• DocumentCode
    1005967
  • Title

    Induction Machine Broken Bar and Stator Short-Circuit Fault Diagnostics Based on Three-Phase Stator Current Envelopes

  • Author

    Da Silva, Aderiano M. ; Povinelli, Richard J. ; Demerdash, Nabeel A O

  • Author_Institution
    Rockwell Autom., Mequon
  • Volume
    55
  • Issue
    3
  • fYear
    2008
  • fDate
    3/1/2008 12:00:00 AM
  • Firstpage
    1310
  • Lastpage
    1318
  • Abstract
    A new method for the fault diagnosis of a broken rotor bar and interturn short circuits in induction machines (IMs) is presented. The method is based on the analysis of the three-phase stator current envelopes of IMs using reconstructed phase space transforms. The signatures of each type of fault are created from the three-phase current envelope of each fault. The resulting fault signatures for the new so-called ldquounseen signalsrdquo are classified using Gaussian mixture models and a Bayesian maximum likelihood classifier. The presented method yields a high degree of accuracy in fault identification as evidenced by the given experimental results, which validate this method.
  • Keywords
    Bayes methods; Gaussian processes; asynchronous machines; fault diagnosis; induction motor protection; short-circuit currents; signal classification; stators; Bayesian maximum likelihood classifier; Gaussian mixture model; fault diagnosis; fault identification; fault signature; induction machine broken bar diagnostic; stator short-circuit fault diagnostic; three-phase stator current envelope; unseen signal; AC motor drive systems; AC motor-drive systems; Fault diagnosis; broken bars; envelope detection and classification; fault diagnosis; induction machines; induction machines (IMs); induction motors; inter-turn short circuits; interturn short circuits;
  • fLanguage
    English
  • Journal_Title
    Industrial Electronics, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0278-0046
  • Type

    jour

  • DOI
    10.1109/TIE.2007.909060
  • Filename
    4401123