• DocumentCode
    2294091
  • Title

    Wavelet Neural Network Based Intelligent System for Oil Pipeline Defect Characterization

  • Author

    Tikaria, Mamta ; Nema, Shikha

  • Author_Institution
    SAKEC, Mumbai, India
  • fYear
    2010
  • fDate
    19-21 Nov. 2010
  • Firstpage
    42
  • Lastpage
    47
  • Abstract
    Wavelet neural network is a new kind of network which fuses advantages of wavelet transform and neural computing. It utilizes the good localize character of the wavelet transformation and combines the self learning function of the neural network. It has the ability of strong adaptive learning and function approach. Wavelet neural network has the simple implementation process and fast convergence rate, therefore it can be used to detect the defect of oil pipe. This paper presents a wavelet neural network approach for detection and characterization of defects from magnetic flux leakage signal.
  • Keywords
    flaw detection; inspection; learning (artificial intelligence); magnetic flux; mechanical engineering computing; neural nets; pipelines; wavelet transforms; adaptive learning; defect detection; intelligent system; magnetic flux leakage signal; neural computing; oil pipeline defect characterization; self learning function; wavelet neural network; wavelet transform; Magnetic Flux Leakage (MFL) technique; Neural Network; Wavelets;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Emerging Trends in Engineering and Technology (ICETET), 2010 3rd International Conference on
  • Conference_Location
    Goa
  • ISSN
    2157-0477
  • Print_ISBN
    978-1-4244-8481-2
  • Electronic_ISBN
    2157-0477
  • Type

    conf

  • DOI
    10.1109/ICETET.2010.88
  • Filename
    5698288