基于安全增强型多智能体强化学习的网络化微电网协同与电池换电站调度

Networked microgrids (NMGs) have garnered significant attention in recent years due to their autonomic ability. With the increasing integration of electric vehicles (EVs) in microgrids, battery swapping stations (BSSs) have emerged as a critical infrastructure to accelerate EV adoption. However, the operational characteristics of BSSs can introduce potential challenges to the secure operation of NMGs, particularly in terms of voltage regulation and line loading. This paper proposes a security-enhanced multi-agent reinforcement learning method for an NMG coordination model utilizing BSSs to address these challenges. Specifically, each NMG agent dynamically captures battery demand and retail price variations to make security-aware decisions, ensuring system operational security while maximizing cost efficiency. The proposed model is validated on a modified IEEE 33-bus distribution network with three NMGs. Results demonstrate that the proposed approach effectively enhances the safe operation of NMGs while preserving the economic benefits of integrating BSSs into NMG operation.