Data compression using 2-DWT power method for power quality detection

Abstract

This paper introduces a method combining 2D discrete wavelet transform (2-DWT) with energy threshold method to solve the data compression of power quality detection data. In this paper the high frequency signals and noise signals are decomposed into three different directions by making use of 2D db wavelet transform to process matrix data with row convolution and line convolution, and the energy of signals are concentrated on a few wavelet coefficients. Through improving energy threshold method, the thresholds are set up by energy correction coefficient. It assures the remainder energy over 99%. Thus, the distortion ratio of reconstructed signals is smaller and the eliminating noise of disturbing signals is adaptive. Results of simulations with six disturbing signals indicate this method has a higher compression ratio and a better ability to eliminate noise compared to compressive results of wavelet packet. (6 pages)