An Improved MPC-based energy management strategy for hydrogen fuel cell EVs featuring dual-motor coupling powertrain

Abstract

Hydrogen fuel cell electric vehicles (HFCEVs) provide significant environmental benefits. Integrating dual-motor coupling powertrains (DMCPs) further enhances efficiency and dynamic performance. This article proposes an energy management strategy (EMS) for the hydrogen fuel cell/battery/super-capacitor system in an HFCEV with DMCP. Model predictive control (MPC) is adopted as the framework to optimize economic performance, defined in this study as the hydrogen consumption cost and fuel cell degradation cost. To improve the prediction horizon and accuracy, the torque split ratio for two varying permanent magnet synchronous motors (PMSMs) and the corresponding mode switching rules of the vehicle are initially established. Subsequently, a combination of Dynamic Programming (DP) and MPC is selected as the framework, utilizing a Dung Beetle Optimizer (DBO)-optimized Bidirectional Long Short-Term Memory (BiLSTM) network to refine the predictive model. Finally, comparisons with other predictive models and commonly used control strategies demonstrate that the proposed EMS notably improves economic performance.