A stochastic sizing approach for sharing-based energy storage applications

In order to foster renewable energy integration, improve power quality and reliability, and reduce hydrocarbon emissions, there is a strong need to deploy energy storage systems (ESSs), which can provide a control medium for peak hour utility operations. ESSs are especially desirable at the residential level, as this sector has the most untapped demand response potential. However, considering their high acquisition, operation, and maintenance costs, isolated deployment of ESSs is not economically viable. Hence, this paper proposes a sharing-based ESS architecture, in which the demand of each customer is modeled stochastically and the aggregate demand is accommodated by a combination of power drawn from the grid and the storage unit when the demand exceeds grid capacity. The optimal size of ESSs is analyzed and an analytical method is developed for a group of customers with a single type of appliances. This framework is also extended to any network size with an arbitrary number of customers and appliance types, where the analytical method provides a tractable solution to the ESS sizing problem. Finally, a detailed cost-benefit analysis is provided, where the results indicate that sharing-based ESSs are practical and yield significant savings in terms of ESS size.