Research On IMS-TOFMS Adaptive Denoising And Spectral Peak Identification Algorithm And Implementation Of Analysis Software

The speedy progress of industrialization in China has made the issue of air pollution increasingly serious,among which volatile organic compounds(VOCs)have been a major cause of contamination,rapid and accurate online detection of VOCs has profound significance for environmental management.Time of Flight Mass Spectromemert(TOF-MS)can obtain the ion-to-mass ratio for rapid detection of the sample to be measured,but it cannot separate the isomers effectively.Ion Mobility Spectrometry(IMS)can separate isomers by obtaining collisional cross-sectional structures,therefore,the combination of the two instruments can fully utilize their advantages and provide important tools for the characterization of isomers and analysis of complex mixtures.In this thesis,the detection requirements of Ion Mobility Spectrometry-Time of Flight Mass Spectromemert(IMS-TOFMS)coupler are used as a background to investigate the spectrum denoising,peak identification algorithm and software development in IMS-TOFMS analysis software.The major components of this thesis are as follows:(1)For the problems that the universal threshold in wavelet threshold denoising algorithm cannot achieve adaptive matching,the soft and hard threshold functions have signal oscillation or distortion,this thesis proposes a wavelet adaptive denoising algorithm which constructs an adaptive threshold and a smoothing threshold function.Adaptive threshold takes into account the influence of the layers of factorization on the threshold level,and adaptively corrects the threshold value by the noise content of each layer;the smoothing threshold function integrates the advantages of the soft and hard threshold functions,overcomes their problems,and has good continuity and smoothness.The experimental outcomes suggest that the wavelet adaptive denoising algorithm not only has good denoising performance,but also can retain the original features of the signal well.(2)The spectral data preprocessing and spectral peak identification algorithms are investigated,and wavelet adaptive denoising algorithm,nonlinear iterative peak stripping algorithm,and maximum intensity normalization,which are most applicable to IMS-TOFMS spectrograms,are selected as preprocessing algorithms.In addition,through the analysis and comparison of commonly used peaking algorithms,this thesis proposes a feature fusion peaking algorithm based on second derivative method,symmetric zero area method and continuous wavelet peaking method.The experimental outcomes indicate that the algorithm of this thesis can increase the accuracy of peak search effectively and can be used as the peak identification algorithm for IMS-TOFMS spectra.(3)The IMS-TOFMS analysis software was designed and developed through preliminary research and analysis of software requirements.The analysis software includes data acquisition,spectrogram display,data memory,data processing and other capabilities.Through system testing and practical application,it was verified that the analysis software meets the application requirements of the coupled instrument,has good stability and scalability,and has been started to be used in the development project of IMS-TOFMS instrument in the group.