The Research Of Spectrum Estimation For The Electircal Harmonic And Interharmonic Based On MUSIC Algorithm

With the development of economy and the improvement of people’s livingstandard，the stable operation of the power system and its power quality level directlyinfluences the development of the national economy, especially when lots of powerelectronic devices and nonlinear load used in electric power network, suppression ofelectric harmonic wave is facing unprecedented challenge. Harmonic analysis is thefocus of power quality analysis for a long time, not only the hamonic in integer timesbut also iner-harmonics,which is hamonic in noninteger times, is existed in the powergrid. In The past few years, harmonic analysis has got good development because ofthe joint efforts of the domestic and foreign experts and scholars, and the introductionof various algorithm infect new technology for Harmonic analysis. Harmonicanalysis is mainly based on detection of frequency,phase,amplitude and power ofelectrical signal, and analysis their curve with time variation. Therefore, accurateharmonic parameter estimation is the premise for suppression and manage ofharmonic and iner-harmonics.The common parsing algorithm of harmonic and iner-harmonics parameterestimation mainly include Fourier transform, Wavelet transform, Neural NetworkAnalysis,and Spatial spectrum. They all have advantages and disadvantagesthemselves,therefore,it is very important to find appropriate parsing algorithm.This article discussed harmonic and iner-harmonics parameter estimation from severalaspect ,such as Fourier transform, Wavelet transform,an modern spectrum. Themainly used analysis method for grid harmonic is based on disperse Fourier transform(DFT). DFT is limited operation, when it is cutting off infinite signal it will formspectral leakage. To Improve the resolution of the analysis, Add window interpolationalgorithm,which reduces leakage on the principle of increasing sampling point,isquoted in the algorithm,however, the spectral leakage problem is not solvedfundamentally.Along with the development of the theory of wavelet, harmonic analysis methodbased on it appeared. Although the wavelet theory has its own advantages, it has manyshortages when applied in harmonic analysis, for example ,it will produce hugeComputational Load in signal decomposition and reconfiguration,and the speed is not high,as well as resolution ratio of phase detection, and spectrum leakage is stillexisted.This article is based on applying super-resolution spectrum estimation theory ofsubspase method in power harmonic and iner-harmonics spectrum analysis.MUSICmethod is used for harmonic analysis in this article, which conductEigendecomposition of a matrix to the covariance matrix of power harmonic signal,and then get subspace of signal or noise, whose orthogonality character are used toconfirm signal direction. Simulated Harmonic signal into array signal, MUSICmethod is using for solving spectral characteristic of hamonic. This method need notto consider leakage effect, and has a high resolution ratio, chapter four of this articleused simulation software to analysis its superiority to Fourier transform, wavelettransform.