Failure rate modeling: A non-parametric data mining approach to MV network field data
ABSTRACT Power distribution fault statistics provides valuable knowledge of system failure rate behavior, as the start point of reliability evaluations. Using this statistics, electric utilities trace and develop their reliability plans based on fault statistics. This paper considers a data mining approach to model momentary failure rate in terms of the most influential factors. A methodology is presented here, for momentary fault cause identification, using a feature selection algorithm applied to MV network of the Greater Tehran Electric Distribution Company (GTEDC). Subsequently, two non-parametric failure rate models; classification and regression tree (CART) and artificial neural network (ANN) are utilized to cope with the high non-linearity of the problem space. Results and comparisons of the characteristics of the proposed methods illustrate the advantages and disadvantages of each model.
- [Show abstract] [Hide abstract]
ABSTRACT: SUMMARY Useful lifetime of power cables and related aging failure rates mostly depends on thermal stresses that they experience because of the loss-resulted heat inside the cable in normal or abnormal currents of the network. Knowing the real values of the cable aging failure rate is a key point, and the network operators should enable to estimate the related aging failure rates of their network cables. The already provided methods for cable transient temperature estimation require several different inputs as well as cable installation configuration data that are difficult to collect for operators. Hence, in this article, an artificial neural network–based approach is applied for estimation a cable maximum temperature, which serves a certain daily load curve. The artificial neural network only requires four inputs that are easy to provide. An experimental equation is then used for estimating the cable minimum temperature, and finally, a three-level temperature curve is formed for cable aging failure rate estimation. The life fractions lost during each level of the predicted temperature curve are evaluated by resorting to an already existing combined electrothermal life model held for cable insulation. This method uses the life model and the probabilistic failure model to predict the failure rate of power cables for a future period. The results show that failure rate estimations are in good accordance with exact results, remarking that the estimated three-level stepwise curve of cable temperature is a good approximation of cable thermal transients during cyclic loads. Copyright © 2012 John Wiley & Sons, Ltd.International Transactions on Electrical Energy Systems 09/2013; 23(6). · 0.63 Impact Factor