• DocumentCode
    3148553
  • Title

    The research of the fault types recognition in transformer by differential protection based on LIBSVM

  • Author

    Wu, Hao ; Fu, ChengHua ; Guo, Hui ; Chen, Chang Zhong

  • Author_Institution
    Autom. & Electron. Inf. Eng., Sichuan Univ. of Sci. & Eng., Zigong, China
  • Volume
    7
  • fYear
    2010
  • fDate
    16-18 Oct. 2010
  • Firstpage
    2999
  • Lastpage
    3003
  • Abstract
    A new method to distinguish various fault types of transformer by differential protection based on LIBSVM is discussed. This paper uses PSCAD/EMTDC to build transformer and differential protection model which is applied to simulate various transformer fault types, the simulation data and characteristic value extracted are preprocessed; then fault identification model based on LIBSVM is established. This paper finds the optimal settings by different kernel functions and parameters through a lot of experiments. Experimental results show that LIBSVM can identify various fault types correctly, it verifies the accuracy of transformer fault types recognition by differential protection based on LIBSVM.
  • Keywords
    fault diagnosis; power engineering computing; power transformer protection; support vector machines; LIBSVM; PSCAD/EMTDC; differential protection; fault identification model; transformer fault type recognition; Accuracy; Fault diagnosis; Kernel; Low voltage; Support vector machines; Surges; Training; differential protection; fault diagnosis; magnetizing inrush current; support vector machine; transformer fault;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Biomedical Engineering and Informatics (BMEI), 2010 3rd International Conference on
  • Conference_Location
    Yantai
  • Print_ISBN
    978-1-4244-6495-1
  • Type

    conf

  • DOI
    10.1109/BMEI.2010.5639826
  • Filename
    5639826