乙醇汽油含量的近红外光谱检测研究  

Detection of ethanol content in ethanol gasoline by near infrared spectroscopy

作　　者：欧阳爱国[1] 刘军[1] 王亚平[1] 

机构地区：[1]华东交通大学机电工程学院,江西南昌330013

出　　处：《激光与红外》2012年第8期901-904,共4页Laser & Infrared

摘　　要：随着石油资源的匮乏和能源需求量的日益增大,乙醇作为一种替代车用燃料越来越受到重视。在调和乙醇汽油的过程中由于其辛烷值的波动较大,影响车辆的动力性能,因此对乙醇汽油中的乙醇含量的准确检测具有重要意义。利用近红外光谱分析技术(NIR)定量分析乙醇汽油中的乙醇含量。实验结果表明,利用PCA在光谱1840～2030 nm范围能够准确分类乙醇汽油和成品汽油;采用PLS在光谱1400～2200 nm范围准确测量乙醇汽油中的乙醇含量,其交叉检验均方根误差(standard error of cross validation,SECV)为1.35(%,V/V),可以满足大部分乙醇汽油生产企业的实际检测需求。因此,采用近红外光谱技术和PCA,PLS分析方法可以有效的对乙醇汽油和成品汽油进行分类,对乙醇含量的定量分析模型也可以达到较高的精度。With the lack of petroleum resources and the increasing energy demand, ethanol gasoline as an alternative vehicle fuel gains more and more attention. As the octane value of ethanol gasoline fluctuates greatly and it affects the dynamic properties of vehicles, so it is important to detect ethanol content in ethanol gasoline accurately. Near infrared （NIR） spectroscopy has been applied to determine ethanol content of ethanol gasoline. The result shows that spectra range from 1840 to 2030 nm is selected to distinguish ethanol gasoline from normal gasoline based on Principal Component Analysis （PCA）. Quantitative analysis model of ethanol is established by Partial Least Squares （PLS） ethanol with spectra ranging from 1400 to 2200 nm. The analysis result shows a Standard Error of Cross Validation （SECV） of 1.35 （ % , V/V）. It is strongly proven that NIR spectroscopy can be used in determining ethanol content of ethanol gasoline and it can meet the practical testing demand of most ethanol gasoline production company very well. So the near infrared spectroscopy and PCA, PLS analysis method can effectively distinguish ethanol gasoline and normal gasoline, and the ethanol gasoline quantitative analysis model ethanol also can achieve high precision.

关 键 词：近红外光谱分析技术 乙醇汽油 定量分析 

分 类 号：TQ015.9[化学工程] TP391.9[自动化与计算机技术—计算机应用技术]