摘要
针对锂电池模型参数辨识不准确以及传统无迹卡尔曼滤波(UKF)无法对噪声进行实时更新,从而导致锂电池荷电状态(SOC)估计误差偏大的问题,提出遗忘因子递推最小二乘法-自适应无迹卡尔曼滤波(FFRLS-AUKF)算法。先利用遗忘因子递推最小二乘法(FFRLS)对电池二阶RC等效电路模型进行在线参数辨识,再将所辨识的各参数传给由UKF和改进的Sage-Husa算法结合得到的AUKF,从而完成对锂电池的SOC估计,并将其与FFRLS-UKF以及离线UKF所估计的结果相比较。从对SOC估计的误差曲线和平均绝对误差以及均方根误差的数值上对比,均可得出FFRLS-AUKF的精度更高,稳定性更好。
Considering the problem that the model parameters of lithium battery is not accurately identified and the traditional unscented Kalman filter(UKF)cannot update the noise in real time,which leads to the large error of state of charge(SOC)calculation of lithium battery,a forgetting factor recursive least square method plus adaptive unscented Kalman filter(FFRLS-AUKF)algorithm was presented.Firstly,the method(FFRLS)was applied to carry out online parameter identification for the second-order RC equivalent circuit model of the battery.Then,the identified parameters were transmitted to AUKF,which was obtained by combining UKF and the improved Sage-Husa algorithm,so as to complete the SOC estimation of the lithium battery.The results are compared with those estimated by FFRLS-UKF and offline UKF.By comparing the error curve of SOC estimation with the value of mean absolute error and root mean square error,it is found that FFRLS-AUKF has higher precision and better stability.
作者
凌六一
吴贤圆
王星凯
邢丽坤
卢路
LING Liuyi;WU Xianyuan;WANG Xingkai;XING Likun;LU Lu(School of Electrical and Information Engineering,Anhui University of Science and Technology,Huainan Anhui 232001,China;School of Artificial Intelligence,Anhui University of Science and Technology,Huainan Anhui 232001,China)
出处
《安徽理工大学学报（自然科学版）》
CAS
2023年第1期1-7,共7页
Journal of Anhui University of Science and Technology:Natural Science
基金
安徽省高校自然科学基金资助项目(KJ2019A0106)。
作者简介
凌六一(1980-),男,安徽枞阳人,教授,博士,研究方向:检测技术与智能信息处理。
