Comparison analysis of different classification methods of power quality disturbances

<span lang="EN-US">Good power quality delivery has always been in high demand system utilities where different types of disturbances are the main obstacles. As these have distinct characteristics and even unique mitigation techniques, their detection classification should be correct effective. In this study, eight were synthetically generated, by using a mathematical approach. Then, continuous wavelet transform (CWT) discrete with multi-resolution analysis (DWT-MRA) applied, which features then extracted from synthesized signals. Three classifiers namely, decision tree (DT), support vector machine (SVM) k-nearest neighbors (KNN) trained to classify disturbances. The accuracy was evaluated analyzed. best classifier integrated full model, performance proposed model observed 50 random signals, without noise. This study found that wavelet-transform effective localize at instant occurrence. On other hand, SVM is superior an overall 94%. Still, need for further optimized crucial ensuring more system.</span>