Thermal Constrained Energy Optimization of Railway Cophase Systems With ESS Integration—An FRA-Pruned DQN Approach

This article investigates the railway cophase traction power supply system (TPSS) with a power flow controller (PFC) to address the power quality and neutral section issues. To collect the regenerative energy and achieve a more flexible power flow, the energy storage system (ESS) is integrated into the cophase system. As the key components, the reliability of power electronics modules in PFC and battery cells in ESS is highly related to their thermal performance. It is, therefore, vital to consider their operational thermal dynamics, leading to the proposal of a deep Q-learning network (DQN)-based thermal constrained energy management strategy in this article. First, the system power flow model and electrothermal models for power electronics modules and battery cells are all established. Then, a DQN method is adopted to learn an optimized policy for peak power shaving while meetings thermal constraints. Finally, an FRA-based pruning method is proposed to reshape the agent to become more compact without sacrificing its performance. Case studies confirm that the proposed strategy can effectively reduce the peak traction power supply by up to 42.0% and achieve up to 94.1% thermal reduction. The FRA-based pruning can achieve up to 89.9% agent size reduction and 87.2% computation savings.