New procedure for gear fault detection and diagnosis using instantaneous angular speed
Title of Paper:
New procedure for gear fault detection and diagnosis using instantaneous angular speed
Journal:
Mechanical Systems and Signal Processing
Summary:
Besidestheextremecomplexityofgeardynamics,thefaultdiagnosisresultsintermsof
vibration signalaresometimeseasilymisledandevendistortedbytheinterferenceof
transmission channelorothercomponentslikebearings,bars.Recently,theresearchfield
of InstantaneousAngularSpeed(IAS)hasattractedsignificantattentionsduetoitsown
advantagesoverconventionalvibrationanalysis.OnthebasisofIASsignal'sadvantages,
this paperpresentsanewfeatureextractionmethodbycombiningtheEmpiricalMode
Decomposition(EMD)andAutocorrelationLocalCepstrum(ALC)forfaultdiagnosisof
sophisticatedmultistagegearbox.Firstly,asapre-processingstep,signalreconstructionis
employedtoaddresstheoversampledissuecausedbythehighresolutionoftheangular
sensor andthetestspeed.ThentheadaptiveEMDisusedtoacquireanumberofIntrinsic
Mode Functions(IMFs).Nevertheless,notalltheIMFsareneededforthefurtheranalysis
since differentIMFshavedifferentsensitivitiestofault.Hence,thecosinesimilaritymetric
is introducedtoselectthemostsensitiveIMF.Eventhough,thesensitiveIMFisstillin-
sufficientforthegearfaultdiagnosisduetotheweaknessofthefaultcomponentrelated
to thegearfault.Therefore,asthefinalstep,ALCisusedforthepurposeofsignalde-
noisingandfeatureextraction.Theeffectivenessandrobustnessofthenewapproachhas
been validatedexperimentallyonthebasisoftwogeartestrigswithgearsunderdifferent
working conditions.Diagnosisresultsshowthatthenewapproachiscapableofeffectively
handlingthegearfaultdiagnosisi.e.,thehighlightedquefrencyanditsrahmonicscor-
responding totherotaryperiodanditsmultiplearedisplayedclearlyinthecepstrum
recordoftheproposedmethod.
Co-author:
Li Bing， ZhangXining， Wu Jili
Papers
