Measurement method of adjacent harmonics or interharmonics based on Gaussian radial basis function frequency domain approximation

Abstract

Interpolated FFT method based on window function is a common method to measure harmonics or interharmonics in power systems. When the frequency resolution is insufficient or the main lobe of the window function is wide, the spectrum of adjacent harmonics or interharmonics may have aliasing of the main lobe, and it is difficult to accurately measure the parameters of harmonics or interharmonics. This paper proposes a modeling method for adjacent harmonics or adjacent interharmonics. The Fourier transform (FT) of the Gaussian function is still a Gaussian function. According to this feature, after adding a Gaussian window in the time domain and calculating the FFT spectrum, the real and imaginary parts of the aliased spectrum are decomposed into a series of linear superpositions of Gaussian radial basis functions (RBF). The center of the RBF is the frequency of the harmonic or interharmonic, and the amplitude and phase can also be calculated by the superposition coefficient of the RBF. The simulation verifies the effectiveness of the method.