基于随机动态规划的燃料电池城市客车能量管理策略优化.pdfVIP

第20 卷第17 期 系 统 仿 真 学 报© Vol. 20 No. 17 2008 年9 月 Journal of System Simulation Sep., 2008 基于随机动态规划的燃料电池城市客车能量管理策略优化 张炳力，代康伟，赵 韩，江 昊，胡先锋 (合肥工业大学机械与汽车工程学院，合肥 230009) 摘 要：设计了一种新型的燃料电池混合超级电容的城市客车动力系统结构，在理论计算和工程分 析的基础上，对该动力系统进行了参数匹配。提出一种基于随机动态规划的燃料电池城市客车能量 管理策略的设计思想，将驾驶员需求功率模拟为一个离散的随机动态过程，并建立相应的马尔可夫 模型，在此基础上采用策略迭代的算法对能量管理策略进行了优化。并对优化后的策略进行仿真， 结果表明该策略下的整车燃料经济性有明显提高。 关键词：燃料电池；城市客车；能量管理策略；马尔可夫决策 中图分类号：U469.79 文献标识码：A 文章编号：1004731X(2008)17-4664-04 Optimized Energy Management Strategy for Fuel Cell City Bus Based on Stochastic Dynamic Programming ZHANG Bing-li, DAI Kang-wei, ZHAO Han, JIANG Hao, HU Xian-feng (School of Machinery and Automobile Engineering, Hefei University of Technology, Hefei 230009, China) Abstract: The powertrain framework of a new model of fuel cell city bus which combines fuel cell with ultracapacitor was designed. The related parameters of driveline were designed on the basis of theory deduction and engineering analysis. A stochastic dynamic programming algorithm was developed for the energy management of this fuel cell city bus. The driver’s power demand was modeled as a discrete -time stochastic dynamic process, then the related Markov chain was set up. The policy iteration algorithm was used to optimize the control strategy. The optimization scheme was validated by simulation. The result indicates that the fuel economy of the bus is improved obviously under this strategy. Key words: fuel cell; city bus; energy management strategy; Markov decision process 引 言1 1 燃料电池城市客车结构选型与参数匹配 燃料电池城市客车的动力性、经济性主要取决于动