三类常用电动汽车电池模型的比较研究

Comparison study of 3 types of battery models for electrical vehicle

为了解决电动汽车仿真时电池模型的选择问题,对神经网络模型、简化的电化学模型和等效电路模型三类常用电池模型的性能和适用范围进行比较研究。基于镍氢电池组试验数据,选取三类模型的代表性模型:反向传播(BP)神经网络模型、简化的电化学组合模型和美国新一代汽车合作伙伴计划(PNGV)模型,从变电流工作状态和恒电流工作状态两方面进行具体分析。研究表明,等效电路模型具有更好的综合性能。

To figure out how to choose battery models for electric vehicle simulation, the paper compared the characteristics and the application fields of 3 types of often used battery models including Neural Network model, Simplified Electric Chemical model and equivalent circuit model. Based on test data of a Ni-MH battery pack, a detailed analysis of 3 typical models, BP（Back Propagation） Neural Network, Simplified Electric Chemical Combined model and PNGV（Partnership for a New Generation of Vehicles） equivalent circuit model were carried out in various current simulation mode and constant current simulation mode. The study indicated that PNGV model has better performance.