Energy management strategy for fuel cell hybrid electric vehicles using Pontryagin’s minimum principle and dynamic SoC planning

A reasonable and reliable energy management strategy (EMS) is crucial for fuel cell hybrid electric vehicles (FCHEV). A nonlinear State of Charge (SoC) trajectory planning with Proportional Integral Controller-based Pontryagin's minimum principle (Nonlinear PI-PMP) EMS is proposed to adjust the power battery SoC consumption rate in response to various power demands and to prevent the fuel cell system (FCS) from operating under unfavorable working conditions. The SoC reference trajectory is updated based on the predicted vehicle speed during the drive, and the PI controller is used to update the co-state in real-time. To comply with the SoC trajectory update framework, a speed prediction method based on LVQ neural network clustering and Improved Markov Velocity Predictor (IMVP) is introduced. Moreover, to balance the economy and durability of FCS, a penalty term is added to the Hamiltonian function to suppress FCS power output fluctuations and maintain the FCS output power within the high-efficiency range. Through the construction of a hybrid driving cycle, the simulation results validate that the proposed nonlinear PI-PMP EMS can significantly reduce hydrogen consumption. Specifically, when the initial SoC is 30%, compared to the Linear PI-PMP, A-PMP, and fuzzy logic strategies, the nonlinear PI-PMP EMS achieves reductions of 4.22%, 4.27%, and 15.75% respectively. Moreover, it efficiently decreases FCS power fluctuations to 1.12.