基于优化VMD与噪声估计的间隔阈值局部放电去噪方法-豆柴文库

您所在位置：网站首页 / 基于优化VMD与噪声估计的间隔阈值局部放电去噪方法.docx / 文档详情

在线预览结束，喜欢就下载吧，查找使用更方便

5 金币

下载文档

/ 2

下载提示

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

文本预览

基于优化VMD与噪声估计的间隔阈值局部放电去噪方法
Introduction
Partialdischarge(PD)isoneofthemostcommonfaultsinpowerequipment,anditoftenleadstotheinsulationdegradation.Therefore,PDdetectionandlocalizationhasbecomeessentialinthemaintenanceofpowerequipmenttoensurethereliabilityofthepowersystems.However,therobustnessofthedetectionandlocalizationmethodsareoftenhamperedbytheexistenceofnoisesources,especiallyinthehighvoltageequipment.Inthispaper,weproposeanintervalthresholdPDdenoisingmethodbasedontheVariationalModeDecomposition(VMD)andnoiseestimationforthede-noisingofPDsignals.
Methods
Firstofall,therecordedPDsignalisdecomposedbyusingVMD.VMDisanovelsignaldecompositiontechniquethatcaneffectivelyseparatethePDsignalfromnoise.Itdecomposesthesignalintoseveralintrinsicmodefunctions(IMFs)whileofferinggoodfrequencyresolution.ThePDsignalusuallylocatesinthehigh-frequencyband,whilethenoiseresidesinthelow-frequencyband.Therefore,thehigh-frequencyIMFsafterdecompositionbyVMDcanreflectthePDsignalwell.
Then,anoiseestimationmethodbasedonthestatisticalcharacteristicsof13-kVpartialdischargesignalsisproposedtoestimatethenoisevarianceincludedinthesignal.Themethodinvolvesa10-foldratioestimatorthatusesthedifferencebetweenthekurtosisofthenoiseandthekurtosisoftheresidualVMDlow-frequencycomponent.Theaccuracyofthenoisevarianceestimationisfurtherimprovedbyusinganiterativeupdatemethodthatreducestheinfluenceofabnormalsamples.
Finally,anintervalthresholddenoisingmethodisproposed.Themethodincludestwokeysteps:thresholddeterminationandthresholdapplication.Inthefirststep,thethresholdisdeterminedbyusingthemeanandstandarddeviationofthePDsignalineachVMDIMF.Inthesecondstep,thethresholdisappliedtotheVMDdecompositionresult,andthedenoisedsignalisreconstructedbytheremaininghigh-frequencyVMDIMFs.TheintervalthresholdmethodcaneffectivelyreducethenoisewhilepreservingthePDsignal.
Results
TheproposedmethodwastestedonbothsimulatedandexperimentalPDsignals.ThesimulationexperimentshowsthattheproposedintervalthresholdPDdenoisingmethodoutperformssev