电气技术2025，Vol.26Issue(7)：40-45,6.
基于海鸥优化算法改进深度极限学习机的锂电池健康状态预测
State of health prediction for lithium batteries based on deep extreme learning machine improved by seagull optimization algorithm
靳灿 1张晓燕 1孙本川1
作者信息
- 1. 国网山东省电力公司嘉祥县供电公司,山东 济宁 272400
- 折叠
摘要
Abstract
The current methods for predicting the state of health of lithium batteries often suffer from low accuracy.This paper introduces a method for state of health prediction using a seagull optimization algorithm optimized deep extreme learning machine.Key health feature parameters,such as constant voltage charging and discharging times during battery cycles,are selected and their correlation with the battery state of health is analyzed using Pearson correlation.The proposed model predicts subsequent state of health values by learning from samples.Experiments conducted with battery data compare the proposed method with single extreme learning machine,single deep extreme learning machine,and other literature.Evaluation metrics,including maximum absolute error and root mean square error,demonstrate that the seagull optimization algorithm optimized deep extreme learning machine model achieves higher accuracy and faster prediction times,with errors below 1.1%,indicating superior prediction accuracy and applicability.关键词
锂电池/健康状态/海鸥优化算法/深度极限学习机Key words
lithium battery/state of health/seagull optimization algorithm/deep extreme learning machine引用本文复制引用
靳灿,张晓燕,孙本川..基于海鸥优化算法改进深度极限学习机的锂电池健康状态预测[J].电气技术,2025,26(7):40-45,6.
