IETE Journal of Research · 2015 · 20 citations · 22 references
EngineeringElectrical EngineeringBuilding DiagnosticsSensor FusionTime Series AnalysisDiagnosticsDigital Signal ProcessingFault DetectionFault DiagnosisFrequency-domain AnalysisVibration MonitoringBearing Defect AnalysisBearing Fault DiagnosisBearing FaultsHilbert FunctionsHilbert TransformMachine Fault DiagnosisSignal ClassificationVibration Spectrum Analysis
This paper presents new methods based on fusion of Hilbert transform and high frequency resolution techniques (estimation of signal parameter by rotational invariance technique (ESPRIT), root multiple signal classification (root MUSIC)) to extract bearing defect frequencies of induction motor bearing faults using vibration signal. Further sliding window concept is introduced to tackle abnormal conditions like load variation and supply frequency variation. It also proposes a decision based approach to classify the severity of bearing fault. A series of results shows that both sliding window Hilbert-ESPRIT and sliding window Hilbert-root MUSIC are capable of extracting bearing defect frequencies accurately but, sliding window Hilbert-ESPRIT takes less computational time than the sliding window Hilbert-root MUSIC.
22
Miguel Delgado-Prieto, Giansalvo Cirrincione, J. A. Ortega et al. · IEEE Transactions on Industrial Electronics · 2012 · 456 citations