Research on harmonic detection algorithm based on extended DFT

Jing ZHANG,Qingsheng ZHAO,Xuping WANG,Zun GUO

DOI: https://doi.org/10.16652/j.issn.1004-373x.2017.21.046

2017-01-01

Abstract:The traditional discrete Fourier transform(DFT)algorithm has fast computational efficiency and is convenient to apply to the embedded system,which is widely used in the harmonic detection. However,the available spectrum leakage and picket fence effect may influence the accuracy of the measuring result under the condition of asynchronous sampling. The extended discrete Fourier transform (EDFT) algorithm is applied to the harmonic detection. Taking the Fourier integral transform as the target of EDFT algorithm,the transform basis is optimized within the extended frequency range,for which the index basis in tra-ditional DFT is replaced by the transform basis function to construct the target function of the signal. The original data is carried out with the iterative approximation fitting to improve the measuring accuracy. The simulation result shows that,under the condi-tions of asynchronous sampling and noise,the algorithm has high frequency resolution and strong anti-interference ability,and is suitable for the high-precision measurement or detection.