化学成分结合统计分析鉴别不同产地绿茶的研究

Identification of green tea origins based on chemical components with statistical analysis

绿茶是我国六大茶类之一,产地分布广,品质差异性大。目前,对茶叶产地的鉴别主要依赖感官审评,缺乏量化的评价指标,评价结果存在着不确定性。本研究对6个省份绿茶主要化学成分进行主成分分析,提取主要成分因子,应用贝叶斯(Bayes)判别结合聚类分析对不同产地绿茶进行鉴别。结果表明,Bayes判别对6个地区27个样品能达100%的正确判别,同时聚类分析结果与原始样品基本相同,为绿茶产地的鉴别提供具体量化模型。

Green tea, one of the six major teas in China, has a wide distribution of origins and a distinct dif-ference in tea quality. Currently, differentiation of green teas from different geographical origins mainly dependson the sensory evaluation. Lack of the quantified evaluation criteria often result in an uncertain result. By analyz-ing the main chemical components of green teas from six provinces, we used the Bayesian discriminant analysistogether with the cluster analysis method to differentiate green teas from six different geographical origins. Re-suits suggested that the Bayesian discriminant analysis can be used to discriminate 27 samples from six regionsperfectly and the cluster analysis also showed the same result. Thus, the Bayesian discriminant analysis can pro-vide a specific quantitative model for differentiating green teas originated from different geographical regions.