Pattern Recognition Applications for Power System Disturbance Classification

Abstract

This paper presents an automated on-line disturbance classification technique. This technique is based on wavelet multiresolution analysis and pattem recognition techniques. The wavelet-multiresolution transform is introduced as a powerful tool for feature extraction in order to classify different disturbances. Minimum Euclidean distance, k-nearest neighbor, and neural network classifiers are used to evaluate the efficiency of the extracted features.