A decision support system based on Machine Learning and Fuzzy Logic Techniques for electrical fault detection and classification

Abstract

The detection and classification of faults in electrical systems are crucial for ensuring uninterrupted power supply. This paper proposes a decision support system that combines Machine Learning and Fuzzy Logic techniques to detect, classify, and diagnose faults in electrical systems. The system comprises Classification Module and Rules Module. The first one employs a Decision Tree to generate a set of rules and classify the current system status. The second one allows the operator to manipulate this knowledge basis to generate more intelligible and interpretable rules using Fuzzy Logic. The results using a well-known data set show the proposal overperforms other classification models and can generate more straightforward and interpretable rules.