The Research And Implementation Of Harmonic Detection Methods In Power System Based On The Improved FFT

With the rapid development of power electronics technology, a number of non-linearpower electronic devices are widely applied in power system, which makes the harmonicpollution is increasingly serious. Therefore, the harmonic detection has become an importantresearch direction in the field of power system.After a brief introduction to some basic conceptsof harmonics and related harmonic detection standard, the improved harmonic detectionalgorithm is realized, taking TMS320F2812as the core.The detailed work is as follows:Firstly, the spectral characteristic of second-order convolution window is discussedcarefully on the basis of the analysis of the spectral properties of the convolution window.Furthemore the formulas of amplitude, frequency and phase are obtained. Especially, theexplicit interpolation coefficients formula of second-order rectangular convolution windowand second-order hanning convolution window are derived.Secondly, an all-phase spectral analysis algorithm is introduced and cited in harmonicanalysis. In addition, a problem of all-phase spectral analysis algorithm is pointed out in thepractical application. And the corresponding correction formula of harmonic parameters isderived. Then, the performance characteristic of the second-order convolution windows andinterpolated FFT algorithm and all phase interpolation FFT algorithm is compared in theory.Simultaneously, the Matlab simulation is completed.Thirdly, the sidelobe characteristics of the Flat-top window are firstly analyzed, and animproved FFT approach based on Flat-top window is proposed in this paper. Through usingsub-correction method, the Flat-top window FFT with small computation is adopted when thefrequency offset is small. Then, the phase difference correction is adopted to rectify theamplitude of harmonic components when the frequency offset is large. Simulation resultsindicate that, spectral leakage and picket-fence effects are effectively suppressed by using theproposed method, and the precision of harmonic parameters is greatly improved.Finally, take TMS320F2812DSP chip as the core and realize the function of dataacquisition using its own internal ADC. Then, editing and testing of the software is completedbased on the improved harmonic analysis algorithm.