A hybrid decision tree and new frequency filtering S-transform for simultaneous power signal disturbance pattern recognition

Abstract

This paper presents a new approach for detection and classification of various power signal disturbances, which constitute an important aspect of power quality assessment. A frequency filtering fast S-transform algorithm is developed with different types of frequency scaling, bandpass filtering and interpolation techniques to reduce the computational cost. The new time-frequency transform based on dyadic scaling has been used for the extraction of relevant features from the power quality disturbance signals. The extracted features are then passed through a decision tree based classifier for the identification of the disturbance patterns. Various simultaneous power signal disturbances have been simulated to prove the efficiency of the technique. The simulation results show superior performance of the new frequency filtering S-transform while classifying overlapping disturbance patterns. Because of the frequency filtering dyadic S-transform algorithm and a relatively simpler classifier methodology, this technique can be used for real time localization, detection, and classification of various power quality events.