预览加载中，请您耐心等待几秒...

基于改进AFSA的参数优化VMD和ELM的轴承故障诊断.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

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

基于改进AFSA的参数优化VMD和ELM的轴承故障诊断 Title:ImprovedAFSA-basedParameterOptimizationforVMDandELMinBearingFaultDiagnosis Abstract: Bearingfaultdiagnosisplaysacriticalroleinensuringthereliabilityandsafetyofrotatingmachinery.Inthispaper,anovelapproachisproposedtoenhancetheperformanceoftheVariationalModeDecomposition(VMD)andExtremeLearningMachine(ELM)forbearingfaultdiagnosis.Toachievethis,animprovedModifiedArtificialFishSwarmAlgorithm(AFSA)ispresentedtooptimizetheparametersofVMDandELM,aimingatimprovingthediagnosisaccuracy.ExperimentalresultsdemonstratethattheproposedapproachshowssuperiorperformancecomparedtotraditionalVMDandELMmethods. 1.Introduction Bearingfaultdiagnosishasgainedsignificantattentioninrecentyearsduetoitsimportanceinpreventingcostlybreakdownsandensuringsafetyinrotatingmachinery.Theidentificationoffaultcharacteristicsaccuratelyandefficientlyiscrucialfortimelymaintenanceanddecision-making.Amongvarioussignalprocessingandmachinelearningtechniques,VMDandELMhaveshownpromisingresultsinbearingfaultdiagnosis.However,theperformanceofthesemethodscanbefurtherimprovedthroughparameteroptimization. 2.VariationalModeDecomposition(VMD) VMDisarecentlyproposedadaptivesignaldecompositionmethodthatseparatesasignalintoasetofcomponentscalledintrinsicmodefunctions(IMFs)withdifferentfrequencyscales.InVMD,theselectionofparameterssuchasthenumberofcomponentsandthebalancingparametergreatlyaffectstheaccuracyoffaultdiagnosis.Theconventionalselectionoftheseparametersisoftensubjectiveandtime-consuming,whichresultsinsuboptimalperformance.Toaddressthisissue,animprovedAFSAisproposedforparameteroptimization. 3.ExtremeLearningMachine(ELM) ELMisapowerfulmachinelearningalgorithmthathasgainedpopularityinvariousapplicationsduetoitsfasttrainingspeedandaccurateprediction.Inbearingfaultdiagnosis,ELMhasbeensuccessfullyappliedtoclassifyfaultpatterns.However,theperformanceofELMcanbeaffectedbyparameterssuchasthenumberofhiddenneuronsandtheregularizationfactor.Theconventionalparameteroptimizationmethodsoftenrequire