1. National Key Lab of Pulp & Paper-Making Engineering，South China University of Technology, Guangzhou 510640, China 2. College of Chemical Engineering, South China University of Technology, Guangzhou 510640, China 3. Department of Biological Chemical Engineering，Guangxi University of Technology, Liuzhou 545006, China
Abstract：An analysis method for separating chromatographic overlapped peaks and purifying infrared spectra is put forward, based on the blind source separation technique and the multi-dimensional data of GC-FTIR, Using various information from hyphenated instruments, this method was used to separate completely a organic mixture, the xylene isomerism system, a problem unable to solve usually. The method can confirm the rationality of theory and algorithm and give integral explanations of the independent component analysis data. The reason for the error in quantitative analysis is discussed.
[1] HU Yun, LIANG Yi-zeng, LI Bo-yan(胡 芸, 梁逸曾, 李博岩). Acta Chimica Sinica(化学学报), 2003，61(9): 1466. [2] Praisler M J, Bocxlaer Van, De Leenheer A, et al. Journal of Chromatography A，2002, 962: 161. [3] LI Yan, WANG Jun-de, CHEN Zuo-ru, et al(李 燕，王俊德，陈作如, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2002, 22(5): 758. [4] Hyvarinen A, Karhunen J, Oja E. Independent Component Analysis. Wiley and Sons Press. 2001. [5] YANG Xing-jun, ZHENG Jun-li(杨行峻, 郑君里). Artificial Neural Network and Blind Siginal Processing(人工神经网络与盲信号处理). Beijing: Tsinghua University Press(北京: 清华大学出版社), 2003. [6] YANG Zhu-qing, LI Yong, HU De-wen(杨竹青, 李 勇, 胡德文). Acta Automatica Sinica(自动化学报)，2002，28: 762.
