基于IACA-SVR的电池SOH预测研究  

Estimation of SOH of Lithium Battery Based on IACA-SVR

作　　者：康道新 李立伟 杨玉新 罗羽 KANG Dao-xin;LI li-wei;YANG Yu-xin;LUO Yu(Qingdao University,Qingdao 266071,China;不详)

机构地区：[1]青岛大学,电气工程学院,山东青岛266071 [2]青岛大学,图书馆,山东青岛266071 [3]潍坊市产品质量检验所,山东潍坊261031

出　　处：《电力电子技术》2020年第9期62-66,共5页Power Electronics

基　　金：山东省科技发展计划(2011GGB01123);山东省重点研发计划(2017GGX50114)。

摘　　要：针对锂电池老化机理复杂、电池健康状态(SOH)估算不准确问题,提出了一种新型电池SOH估计方法。首先,建立基于改进蚁群算法优化支持向量回归(IACA-SVR)的电池SOH预测模型,以电池放电过程平均电压和平均温度为输入变量,模拟电池老化。其次,利用样本数据训练SVR,并用IACA来优化SVR的参数。在数据集上的实验结果表明,经过优化的SVR预测结果与遗传算法SVR(GA-SVR)预测结果相比更为精确,稳定性更好,从而验证了IACA-SVR预测模型的可行性,能为电动汽车锂电池的安全使用提供准确电池数据。In view of the complex aging mechanism of lithium battery and the inaccurate estimation of battery state of health(SOH),a new estimation method of battery SOH is proposed.Firstly,a prediction model of battery SOH based on improved ant colony algorithm support vector regression(IACA-SVR)is established,and the average voltage and temperature during battery discharge are used as input variables to simulate battery aging.Secondly,SVR is trained by sample data,and parameters of SVR are optimized by IACA.The experimental results on the data set show that the optimized SVR prediction results are more accurate and more stable than genetic algorithm-SVR(GA-SVR)prediction results,thus verifying the feasibility of IACA-SVR prediction model,which can provide accurate battery data for the safe use of lithium battery in electric vehicles.

关 键 词：锂电池 健康状态 蚁群算法 支持向量回归 

分 类 号：TM912[电气工程—电力电子与电力传动]