Skip to Main Content
In this paper, a wavelet based artificial neural network classifier for recognizing power quality disturbances is implemented and tested. Discrete wavelet transforms based multi-resolution signal decomposition technique is integrated with the feed-forward neural network model to develop the power quality problem classifier. Classification of the power quality problems has been carried out in two parts. In first part, multi-resolution signal decomposition analysis with Parseval's energy theorem is used to extract the energy features of the power quality signal. In the second part, this feature information is used to develop neural network classifier. The classifier has been tested on various disturbances viz. voltage sag, swell, momentary interruption, capacitor switching and single line to ground fault. Results obtained show the versatility of the classifier for classifying the most commonly power quality problems.