• DocumentCode
    493213
  • Title

    Failure Probabilities of Existing Overhead Shield Wires

  • Author

    Hathout, Ibrahim ; Vu, Linda

  • Author_Institution
    Lines Eng., Hydro One Inc., Toronto, ON
  • fYear
    2008
  • fDate
    25-29 May 2008
  • Firstpage
    1
  • Lastpage
    7
  • Abstract
    The objective of this paper is to introduce analytical models for accurate prediction of failure probabilities of existing shield wires and to recommend a time window for wire replacement at an acceptable failure threshold. Ductile and strength tests results and failures records of shield wires were examined to determine the trend of wires deteriorations as they age. Since most of failures occurred due to ice accretion during winter storms and wire deteriorations due to corrosion; an ice accretion model and corrosion models were used to predicts wire tensions and wire deteriorations during service life. To calculate the probability of failures, the strength and cross section area were assumed to be random variables modeled using normal distributions. The wire tensions due to ice accretions were calculated from the last 50 years of recorded winter storms. Anderson-Darling goodness of fit test was used to determine the best statistical model to fit the maximum annual wire tensions. The Gumbel distribution passes the test and was found to give stable results over wide range of tested conditions. Limit state function was developed and a computer program was written to determine the failure probabilities of shield wires using Rackwitz-Fiessler algorithm.
  • Keywords
    ageing; failure analysis; power transmission lines; statistical analysis; tensile strength; wires (electric); Anderson-Darling goodness; Gumbel distribution; Rackwitz-Fiessler algorithm; corrosion models; ductile test; failure probabilities; high voltage lines; ice accretion model; overhead shield wires; statistical model; strength tests; transmission lines; wire deteriorations; Analytical models; Corrosion; Gaussian distribution; Ice; Predictive models; Probability; Random variables; Storms; Testing; Wire;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Probabilistic Methods Applied to Power Systems, 2008. PMAPS '08. Proceedings of the 10th International Conference on
  • Conference_Location
    Rincon
  • Print_ISBN
    978-1-9343-2521-6
  • Electronic_ISBN
    978-1-9343-2540-7
  • Type

    conf

  • Filename
    4912651