基于局部异常因子的锂离子电池储能系统故障诊断  

Fault diagnosis of lithium-ion battery energy storage systems based on local outlier factor

作　　者：彭鹏[1,2,3] 林达[4] 汪湘晋 丘意书 董缇 蒋方明 PENG Peng;LIN Da;WANG Xiangjin;QIU Yishu;DONG Ti;JIANG Fangming(Guangzhou Institute of Energy Conversion,Chinese Academy of Sciences,Guangzhou 510640,China;CAS Key Laboratory of Renewable Energy,Guangzhou 510640,China;Guangdong Provincial Key Laboratory of New and Renewable Energy Research and Development,Guangzhou 510640,China;State Grid Zhejiang Electric Power Co.,Ltd.Research Institute,Hangzhou 310014,China)

机构地区：[1]中国科学院广州能源研究所,广州510640 [2]中国科学院可再生能源重点实验室,广州510640 [3]广东省新能源和可再生能源研究开发与应用重点实验室,广州510640 [4]国网浙江省电力有限公司电力科学研究院,杭州310014

出　　处：《浙江电力》2023年第5期11-17,共7页Zhejiang Electric Power

基　　金：国网浙江省电力有限公司科技项目(B311DS210015)。

摘　　要：锂离子电池在过充、高温及外短路等滥用工况下工作会导致火灾等事故的发生。通过对锂离子电池进行早期故障诊断,定位故障的具体位置并及时采取相应措施,可以有效避免故障进一步升级为热失控。为此,基于锂离子电池储能系统的运行数据,采用局部异常因子算法对其进行故障诊断分析。通过计算单日及多日的电压运行数据,确定故障电池的具体位置,分析故障电池的异常情况。研究结果验证了局部异常因子算法应用于锂离子电池储能系统故障诊断的有效性。Lithium-ion batteries may lead to fire and other accidents when working under overcharge,high tempera⁃ture,and external short circuits.The faults can be prevented from escalating to thermal runaway through early fault diagnosis and fault location of lithium-ion batteries and corresponding measures in time.To this end,based on the operational data of the lithium-ion battery energy storage system,the local outlier factor(LOF)algorithm is used for fault diagnosis and analysis.By calculation of the single-day and multi-day voltage operation data,the specific loca⁃tion of the faulty battery is determined,and the abnormal condition of the battery is analyzed.The research results verify the effectiveness of the LOF algorithm applied to the fault diagnosis of the lithium-ion battery energy storage system.

关 键 词：锂离子电池 储能系统 运行数据 局部异常因子 故障诊断 

分 类 号：TM912[电气工程—电力电子与电力传动]