Repair models of power distribution components

For reliability assessments of power distribution systems it has been customary to represent the failure and repair processes of the components by exponential models. A problem with this practice is that in many cases it is not checked if component operating data really fits to exponential models. Regarding repair times, several references have claimed they are generally not exponentially distributed but lognormally. For the case of power system components, a review of the literature shows this subject is not treated in dept and the most common information about repair times is given in the form of mean values i.e. point estimators not probabilistic models. Thus, using real data, a study on the modeling of repair times for 46 classes of power distribution components was carried out. The main results are: 1. Repair times have a very high variability; thus, results of analysis based only on their mean values should be used with caution. 2. Only for a half of the studied classes the exponential model is valid, but in contrast, the log-normal distribution is valid for all them; this means, if a model for repair times of power distribution components is to be assumed the lognormal distribution is the one to be chosen, and, for system reliability assessments, analysts should consider the Montecarlo simulation method that is not restricted to exponential modeling.

[1]  N. Balijepalli,et al.  Modeling and analysis of distribution reliability indices , 2004, IEEE Transactions on Power Delivery.

[2]  Harold E. Ascher.  Repairable Systems Reliability , 2008 .

[3]  D. W. Jacobson,et al.  A nonexponential approach to availability modeling , 1995, Annual Reliability and Maintainability Symposium 1995 Proceedings.

[4]  Magdy M. A. Salama,et al.  Resistance formulas of grounding systems in two-layer earth , 1996 .

[5]  K. Sand,et al.  RELRAD-an analytical approach for distribution system reliability assessment , 1991, Proceedings of the 1991 IEEE Power Engineering Society Transmission and Distribution Conference.

[6]  Peng Wang,et al.  Teaching distribution system reliability evaluation using Monte Carlo simulation , 1999 .

[7]  Hassan Zahedi,et al.  Repairable Systems Reliability: Modeling, Inference, Misconceptions and Their Causes , 1989 .

[8]  C. K. Hansen,et al.  Spurious exponentiality observed when incorrectly fitting a distribution to nonstationary data , 1998 .

[9]  Richard L. Scheaffer,et al.  Probability and statistics for engineers , 1986 .

[10]  Daniel R. Jeske.  Estimating the cumulative downtime distribution of a highly reliable component , 1996, IEEE Trans. Reliab..

[11]  I. Miller.  Probability, Random Variables, and Stochastic Processes , 1966 .

[12]  Mo-Yuen Chow,et al.  Time of outage restoration analysis in distribution systems , 1996 .

[13]  Christos Douligeris,et al.  An integrated framework for managing emergency-response logistics: the case of the electric utility companies , 1998 .

[14]  R. Billinton.  System and equipment performance assessment [power systems] , 1995 .

[15]  Richard Brown,et al.  Electric Power Distribution Reliability, Second Edition , 2008 .

[16]  Averill M. Law,et al.  Simulation Modeling and Analysis , 1982 .

[17]  Bruno O. Shubert,et al.  Random variables and stochastic processes , 1979 .

[18]  S. D. Singh,et al.  System reliability modeling and evaluation , 1989 .

[19]  Ron Allan,et al.  Sequential probabilistic methods for power system operation and planning , 1998 .

[20]  C.J. Zapata,et al.  Reliability Assessment of Unbalanced Distribution Systems using Sequential Montecarlo Simulation , 2006, 2006 IEEE/PES Transmission & Distribution Conference and Exposition: Latin America.