Mining Association Rules of Distribution Network Equipment Based on Genetic Algorithm

Abstract

Due to the wide variety of 10kV distribution network equipment and the complex operating environment, various defects have emerged frequently. Because the distribution network is directly connected to users, the distribution network equipment defects will lead to the occurrence of distribution network failures if not processed in time, affecting the safe and stable operation of the power grid. In this paper, a genetic algorithm(GA)-based method is used to find useful association rules from the defect data set of distribution network equipment. In this method, genetic algorithm is used to avoid scanning transaction database multiple times and improve the efficiency of mining association rules. Moreover, the confidence factor is used to objectively measure the interest of the association rules without providing a minimum support threshold. In this paper, the defect association analysis of the main distribution equipment in a region is carried out, and the association results are obtained, which can help the maintenance personnel to find the family defects of the equipment and speculate the type of defects, providing support for the maintenance and repair of equipment.