This paper is focused on the energy-saving control issues of Fuel Cell Hybrid Electric Tractors(FCHET),presenting a rapidly trainable and real-time applicable energy management strategy.The strategy derives a computational model capable of achieving rate allocation in a single matrix computation by learning from the optimal power distribution control sequences of past similar operating conditions.To validate the optimization performance and computational efficiency of the proposed method,a standard cycle condition is constructed based on Markov theory for simulation verification.Simulation results indicate that the hydrogen consumption of the proposed method is only 1.61％higher than strategies based on Dynamic Programming(DP),while saving 1.54％equivalent hydrogen consumption compared to traditional Model Predictive Control(MPC)strategies,with significantly lower computational time than MPC strategies.
关键词
能量管理策略/多目标优化/动态规划/极限学习机
Key words
Energy management strategy/Multi-objective optimization/Dynamic programming/Extreme learning machine
维普
万方数据