• DocumentCode
    911939
  • Title

    An equivalent linear model for magnetostatic nondestructive evaluation

  • Author

    Zhang, Zuorong ; Udpa, Lalita ; Udpa, Satish S. ; Sun, Yushi ; Si, Jiatun

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Iowa State Univ., Ames, IA, USA
  • Volume
    32
  • Issue
    3
  • fYear
    1996
  • fDate
    5/1/1996 12:00:00 AM
  • Firstpage
    718
  • Lastpage
    721
  • Abstract
    Numerical models capable of modeling magnetic flux leakage (MFL) methods of nondestructive testing are of significant interest to industry. The nonlinear nature of the MFL problem necessitates the use of an iterative model, thereby resulting in excessive computational effort. This paper describes an approach for developing an equivalent linear model (ELM) where the ferromagnetic region is appropriately partitioned into different domains with each domain being assigned a constant permeability value depending on the magnetization level and the flaw size. The nonlinear behavior of the multi-layered object is then modeled using a linear MFL model. The strategy results in significant computational savings without a substantial loss in accuracy. Results supporting the validity of the approach have been obtained using a 3D magnetostatic finite element (FE) model
  • Keywords
    finite element analysis; magnetic leakage; magnetostatics; nondestructive testing; 3D finite element model; domains; equivalent linear model; ferromagnetic region; magnetic flux leakage; magnetization; magnetostatic nondestructive evaluation; multi-layered object; nonlinearity; numerical model; permeability; Computer industry; Finite element methods; Geometry; Magnetic domains; Magnetic flux leakage; Magnetization; Magnetostatics; Nondestructive testing; Numerical models; Permeability; Saturation magnetization; Sun; Testing;
  • fLanguage
    English
  • Journal_Title
    Magnetics, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9464
  • Type

    jour

  • DOI
    10.1109/20.497340
  • Filename
    497340