分析测试学报2025，Vol.44Issue(3)：402-410,9.DOI:10.12452/j.fxcsxb.240712214
基于IWOA-SVR的锂离子电池健康状态在线快速检测
Online Rapid Detection of Lithium-ion Battery State of Health Based on IWOA-SVR
摘要
Abstract
In this paper,a detection and evaluation method of lithium-ion battery state of health(SOH)is proposed by integrating the improved whale optimization algorithm and support vector re-gression(IWOA-SVR).Firstly,the charging/discharging data under different charging/discharging strategies are collected,and the critical battery aging characteristic parameters are extracted.Then Pearson correlation analysis is applied to verify the strong correlation between the parameters and the SOH,and the algorithm integrates the adaptive weight adjustment mechanism and Levy flight strate-gy into the traditional whale optimization algorithm,which effectively overcomes the problem of the large error of the traditional method when evaluating the SOH online.Finally,in order to verify the effectiveness of the algorithm,experimental test data under two typical operating conditions of con-stant-current and constant-voltage charging and constant-current charging are used for validation,and the results show that the IWOA-SVR detection method has higher stability and accuracy,and the maximal error can be controlled within 1.4%.Meanwhile,IWOA-SVR significantly outperforms the comparison algorithms in two key evaluation metrics,namely,mean absolute percentage error(MAPE)and root mean square error(RMSE),which fully proves its high accuracy and strong robust-ness in the online detection of SOH in lithium-ion batteries.关键词
锂离子电池/改进鲸鱼优化算法/支持向量回归/电池健康状态检测Key words
lithium-ion batteries/improved whale optimization algorithm/support vector regres-sion/state of health detection分类
化学引用本文复制引用
陈洋,黄江东,余春雷,谢基,姜伟..基于IWOA-SVR的锂离子电池健康状态在线快速检测[J].分析测试学报,2025,44(3):402-410,9.基金项目
国家重点研发计划(2022YFF0607201) （2022YFF0607201）
国家自然科学基金资助项目(52171331,51907082) （52171331,51907082）
广州市科技计划项目(2023A03J0646,2023A03J0120,2023A04J1013) （2023A03J0646,2023A03J0120,2023A04J1013）
海关总署课题(2022HK062) （2022HK062）
