基于EKF-SVSF的锂离子电池SOC和SOH准确估计  

Accurate estimation of SOC and SOH of Li-ion battery based on EKF-SVSF

作　　者：陈剑 肖振锋 刘顺成 罗磊鑫 夏向阳[3] CHEN Jian;XIAO Zhen-feng;LIU Shun-cheng;LUO Lei-xin;XIA Xiang-yang(Hunan Key Laboratory of Energy lnternet Supply-demand and Operation,Slate Grid Hunan Fconomy lnstitute,Changsha Hunan 410004,China;Hunan Eiconomy Instite Electri Power Design Co.,Ld.Changsha Hunan 410004,China;School of Electrical and lnformation Enginecring,Changsha University of Science&Technology,Changsha Hunan 410004,China)

机构地区：[1]国网湖南省电力有限公司经济技术研究院能源互联网供需运营湖南省重点实验室,湖南长沙410004 [2]湖南经研电力设计有限公司,湖南长沙410004 [3]长沙理工大学电气与信息工程学院,湖南长沙410004

出　　处：《电源技术》2020年第10期1483-1487,共5页Chinese Journal of Power Sources

基　　金：国网湖南省电力有限公司科技项目(5216A220000F);国家自然科学基金资助项目(51977014)。

摘　　要：提升锂离子电池荷电状态(SOC)估计精度并准确估计健康状态(SOH)对于设计高性能和安全的电池管理系统(BMS)至关重要。以二阶RC锂离子电池电路模型为研究对象,提出了一种由扩展卡尔曼滤波(EKF)算法和平滑可变结构滤波(SVSF)算法组成的混合滤波算法(EKF-SVSF),其中扩展卡尔曼滤波算法用于锂离子电池参数辨识,平滑可变结构滤波算法用于SOC估计,并采用改进的粒子群优化算法(PSO)对混合滤波算法迭代过程中的噪声协方差矩阵进行修正,电池容量实时在线估计的结果用于SOH参数的预测。实验和仿真结果表明,所提出的EKF-SVSF算法对电池的欧姆电阻和容量有一个较好的估计,进一步提升了电池状态的估计精度。It is very important to improve the accuracy of state of charge(SOC)estimation and estimate the state of health(SOH)accurately for the design of high-performance and safe battery management system(BMS).A hybrid filtering algorithm(EKF-SVSF)was proposed,which was composed of the extended Kalman filtering algorithm and the sliding variable structure filtering algorithm.The EKF algorithm was used to identify the parameters of the lithium-ion battery,and the smooth variable structure filtering algorithm(SVSF)was used to estimate the state of charge(SOC).The noise covariance matrix in the iterative process of the hybrid filter algorithm was modified by the improved particle swarm optimization(PSO)algorithm.The results of on-line estimation of battery capacity were used to predict SOH parameters.Experimental and simulation results show that the ekf-svsf algorithm has a good estimation of the ohmic resistance and capacity of the battery,and further improves the estimation accuracy of the battery state.

关 键 词：荷电状态 健康状态 扩展卡尔曼滤波 平滑可变结构滤波 粒子群优化算法 

分 类 号：TM912[电气工程—电力电子与电力传动]