Power quality estimation and classification using wavelets

Abstract

Due to increase in the number of electric and electronic equipment along with the fast controlling power electronic devices, which has affected the main power quality events(PQ).These are namely short-circuits , notching, voltage sag and swell, harmonics and transient due to load switching. Whenever these disturbances occur, these last for few cycles and simple observation of waveforms in the bus-bar will not help to recognise the problem in there and henceforth will not be able to identify and sort out the problem. If such events occur for few more cycles/minute it may result to overvoltage and under-voltage, or long time power interruption or any other problem. Hence, an approach is developed for the detection and location of time and finally classifies the different power quality events including both the transients and steady state signals. By using one of the signal processing we can decompose namely wavelet decomposition by using “DISCRETE WAVELET TRANSFORM (DWT)”.In a sampling frequency, we can samples those signals and any change on the smoothness is detected at finer resolution levels of decomposition. By the decomposition process we get the power coefficients or wavelet packets which are necessary performance indices of the signal. We are in generally using the a) THD (Total Harmonic Distortion) and b) Energy of the signal to classify them in the PQ event analysis. The main purpose of the paper also focuses on using the MATLAB WAVELET TOOL GUIDE, for the study of De-noising and other purpose. Also we have to understand the mathematical tools used in the discrete wavelet transform (DWT) starting from the decomposition algorithm to the de-noising as well as thresh-holding of the PQ signals which are necessary in monitoring the classification scheme