Abstract: Tobacco quality evaluation has been a research spot through the indicator of tobacco chemical components. A large number of statistical methods have been widely used in the evaluation of tobacco quality. Grey incidence analysis is a multiple factors method, and the closeness of relationship of the multiple factors which is judged based on the similarity level of the geometric patterns of sequence curves. Therefore, the slope ratio of reference sequences and relevant sequences on every corresponding interval is used to reflect the closeness of the sequences. Grey correlation analysis is used as a novel factor analysis method which has a broad application prospect in digging the information of tobacco chemical components. Focused on tobacco chemical components, this paper summarized the current situation of tobacco quality evaluation, sorted out the application achievements of grey relation analysis in tobacco chemical components evaluation, compared several different grey relative degree models and their properties, and estimated the superiority of grey incidence model in digging the information from tobacco chemical components.

Key words: grey incidence analysis, tobacco leaf, chemical components, quality evaluation

摘要： 利用烟叶化学成分评价烟叶质量一直是烟草研究的热点，大量统计学方法已广泛应用于评判烟叶等级。灰色关联分析作为一种新颖的因素分析方法，通过对序列几何关系的度量来比较系统中多因素关系的紧密程度，其紧密程度可由序列曲线间几何形状的接近程度来度量，灰色关联分析在挖掘烟叶化学组分信息中具有广阔的应用前景。以烟叶化学成分作为主线，综述了烟叶品质评价的现况，并对灰色关联分析模型在烟叶化学成分评价上的应用进行了系统分析，同时比较了几种经典灰色关联度模型的算法和性质，以期为卷烟工业构建烟叶质量评价方法体系提供参考。

关键词: 灰色关联分析, 烟叶, 化学成分, 质量评价