期刊名称：International Journal of Grid and Distributed Computing
印刷版ISSN：2005-4262
出版年度：2015
卷号：8
期号：4
页码：317-326
DOI：10.14257/ijgdc.2015.8.4.31
出版社：SERSC
摘要：In the detection of drugs in the field, because the inclusion complex environment exists, the obtained signal often has a poor signal to noise ratio, and the intrinsic characteristics of the target material is often drowned signal formed in the package material in the background. According to the obtained experimental data, proposed one kind based on the principal component analysis and support vector machine algorithm of gas chromatography identification sensor signal processing and recognition; the method used for detection and identification of the air in the precursor gases combine tester self-developed, obtained very good result. This paper designed and developed a chromatographic separation and sensor based on the combination of gas detection instruments, to multi gas detection instrument. On separation characteristics using chromatography, to solve the traditional single common precursor gas detection. The use of a preprocessing based on domestication, principal component analysis for feature extraction method of all kinds of gas data. This effectively avoids the sensor substrate fluctuation and gas concentration effects on body recognition, and reduces the gas sample feature vector dimension.
关键词：Gas detection; Chromatogram sensor; PCA; Support vector machine
