Detection and classification of power quality disturbances with wavelet transform, decision tree algorithm and support vector machines

The main goal in this article is to present a comparative analysis of two well-known classification methods applied to the detection and classification of power quality disturbance signals. The Decision-Tree algorithm and Support Vector Machines are two of the most valuable and popular techniques used in the pattern recognition field. In this paper, power quality signals are generated through the Matlab/Simulink software and then classified in the trained model built. The method chosen for the characteristics extraction is the discrete Wavelet transform due to its great advantages of reducing the amount of data for computation and its perfect reconstruction. The analysis is performed for different kinds of Wavelet filters, noise levels and characteristic vectors.