摘要
随着使用工况、衰退状况以及工作温度的改变,动力锂离子电池内部参数也会随之变化,因此基于模型估算电池荷电状态(SOC)的方法中模型参数是时变的。然而,由于采用离线的方式获取,传统的扩展卡尔曼滤波(EKF)算法中的模型参数却是不变的,由此导致随时间的推移SOC估算精度会下降。为了解决这一问题,建立带遗忘因子的递推最小二乘法来实时更新模型参数,联合改进的自适应扩展卡尔曼滤波(AEKF)算法进行SOC估算。使用混合脉冲功率特性(HPPC)测试数据对比该方法与传统EKF表现,结果表明该方法具有更高的精度,其相对误差小于1%,且对电池充放电的动态特性有更好的模拟效果。
With changes in operating conditions,decay conditions,and temperature,the internal parameters of the power lithium-ion battery will change accordingly,so the model parameters in the method of estimating the state of charge(SOC)based on the model should be time-varying.However,due to the offline method,the model parameters in the traditional extended Kalman filter(EKF)do not change with time,which will cause the accuracy of the SOC estimation to decrease over time.In order to solve this problem,a recursive least square method with forgetting factor is established to update the model parameters in real time,and the improved adaptive extended Kalman filter(AEKF)algorithm is used to estimate the SOC.Using hybrid pulse power characteristic(HPPC)test data to compare the performance of this method with traditional EKF,the results show that the method has higher accuracy.The relative error is less than 1%and it has better simulation effects on the dynamic characteristics of battery during charging and discharging.
作者
秦鹏
王振新
康健强
王菁
朱国荣
向馗
Qin Peng;Wang Zhenxin;Kang Jianqiang;Wang Jing;Zhu Guorong;Xiang Kui(Hubei Research Center for New Energy&Intelligent Connected Vehicle,Wuhan University of Technology,Wuhan 430070,China;School of Automation,Wuhan University of Technology,Wuhan 430070,China)
出处
《电子测量技术》
2020年第10期30-35,共6页
Electronic Measurement Technology
关键词
锂离子电池
自适应卡尔曼滤波
荷电状态
参数辨识
lithium-ion batteries
adaptive extended Kalman filter
state of charge
parameter identification
作者简介
秦鹏,硕士,主要研究方向为新能源汽车电池管理系统、电池状态估计算法、电池性能测试等。E-mail:qincpeng@163.com
