电动汽车动力电池健康状态在线动态估算方法  

On-line dynamic estimation method of power battery health status for electric vehicle

作　　者：肖业 刘芳[2] 刘欣怡 林辉 XIAO Ye;LIU Fang;LIU Xin-yi;LIN Hui(CRRC Electric Vehicle Co.,Ltd.,Zhuzhou Hunan 412007,China;School of Computer Science and Technology,Tiangong University,Tianjin 300387,China;Automotive Research Institute,Neusoft Reach Automative Technology Co.,Ltd.,Shenyang Liaoning 110179,China)

机构地区：[1]中车时代电动汽车股份有限公司,湖南株洲412007 [2]天津工业大学计算机科学与技术学院,天津300387 [3]东软睿驰汽车技术有限公司汽车电子研究院,辽宁沈阳110179

出　　处：《电源技术》2021年第2期177-180,235,共5页Chinese Journal of Power Sources

基　　金：国家重点研发计划(2017YFB1103003);国家自然科学基金青年(61802280,61806143,61772365,41772123);天津市自然科学基金(18JCQNJC77200);天津市教委科研计划项目(2017KJ094)。

摘　　要：针对电动汽车领域电池SOH在线估算的问题,提出了一种以戴维南(Thevenin)等效电路模型为框架,以BMS在线监测的过程数据为基础的SOH在线估算方法。此方法首先提出构建以电池使用时间t为自变量,SOH为隐变量的数学模型,并在此基础上,提出错时参数更新策略,有效降低单采样周期内的计算复杂度,使其更适用于电动汽车BMS控制单元。其次,提出利用非线性最小二乘初始化遗传算法初始种群的方式加快辨识速度。此SOH估算方法的优势在于无需动力电池前期实验室实验数据支撑,仅依靠电池管理系统实时测得的过程数据便可实现对电池SOH的估算,因此算法具有较好的动态特性。经验证证明所提出的SOH估算方法在电动汽车动力电池领域具有很好的适用性并且算法具有较高的估计精度。Aiming at the problem of online estimation of power battery SOH in the field of electric vehicles,this paper proposed a SOH online estimation method based on Thevenin EMC and monitoring process data from BMS.A mathematical model with the power battery usage time t as the independent variable and SOH as the hidden variable was built,and a strategy for updating the time-varying parameters was proposed to effectively reduce the computational complexity within the single sampling period,making it more suitable for on-board BMS control units of electric vehicles.Secondly,for parameter identification,a method of initializing the initial population of genetic algorithm using nonlinear least squares was proposed to speed up identification.The advantage of this SOH estimation method is that it does not require the support of preliminary laboratory test data of the power battery,and can only estimate the battery SOH by relying on the process data measured in real time by the battery management system,so the algorithm has good dynamic characteristics.The verification proves that the proposed SOH estimation method has good applicability in the field of electric vehicle power batteries and the algorithm has a high estimation accuracy.

关 键 词：SOH估算 电池模型 AR模型 GA算法 非线性最小二乘算法 

分 类 号：TM912[电气工程—电力电子与电力传动]