Predictive energy management strategy of dual-mode hybrid electric vehicles combining dynamic coordination control and simultaneous power distribution

For the energy management, the energy conversion usually attracts focus of the researches in the control strategy design of hybrid electric vehicle (HEV), but the computational efficiency and dynamic coordination problem are often ignored, especially for the multi-mode HEV. Thus, this paper proposes a model predictive control (MPC)-based predictive energy management strategy for dual-mode HEV. In this strategy, the future vehicle speed is predicted in the given horizon, and then, an improved sequence quadratic programming algorithm (ISQP) that combines the deep Q-learning is designed to solve MPC problem, which effectively improves the computational efficiency and optimality of original SQP in iterative optimization. Meanwhile, a dynamic process coordination control algorithm is developed to take the torque coordination problem and balance relationship of mode shift dynamic process into the energy management problem. Eventually, the DP, SQP-MPC and rule-based energy management strategy are designed as the benchmark strategies to compare with the proposed method, and they are conducted in the three different test cycles. The results verify that the proposed strategy presents the desirable performance in fuel saving, real-time capability and robustness.