Ensemble classifier selection using multi-objective PSO for fault diagnosis of power transformers

This paper presents a binary version of the Multi-Objective Particle Swarm Optimization (bi-MOPSO) algorithm to classify the faults of power transformers. The proposed method selects the most accurate and diverse classifiers, simultaneously. Then, the selected classifiers are combined to diagnose the actual faults of power transformers using dissolved gas analysis (DGA) performed on the oil of power transformers. The obtained results are compared to other scenarios such as combining the outputs of all classifiers or using only the most accurate classifier to diagnose the faults. The comparison reveals that the proposed method is highly reliable and useful for diagnosing the faults of power transformers.