基于PCA-MD分类法的云烟系列卷烟风格表征及品质维护

Application of PCA-MD Modeling in Cigarette Style Characterizing and Quality Maintenance of Yunyan Brand Series

摘要: 基于FT-NIR(Fourier transform near infrared spectrometry)分析技术,结合主成分分析-马氏距离(Principal component analysis-Mahalanobis distance,PCA-MD)分类法,对云烟系列卷烟配方烟丝的多变量评价进行了探讨。结果表明,通过建立PCA-MD分类模型,采用马氏距离这一统计量能够表征卷烟风格及品质的整体稳定性与差异性,方法简单,操作容易。

Abstract: Basing on Fourier transform near infrared spectrometry(FT-NIR) and principal component analysis-Mahalanobis distance(PCA-MD) grouping,the multivariate evaluation for cut filler of Yunyan brand series was discussed.The results showed that the consistence and diversity of cigarette style and quality could be characterized as a whole by the statistics of Mahalanobis distance through the establishment of PCA-MD model;and this method was simple and easy of operation.