Evaluating Mean Life of Power System Equipment With Limited End-of-Life Failure Data

Abstract

This paper presents two methods to estimate the mean life and its standard deviation of a power system equipment group with limited end-of-life or aging failure data. One is for the normal distribution model and another for the Weibull distribution model. Unlike the conventional sample mean technique which only uses ages of died components, the presented methods are based on all the information in an equipment group including both died and surviving components and provide a more accurate estimation. In the method for the normal distribution, the estimate is obtained from a set of simple calculation formulas while for the Weibull distribution, an optimization technique is used to obtain the estimates of the mean and its standard deviation as well as the shape and scale parameters. An equipment group containing 100 reactors with only four retired units was used as an application example to illustrate the procedure.