Reliability indices estimation in PEA distribution system using ANNs

Abstract

The purpose of this paper is to present the methodology of SAIFI, SAIDI and CAIDI estimation of Provincial Electricity Authority (PEA) distribution network in Thailand using Artificial Neural Networks (ANNs). The data used in this study was obtained from the Reliability Program of PEA. The feeder data from Singburi substation was used as the samples of ANNs in this research from January 2010 to October 2012. The three inputs of ANNs for reliability indices (SAIFI, SAIDI and CAIDI) consist of the number of customers behind the protective equipment, the interruption frequency of protective equipment per month and the total time interruption per month. ANNs gave the outputs (SAIFI, SAIDI and CAIDI) faster with accurate results.