Improved algorithm for harmonic parameters estimation under frequency offset
MA Ye chi, CHEN Long dao
(College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China)

Abstract: Aiming at the problem that the traditional phase difference correction method has a large error in the measurement of the fundamental frequency offset of the power grid signal, it may even produce the problem of measurement failure. An improved algorithm based on the traditional phase difference was proposed. The voltage signal of the grid was added to the Blackman Harris window. By analyzing the spectral expression of the windowed signal, the error source of the electrical parameter estimation was studied, and the spectral expression was polynomial transformed to accelerate the sidelobe decay rate, further reduce the spectral leakage and the spectrum Line, and then re estimate the electrical parameters according to the new spectral expression obtained from the estimation formula of the traditional phase difference method and the polynomial transformation. Respectively, using the traditional phase difference method and the polynomial transformation of the improved phase difference method for numerical simulation comparison. The results indicate that the improved algorithm is improved by at least one order of magnitude compared with the traditional phase difference method, and it is suitable for the high accuracy estimation of the harmonic parameters of the power system under the frequency offset. Even under the noise condition, The advantages of the algorithm is also more obvious.

Key words: harmonic analysis; frequency offset; windowed Fourier transform; phase difference; polynomial transformation