Optimal BESS Allocation in Large Transmission Networks Using Linearized BESS Models

Abstract: The most commonly used model for battery energy storage systems (BESSs) in optimal BESS allocation problems is a constant-efficiency model. However, the charging and discharging efficiencies of BESSs vary non-linearly as functions of their state-of-charge, temperature, charging/discharging powers, as well as the BESS technology being considered. Therefore, constant-efficiency models may inaccurately represent the non-linear operating characteristics of the BESS. In this paper, we first create technology-specific linearized BESS models derived from the actual non-linear BESS models. We then incorporate the linearized BESS models into a mixed-integer linear programming framework for optimal multi-technology BESS allocation. Studies carried out on a 2,604-bus U.S. transmission network demonstrate the benefits of utilizing the linearized BESS models from the model accuracy, convexity, and computational performance viewpoints.