Time Frequency Analysis Method Programming And Application Based On Visual C++

At present, time frequency software based on Matlab have many problems in the processing oflarge data capacity, such as slow calculation, low efficiency of implementation, interface delay issues,more suitable algorithm research and theory verification, it is difficult to meet the actual demand forengineering application. In order to adapt to fast analysis of large amounts of data and parameteridentification, on the basis of studing real-time data processing and display key technologies, thispaper uses multithreading and double buffering advanced programming techniques to develop thetime frequency software based on Visual C++. The software implements wavelet time frequencyfiltering, wavelet modal parameter identification and main frequency analysis function, with fastcalculation, high efficiency of inmplemetation, simple and reliable performance. Each functionalmodule is effectiveness and practicality which is validated through actual data, to provide a powerfultoll for time frequency analysis. The main contents of this paper are as follows:（1）Analysis and comparision of the existing common numerical calculation library. In order tosolve the problem of lost response to software interface for a long time calculation, multi-threadingtechnology is adopted. The method improves the interactivity and friendlinesss.（2）A practical fast graphics display control CPlotChartControl is developed using doublebuffering technology based on Visual C++, the control is able to plot a large amount of data rapidlyand display curve graph, tow dimensional time frequency diagram and scatter diagram, with graphiccopy, coordinate selection and other functions.（3）According to the software development process, firstly, this paper describes softwarerequirements analysis, to determine the software development goals and performance factors.Secondly, determine interface design and functional modules division for overall structural design,and then design the various functional modules, eventurally forming a correct and efficient, goodhuman-computer interaction capability analysis software. Finally, applied to testing data, the resultsshow that various functional modules are valid and effective.（4）Applied to big noise data filtering, modal parameter identification of semi-circular archstructure model and analysis of bearing conditions, the results verifies that the software is practicality.