Novel Frequency Estimator for Distorted Power System Signals Using Two-Point Iterative Windowed DFT

Tracking the frequency of power systems accurately and promptly is the cornerstone of power industry applications. The excellence of windowed interpolation discrete Fourier transform (DFT)-based methods has been proved in frequency estimation by suppressing both short- and long-range spectral leakage. However, due to the nonnegligible image component, their behavior deteriorates dramatically when short time windows are adopted. To this purpose, a novel frequency estimator based on two-point DFT and Hanning window is proposed for estimating distorted power signals. The method starts with an accurate spectral model that fully considers the effects of the image component. An iterative procedure is then designed to approach the minimum estimation error by using the golden section search. Simulation and experimental results demonstrate that the method offers fast and accurate estimation within a short time window (even as short as 0.5-1.5 cycles).