Research On Power Harmonic Detection And Measurement Algorithm

The power harmonic detection and measurement algorithm based on Fast Fourier Transform(FFT)is used as the main method of harmonic detection because of its high detection precision,simple principle and easy implementation in hardware.However,when using this method for harmonic analysis,spectral leakage and fence effects are often generated,resulting in harmonic detection accuracy that does not meet the relevant standards.The study found that the interpolation algorithm can effectively reduce the error caused by the fence effect,and the signal windowing can reduce the error caused by the spectrum leakage.But the sidelobe characteristics of the window function directly affect the improvement effect.Therefore,a combined cosine window function parameter optimization method based on genetic algorithm(GA)is proposed.By using this method,the combined cosine window function of six terms is optimized,and a six-term fifth-order combined cosine window function is obtained.The four-line interpolation algorithm of the window is derived,and the correction formula is obtained by curve fitting function.Furthermore,a power harmonic detection and measurement algorithm based on six-term fifth-order combined cosine window four-line interpolation FFT is proposed.The simulation shows that the sidelobe characteristics of the six-term fifth-order window are better than the common window functions such as Hanning window,Blackman window and Nuttall window.The power harmonic detection and measurement algorithm of the four-line interpolation FFT realized by this window has higher precision.The effectiveness of the algorithm is verified when considering the fundamental frequency fluctuation of the power grid,different intensity white noise and inter-harmonic.At the same time,considering that the FFT can not analyze the transient signal and the wavelet transform has localized characteristics in the time domain.The power harmonic detection and measurement algorithm based on wavelet transform is studied.The power harmonic detection algorithm based on wavelet packet transform is realized by Matlab.Based on the simulation of power harmonic detection and measurement algorithm based on FFT windowed interpolation or wavelet packet transform,a power harmonic hybrid detection algorithm based on FFT and wavelet packet transform is proposed.The algorithm is verified by simulation.Finally,the power harmonic detection algorithm based on six-term fifth-order combined cosine window four-line interpolation FFT is realized by programming DSP28335 integrated development board,and the algorithm is verified by experiments.