Optimal scheduling for multi-ESS considering life-cycle battery degradation

In this study, we propose an optimal strategy for managing multiple energy storage systems (ESS) to reduce both electricity expenses and battery degradation costs. Our approach uses a two-stage optimization process for battery management. In the first stage, reinforcement learning (RL) identifies the aggregated optimal amounts for charging and discharging. Then, quadratic programming (QP) distributes these aggregated amounts across multiple batteries. Experiments conducted under different test conditions, including variations in the number of batteries and their remaining lifespans, show that ESS with reused batteries can improve operational efficiency and achieve total cost savings of 1.7% to 11.2% compared to ESS with new batteries.