With declining costs, improved technologies, and increasing deployment, energy storage is poised to become a growing part of the evolving U.S. power system. But measuring the value of energy storage is inherently complex—and future systems will likely include multiple storage technologies, adding new complexity.
To answer the big questions around the role of storage in our future grid, the National Renewable Energy Laboratory (NREL) has launched the multiyear Storage Futures Study (SFS). Supported by the U.S. Department of Energy’s (DOE’s) Energy Storage Grand Challenge, the study explores how energy storage technology advancement could impact the deployment of utility-scale storage and adoption of distributed storage, as well as future power system infrastructure investment and operations.
In the first report in the series, The Four Phases of Storage Deployment: A Framework for the Expanding Role of Storage in the U.S. Power SystemPDF, NREL analysts Paul Denholm, Wesley Cole, Will Frazier, and Nate Blair, along with DOE analyst Kara Podkaminer, outline a visionary framework for the possible evolution of the stationary energy storage industry—and the power system as a whole. The framework presents a value proposition of cost-competitive storage deployment in four phases, potentially resulting in hundreds of gigawatts of installed capacity and a significant shift in our electric grid.
| Phase | Primary Services | National Deployment Potential (Capacity) in Each Phase | Duration | Response Speed |
|---|---|---|---|---|
| Deployment prior to 2010 | Peaking capacity, energy time-shifting and operating reserves | 23 gigawatts of pumped storage hydropower | Mostly 8–12 hr | Varies |
| 1 | Operating reserves | <30 gigawatts | <1 hr | Milliseconds to seconds |
| 2 | Peaking capacity | 30–100 gigawatts, strongly linked to photovoltaics deployment | 2–6 hr | Minutes |
| 3 | Diurnal capacity and energy time shifting | 100+ gigawatts. Depends on both Phase 2 and deployment of variable renewable energy resources | 4–12 hr | Minutes |
| 4 | Multiday to seasonal capacity and energy time-shifting | Zero to more than 250 gigawatts | >12 hr | Minutes |
Phase One: Short-Duration Storage To Provide Operating Reserves
Assuming the cost of most storage technologies increases with duration, the analysts make a case for deployments following a natural progression from shorter to longer duration over time, particularly aligned with current and anticipated growth in photovoltaics and wind power. Phase one builds on the long history of energy storage on the grid that has been primarily supplied by pumped storage hydro, and which is by far the dominant source of energy storage today with 22 gigawatts installed on the U.S. electric grid.
In phase one of the proposed framework, short-duration storage provides operating reserves to the grid—increasing or decreasing output for a short amount of time to help maintain the balance of supply and demand on the grid.
Monetization of existing frequency response requirements, additional market products, and growth in regulating reserves from variable generation could increase the deployment of storage in phase one.
Phase Two: The Rise of Battery Peaking Power Plants
Falling battery prices have introduced the opportunities for phase two—the deployment of batteries with 2–6 hours of duration to meet peak demand on hot summer days or in extreme cold.
Batteries’ cost-competitiveness is based on their ability to provide the same level of peaking capacity compared to traditional resources such as gas turbines. NREL analysts found significant opportunities for batteries with 4-hour durations.
In this phase, the cost-competitiveness of battery storage increases with hybrid system configurations, where storage is co-located with renewable generation sources, particularly solar photovoltaics, to receive an investment tax credit. In addition, cost-competitiveness increases when demand for additional capacity increases or traditional sources retire.
Phase Three: The Age of Low-Cost Diurnal Storage
Decreased value of shorter-duration capacity marks the transition to phase three—the advent of storage technologies that can provide additional or lower-cost services to meet longer peak periods.
Phase three largely depends on how much storage costs decline and variable renewable energy deployment increases. At higher levels of variable generation deployment, cost-effective storage with up to 12 hours of duration could potentially provide significant amounts of capacity-related services.
Longer-duration storage can provide other services that add flexibility to the grid, like supporting transmission from remote wind generation sites.
Phase Four: The End Game—Multiday to Seasonal Storage
The final phase of NREL’s vision for storage deployment introduces seasonal storage, driven by the seasonal mismatch of variable resource supply and demand, particularly in a deep-decarbonization future that relies heavily on renewable resources.
Seasonal storage would mitigate this imbalance by shifting excess renewable energy available in the spring to times of higher demand and/or lower availability of renewable resources.
There are a variety of technologies that can be built for seasonal or long-duration storage. Compressed air and pumped storage hydropower have already been deployed at scale. Other technologies involve producing hydrogen, methane, or other fuels that can be stored underground for months or years. In this phase, when and how much seasonal storage is deployed largely depends on the competitiveness and declining costs of these or other emerging technologies.
Coming Next: More from the Storage Futures Study
Many concepts presented in this vision will be further explored in upcoming studies, including detailed results of the modeling and analysis of power system evolution scenarios and their operational implications.
An improved characterization of various grid services needed, including capacity and duration, could support a deeper understanding of the trade-offs between various technologies and non-storage resources such as responsive demand. This would help ensure the mix of flexible technologies deployed is robust to an evolving grid, which will ultimately determine the amount of storage and flexibility the power system will need.
