Investigation of the Effectiveness of Averaging Frequency Measurements Using the Spectrum Analysis Method

Abstract

Accurate measurement of the power system signal frequency remains an important and relevant task, since it is one of the key indicators of the stability of electric power system operation. At present, one of the widely used methods for measuring the power system signal frequency is the method based on amplitude spectrum analysis. The essence of this method consists in determining the signal amplitude spectrum using the discrete Fourier transform, followed by identifying the position of the maximum spectral component on the frequency axis. The spectral leakage effect is the main reason for the reduction in measurement accuracy. The application of window functions and interpolation of spectral samples makes it possible to reduce the influence of this effect. The paper considers the application of the Hann window with interpolation based on two, three, and five spectral samples. This modification has finite accuracy and does not allow complete elimination of spectrum distortion. To improve the accuracy of this technique, the paper considers dividing the original measurement time into a certain number of intervals during which frequency measurement is performed using the method described earlier. Subsequently, averaging of the obtained frequency values is applied to determine the final measurement result. Both the division of the original observation time into equal intervals and into intervals of different lengths are considered. Using simulation in Matlab, a comparison is performed between the proposed modification and the approach in which the frequency is measured over a single long observation interval.