基于SSA-BP改进EKF算法的锂电池SOC估算  

SOC estimation of lithium battery with SSA-BP improved EKF algorithm

作　　者：张淞 周永军 蒋淑霞[1] 梁杨 ZHANG Song;ZHOU Yongjun;JIANG Shuxia;LIANG Yang(School of Mechanical and Electrical Engineering,Central South University of Forestry and Technology,Changsha Hunan 410000,China)

机构地区：[1]中南林业科技大学机电工程学院,湖南长沙410000

出　　处：《电源技术》2023年第8期1050-1054,共5页Chinese Journal of Power Sources

基　　金：湖南省自然科学基金项目(2019JJ60076)。

摘　　要：针对EKF算法与BP神经网络的缺陷,搭建一阶RC等效电池模型模拟锂电池的实际动态特性,提出在模型基础上利用麻雀搜索算法优化BP神经网络的初始权值与阈值,离线训练优化后的BP神经网络可在线补偿EKF算法估计出的SOC,得到SOC最佳估计值。根据实验数据在MATLAB/Simulink中搭建仿真模块验证算法精度。结果表明提出的算法具有优于EKF算法和BP-EKF算法的准确性与收敛性,可有效提高锂电池SOC估算精度,具有一定的实际应用价值。In order to improve the defects of EKF algorithm and BP neural network,a first-order RC equivalent battery model was built to simulate the actual dynamic characteristics of lithium batteries,on the basis of the model,sparrow search algorithm was adopted to optimize the initial weights and thresholds of BP neural network,and the optimized BP neural network was trained offline to compensate the SOC estimated by EKF algorithm online to get the best estimate of SOC.A simulation module was built in MATLAB/Simulink to verify the accuracy of the algorithm based on the experimental data.The results show that the proposed algorithm has better accuracy and convergence than the EKF algorithm and BP-EKF algorithm,which can effectively improve the SOC estimation accuracy of lithium batteries and has certain practical application value.

关 键 词：麻雀搜索算法 荷电状态 扩展卡尔曼滤波 BP神经网络 

分 类 号：TM912.9[电气工程—电力电子与电力传动]