基于改进支持向量机算法的高压断路器故障诊断技术

打开文本图片集

中图分类号：R129 文献标志码：A 文章编号：2095-2945（2025）34-0078-04

Abstract：Theapplicationof thePVA-SVMclasifierbasedontheAPSOalgorithminthefaultdiagnosisofhigh-voltage circuit breakers is explored. Taking the ZW10-40.5kV vacuum circuit breaker as the research object，the opening and closing current dataofthecircuitbreakerandthemovementdisplacementdataofthemovingcontactswerecollcted，andthePCA methodwasusedfordatadimensionreductionprocessing.Theproceseddataisdividedintothetrainingsetandthetestset， whicharerespectivelyusedforthetrainingandtestingofdiferentclasifers.Fromthesimulationresultsoffaultdiagnosis， comparedwiththesingleSVMclasifier，thefaultdiagnosisaccuracyratesofthePCA-SVMclasifier，PSO-PCA-SVMclassifier andAPSO-PCA-SVMclasifierafterPCAdimensionreductionhaveallincreased.Among them，thediagnosis timeof theAPSOPCA-SVM clasifier isrelativelyshort，only O.924 seconds.The diagnostic accuracy is the highest，reaching 97.12%

Keywords：supportvectormachine；high-voltagecircuitbreaker;faultclasification；multi-sourcefeatureoptimization; classifier

高压断路器在运行中可能出现触头烧损、合闸接触不良、无状况自动分闸以及触头松动脱落等多种类型的故障，进而影响整个电气系统的正常运行，带来严重的经济损失。（剩余3897字）