关键词: 车速预测；工况类型；能量管理；工况自适应；氢耗

Abstract: Common operating scenarios for fuel cell heavy-duty trucks include low-speed operations within yards, medium-speed urban driving, plain national highways, mountainous national highways, and highway operations. Different operating scenarios impose varying requirements on the output power of the fuel cell and the discharge capability of the power battery, presenting challenges in optimizing energy efficiency and durability for fuel cell vehicles under complex and changing conditions. This paper designs a driving condition-adaptive energy management strategy for fuel cell vehicles. Firstly, a machine learning model is constructed based on historical data to predict the future vehicle speed. Secondly, the vehicle's operating state is classified by combining predicted data with real-time data. Subsequently, energy management strategies with optimal energy consumption under different operating conditions are defined based on experimental results to ensure optimal vehicle economy under corresponding conditions. Finally, the output power of the fuel cell is filtered. According to simulation comparative analysis, the condition-adaptive energy management strategy reduces hydrogen consumption by 1.8% compared to the original strategy, while maintaining good vehicle following performance.

Key words: speed prediction; driving condition type; energy management; driving condition adaptation; hydrogen consumption