Multi-time Scale Energy Management Strategy for Plug-in Hybrid Electric Vehicle with DMP Machine Considering Electromagnetic Coupling

The dual mechanical port (DMP) machine demonstrates potential for plug-in hybrid electric vehicle (PHEV), however, its electromagnetic (EM) coupling compromises the fuel economy and dynamic performance. To coordinate energy saving and drivability improvement, a multitime scale energy management strategy (EMS) for PHEV with DMP machine considering EM coupling is proposed in this paper. In the proposed EMS, the back propagation neural network optimized by particle swarm optimization is designed first to predict the future possible velocity. Then, the multi-time scale structure decouples the trade-off between fuel economy and dynamic performance through temporal decomposition, and a multi-time scale model of DMP machine replaces the conventional motor model to account for its EM coupling effects. In coarse-grained time scale, an alternating direction method of multipliers embedded nonlinear model predictive control (NMPC) is adopted to improve computational efficiency and fuel economy improvement. In fine-grained time scale, a compound torque coordination control is designed to consider such fast system behavior as the torque ripple of engine and torque reduction due to EM coupling. Finally, the performance of the proposed strategy is verified by simulation and hardware-in-loop test. The results indicate that the proposed strategy outperforms the existing rule-based and NMPC-based strategies and reaches above 96% energy-saving rate of the offline global optimum. Meanwhile, compared with single time scale EMS, the torque ripple of output shaft is reduced by 45.24%, highlighting its anticipated dynamic performance.