基于IBA-PF的锂电池SOC估算  

SOC of estimation of lithium battery based on IBA-PF

作　　者：田冬冬 李立伟 杨玉新 王凯 TIAN Dongdong;LI Liwei;YANG Yuxin;WANG Kai(School of Electrical Engineering,Qingdao University;Library of Qingdao University,Qingdao 266071,Shandong,China)

机构地区：[1]青岛大学电气工程学院 [2]青岛大学图书馆,山东青岛266071

出　　处：《储能科学与技术》2020年第5期1585-1592,共8页Energy Storage Science and Technology

基　　金：山东省自然科学基金(Y2008F23);山东省科技发展计划项目(2011GGB0112);山东省重点研发计划项目(2017GGX50114)。

摘　　要：针对传统粒子滤波估算电池荷电状态(SOC)出现的粒子贫化问题,本工作提出了一种改进蝙蝠算法(IBA)优化粒子滤波(PF)来估算SOC的方法。将粒子表征为蝙蝠个体,模仿蝙蝠族群捕食过程,解决粒子滤波技术中粒子贫化问题;结合二阶Thevenin电池模型构建电池状态空间理论模型并对电池进行相关参数辨识;利用IBAPF算法与标准PF算法在脉冲电流工况和DST工况电流下进行SOC估算。实验结果证明,与传统的PF算法比较,基于IBA-PF的锂电池SOC估算精度在2%以内,对非线性和非高斯特性的锂电池SOC估算具有良好的适应性和稳定性。Aiming at the problem of particle dilution in traditional particle filter(PF)estimations of the state of charge(SOC),an improved bat algorithm(IBA)is proposed to optimize the PF algorithm to estimate the SOC.The particles are represented as bat individuals,which imitate the predatory process of a bat population and solve the problem of particle dilution in PF technology.The theoretical model of the battery state space is built using a second-order Thevenin battery model,and the relevant parameters of the battery are identified.The SOC estimation experiment is carried out based on the IBA-PF algorithm and the standard PF algorithm under pulse current operating conditions and dynamic stress test operating conditions.The experimental results show that,compared to the traditional PF algorithm,the estimation accuracy of the lithium battery SOC based on IBA-PF is within 2%,which indicates good adaptability and stability for non-linear and non-Gaussian SOC estimations of lithium batteries.

关 键 词：锂电池 荷电状态 蝙蝠算法 粒子滤波 

分 类 号：TM912[电气工程—电力电子与电力传动]