High-Precision Harmonic Analysis Algorithm Based on Five-Term MSD Second-Order Self-Convolution Window Four-Spectrum-Line Interpolation

For the frequency spectrum leakage and fence effect generated in Fast Fourier Transform (FFT) during asynchronous sampling and integral period truncation affect the precision of harmonic detection, a new algorithm for harmonic analysis based on four-spectrum-line interpolation FFT with second-order self-convolution window with five-term Maximum-Sidelobe-Decay (MSD) was proposed and the polynomial fitting method was used to construct the four-spectrum-line interpolation correction formula. The simulation results showed that this algorithm could improve the detection accuracy of amplitude, phase and frequency by 1-2 orders of magnitude compared with other commonly used windowed interpolation algorithms.