动力电池动态模型参数辨识的研究  

Research on identification of dynamic model parameters for power batteries

作　　者：张彦琴[1] 郭凯[1] 刘汉雨[1] 吴斌[1] 李怡霞[1] 

机构地区：[1]北京工业大学环境与能源工程学院,北京100124

出　　处：《电池工业》2013年第1期29-32,共4页Chinese Battery Industry

基　　金：国家自然科学基金项目(11005012201001)

摘　　要：为实现动力电池状态的精确估计,需要提高电池模型的精确性。对选定的电池模型进行在线参数辨识能够实时反映电池真实的工作状态。采用含遗传因子的递推最小二乘算法(FFRLS)对二阶RC模型进行参数辨识,并在Matlab/Simulink环境下搭建仿真模型用于验证辨识结果。仿真验证结果表明:由FFRLS算法得到的模型参数较为精确,可以作为模型参数在线辨识的一种有效手段。另外,FFRLS算法还能完成电池开路电压的估计,可作为估计电池SOC的参考,有利于了解电池性能状态。To achieve the accurate estimation of the state of power batteries, it is necessary to im- prove the accuracy of the battery model. Using the method of on-line parameter identification with selected battery model, the real battery state could be reflected in real time. In this paper, Forgetting factor recursive least square algorithm （FFRLS） was adopted to identify the parame- ters of Double-RC battery model, meanwhile, a simulate model with Matlab/Simulink was ap- plied to verify identification result. The verified results indicated that the model parameters got from FFRLS was accurate and FFRLS was an effective method to identify model parameter on- line. Additional, the above-mentioned method could be used to estimate the OCV to understandthe performance and state of batteries.

关 键 词：动力电池 二阶RC模型 参数辨识 最小二乘法 

分 类 号：TM912[电气工程—电力电子与电力传动]