| 基于支持向量机的汽油族组成近红外光谱分析方法研究 |
| |
| 引用本文： | 史永刚,刘绍璞,宋世远,张洁,李子存. 基于支持向量机的汽油族组成近红外光谱分析方法研究[J]. 分析测试学报, 2007, 26(3): 343-346,351 |
| |
| 作者姓名： | 史永刚 刘绍璞 宋世远 张洁 李子存 |
| |
| 作者单位： | 西南大学,化学博士后流动站,重庆,400715;后勤工程学院,油料应用工程实验中心,重庆,400016;西南大学,化学博士后流动站,重庆,400715;后勤工程学院,基础部,重庆,400016 |
| |
| 摘 要： | 提出一种基于最小二乘支持向量机(LS-SVM)的汽油族组成近红外光谱分析方法。采用国家标准方法(GB 11132-1989液体石油产品烃类测定法——荧光指示剂吸附法)测定了重庆地区销售的汽油族组成,并采用主成分分析-最小二乘支持向量机建立汽油族组成的预测模型。预测模型对汽油中芳烃和烯烃含量的RMSEC分别为0.2090和0.2142。实验结果表明所建模型具有计算量小,预测准确、可靠,而且操作简单、维护费及测试费用低等特点。
|
| 关 键 词： | 主成分分析 最小二乘支持向量机 近红外光谱 汽油 族组成 |
| 文章编号： | 1004-4957（2007）03-0343-04 |
| 收稿时间： | 2006-04-10 |
| 修稿时间： | 2006-04-102006-07-27 |
Near Infrared Analysis of Major Classes of Hydrocarbon Constituents in Gasoline with Least Square-Support Vector Machine |
| |
| SHI Yong-gang,LIU Shao-pu,SONG Shi-yuan,ZHANG Jie,LI Zi-cun. Near Infrared Analysis of Major Classes of Hydrocarbon Constituents in Gasoline with Least Square-Support Vector Machine[J]. Journal of Instrumental Analysis, 2007, 26(3): 343-346,351 |
| |
| Authors: | SHI Yong-gang LIU Shao-pu SONG Shi-yuan ZHANG Jie LI Zi-cun |
| |
| Affiliation: | 1. Post-doctorial Research Station of Chemistry, Southwest Normal University, Chongqing 400715, China; 2. Experimental Center for Oil Application, Logistic Engineering College, Chongqing 400016, China; 3. Department of Education, Logistic Engineering College, Chongqing 400016, China |
| |
| Abstract: | A near infrared analysis of major classes of hydrocarbon constituents in gasoline with least square-support vector machine(LS-SVM) was developed.The major classes of hydrocarbon constituents from different gas stations in Chongqing were determined by GB11132-1989 method(i.e.standard test method for hydrocarbon types in liquid petroleum products by fluorescent indicator adsorption).A prediction model was then established using the Principal Component Analysis(PCA)-LS-SVM.With the prediction model,the RMSEC(root mean square error of calibration) for aromatics and olefins in gasolines were 0.209 0 and 0.214 2,respectively.Experimental results showed that the developed model was easy to operate,reliable and less analysis cost. |
| |
| Keywords: | PCA LS-SVM Near infrared analysis Gasoline Major classes of hydrocarbon constituent |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
