• DocumentCode
    2445047
  • Title

    Fault location using wavelet packets

  • Author

    Yan, Feng ; Chen, Zhiye ; Liang, Zhirui ; Kong, Yinghui ; Li, Peng

  • Author_Institution
    Dept. of Electr. Power Eng., North China Electr. Power Univ., Baoding, China
  • Volume
    4
  • fYear
    2002
  • fDate
    2002
  • Firstpage
    2575
  • Abstract
    A technique using wavelet packets is presented for accurate fault location on power lines with branches. It relies on detecting fault-generated transient traveling waves and identifies some waves reflected back from discontinuities and the fault point. Wavelet packets analysis is used to decompose and reconstruct high-frequency fault signals. An eigenvector matrix consists of local energies of high-frequency content of the fault signal; the faulty section is determined by comparing local energies in the eigenvector matrix with a given threshold. With the faulty section determined, the time ranges of two reflected waves related with fault point would be found out, then the fault point is located. The paper shows the theoretical development of the algorithm, and together with the results obtained using EMTP simulation software modeling, a simple 10 kV overhead line circuit.
  • Keywords
    EMTP; digital simulation; eigenvalues and eigenfunctions; matrix algebra; power overhead lines; power system faults; power system simulation; signal reconstruction; wavelet transforms; 10 kV; EMTP simulation; eigenvector matrix; fault location; fault point; fault point discontinuities; fault-generated transient traveling waves; high-frequency fault signals; overhead line circuit; power lines; reflected waves; signal decomposition; signal reconstruction; wavelet packets; Circuit faults; EMTP; Fault detection; Fault diagnosis; Fault location; Matrix decomposition; Signal analysis; Software algorithms; Wavelet analysis; Wavelet packets;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Power System Technology, 2002. Proceedings. PowerCon 2002. International Conference on
  • Print_ISBN
    0-7803-7459-2
  • Type

    conf

  • DOI
    10.1109/ICPST.2002.1047251
  • Filename
    1047251