Research On Harmonic Detection Algorithm Based On ApFFT And Wavelet Transform

With the development of modern industrial economy,a lot of power electronics device is applied in power system,transportation,industry,and family,the move has brought convenience,but also brings serious harmonic pollution in power system,the harmonic detection is the foundation of research on harmonic problems,so the new harmonic detection algorithm for harmonic governance has important significance.This paper carries out the work from the following five aspects:1.Based on the basic theory of harmonics,a harmonic analysis framework is constructed,and the advantages and disadvantages of several commonly used harmonic detection methods at home and abroad are discussed.The basic theories of FFT algorithm,wavelet transform and windowed interpolation algorithm are introduced,and the advantages and disadvantages of each algorithm are compared and verified.2.In view of the defects of the traditional method is proposed based on wavelet and FFT spectrum wave detection algorithm,the wavelet transform as an extension of the FFT,it has the dual resolution of time domain and frequency domain,can signal with discontinuous points are dealt with,the FFT is unable to do,so wavelet transform can not only get the signal frequency domain information,also can get the signal of time domain information.The algorithm combines the advantages of FFT and wavelet transform.FFT algorithm has certain advantages in analyzing steady signals,but it cannot detect unsteady signals.Wavelet transform is suitable for detecting unsteady signals,but not for detecting steady signals.The joint algorithm of FFT and wavelet combines the advantages of both,and can detect the steady and unsteady signals at the same time.However,it is not practical and accurate,so the method based on APFFT is adopted to analyze the harmonics.3.In the case of asynchronous sampling,the use of fast Fourier transform(FFT)for power system harmonic analysis,will bring the spectrum leakage phenomenon and fence effect,affecting the measurement accuracy of the signal.Therefore,a Hanning double window all phase FFT trispectral interpolation algorithm for harmonic detection is proposed.Based on the Hanning double-window all-phase FFT analysis,the amplitude of three adjacent spectral lines near the fundamental frequency point is compared.Combined with the phase invariance of all-phase FFT analysis,the phase of the spectral line with the largest amplitude at the sampling point is taken as the initial phase of the signal,so as to further calculate the parameters of the harmonic signal.Simulation results show that compared with other interpolation algorithms,the proposed algorithm can reduce the detection error of harmonic parameters more effectively.However,the proposed algorithm usually gets the wrong frequency spectrum and the wrong initial value under the condition of noise and attenuation signals,which leads to the failure of the proposed algorithm.4.On the basis of the above,this paper proposes an algorithm combining wavelet transform and APFFT.This algorithm combines the advantages of wavelet transform and improved APFFT algorithm respectively.For low frequency band signals,APFFT analysis is used to obtain all the steady-state components,and the precise amplitude,frequency and phase of integer harmonics and inter-harmonics are obtained by training.In this way,combining the advantages of the wavelet transform algorithm and the improved APFFT algorithm,it not only completes the detection of the unsteady signal,but also for the detection of the steady-state integer harmonics and inter-harmonics,its accuracy,anti-noise,practicability and real-time performance have been greatly improved.5.For the traditional FFT and its interpolation algorithm,wavelet transform,based on the improved APFFT algorithm,and the joint WT-APFFT algorithm,this paper constructed a number of signal models,in the MATLAB software environment to write a program for simulation research.The simulation results show that the amplitude and frequency measurement of the proposed algorithm are within the error range,and the detection accuracy is better than that of the traditional FFT algorithm,which has a certain practical application and feasibility.