Detection Method Research Of Harmonic And Inter-harmonic In Power System Based On MUSIC

In today's society more and more non-linear power equipment widely used to bring civilization and development of the community also has a greater impact on the performance of the power system.How to test and evaluate the merits and demerits of power system performance is widely concerned by both provider or users and researchers.Harmonic/inter-harmonics due to the wide application of non-linear,impact and volatility loads in the grid can cause severe distortion of the voltage and current wave forms,resulting in malfunction of the relay protection device,equipment failure caused by it,and safety accidents occur frequently,harmonic/inter-harmonic serious threat to the safe and stable operation of power systems.Therefore,it is very important and urgent to detect the harmonic and inter-harmonics in the power grid,and the key is to detect and accurately measure the parameters of the harmonics/inter-harmonic components of the power grid in real time.When the modern spectral estimation method is applied,time domain sampling signal for high frequency resolution detection,even if the frequency difference between the two component signals also do not affect each other,avoiding the interference caused by the DFT method spectrum leakage and fence effect.In this paper,we mainly study the application of multiple signal classification in harmonic and inter-harmonics analysis and detection.There are several main research contents:The MUSIC method of modern spatial spectrum estimation is discussed and analyzed.The classical MUSIC algorithm is applied to the harmonic and inter-harmonics detection process of power system.The simulation results show that the method has high frequency resolution and is not affected by leakage and fence effect,but the signal to noise ratio requirements are high,when the increase in system noise the algorithm performance degradation.In order to solve this problem,the classical MUSIC algorithm is improved,that is,the bidirectional sequence smoothing MUSIC algorithm,which combines the sampling sequence multiple smoothing before the noise subspace,and carries on the forward and backward sampling sequence integration respectively.That is,the source sampling sequence is decomposed into several small sub-sampling sequences,and the corresponding autocorrelation functions are obtained from these sub-sequences,and the auto-correlation matrix of each sub-sequence is obtained.The weighted average of the correlation data is the bidirectional sequence smoothing method,which combines the advantages of forward and backward sequence smoothing,which ensures the relationship between the sampled data of the signal is more robust and improves the anti-noise performance.The classical MUSIC method is used to know that the relationship between the obtained noise subspace and the signal subspace is the key problem that affects the parameter measurement.Therefore,the MUSIC algorithm is weighted by the minimum modulus method.The main work is reflected in the auto-correlation Matrix eigenvalue decomposition after the noise subspace minimum mode of linear transformation combination,based on the orthogonal relationship between the noise subspace and the signal vector,the harmonic and inter-harmonics detection method of the power system based on the minimum modulus(MNM)MUSIC algorithm is obtained,and the spectral estimation of the signal is obtained.The extended Prony method is used to estimate the signal correlation parameters and improve the detection information.The simulation results show that the method has high detection accuracy and higher anti-noise.This paper designs a harmonic detection and measurement system with LabVIEW2015 as the development platform.First of all,the LabVIEW development environment,including its characteristics and advantages,LabVIEW has a mixed programming interface with MATLAB software,LabVIEW bounded with MathScript window and MatlabScript node application and features were introduced.The MatlabScript nodes are used in the functional modules and programming of the system.The MUSIC algorithm and MNMMUSIC algorithm are used to detect the number of signal sources in different noise cases,and all kinds of information parameters are displayed in the form of graphical interface.Finally,the percentage of the effective ratio of the number of tests is calculated,and the accuracy of the two algorithms is analyzed in the form of table and graph.