Fault Classification from Dissolved Gas Analysis Results Using Machine Learning

Transformers are essential devices in the distribution system. If the transformer is damaged, It will cause a prolonged power outage and affect many customers. Therefore, inspecting and analyzing the transformer’s condition for maintenance planning is essential. Dissolved gas analysis is extensive use because sampling of transformer oil for analysis can be carried out without shutting down the transformer and at a low cost. This paper discusses the application of machine learning methods to classify fault types from the results of dissolved gas analysis. The significant gases dissolved in transformer oil are Hydrogen (H
2
), Methane (CH
4
), Ethane (C
2
H
6
), Ethylene (C
2
H
4
), and Acetylene (C
2
H
2
). Fault type analysis using Classification Machine Learning Algorithms such as Decision Tree, Naive Bayes, k-nearest Neighbor, and Support Vector Machine compared to Ensemble Machine Learning Algorithms such as Bootstrap Aggregation (Bagging), Random Forest, and Adaboost. Ensemble Machine Learning Algorithms make prediction results more accurate than Classification Machine Learning algorithms.