项目名称: 基于庞特里亚金最小值原理的混合动力汽车能量管理策略研究
项目编号: No.51305437
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 机械、仪表工业
项目作者: 郑春花
作者单位: 中国科学院深圳先进技术研究院
项目金额: 25万元
中文摘要: 混合动力汽车的燃油经济性在很大程度上依赖于其能量管理策略。目前,基于启发式概念的能量管理策略需根据丰富的经验知识制定控制规则,而且更重要的是这种策略很难达到混合动力汽车能量最优化的目标。另外,基于动态规划算法的策略虽能在理论上保证全局优化的结果,但具有不能应用于实际的缺点。面对这种现状,本项目提出一种基于庞特里亚金最小值原理的混合动力汽车能量管理策略。这种策略既能保证能量管理的最优化,又具有实用性。在庞特里亚金最小值原理的应用中有几个关键问题,如控制参数的选定、全局优化的讨论、向多状态变量多控制目标的扩充等。本项目将这一能量管理策略分别应用于混合动力电动汽车(油电混合)和燃料电池混合动力汽车,而且将这一能量管理策略的结果与其他策略的进行比较;对上述几个关键问题进行研究;将这一能量管理策略应用于整车控制器进行整车测试。本项目的研究将为基于庞特里亚金最小值原理的能量管理策略的实现奠定基础。
中文关键词: 燃料电池混合动力汽车;混合动力电动汽车;能量管理策略;最优控制理论;庞特里亚金最小值原理
英文摘要: The fuel economy of hybrid vehicles strongly depends on the energy management strategy. Currently, heuristic-concept-based energy management strategies need expert knowledge to construct control rules, and more importantly, these energy management strategies hardly achieve optimal power distribution results. In addition, Dynamic Programming (DP) can obtain global optimal results theoretically, but it cannot be applied to realistic cases. Being confronted with this situation, Pontryagin's Minimum Principle (PMP)-based energy management strategy is proposed in this project. This energy management strategy not only guarantees optimal power distribution results, but also has the probability of realistic application. There are several key points when applying the PMP-based energy management strategy to hybrid vehicles including selecting control parameters, discussing global optimality, and extending to multi-state variable and multi-optimization. In this project, the PMP-based energy management strategy is applied to hybrid electric vehicles (HEV) and fuel cell hybrid vehicles (FCHV), and the power distribution results of this strategy are compared to those of other strategies. This project also deals with the key points stated above, and the PMP-based strategy is applied to a vehicle controller and its performance
英文关键词: Fuel Cell Hybrid Vehicle;Hybrid Electric Vehicle;Energy Management Strategy;Optimal Control Theory;Pontryagin's Minimum Principle