基于改进EKF算法的锂电池SOC估计方法  

State of Charge Estimation of Lithium Battery based on Improved EKF algorithm

作　　者：谭威 蒋瞻 刘勇[2] 任芳[2] Tan Wei;Jiang Zhan;Liu Yong;Ren Fang

机构地区：[1]中车时代电动汽车股份有限公司,湖南株洲412007 [2]湘潭大学自动化与电子信息学院,湖南湘潭411105

出　　处：《时代汽车》2024年第24期101-103,共3页Auto Time

摘　　要：准确的荷电状态估计(SOC)对于提升车辆性能、续航里程和整体效率至关重要,同时也有助于确保电池健康和使用寿命。传统的扩展卡尔曼滤波(EKF)算法被广泛应用,但其精度易受噪声协方差矩阵影响。为解决此问题,文章提出一种基于灰狼优化算法(GWO)改进的EKF算法,旨在提高锂电池SOC估计精度。该算法在锂电池测试平台上,使用HPPC动态工况电流数据进行验证。结果表明,与传统EKF算法相比,改进算法的SOC估计误差显著降低,大幅提升了估计精度。Accurate state-of-charge(SOC)estimation is crucial for enhancing vehicle performance,range,and overall effi ciency,as well as ensuring battery health and longevity.While the traditional extended Kalman fi lter(EKF)algorithm is widely employed,its accuracy is susceptible to the noise covariance matrix.To address this issue,this paper proposes an improved EKF algorithm based on the Gray Wolf Optimization(GWO)algorithm,aiming to enhance the accuracy of SOC estimation for lithium batteries.The algorithm is validated on a lithium battery test platform using Hybrid Pulse Power Characterization(HPPC)dynamic operating current data.The results demonstrate that compared to the traditional EKF algorithm,the SOC estimation error of the improved algorithm is signifi cantly reduced,substantially improving the estimation accuracy.

关 键 词：荷电状态估计 EKF 算法 灰狼算法 噪声协方差矩阵 

分 类 号：TM9[电气工程—电力电子与电力传动]