| Along with the development of chromatographic techniques, the resolving power of the chromatographic instruments has been improved, which are used broadly in the analysis field. In order to improve the quality of the tobacco, it is required to know the species and contents of the chemical compounds in the tobacco. As the main method in chemical analysis, chromatographic analysis takes important role in the tobacco chemistry. However, it is intractable problem to resolve the abundant data. So, it is a necessary to design a system to resolve the chromatographic signals automatically.First, the main methods about the separation and analysis in the tobacco science were reviewed. Based on this, a new method was proposed to determine the component number in the overlapping chromatographic peaks rapidly. Meanwhile, we achieved the expedient realization of the EMG model. With some experience, our target was moved to the two basic problems in the resolution of chromatographic signals: baseline drifting and peak overlapping.According to the property of the moving average method, multiple-pass moving average based method was proposed to deal with the baseline drift. It was proved that this method was effective. In order to determine the window size of the multiple-pass moving average method, decay curve was used, which made the algorithm to eliminate the baseline automatically.The realization of baseline elimination made it possible to resolve the complicated chromatographic signals automatically. Area reproduction method was introduced in combination with peak recognition algorithm based on high-order derivatives to automate the chromatograms division, peak recognition and rapid resolution. It was proved that the method was a useful tool by applying it to the model and smoke chromatographic signals.Furthermore, it is worthy to note that the method proposed in this paper was also capable to other signals, for example: CE, NMR, XRD, et al.
|
PDF Full Text Request