A fault diagnosis model of power transformer using Imperialistic Competitive Algorithm and Support Vector Machine

ICA, firstly introduced by Esmaeil Atashpaz Gargari in 2007[8], is an effective optimization algorithm because of its ability of the efficient local search control and good convergence solution to the global optimum. Therefore, in this paper, a multiclass ICA-SVM integrated approach was proposed to find out whether the approach is

1 INTRODUCTION

Power transformers are considered to be the most essential equipment in a power system. A fault in a transformer can result in not only substantial repair cost but also power interruptions to thousands of customers. Therefore, it is essential to assess the working condition and conduct fault diagnosis of power transformers[1].