基于改进Sage-Husa的自适应无迹卡尔曼滤波的锂离子电池SOC估计  

States of charge estimation of lithium-ion battery based on improved Sage-Husa adaptive unscented Kalman filters

作　　者：杨海学[1,2] 张继业[2] 张晗[1,2] 

机构地区：[1]西南交通大学信息科学与技术学院,四川成都610031 [2]西南交通大学牵引动力国家重点实验室,四川成都610031

出　　处：《电工电能新技术》2016年第1期30-35,共6页Advanced Technology of Electrical Engineering and Energy

基　　金：国家自然科学基金(11172247);牵引动力国家重点实验室开放课题(TPL1211)资助项目

摘　　要：扩展卡尔曼滤波(EKF)算法估计锂离子动力电池荷电状态(SOC)时,由于系统噪声的不确定,可能导致估计算法不精确,并且算法中的线性化处理受电池模型的影响很大。为了解决上述两个问题,本文采用改进Sage-Husa的自适应无迹卡尔曼滤波法(AUKF)来动态地估计多元复合锂离子电池的SOC。与EKF相比,改进Sage-Husa的自适应卡尔曼滤波法提高了SOC估计的精度,并能够实时修正微小的模型误差带来的SOC估计误差,实时的工况模拟证明了该算法更适用于多元复合锂离子电池的动态SOC估计。When using the extended Kalman filter to estimate the electric vehicle lithium-ion power battery state of charge（ SOC）,the uncertainty of the system noise can cause algorithm convergence,and the algorithm of linearization will be much affected by the model. In order to solve the two problems,this paper uses the improved Sage-Husa adaptive unscented Kalman filters to dynamically estimate the multiple lithium battery SOC. Simulation results show that compared with the EKF,the improved Sage-Husa adaptive Kalman filters can improve the accuracy of the SOC estimation,and can in real-time correct SOC estimation error caused by tiny model error online,and the operation condition of the real time simulation proved that the algorithm is more applicable to dynamic multiple lithium battery SOC estimate.

关 键 词：多元锂离子动力电池 SOC 状态估计 自适应 无迹卡尔曼滤波 Sage-Husa 

分 类 号：TM912.6[电气工程—电力电子与电力传动]