Effect of Feature Clubbing with Classifier in Power Quality Disturbance Classification

In order to determine the most effective wavelet-based feature extraction technique, we analyze the efficacy of certain wavelet-based features utilized in the PQD classification and compare the effect of their combinations with each other when applied with different classifiers. In this research three machine learning (ML)-based classifiers are used along with features obtained with discrete wavelet transform (DWT) to classify 15 different PQDs. Three different ML-based classifiers used for the purpose are random forest (RF), support vector machine (SVM) and random forest (RF). It is demonstrated that specific feature’s efficacy is not universal but rather depends on the other features it is combined with and the type of categorization technique being employed.