【摘要】 本文针对传统模型预测控制的燃料电池船舶能量管理策略，存在预测区间内负载功率信息缺失，导致等效氢耗增加的问题，提出基于随机模型预测控制的能量管理策略。依托收集的船速数据构建马尔可夫链精准预测船速以实现负载功率预测，提出以等效氢耗最小化、锂电池荷电状态维持合理区间、燃料电池效率最大化为多目标函数的随机模型预测控制策略，将其与有限状态机策略、传统模型预测控制策略对比，并深入分析不同权重参数组合下的优化效果。仿真结果表明，与传统模型预测控制策略相比，等效氢耗减少了1.7%，燃料电池工作在高效率区间的时间更长，锂电池荷电状态与参考值误差为1.32%，并进行硬件在环实验验证。本文提出的随机模型预测控制策略，显著提升系统能效与动态适应性，为绿色航运与“双碳”战略深度融合提供创新实践支撑。更多还原

【Abstract】 Aiming at the issue of missing load power information within the prediction horizon in traditional model predictive control-based energy management strategies for fuel cell ships, which leads to increased equivalent hydrogen consumption, this paper proposes an energy management strategy based on stochastic model predictive control. Leveraging collected ship speed data to construct a Markov chain for accurate ship speed prediction and load power forecasting, this study proposes a stochastic model predictive control strategy with multi-objective functions: minimizing equivalent hydrogen consumption, maintaining the lithium battery state of charge(SOC) within a reasonable range, and maximizing fuel cell efficiency. The proposed strategy is compared with finite state machine strategies and traditional model predictive control strategies, while an in-depth analysis is conducted to evaluate optimization effects under different combinations of weighting parameters. The simulation results indicate that, compared to the model predictive control strategy, the equivalent hydrogen consumption is reduced by 1.7%, the fuel cell operates in the high-efficiency range for a longer duration, and the lithium battery SOC exhibits a 1.32% error relative to the reference value, and hardware-in-the-loop experimental verification. The stochastic model predictive control(SMPC) strategy proposed in this study significantly enhances system energy efficiency and dynamic adaptability, offering innovative practical support for the deep integration of green shipping and the "dual-carbon" strategic goals.更多还原