基于PMP算法的燃料电池混合动力车辆实时能量管理策略

A Real Time Management Strategy for Fuel Cell Hybrid Vehicles Based on Pontryagin’s Minimum Principle

摘要: 为了保证动力电池荷电状态(SOC, State of Charge)处于高效工作区间，针对燃料电池混合动力车辆(FCHEV, Fuel Cell Hybrid Vehicle)设计了一种实时能量管理策略。基于庞特里亚金极小值原理(PMP, Pontryagin’s Minimal Principle)对能量管理问题进行离线求解，得到三种典型工况下协态变量与SOC的关系。根据学习矢量量化(LVQ, Learning Vector Quantization)神经网络，对行驶工况进行识别，进而根据协态变量与SOC关系选取最优协态变量。构建测试工况进行仿真，结果表明，该策略可以有效控制电池SOC在一定范围内，维持电池健康状态。

Abstract: In order to ensure that the power battery state of charge (SOC) is stable in the efficient operating range, an adaptive energy management strategy is proposed for a fuel cell hybrid electric vehicle (FCHEV). The energy management problem of the FCHEV is solved off-line based on Pontryagin’s Minimal Principle (PMP), and the relationship between the co-state variable and the SOC is obtained under three typical drive cycle. According to the learning vector quantization (LVQ) neural network, the driving conditions are identified, and then the optimal co-state variable are selected according to the relationship between co-state variable and SOC. The result under the test driving cycle shows that the strategy can effectively control the battery SOC within a certain range and maintain the battery’s health status.