• DocumentCode
    2530820
  • Title

    Artificial neural network based load blinder for distance protection

  • Author

    Zadeh, H. Khorashadi ; Li, Z.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Illinois Inst. of Technol., Chicago, IL
  • fYear
    2008
  • fDate
    20-24 July 2008
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    Distance relay can provide remote backup protection by zone 2 and zone 3, but it may mal-operate under heavy loading conditions and cause cascading trips in the network, which could further lead to a wide spread blackout. To prevent cascading outages, load blinders are generally used to block distance relay when there is heavy load in the system. However, conventional load blinders are not always able to discriminate between heavy loading conditions and fault conditions, especially for heavy loads with low power factor or faults with fault resistance. This paper proposes a novel load blinder scheme for distance protection by using artificial neural network (ANN). Test results show that the proposed ANN-based load blinder scheme is able to discriminate between different heavy loads with a wide range of power factors and different faults with fault resistance.
  • Keywords
    fault diagnosis; neural nets; power engineering computing; power factor; relay protection; ANN-based load blinder scheme; artificial neural network; cascading outage prevention; distance relay protection; fault resistance; power factor; remote backup protection; wide spread blackout; Artificial neural networks; Cities and towns; Impedance; Power system faults; Power system protection; Power transmission lines; Protective relaying; Reactive power; Relays; Testing; Distance relay; artificial neural network; blackout; load blinder; loadability;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century, 2008 IEEE
  • Conference_Location
    Pittsburgh, PA
  • ISSN
    1932-5517
  • Print_ISBN
    978-1-4244-1905-0
  • Electronic_ISBN
    1932-5517
  • Type

    conf

  • DOI
    10.1109/PES.2008.4596056
  • Filename
    4596056