POWER TRANSFORMER DIAGNOSTIC MODEL USING MACHINE LEARNING ALGORITHMS

The use of machine learning techniques in the process of power transformer fault diagnosis represents a significant advancement in technology. This abstract of this project provides an overview of the key features of this work, displaying the profound benefits of machine learning in power transformer fault detection and prediction.
In recent years experiment have been conducted in this area using traditional methods like the Key Gas Method, Duval triangle method and others and utilizing the data gotten from DGA (Dissolved Gas Analysis) to detect faults in the power transformers. These traditional methods had a draw back of poor accuracy in detection and fault prediction. In this project machine learning techniques like the Naive Bayes, Support vector Machine, C4.5 Decision Tree and their ensembles also included in this is the Majority voting were introduced to solve the problems.
In this project a model is created to train and test data using this machine learning techniques. The results gotten from these experiments show that the machine learning techniques presents a brighter future for professionals and experts in this field of work.