基于静态EIS模型的锂离子电池SOC估计  

SOC estimation method of Li-ion battery based on static EIS model

作　　者：朱一昕 吴昊[1] 黎莞伟 刘宇凡 ZHU Yixin;WU Hao;LI Guanwei;LIU Yufan(School of Internet of Things Engineering,Jiangnan University,Wuxi 214122,Jiangsu,China)

机构地区：[1]江南大学物联网工程学院,江苏无锡214122

出　　处：《电池》2025年第2期267-272,共6页Battery Bimonthly

基　　金：国家自然科学基金优秀青年科学基金(62222307)。

摘　　要：针对传统估算方法难以确定合适的等效模型结构导致误差较大的问题,提出基于静态电化学阻抗谱(EIS)确定电池模型的方法。对EIS进行分段分析,选择合适的分数二阶模型,利用分数降阶理论和带遗忘因子的递归最小二乘法进行参数辨识。为解决扩展卡尔曼滤波(EKF)算法线性化后估计误差较大的问题,引入粒子滤波算法,根据上一时刻的观测数据计算粒子均值和协方差,进行本时刻的状态估计和粒子更新。根据混合功率脉冲特性(HPPC)测试的电池放电数据,对比所提算法与传统整数二阶模型。采用分数阶模型的误差均值、误差最大值分别仅为整数阶模型的46.88%、3.75%。In order to solve the issue that the traditional estimation method has a large error due to the difficulty in determining the appropriate equivalent model structure,a method is proposed to determine the battery model based on static electrochemical impedance spectroscopy(EIS).The appropriate fractional second-order model is selected through the piecewise analysis of the EIS,the fractional reduction theory and the recursive least squares method with forgetting factor are used for parameter identification.In order to solve the issue of large estimation error after linearization of the extended Kalman filter(EKF)algorithm,the particle filter algorithm is introduced,the particle mean and covariance are calculated according to the observation data of the previous time,so as to estimate the state and update the particles at this time.According to battery discharge data under the hybrid pulse power characteristic(HPPC)test,the proposed algorithm is compared with the traditional integer second-order model.The mean error and maximum error value of the fractional model are only 46.88%and 3.75%of the integer order model.

关 键 词：电化学阻抗谱(EIS) 分数阶模型 锂离子电池 荷电状态(SOC) 扩展卡尔曼滤波(EKF)算法 

分 类 号：TM912.9[电气工程—电力电子与电力传动]