Research On Harmonic Detection Method Based On Corrected FFT And Symplectic Geometric Mode Decomposition

With the continuous increase of nonlinear loads in the power system,the harmonic problem in the power system is becoming more and more serious.Harmonics will affect the normal operation of the power system.Therefore,detecting and analyzing the harmonic content and start-stop time is the basis of harmonic control in the power system.It is of great practical significance to maintain the safe operation of the power system.At present,there are many harmonic detection methods,but many detection methods only have advantages in detecting a specific harmonic,and have limitations in the face of a variety of complex harmonics.In this thesis,the research contents of steady-state harmonic and transient harmonic signals in power system are as follows :(1)Firstly,the fast Fourier transform(FFT)and symplectic geometric mode decomposition(SGMD)are introduced,and the shortcomings of FFT-based harmonic detection method and SGMD-based harmonic detection method in application are expounded,which lays a theoretical and analytical foundation for the subsequent improvement of harmonic detection method.(2)Aiming at the requirements of harmonic detection methods for accuracy and rapidity,an improved harmonic detection method based on corrected FFT is proposed.Firstly,the sliding window function is used to truncate the current signal,and the frequency,amplitude and phase information of the fundamental component are obtained by correcting FFT.Then,based on the obtained three-element information,the fundamental component in the time-domain waveform is reconstructed and subtracted from the original signal to obtain all harmonic components except the fundamental component.Finally,the effectiveness and feasibility of the proposed method are verified by simulation and experiment.(3)Aiming at the problem of accurate detection of steady-state harmonics and transient harmonics,a harmonic detection method based on improved symplectic geometric mode decomposition(ISGMD)is proposed.Firstly,the trajectory matrix is improved to improve the endpoint effect.Then,the initial single-component signal is selected by solving the cosine similarity,and it is added to obtain the symplectic geometric component(i.e.,a certain harmonic).Then the component is subtracted from the original signal,and the same operation is performed on the remaining component until the power spectrum entropy difference between the last two remaining components is small enough,and the iteration is stopped to finally obtain each harmonic component.On this basis,the Hilbert transform and the transient harmonic disturbance time positioning method based on the instantaneous amplitude curve of the harmonic component are introduced to calculate the frequency,amplitude,phase and transient positioning information for each harmonic.Finally,the effectiveness and feasibility of the proposed method are verified by simulation and experiment.The thesis has 41 drawings,43 tables,and 91 references.