预览加载中,请您耐心等待几秒...
1/3
2/3
3/3

在线预览结束,喜欢就下载吧,查找使用更方便

如果您无法下载资料,请参考说明:

1、部分资料下载需要金币,请确保您的账户上有足够的金币

2、已购买过的文档,再次下载不重复扣费

3、资料包下载后请先用软件解压,在使用对应软件打开

基于混沌果蝇-最小二乘支持向量机的变压器DGA故障模式预测 Abstract DissolvedGasAnalysis(DGA)isoneofthemostimportantmethodsfortransformerinsulationdiagnosis.Inthispaper,wefocusonpredictingthefaultmodeoftransformerbasedonDGA.Toachievethis,ahybridmethodcombiningchaosfruitflyoptimization(CFO)andleastsquaressupportvectormachine(LSSVM)isproposed.TheCFOisusedtooptimizetheparametersoftheLSSVMandimproveitspredictionaccuracy.Theexperimentalresultsshowthattheproposedmethodoutperformsthetraditionalmethodsandachieveshighaccuracyinpredictingthefaultmodeoftransformer. Introduction Transformersareessentialcomponentsinelectricpowersystems,andanyfaultintransformersmayresultinthefailureorevencollapseoftheentirepowersystem.Therefore,timelyandeffectivedetectionanddiagnosisoftransformerfaultshavealwaysbeenimportanttopicsinthepowersystemfield.Dissolvedgasanalysis(DGA)isawidelyusedmethodfortransformerfaultdiagnosis.Thefaultgasgeneratedbythetransformerduringthefaultprocesswilldissolveintheoilandcanbeanalyzedtoinferthefaultmodeofthetransformer. Inrecentyears,variousmachinelearningalgorithmshavebeenappliedtothetransformerDGAfaultdiagnosis.Supportvectormachine(SVM)isapopularalgorithmamongthemduetoitshighaccuracyandgoodgeneralizationability.However,thedisadvantageofSVMisthatitrequirestime-consuminghyper-parametertuningtoachievegoodperformance.Thismakesitdifficulttoapplytoreal-timefaultdiagnosis. Toaddressthisissue,ahybridmethodcombiningthechaosfruitflyoptimization(CFO)andleastsquaressupportvectormachine(LSSVM)isproposedinthispaper.CFOisapowerfuloptimizationalgorithmthatcanquicklyfindtheoptimalsolution.LSSVMisafastandefficientpredictionmodelthatcanachievegoodaccuracyinfaultdiagnosis.TheoptimizationoftheLSSVMmodelparameterswiththeCFOalgorithmcanimproveitspredictionaccuracyandreducethetimerequiredforparametertuning. Methodology Theproposedmethodiscomposedoftwomainparts:parameteroptimizationwithCFOandfaultmodepredictionwithLSSVM.TheworkflowoftheproposedmethodisshowninFigure1. ![workflow.png](attachment:workflow.png) Figure1Wo