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。
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.
出处
《光谱实验室》
CAS
CSCD
北大核心
2010年第3期1208-1212,共5页
Chinese Journal of Spectroscopy Laboratory
基金
国家863项目(2009AA04Z315)
