co-located plants including the status of interconnection queues across the US.
The focus was on larger, 1 MW plus systems. Smaller, often behind-the-meter, projects are increasingly common but were not included in the LBL study. According to the LBL report, there are at least 125 co-located hybrid plants greater than 1 MW already operating in the US at the end of 2019, totalling over 14 GW of capacity.
The most common configurations include:
- Wind plus storage: 13 projects, 1290 MW wind, 184 MW storage;
- Solar PV plus storage: 40 projects, 882 MW PV, 169 MW storage; and
- Fossil plus storage: 10 projects, 2414 MW fossil, 91 MW storage.
These numbers, however, are dwarfed by what is in the nation’s interconnection queues. At the end of 2019, there were 367 GW of solar plants in the queues; 102 GW (approx. 28%) of this capacity was proposed as a hybrid, most typically pairing PV with battery storage.
The corresponding number for wind is 225 GW, with 11 GW (approx. 5%) proposed as a hybrid, again most-often pairing wind with storage. The hybrid plants are located throughout the US with California and the non-ISO West being the most prominent.
In time, storage will begin to challenge natural gas-fired peakers. A recent study of the effective load carrying capability (ELCC) of different renewable technologies by California’s three big investor-owned utilities (IOUs) concluded that the days of peaker gas plants may be numbered, especially if the aim is to phase out fossil-fuel emissions.
The study examined the effective load carrying capability (ELCC) of various energy technologies. ELCC is a measure of how much of a wind or solar plant’s nameplate capacity can be counted on when the grid needs it the most – an important issue for a state like California contemplating a 100% carbon-free electricity future. It found that wind has a 19% ELCC, which means that a wind park with a capacity of 100 MW could be counted on to reliably supply 19 MW of energy. Not good.
Add four hours of energy storage and the ELCC improves a little but still remains well below that of a 100 MW natural gas-fired power plant. Not good. Utility-scale solar, according to the study, is “close to useless” at reliably meeting load when needed; distributed residential-scale solar is even worse with an ELCC of 4%. But solar’s performance gets a significant boost – to 99,8% – by adding four hours of storage.
As more solar is added over time, that value declines slightly, but by not much. The stunning difference between wind and solar is due to the fact that solar can be counted on to charge a battery so that it’s available when needed 99,8% of the time while wind isn’t as consistent, These findings are good news for solar-plus-storage developers. It should come as no surprise that virtually all new solar plants are now paired with storage. This is critical given what happened in mid-August at CAISO.
Historically, natural gas peaking plants were built to operate for relatively short periods of time to meet spikes in demand, say to meet air conditioning load during a few afternoons on hot summer days. With the rise of variable renewable resources, they fill in the gaps when the former’s generation drops, such as during the late afternoon when the sun sets, and solar generation drops to zero.
But if solar-plus-storage is highly predictable and reliable, who’ll need natural gas peaker? California, for instance, has 52 peaker plants, the majority of which typically run for less than four hours at a time, and only as needed. A solar plant with four hours of storage, according to the study, will make new gas peakers a non-starter while competing withthe existing ones.
As with any study, one has to interpret the results of the ELCC study with a grain of salt. Most experts believe that there will be room for some gas-fired power generators – as long as they can earn high prices at times of peak demand or renewable generation shortfalls. The price of electricity in all competitive wholesale markets spikes now and then even if not for too many hours. In the ERCOT market in Texas, for instance, prices are allowed to reach $9000/MWh. Prices above $1500/MWh are not uncommon even though for most hours’ prices are around $20/MWh.
We can be sure, however, that solar plus storage is likely to make life a lot tougher for natural gas peakers over time.
This article was first published in the November 2020 issue of EEnergy Informer and is republished here with permission.
Contact Fereidoon Sioshansi, Menlo Energy, firstname.lastname@example.org