• DocumentCode
    376837
  • Title

    Detection of a faulty power distribution apparatus by using thermal images

  • Author

    Ishino, Ryuichi

  • Author_Institution
    Central Res. Inst. of Electr. Power Ind., Japan
  • Volume
    2
  • fYear
    2002
  • fDate
    2002
  • Firstpage
    1332
  • Abstract
    A thermal image has often been used to inspect a power distribution apparatus. The development of an automatic diagnosis method for the apparatus, which utilizes thermal images, will facilitate the inspection. To this end, a new method of diagnosing a power distribution apparatus by means of thermal images has been developed as one of the functions incorporated into a monitoring device that moves on an overhead ground wire. The method consists of both extraction of the apparatus from a thermal image and diagnosis of the apparatus in terms of the local temperature gradient. Experiments using the proposed method were conducted under different weather conditions, times, and seasons. An error rate of 3% was obtained from the experiments on the extraction of apparatus, and an error rate of 17% was obtained from the experiments on the detection of faulty apparatus. The proposed method can extract and detect faulty apparatuses except the case in which the distance between the pole and the infrared camera is so large that the sensitivity is insufficient.
  • Keywords
    automatic optical inspection; distribution networks; fault diagnosis; infrared imaging; power apparatus; power system measurement; automatic diagnosis method; error rate; faulty power distribution apparatus detection; inspection automation; local temperature gradient; overhead ground wire; thermal images; Cameras; Electrical fault detection; Fault detection; Image processing; Infrared detectors; Inspection; Insulation; Land surface temperature; Power distribution; Wire;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Power Engineering Society Winter Meeting, 2002. IEEE
  • Print_ISBN
    0-7803-7322-7
  • Type

    conf

  • DOI
    10.1109/PESW.2002.985233
  • Filename
    985233