A paper recently published by Berkeley Lab researchers in the journal iScienceexplores this question. The article, entitled "Private vs. public value of U.S. residential battery storage operated for solar self-consumption," quantifies the value of using residential battery storage to maximize solar self-consumption from the perspectives of both the individual solar customer and the larger power system. 

This question is becoming increasingly relevant as states replace traditional net metering rules with "net billing" structures. Net billing allows customers to offset their own consumption with solar on an instantaneous or hourly basis, receiving full retail rate credit for that portion of their solar generation (area B in the schematic to the right), but any solar generation exported to the grid is compensated at a reduced rate (area A). This kind of asymmetric pricing structure incentivizes solar customers with battery storage to charge their batteries with surplus solar and discharge that stored solar energy to serve load during evening and nighttime hours (area C).

Relying on hourly metered loads from roughly 1,800 residential customers across six utility service territories, this study assesses the value of this arbitrage behavior. Key findings from the analysis include the following:

• Even under a relatively optimistic set of assumptions, the bill savings from using storage in this manner would not be nearly enough on their own ($14-27 per kWh of storage capacity, annually, across the sample of customers) to justify the current up-front cost of residential battery storage.
• Storage used for solar self-consumption yields virtually no value to the bulk power system in terms of reduced wholesale energy costs. This is primarily due to a misalignment between the temporal profiles of storage dispatch and wholesale energy prices. The analysis shows that even in a future with high solar penetration, where wholesale prices resemble the proverbial "duck curve", the energy value of storage dispatched for solar self-consumption remains highly suboptimal.
• Storage used for solar self-consumption also yields virtually no value in terms of reduced peak-related costs, such as those related to generation, transmission, and distribution system capacity. This is primarily due to the fact almost all solar generation on peak-load days is used to directly serve onsite customer load, resulting in little surplus solar energy (area A in the schematic above) available to fuel storage discharge during peak-load hours later in the day. As a result, battery storage largely sits idle on those days, barring some other incentive to operate for system-peak reduction purposes.

The analysis compares the results above to what would occur if storage were, instead, operated optimally from the power system perspective, independent of customer load. For the particular locations and time periods analyzed, the resulting system value equates to $16-33 per kWh of storage capacity. The paper then shows how net billing tariffs could be designed or coupled with other incentives in a manner that could capture a large fraction of that potential system value, without degrading solar self-consumption levels or imposing significant additional stress on the local distribution network. In part, that can be achieved by allowing and incentivizing limited storage discharge to the grid during infrequent, high-value hours.

For further details, please download the paper and accompanying summary slide deck. The authors will also be hosting a webinar to discuss the results on August 4th at 10 am Pacific / 1 pm Eastern. Please register here: https://lbnl.zoom.us/webinar/register/WN_J566sElLS02fdUvHRKxx5g.

We thank the U.S. Department of Energy Solar Energy Technologies Office for their support of this work.

Berkeley Lab | emp.lbl.gov