Optimal Energy Management Strategy Based on Neural Network Algorithm for Fuel Cell Hybrid Vehicle Considering Fuel cell Lifetime and Fuel Consumption

Abstract

Abstract In order to improve the power performance, fuel cell lifetime, and fuel consumption of fuel cell/battery/ supercapacitor powered-vehicle, this paper proposes a new framework of energy management strategy (EMS) consisting of neural network optimization algorithm (NNOA) optimized fuzzy logic controller-based frequency decoupling and adaptive super-twisting sliding mode control based on nonlinear disturbance observer (ASTSMC-NDOB). In the proposed EMS, frequency decoupling based on adaptive low-pass filter and Harr wavelet transform (HWT) using fuzzy logic controllers (FLCs) are employed to decouple the required power into low, medium, and high-frequency components for fuel cell, battery, and supercapacitor, respectively. The proposed frequency decoupling-based strategy can improve the power performance of the vehicle as well as reduce load stress and power fluctuation on fuel cell. In order to precisely optimize membership functions of suggested FLCs, NNOA is adopted to tune them while minimizing the objective function, considering the hydrogen consumption and constraints on the battery/supercapacitor SOC. Furthermore, in order to achieve robustness and high-precision control, the ASTSMC-NDOB controllers are developed to stabilize the DC bus voltage and force currents of the fuel cell, battery, and supercapacitor to track their obtained reference values. The fuel cell hybrid electric vehicle with proposed EMS is modelled on MATLAB/Simulink, and three driving conditions such as HWFET, UDDS, and WLTP driving schedules are used for evaluation. The findings exhibit that the proposed EMS can effectively improve the fuel economy, reduce power fluctuation on fuel cell, and prolong its lifetime compared to state machine strategy and fuzzy logic control-based EMS.