Classification of transient power quality disturbance based on S-transform and fuzzy KNN

Abstract

This paper proposed a new method to classify transient power quality disturbances using S-transform and K-Nearest Neighbor classifier. First, S-transform was applied to perform time-frequency analysis on the power quality disturbance signals, and the S-transform mode time-frequency matrix was achieved from the S-transform, and we got 8 valid features extracted from the matrix, then the feature vectors who consisted of the former features were input into the Fuzzy KNN classifier to perform the disturbing signals intelligent classification, then completed the intelligent classification for six types of transient power quality disturbances signals. The result showed that the structure of the proposed method was simple, the number of the needed features was little and it was doing well in classifying these types of transient power quality disturbances.