储能科学与技术 ›› 2022, Vol. 11 ›› Issue (11): 3603-3612.doi: 10.19799/j.cnki.2095-4239.2022.0277

• 储能测试与评价 • 上一篇    下一篇

基于Sage-Husa EKF算法的锂离子电池能量状态估计

李晓涵1(), 孙磊1, 马勇1, 郭东亮1, 肖鹏1, 刘建军1, 吴鹏1, 张志行2, 韩雪冰2()   

  1. 1.国网江苏省电力有限公司电力科学研究院，江苏 南京 211103
    2.清华大学车辆与运载学院，北京 100084
  • 收稿日期:2022-05-24 修回日期:2022-06-11 出版日期:2022-11-05 发布日期:2022-11-09
  • 通讯作者: 韩雪冰 E-mail:lxhsgcc@163.com;hanxuebing@mails.tsinghua.edu.cn
  • 作者简介:李晓涵（1994—），男，博士研究生，研究方向为电化学储能技术应用研究和电力系统输变电设备状态评估技术，E-mail：lxhsgcc@163.com；
  • 基金资助:
    国网江苏省电力有限公司科技项目“电动汽车动力电池状态检测与安全评估技术研究”(J2021042)

Energy state estimation of lithium-ion batteries based on sage-husa EKF algorithm

Xiaohan LI1(), Lei SUN1, Yong MA1, Dongliang GUO1, Peng XIAO1, Jianjun LIU1, Peng WU1, Zhihang ZHANG2, Xuebing HAN2()   

  1. 1.Electric Power Scientific Research Institute of State Grid Jiangsu Electric Power Co. , Ltd. , Nanjing 211103, Jiangsu, China
    2.School of Vehicle and Mobility, Tsinghua University, Beijing 100084, China
  • Received:2022-05-24 Revised:2022-06-11 Online:2022-11-05 Published:2022-11-09
  • Contact: Xuebing HAN E-mail:lxhsgcc@163.com;hanxuebing@mails.tsinghua.edu.cn

RichHTML

31

PDF

230

摘要：

为充分利用电池内存储的能量，防止过放电和过充电，需要精准地估计电池的能量状态(SOE)，通常SOE定义为剩余能量与标准额定能量的比值。现有的SOE估计算法尚未充分考虑温度、工况等的影响，导致估计结果精确度较低，为准确估计电池的能量状态，本工作提出了基于Sage-Husa自适应扩展卡尔曼滤波(S-H EKF)算法的锂离子电池能量状态估计方法，并与传统的基于扩展卡尔曼滤波(EKF)的SOE估计算法精度进行对比分析。首先，分析了温度对电池能量特性的影响，获取不同温度下电池的标准能量。之后，建立考虑参数值随温度和SOE值变化的二阶RC等效电路模型，结合混合脉冲功率特性实验，使用最小二乘法进行模型参数辨识，并对模型精度进行验证，验证结果表明模型能够较好地仿真电池的端电压，具有较高精度。最后，使用S-H EKF和EKF对动态工况和间歇大倍率充电工况进行SOE估计，对比结果表明：以绝对误差均值为对比标准，S-H EKF的估计精度相比EKF高20.72%，SOE估计的最大绝对误差小于3%，更加适用于锂离子电池的能量估计。

关键词: 锂离子电池, 能量状态, 等效电路模型, 扩展卡尔曼滤波, Sage-Husa自适应扩展卡尔曼滤波

Abstract:

It is necessary to accurately estimate the state of energy (SOE) of a battery to make full use of the energy stored in the battery and prevent over-discharge and over-charge. Generally, SOE is defined as the ratio of the remaining energy to the standard rated energy. The current SOE estimation algorithms have not fully considered the influence of temperature, operating conditions, etc., leading to low accuracy of the estimation results. This study proposes an adaptive extended Kalman filter (S-H EKF) based on Sage-Husa to accurately estimate the energy state of a battery. The algorithm's energy state estimation approach for lithium-ion batteries is compared with the traditional SOE estimation algorithm based on the EKF. First, the influence of temperature on the battery's energy characteristics was examined, and the standard energy of the battery at various temperatures was obtained. Then, a second-order RC equivalent circuit model considering the variation of parameter values with temperature and SOE value is established. Combined with an experiment of mixed pulse power characteristics, the least squares approach is employed to identify the model parameters, and the model accuracy is verified. With a good simulation of the battery's terminal voltage having high accuracy, the verification findings reveal that the model can be compared better. Finally, by employing S-H EKF and EKF to estimate the SOE under dynamic conditions and intermittent high-rate charging conditions, the comparison results demonstrate that: taking the mean absolute error as the comparison standard, the estimation accuracy of S-H EKF is 20.72% higher than that of EKF, and the maximum SOE estimation is the absolute error is less than 3%, which is more appropriate for energy estimation of lithium-ion batteries.

Key words: lithium ion battery, energy state, equivalent circuit model, extended Kalman filter, adaptive extended Kalman filter

中图分类号: