Compression Method of Power Quality Disturbances Based on Independent Component Analysis and Fast Fourier Transform

In power quality monitoring systems, a large volume of data is generated. A compression method should be used to perform the storage of data more efficiently. The present study proposes a compression method of power quality disturbances based on independent component analysis, Fast Fourier Transform and an adaptive threshold obtained using mathematical morphology. The independent component analysis algorithm returns three statistically independent components among themselves. Considering the noise and disturbances are statistically independent of the fundamental component, the independent component analysis algorithm isolates the noise and disturbances in the third component, where a threshold is applied, while the first and second components are presented to the Fast Fourier Transform in order to characterize the fundamental component. The results were compared with a Wavelet-based method in which the proposed method achieved compression rates slightly higher than the Wavelet-based method.