基于局部均值分解与局部离群因子动力电池故障诊断
Fault Diagnosis of Power Batteries Based on Local Mean Decomposition and Local Outlier Factor
胡杰 1贾超明 1程雅钰 1余海1
作者信息
- 1. 武汉理工大学 现代汽车零部件技术湖北省重点实验室,武汉 430070;武汉理工大学 汽车零部件技术湖北省协同创新中心,武汉 430070;新能源与智能网联汽车湖北省工程技术研究中心,武汉 430070
- 折叠
摘要
动力电池故障诊断是保证电动汽车正常运行的关键.提出一种基于局部均值分解和局部离群因子的动力电池故障诊断方法,用于电池组故障识别与定位.通过局部均值分解对电压信号预处理,并根据相关系数高低重构电压信号.进一步提取重构信号的峭度因子作为故障特征输入到局部离群因子算法中,根据局部离群因子算法自适应阈值输出故障电池.采用实车数据验证了所提方法能有效、准确地检测出故障,具有较好的可靠性与鲁棒性.
Abstract
The diagnosis of power battery faults is crucial for the normal operation of electric vehicles.In response,this paper proposes a power battery fault diagnosis method using local mean decomposition and the local outlier factor,aimed at fault recognition and localization within battery packs.Firstly,the voltage signal is preprocessed through local mean decomposition,followed by the reconstruction of the voltage signal according to the correlation coefficient.Furthermore,the kurtosis factor of the reconstructed signal is extracted as the fault feature input to the local outlier factor algorithm,which then identifies the faulty battery based on an adaptive threshold.Finally,the proposed method is validated on a real vehicle,effectively and accurately detecting faults while demonstrating the reliability and robustness of the method.
关键词
局部均值分解/峭度/故障诊断/局部离群因子/动力电池Key words
local mean decomposition/kurtosis/fault diagnosis/local outlier factor/power battery引用本文复制引用
基金项目
广西壮族自治区科技重大专项(23062062)
出版年
2024
国家科技期刊平台
NSTL
维普
万方数据