A novel DFT algorithm used in active power filter under frequency distortion

Abstract

The discrete Fourier transform (DFT) can be used to obtain the harmonic spectrum of input signal which makes it a key technology in active power filter (APF). The computation of conventional DFT algorithm is too complex to implement in a digital signal processor (DSP). By making use of the periodic characteristic of the input signal, the sliding DFT (SDFT) can greatly simplify the operation. But the SDFT will produce cumulative error under frequency distortion, which makes aberrance in the phase information of extracted harmonics. In this paper, a novel DFT algorithm is presented. The key idea of this algorithm is to amortize the operations of DFT to each control cycle. By measuring the zero crossing point of grid voltage, the influence of frequency distortion is avoid. The reliability of this algorithm is verified by both simulation and experiment.