Comparative performance of different wavelets in Power Quality disturbance detection and quantification

Power Quality and Power Quality issues have become proliferent catchwords today in the power industry due to the different power quality aberrations caused by the increased use of power electronic devices and nonlinear loads. Detection of power quality disturbances is essential in order to mitigate them and to have increased efficiency of the power system. This paper presents a method of detection and quantification of power quality disturbances using different wavelets and a neural network classifier. Different wavelets have been used to extract features from the raw signal. Neural network classifier is employed to detect the type of power quality problem. The input to the neural network are the wavelet coefficients. The disturbance of interest include Voltage sag, Voltage swell, Harmonics, Interruption, Sag with harmonics and Swell with harmonics. Fault level and THD in the disturbances are also estimated by the trained Neural Network.