电源技术2025，Vol.49Issue(6)：1192-1200,9.DOI:10.3969/j.issn.1002-087X.2025.06.014
基于分布熵和MCPO-RF的锂离子电池故障诊断方法
Fault diagnosis method of lithium-ion battery based on distribution en-tropy and MCPO-RF
摘要
Abstract
The fault diagnosis of lithium-ion battery is crucial for ensuring the safe operation of elec-tric vehicles.For the progressive failure in lithium-ion batteries,a fault diagnosis method was pro-posed based on distribution entropy(DE)and improved crested porcupine optimization algorithm(MCPO)optimizing random forest model(RF).The fault battery data was obtained through internal short circuit experiments,the distribution entropy was extracted from the battery voltage signals as the feature vectors,and the lithium battery internal short circuit fault diagnosis model was estab-lished based on the random forest algorithm.The improved crested porcupine optimization algo-rithm was used to adaptively optimize the model parameters,and the experimental data was em-ployed to test the model.The results show that the distribution entropy can effectively reflect the bat-tery faults,and the proposed MCPO-RF method has high accuracy,effectively identifying the pro-gressive failure in lithium batteries.关键词
锂离子电池/故障诊断/分布熵/冠豪猪优化算法/随机森林Key words
lithium-ion battery/fault diagnosis/distribution entropy/crested porcupine optimization algorithm/random forest分类
信息技术与安全科学引用本文复制引用
何山,唐文俊,赵宇明,姜久春,吕露..基于分布熵和MCPO-RF的锂离子电池故障诊断方法[J].电源技术,2025,49(6):1192-1200,9.基金项目
中国南方电网有限责任公司创新项目(090000KK52222142) （090000KK52222142）
