| 基于不同波段对成品汽油的模式识别分析 |
| |
| 引用本文： | 姜黎,张军,陈哲,余谦,王京华.基于不同波段对成品汽油的模式识别分析[J].光谱实验室,2010,27(3):1208-1212. |
| |
| 作者姓名： | 姜黎 张军 陈哲 余谦 王京华 |
| |
| 作者单位： | 1. 广东省高等学校光电信息与传感技术重点实验室(暨南大学),广州市,510632;暨南大学理工学院光电工程系,广州市黄埔大道西601号,510632
2. 中国石化总公司广州分公司,广州市,510726 |
| |
| 摘 要： | 为了实现对输油管道上顺序输送不同牌号汽油(90#,93#,97#)时汽油间界面的区分,根据近红外光谱分析的基本原理,采用主成分分析-马氏距离(PCA-MD)模式识别方法,分析了700—1100nm波段和1100—1700nm波段3种不同牌号成品汽油的近红外光谱,并根据分析结果对汽油进行了分类。结果表明,使用1100—1700nm波段的汽油光谱分类结果较好,该方法可进一步鉴别汽油的质量。
|
| 关 键 词： | 输油管道界面检测 主成分分析-马氏距离 模式识别 成品汽油 |
Study on Pattern Recognition Analysis Method for Gasoline Based on Different Spectral Bands |
| |
| JIANG Li,ZHANG Jun,CHEN Zhe,YU Qian,WANG Jing-Hua.Study on Pattern Recognition Analysis Method for Gasoline Based on Different Spectral Bands[J].Chinese Journal of Spectroscopy Laboratory,2010,27(3):1208-1212. |
| |
| Authors: | JIANG Li ZHANG Jun CHEN Zhe YU Qian WANG Jing-Hua |
| |
| Institution: | ca[Key Laboratory of Optoelectronic Information and Sensing Technologies of Guangdong Higher Educational Institutes,(Jinan University),Guangzhou 510632,P.R.China]b(Department of Optoelectronic Engineering,Jinan University,Guangzhou 510632,P.R.China)c(China Petrochemical Corporation,Guangzhou Branch,Guangzhou 510726,P.R.China) |
| |
| Abstract: | The near infrared spectroscopy of gasoline has been analyzed in order to research the interfaces of gasoline （90#,93#,97#） in pipeline.The principal component analysis-Mahalanobis Distance（PCA-MD） pattern recognition method is used according to the basic principles of near-infrared spectroscopy.The calibration models of three kinds of gasoline have been built on 700—1100nm and 1100—1700nm spectra respectively.The results show that the classification ability of model on 1100—1700nm range is better than the model on 700—1100nm spectrum.The calibration method can further identify the quality of gasoline. |
| |
| Keywords: | Pipe Laying Interface Detection PCA-MD Pattern Recognition Gasoline |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
