Power Quality Data Compression Based on Pattern Similarity and LZW Compressed Encoding
LIU Xiaosheng,WANG Xinku,HUANG Nantian,XU Dianguo
DOI: https://doi.org/10.3969/j.issn.1006-6047.2012.03.011
2012-01-01
Abstract:According to the traits of power quality data in symmetry,parity and periodicity,a data compression algorithm based on pattern similarity and LZW compressed encoding is introduced,which normalizes the distance of pattern similarity to identify the distortion for the lossy compression of undistorted data and the lossless compression of distorted data,and adopts reasonably the LZW compressed encoding to increase the compression ratio.Models are established in MATLAB7.10 environment for six kinds of typical power quality disturbance signal,simulative experiments are designed to check the distortion detection capability of normalized distance,and the comparison of compression effect between the proposed algorithm and the one-dimension discrete wavelet transform show its performance and feasibility.