基于IBOA-DKF算法的锂电池SOC估计  

SOC Estimation of Lithium Battery Based on IBOA-DKF Algorithm

作　　者：刘意期 王聪[1,2] 黄建宇 LIU Yiqi;WANG Cong;HUANG Jianyu(Guangzhou Marine Geological Survey,Guangzhou 511458,China;Key Laboratory of Marine Mineral Resources,Ministry of Natural Resources,Guangzhou 511458,China)

机构地区：[1]广州海洋地质调查局,广东广州511458 [2]自然资源部海底矿产资源重点实验室,广东广州511458

出　　处：《自动化仪表》2025年第3期30-37,共8页Process Automation Instrumentation

基　　金：中国地质调查局基金资助项目(DD20230642);2023年广东省海洋经济发展专项基金资助项目(GDNRC[2023]40)。

摘　　要：应用传统卡尔曼滤波(KF)算法估计锂电池荷电状态(SOC)时,噪声往往假设为一个固定值的零均值白噪声,从而导致锂电池SOC估计值误差随着迭代次数的增加而不断增大。对此,提出了一种改进蝴蝶优化算法-双卡尔曼滤波(IBOA-DKF)算法。将反向学习策略及动态调整转换概率策略引入蝴蝶优化算法(BOA),可以提高收敛速度、均衡全局搜索及局部开发能力,从而对KF算法的噪声协方差矩阵进行迭代更新。在二阶电阻电容(RC)等效电路模型基础上,利用IBOA-DKF算法分别对内阻Rs与锂电池SOC进行估计。同时,通过两种动态工况测试数据进行仿真,验证了IBOA-DKF算法对锂电池SOC估计绝对值误差在1%以内,因而具备更高的精度、更好的收敛性及鲁棒性。该研究为锂电池SOC更高精度的估计提供了理论依据。When applying the traditional Kalman filtering(KF)algorithm to estimate the state of charge(SOC)of lithium batteries,the noise is often assumed to be a fixed-value zero-mean white noise,which leads to an increasing error in the SOC estimation value of lithium batteries as the number of iterations increases.In this regard,an improved butterfly optimization algorithm-double Kalman filtering(IBOA-DKF)algorithm is proposed.Introducing the inverse learning strategy and the dynamic adjustment of the conversion probability strategy into the butterfly optimization algorithm(BOA)can improve the convergence speed,balance the global search and local development ability,to iteratively update the noise covariance matrix of the KF algorithm.Based on the second-order resistance capacitance(RC)equivalent circuit model,the internal resistance Rs and lithium battery SOC are estimated using the IBOA-DKF algorithm,respectively.At the same time,the simulation of the two dynamic operating conditions test data verifies that the IBOA-DKF algorithm estimates the SOC of lithium batteries with an absolute value error of less than 1%,and thus has higher accuracy,better convergence and robustness.This study provides a theoretical basis for the higher precision estimation of SOC for lithium batteries.

关 键 词：锂电池 荷电状态 卡尔曼滤波 蝴蝶优化算法 等效电路模型 

分 类 号：TH701[机械工程—仪器科学与技术]