Long-run system value of battery energy storage in future grids with increasing wind and solar generation

With declining costs of battery storage, there is growing interest to deploy them in power systems to provide multiple grid services that directly support integration of variable renewable energy (VRE) generation. Here, we assess the holistic system value of energy storage in future grids with increasing wind and solar generation. We also identify the major sources of storage value and their dynamics under different system settings and at increasing storage and wind and solar penetration levels. We use a high temporal resolution capacity expansion model to study least-cost integration of storage in two variants of an abstract power system that is populated with load and VRE profiles consistent with the U.S. Northeast (North) and Texas (South) regions. For both systems, storage value originates primarily from deferring investments in generation capacity (VRE, natural gas) and transmission, and generally declines with increasing storage penetration. Increasing VRE penetration from 40% to 60% increases the value of storage, but only enough to make storage capacity up to 4% of peak demand cost-effective at current Lithium-ion capital costs. With future capital costs of $150/kWh for 4 h duration storage, the cost-effective storage penetration ranges between 4% and 16% of peak demand across the system scenarios studied here. Storage substitution of natural gas capacity is dependent on the VRE resource mix and penetration level, but is less than 1 GW per GW of storage added for the durations (2, 4 or 8 h) considered here. Increasing storage duration increases storage value in some cases, but this increase in value may be insufficient to compensate for the increase in capital cost per kW even under the future cost scenario.