Actor-Critic TD3-based Deep Reinforcement Learning for Energy Management Strategy of HEV

Yazar, OzanCoskun, SerdarLi, LinZhang, FengqiHuang, Cong2025-03-172025-03-172023979-835033752-5https://doi.org/10.1109/HORA58378.2023.10156727https://hdl.handle.net/20.500.13099/13995th International Congress on Human-Computer Interaction, Optimization and Robotic Applications, HORA 2023 -- 8 June 2023 through 10 June 2023 -- Istanbul -- 190025In the last decade, deep reinforcement learning (DRL) algorithms have been employed in the design of energy management strategy (EMS) for hybrid electric vehicles (HEVs). Investigation of the real-time applicability of DRL algorithms as an EMS is critical in terms of training time, fuel savings, and state-of-charge (SOC) sustainability. To this end, we propose a twin delayed deep deterministic policy gradient (TD3) algorithm that is an improved version of the deep deterministic policy gradient (DDPG) algorithm for HEV fuel savings. Compared to the existing Q-learning-based reinforcement learning and the deep Q-network-based and DDPG-based deep reinforcement algorithms, the proposed TD3 provides stable training efficiency, promising fuel economy, and a lower variation range of SOC charge sustainability under various drive cycles. © 2023 IEEE.eninfo:eu-repo/semantics/closedAccessactor-critic networkDeep reinforcement learningenergy managementhybrid electric vehiclesTD3 algoritmActor-Critic TD3-based Deep Reinforcement Learning for Energy Management Strategy of HEVConference Object10.1109/HORA58378.2023.101567272-s2.0-85165707248N/A