Interpolated-DFT-Based Fast and Accurate Frequency Estimation for the Control of Power

Abstract

The energy produced by renewable energy systems must fulfill quality requirements as defined in the respective standards and directives. Improvement of the quality could be achieved through a more accurate estimation of the frequency of the grid's signal that is used to control an inverter. This paper presents an overview of a method for spectrum interpolation and frequency estimation, and a generalized method for very accurate frequency grid estimation using the fast Fourier transform procedure coupled with maximum decay sidelobe windows. An important feature of this algorithm is the elimination of the impact associated with the conjugate's component on the estimation's outcome (i.e, the possibility of designating the frequency even if the signal's measurement time is on the order of 2.5 periods), and the implementation of the algorithm is straightforward. The results of the simulation show that the algorithm could be successfully used for a fast and accurate estimation of the grid signal frequency. The systematic frequency estimation error is approximately 5·10 -11 Hz for a 5-ms measurement window. The algorithm could be used not only for a single sinusoidal signal, but also for a multifrequency signal. This is assuming that the appropriate spectrum leakage reduction (by a time window) will be performed.