Cooperative Control Strategy of Shared Energy Storage System for Smart Grid: A Case Study of Campus Building Energy Management

With the advancement of technology in energy storage systems (ESS) coupled with photovoltaics (PV), research on energy management systems is actively being conducted. However, due to the high investment costs associated with ESS, shared ESS used by multiple consumers has emerged as a current solution. Additionally, the allocation of capacity in shared ESS must be carefully considered so that each consumer can maximize its utilization. Therefore, we propose a multi-agent deep reinforcement learning-based energy management system that considers cooperative sharing of ESS between consumers. Furthermore, we conducted two additional cases (fair capacity allocation and a single-agent case) to verify the cooperative agent's performance. Finally, from the result of the experiment, the proposed model has been shown to maximize cost reduction effectively compared with both other cases.