An integrated energy management strategy for plug-in hybrid electric buses based on receding horizon control and TD3 algorithm

Due to its exceptional performance in terms of fuel efficiency, emissions reduction and driving convenience, the plug-in hybrid electric vehicle (PHEV) possesses a broad application market and development potential. However, the utilization of multiple power sources necessitates a rational and efficient energy management strategy (EMS) to coordinate multiple power sources to achieve efficient power output. In this study, an EMS within the receding horizon control (RHC) framework is proposed for the plug-in hybrid electric bus (PHEB), and a strategy based on twin delayed deep deterministic policy gradient (TD3) algorithm is introduced as a complementary strategy to enhance the robustness of the EMS. First, a vehicle velocity prediction model is constructed based on the gated recurrent unit (GRU) neural network with an attention mechanism to enable accurate prediction of future velocity in a finite horizon. Subsequently, a multi-objective RHC framework is established to effectively coordinate the objectives of vehicle fuel economy improvement and battery degradation mitigation. The power allocation problem is formulated as a rolling optimization issues over a finite prediction horizon, and the optimal control sequence is solved by the alternating direction method of multipliers (ADMM) algorithm. Additionally, the real-time monitoring of velocity prediction error enables the control system to timely switch to the TD3-based EMS when the error exceeds the preset range, so as to cope with unexpected situations. The simulation results demonstrate that the proposed EMS ensures reasonable battery charging and discharging under different initial state of charge (SOC) and driving distances, and mitigates battery degradation. Meanwhile, in comparison to the separate RHC strategy and TD3-based strategy, the proposed integrated EMS reduces the total cost of PHEB by 5.50% and 7.03%, respectively, thereby highlighting the superior fuel efficiency and adaptability to different driving conditions of the EMS. • An integrated EMS considering fuel economy and battery aging is proposed. • RHC-based EMS serves as the main strategy for efficient management. • TD3 improves robustness and adaptability against uncertainties.