基于分形-小波的低速軸承磨損故障物理特征研究的中期報告

基于分形--小波的低速軸承磨損故障物理特征研究的中期報告Abstract:Thisisamid-termreportonthephysicalcharacteristicsoflow-speedbearingwearfaultsbasedonfractalwaveletanalysis.Takingthelow-speedbearingastheresearchobject,firstly,weanalyzedthewearfaultcharacteristicsoflow-speedbearings,andthenconstructedwaveletreconstructionmodelsfordifferentdegradationstagesofthebearing.Thereconstructedwaveletcoefficientswerethenusedtocalculatethefractaldimensionofthesignals,andthefractalcharacteristicsofthesignalsatdifferentdegradationstageswereanalyzed.Basedontheaboveanalysis,wefoundthatthefractaldimensionoftheoriginalsignalincreasedgraduallywiththedegradationofthebearing,andthatthefractalcharacteristicsofthereconstructedsignalsweresimilartothoseoftheoriginalsignals.Inaddition,wealsofoundthatthefractaldimensionofthehigh-frequencycomponentofthesignalswasmoresensitivetothewearfaultthanthatofthelow-frequencycomponent.Theresultsofthisstudyprovideatheoreticalbasisfortheearlywarninganddiagnosisoflow-speedbearingwearfaultsbasedonfractalwaveletanalysis.Theyalsoprovideareferenceforthedevelopmentofintelligentfaultdiagnosissystemsforbearings.Introduction:Thelow-speedbearingisanimportantcomponentinrotatingmachinery,anditsperformancedirectlyaffectstheefficiencyandreliabilityofthemachinery.Intheactualoperationprocess,thelow-speedbearingiseasilydamagedduetovariousfactorssuchasload,vibration,andlubrication,resultinginfaultssuchaswearandtear.Earlywarninganddiagnosisofthesefaultscaneffectivelyreducethemaintenancecostoftheequipmentandimprovethesafetyandreliabilityofthemachinery.Thefractalwaveletanalysismethodisanewmethodforsignalanalysis,whichhasbeenwidelyusedintheanalysisanddiagnosisoffaultsignalsinmachinery.Fractaldimensionisanimportantcharacteristicparameterofsignalsinfractaltheory,whichcanreflectthecomplexityofsignals.Wavelettransformisamulti-scaleanalysismethod,whichcandecomposesignalsintodifferentfrequencycomponentsandanalyzetheircharacteristics.Inthisstudy,takingthelow-speedbearingastheresearchobject,weanalyzedthewearfaultcharacteristicsoflow-speedbearingsandconstructedwaveletreconstructionmodelsfordifferentdegradationstagesofthebearing.Thereconstructedwaveletcoefficientswerethenusedtocalculateandanalyzethefractaldimensionofthesignalsatdifferentdegradationstages,providingatheoreticalbasisforearlywarninganddiagnosisofbearingwearfaults.Methodology:Experimentalsetup:Theexperimentalsetupincludesalow-speedbearing,amotor,aloadcell,andadataacquisitionsystem.Themotordrivesthelow-speedbearingtorotate,andtheloadcellisusedtomeasuretheloadofthebearing.Thedataacquisitionsystemcollectsthevibrationsignalsofthebearingduringtheoperationprocess.Signalprocessing:Thecollectedvibrationsignalwasprocessedbywavelettransform.Thedecompositionlevelwassettofive,andthedb4waveletwasusedasthemotherwavelet.Thewaveletreconstructionmodelwasconstructedfortheoriginalsignalandthesignalsatdifferentdegradationstagesofthebearing.Thereconstructedwaveletcoefficientswerethenusedtocalculatethefractaldimensionofthesignalsusingthebox-countingmethod.ResultsandDiscussion:Thewearfaultcharacteristicsofthelow-speedbearingwereanalyzed,andthewaveletreconstructionmodelswereconstructedfortheoriginalsignalandthesignalsatdifferentdegradationstagesofthebearing.Thereconstructedwaveletcoefficientswereusedtocalculatethefractaldimensionofthesignals,andthefractalcharacteristicsofthesignalsatdifferentdegradationstageswereanalyzed.Theresultsshowedthatthefractaldimensionoftheoriginalsignalincreasedgraduallywiththedegradationofthebearing,indicatingthatthesignalbecamemorecomplex.Thefractalcharacteristicsofthereconstructedsignalsweresimilartothoseoftheoriginalsignals,indicatingthatthewaveletreconstructionmodelwaseffective.Inaddition,wefoundthatthefractaldimensionofthehigh-frequencycomponentofthesignalswasmoresensitivetothewearfaultthanthatofthelow-frequencycomponent.Thisisbecausethehigh-frequencycomponentcontainsmoredetailedinformationofthesignal,whichisaffectedmorebythewearfault.Conclusion:Inthisstudy,thephysicalcharacteristicsoflow-speedbearingwearfaultsbasedonfractalwaveletanalysiswerestudied.Thewearfaultcharacteristicsofthelow-speedbearingwereanalyzed,andthewaveletreconstructionmodelswereconstructedfortheoriginalsignalandthesignalsatdifferentdegradationstagesofthebearing.Thefractaldimensionofthesignalswascalculated,andthefractalcharacteristicsofthesignalsatdifferentdegradationstageswereanalyzed.Theresultsshowedthatthefractaldimensionoftheoriginalsignalincreasedgraduallywiththedegradationofthebearing,andthatthefractalcharacteristicsofthereconstructedsignalsweresimilartothoseoftheoriginalsignals.Inaddition,thefractaldimensionofthehigh-frequencycomponentofthesignalswasmoresensitivetothewearfaultt