Estimation of the Power Quantities Below One Signal Period Using DFT Coefficients

Abstract

This paper proposes a method to estimate power quantities based on IEEE 1459–2010 using measurement window with length as close as possible to the length of the changing signal period and the nonparametric estimation approach in the frequency domain. Parameters are calculated from the largest local discrete Fourier transformation (DFT) coefficients of the particular frequency components. The improvement in reducing the influences of the harmonic components is the estimation around one signal period. When the time is shortened below the signal period, the dc coefficients of the DFT are taken into account using two Rife–Vincent windows class I with different orders. First, the frequency is estimated by the quotient of the dc coefficients on the sampling points from the measurement interval of around half of the signal period, and upon this, the fine estimation of the signal period, number of samples in this period, and belonging relative frequency with displacement can be effectively estimated. Effectiveness of estimations and tracking of the power quantities is demonstrated by simulations and on the real signals of the short-circuit conditions in the power distribution system.