Font Size: a A A

Research On The Time-frequency Analysis Instrument Based On VMIDS Developing System

Posted on:2013-09-16Degree:MasterType:Thesis
Country:ChinaCandidate:J T LiFull Text:PDF
GTID:2232330362974391Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
Time frequency analysis is widely used in automatic control, pattern recognition,acoustic, mechanical vibration, fault diagnosis, seismic surveys and BiomedicalEngineering. The main methods of Time-frequency analyzing and processing non-stationary signals include the short-time Fourier transform, the Wigner-Villedistribution and the wavelet analysis and so on. Hilbert-Huang Transform is a newnonlinear and non-stationary signal analysis method,the method consists of twosuccessive parts: the empirical mode decomposition (EMD) and the Hilbert spectralanalysis.For multi-component non-stationary signal analysis, there is a serious cross-terminterference in Wigner-Ville distribution (WVD).The Gabor Wigner transform (GWT)avoid cross-term interference of the Wigner-Ville distribution and with good frequencyconcentration. However, due to poor time-frequency clusters of Gabor transform, itwill have frequency aliasing when dealing with multi-component signal, that GWT cannot get the desired results. Therefore, we propose an improved algorithm ofSTFT-Wigner (SWT), can effectively inhibit the cross-term interference, and maintainhigh frequency concentration.In this paper, the work is summarized as follows:①We proposed and improved Gabor-Wigner transform, STFT-Wigner transform,Hilbert-Huang Transform algorithm, Study of existing algorithms including short-timeFourier transform, Wigner-Ville distribution, adaptive transform, Gabor transform,continuous wavelet transform, analysis wavelet transform,Choi-Williams distribution,and Cone-shaped distribution. We created VI in LabVIEW to achieve these algorithms.②In-depth study the principle of VMIDS4.0system, and interface definitionbetween VMIDS4.0system with LabVIEW. Match the data types between the twosystems and design the corresponding interface function. we added STFT,WVD, ADT,GT, CWT, AWT, CWD, CSD, GWT, SWT and HHT functions in VMIDS4.0virtualinstrument development system.③Comparative analysis of the advantages and disadvantages of the three virtualinstrument software: LabVIEW, Agilent VEE and Visual Studio6.0.And We adoptionLabVIEW to generate the desired dynamic link library of time-frequency analysiseventual. We have described in detail of the process of VIMDS4.0system call dynamic link library.④Study of input and output data preprocessing methods in LabVIEW, such astwo-dimensional array (the time-frequency amplitude) converted to the desiredone-dimensional array, one-dimensional array (kernel function) to convert the requiredtwo-dimensional array.⑤Debug the new time-frequency analysis functions in VMIDS4.0virtualinstrument development system and design a multi-functional three-dimensionaltime-frequency spectrum instrument. Using the instrument analysis simulated andpractical signals achieved good results shows that the time-frequency analysisinstrument is correctness and usability, the experimental results can be seen on Gabor-Wigner transform and STFT-Wigner transform than the existing time-frequencyanalysis method have a higher resolution, and no cross terms.
Keywords/Search Tags:Qin’s model, time-frequency analysis, time-frequency analysis instrumen, dynamic link library, time-frequency spectrumt
PDF Full Text Request
Related items