Classification of Single and Combined Power Quality Disturbances Using Stockwell Transform, ReliefF Feature Selection Method and Multilayer Perceptron Algorithm

: In this study, a method based on Stockwell transform (ST), ReliefF feature selection and Multilayer Perceptron Algorithm (MPA) algorithm was developed for classification of Power Quality (PQ) disturbance signals. the method, firstly, ST applied to different PQ signals obtain features. A total 30 features were obtained by taking entropy values matrix after The use all causes be complicated training/testing times prolonged. Therefore, so as determine effective ones among ensure high success with less features, used in study. disturbances classified using 8 determined MPA. simulation results show that provides shorter time. At same time, have shown successful testing data noise levels 35 dB above only one training.