TD3-Based EMS Using Action Mask and Considering Battery Aging for Hybrid Electric Dump Trucks

The hybrid electric dump truck is equipped with multiple power sources, and each powertrain component controlled by an energy management strategy (EMS) to split the demanded power. This study proposes EMS based on deep reinforcement learning (DRL) algorithm extend battery life reduced total usage cost for vehicle, namely twin delayed deterministic policy gradient (TD3) EMS. Firstly, vehicle model constructed optimization objective function, including aging fuel consumption cost, designed. Secondly, TD3-based used continuous action control of ICE state, mask applied filter out invalid actions. Thirdly, simulations EMSs are trained under CHTC-D driving cycle C-WTVC cycle. results show that improves convergence efficiency strategies, proposed outperforms (DDPG) Meanwhile, extended 36.17% 35.49% C-WTVC, 4.30% 2.49% when considers aging. In summary, can reduce provides a method solve problem systems.