• DocumentCode
    856490
  • Title

    Modelling losses in electrical steel laminations

  • Author

    Zirka, S.E. ; Moroz, Y.I. ; Marketos, P. ; Moses, A.J.

  • Author_Institution
    Dnepropetrovsk Nat. Univ., Ukraine
  • Volume
    149
  • Issue
    5
  • fYear
    2002
  • fDate
    9/1/2002 12:00:00 AM
  • Firstpage
    218
  • Lastpage
    221
  • Abstract
    The possibility of obtaining a voltage-driven solution of the diffusion equation by using a history-dependent hysteresis model based on the transplantation principle and a model for excess loss simulation derived from the Landau-Lifshitz-Gilbert equation for magnetic viscosity is presented. The proposed solver appears to be more versatile than existing models and can predict the total losses under arbitrary magnetising conditions as well as reproduce any frequency dependence of the excess losses. The model was tested on nonoriented (NO) and grain-oriented (GO) electrical steels under sinusoidal flux density over a frequency range from 50 to 400 Hz. Initial calculations, performed without viscosity being taken into account, gave underestimated losses with reduced RMS errors which varied from 5% for NO to 10% for GO steels. By introducing viscosity, this error reduced to 1% for both materials. The model parameters are thought to be material dependent so it is reasoned that the model can be used to accurately predict losses in a wide range of materials magnetised under sinusoidal as well as nonsinusoidal flux waveforms.
  • Keywords
    laminations; losses; magnetic flux; power apparatus; steel; 50 to 400 Hz; Landau-Lifshitz-Gilbert equation; electrical steel laminations; excess loss simulation model; grain-oriented electrical steels; history-dependent hysteresis model; losses modelling; magnetic viscosity; nonoriented electrical steels; sinusoidal flux density; transplantation principle;
  • fLanguage
    English
  • Journal_Title
    Science, Measurement and Technology, IEE Proceedings -
  • Publisher
    iet
  • ISSN
    1350-2344
  • Type

    jour

  • DOI
    10.1049/ip-smt:20020619
  • Filename
    1044816