汽车工程 ›› 2022, Vol. 44 ›› Issue (4): 505-513.doi: 10.19562/j.chinasae.qcgc.2022.04.006

所属专题: 新能源汽车技术-动力电池&燃料电池2022年

• • 上一篇    下一篇

基于FFRLS-EKF联合算法的锂离子电池荷电状态估计方法

孙金磊(),邹鑫,顾浩天,崔凯,朱金大   

  1. 国网电力科学研究院有限公司,南京  211106
  • 收稿日期:2021-10-21 修回日期:2022-01-29 出版日期:2022-04-25 发布日期:2022-04-22
  • 通讯作者: 孙金磊 E-mail:jinlei.sun@njust.edu.cn
  • 基金资助:
    国家电网有限公司科技项目(1500V高压电池系统及其功率变换关键技术研究,合同号:5419-202140235A-0-0-00)资助

State of Charge Estimation for Lithium-ion Battery Based on FFRLS-EKF Joint Algorithm

Jinlei Sun(),Xin Zou,Haotian Gu,Kai Cui,Jinda Zhu   

  1. State Grid Electric Power Research Institute,Nanjing  211106
  • Received:2021-10-21 Revised:2022-01-29 Online:2022-04-25 Published:2022-04-22
  • Contact: Jinlei Sun E-mail:jinlei.sun@njust.edu.cn

摘要:

针对现有基于电池恒定参数模型的SOC估计方法忽略了工况和SOC对电池模型参数的影响而导致SOC估计误差偏大的问题,本文提出一种将带有遗忘因子递推最小二乘算法与扩展卡尔曼滤波算法相结合的联合SOC估计方法。该方法先利用FFRLS算法在线辨识电池等效电路模型参数并实时修正电池模型,再利用EKF算法和实时修正的电池模型估计电池SOC。实验结果表明,本文所提的SOC估计方法能有效减小电池模型参数变化所带来的SOC估计误差。在脉冲放电、脉冲充电和动态应力测试实验中,最终电池SOC估计的最大误差分别为1.01%、0.87%和1.59%。

关键词: SOC估计, 参数辨识, 递推最小二乘算法, 扩展卡尔曼滤波

Abstract:

In view of the problem of a bit too large error of battery SOC estimation due to the existing SOC estimation method based on the battery model with constant parameters ignores the effects of working conditions and battery model parameters, a SOC joint-estimation method for lithium-ion batteries combining the forgetting factor least squares algorithm (FFRLS) and the extended Kalman filter (EKF) algorithm is proposed in this paper. The method utilizes the FFRLS algorithm to online identify the parameters of battery equivalent circuit model and correct the battery model in real-time first. Then, the EKF algorithm and real-time corrected battery model are used to estimate battery SOC. The experimental results show that the SOC estimation method proposed can effectively reduce the SOC estimation error caused by the variation of battery model parameters. In the experiments of pulse discharging test, pulse charging test and dynamic stress test, the final maximum errors of SOC estimation are 1.01%, 0.87% and 1.59%, respectively.

Key words: SOC estimation, parameter identification, RLS, EKF