Dissolved Gas Analysis of power transformer using K-means and Support Vector Machine

Abstract

The power transformer is ranked as one of the most important and expensive components in the electricity sector. However, the sudden failure of the power transformer places the system into serious or critical conditions. This paper utilizes artificial intelligence techniques to detect and predict transformer faults based on Dissolved Gas Analysis method and presents an intelligent methodology KMSVM (k-means and support vector machine) based on optimization technique to properly monitor, diagnose and predict the faults in the power transformer. Furthermore, the proposed technique helps in finding an effective and reliable monitoring technique to address transformer conditions at a much faster rate and hence minimizes the challenges.