HEALTH INDEX OF TRANSFORMER’S MONITORING USING ARTIFICIAL INTELLIGENT

Abstract

Dissolve Gas Analysis (DGA) for transformers is used to differentiate between a transformer in good condition and one which needs to schedule for maintenance. The main goal of DGA is to identify more precisely problems caused by the various gas formations in the transformer encountered. Key Gas Method (KGM) analysis is one of the DGA techniques often used. KGM is used in forecasting the health index of the transformer based on the formational of gases in the transformer. KGM’s classified the transformer health index in several conditions, which are Condition 1, Condition 2, Condition 3, or Condition 4. The multilayer perceptron (MLP) network outperforms K-Nearest Neighbourhood (KNN), Linear Discriminant Analysis (LDA), and Support Vector Machine (SVM) classifiers with 90.02 % accuracy. On the other hand, Bayesian Regularization (BR) training algorithm gives the best accuracy results among Levenberg Marquardt and Backpropagation training algorithms with 95.10 % accuracy.