The optimization algorithm of reducing dimensional FFT for harmonics analysis of power system

In order to obtain higher precision, FFT (Fast Fourier Transform) algorithm requires that sampling points (N) in time-domain are largely enough, if the number of points increases, the amount of calculation will increase further more. For meeting the need of calculating speed, in this paper an optimization algorithm of reducing dimensional FFT that is suitable for harmonic analysis is proposed, the algorithm is based on characteristics that the number of points in frequency domain is far less than that of time domain for harmonics analysis of power system, and the application of windowed reducing dimensional FFT algorithm is also studied in the power system harmonic analysis. The harmonic measurement software of power system based on this algorithm has been tested through simulative experiment. The results show that the precision of optimization algorithm of reducing dimensional FFT is identical to that of standard FFT algorithm, but the amount of calculation and occupied memory in the computer of the former are much less than those of latter. So the presented algorithm greatly reduces the amount of calculation under the premise of ensuring accuracy.