Research On A Segmented FFT Algorithm Of Power Harmonic Detection Based On 1st-Order Self-Convolution Window

With the development of hybrid multi-terminal HVDC transmission systems and the integration of various new energy sources such as wind energy and solar energy into the grid,the harmonic characteristics of power electronics show clear trends.Significant changes have occurred such as widening harmonic bandwidth,and increasing noise interference,which has led to new challenges in power harmonic detection.Traditional harmonic detection algorithms are no longer fully applicable.Harmonic parameter estimation is a prerequisite for effective harmonic control.Therefore,a new harmonic detection algorithm suitable for electronic power systems has a great significance and value.In this regard,this article comprehensively considers the harmonic characteristics of the current power system,and this article studies the shortcomings of FFT-based harmonic detection method and the behaviors of existing improvement measures.Finally,the FFTbased harmonic detection algorithm is improved to meet the requirements of harmonic detection of power electronics.The main contents of article are as follows:(1)By studying the influence of FFT defects on the estimation of harmonic parameters,including frequency aliasing,spectrum leakage and fence effect.Secondly,this article studies the pros and cons of the solutions for FFT defects,including the spectral characteristics of the window function and spectral correction measures,and mathematically analyzes the influence of the allphase data preprocessing algorithm on the performance of FFT harmonic parameter estimation.(2)Firstly,the article reforms the Kaiser window function by self-convolution operation,and further analyzes the influence of self-convolution operation on the spectral characteristics.Secondly,the 1st-order Kaiser self-convolution window is used to realize the complementary advantages of the spectral interpolated algorithm and the all-phase preprocessing algorithm,which solves the problem of the accumulation of theoretical errors in the frequency estimation of the spectral interpolation algorithm and the increase of the fence effect of the all-phase data preprocessing algorithm.Therefore,a power harmonic detection FFT algorithm based on the 1storder Kaiser self-convolution window is proposed.(3)Aiming at the redundancy problem of calculation time of high-frequency harmonic detection under harmonic characteristics of power electronics,the sampling frequency and sampling time window data of the algorithm are optimized,and the idea of segmented harmonic detection is proposed.For the harmonics of different frequency bands,different sampling frequencies and sampling time window data are used to analyze the signal by the FFT algorithm based on 1st-order Kaiser self-convolution window.In the case of ensuring the detection accuracy,the calculation speed of the detection algorithm based on a 1st-order Kaiser self-convolution window is improved.The paper achieves better frequency,amplitude,and initial phase estimation without significantly increasing the amount of calculation,and at the same time improves the algorithm’s anti-noise performance and stability.Finally,a case analysis is carried out based on 50-fold fundamental frequency wideband signal model,the introduction of fundamental frequency fluctuations and white noise interference tests verify the effectiveness of the proposed algorithm and sub-band ideas for broadband harmonic detection.