Classification of Stockwell Transform Based Power Quality Disturbance with Support Vector Machine and Artificial Neural Networks

The detection and classification of power quality events that disturb the voltage and/or current waveforms in electrical distribution networks is very important to generate energy deliver this end-user equipment at an acceptable voltage. Various property extraction methods are used determine type disturbances signal. In study, seven distortions including sag, swell, harmonics, sag with swell flicker, transient signals pure sine as a reference signal used. Synthetic data produced MATLAB using parametric equations based on TS EN 50160 standard. Four kinds feature frequency-amplitude, time-amplitude, geometric mean standard deviation made Stockwell Transform (ST), which one for determined GKB. Detection interpreted through these properties. 640 simulation entered into classifier by Support Vector Machines (SVM) Artificial Neural Networks (ANN) their performance compared.