Forecast of Power Quality Index Based on the Discrete Fourier Decomposition and AR Model

Abstract

A method of autoregressive (AR) based on discrete Fourier analysis is proposed to forecast the indexes of unbalance factor and total harmonic distortion. The discrete Fourier transform of the index sequence is analyzed and the low frequency components are extracted. Autoregressive method is applied to forecast each low frequency component. Through inverse discrete Fourier transform, the forecasting low frequency components are inverted to forecasting sequence of the index in time domain. Actual data is used to test this method and the results show that discrete Fourier analysis is possible to reduce the influence of high frequency noise and that the AR method based on Fourier transform can effectively forecast the indexes of unbalance factor and total harmonic distortion.