电子鼻在六堡茶陈化年份识别中的应用

[目的]探索电子鼻(EN)技术在六堡茶陈化年份识别上的应用,为建立六堡茶品质评价新技术体系提供参考依据.[方法]采用PEN3型便携式电子鼻气味分析系统提取不同厂家、不同陈化年份六堡茶茶干、茶汤和茶底的响应特征值,用主成分分析法(PCA)和线性判别分析法(LDA)分析其陈化年份和预测模型,以马氏距离(MA)、欧氏距离(EU)和判别函数分析(DFA)法验证预测模型的准确性.[结果]不同陈化年份六堡茶对EN产生不同的响应信号,分别选取171~175、166~170和155~160 s的EN响应值作为建立干茶、 茶汤和茶底识别模型的特征值;LDA较PCA能更好地区分不同陈化年份的六堡茶,不同陈化年份样品分布在不同区域,干茶、茶汤和茶底的判别式LD1和LD2的累计贡献率分别为93.00%、88.31%和80.52%;EU、MA和DFA法对干茶识别模型的验证准确率分别达88.23%、88.23%和94.12%,对茶汤识别模型的验证准确率均为82.35%,对茶底识别模型的验证准确率分别为70.58%、82.35%和94.12%.[结论]EN可识别不同陈化年份六堡茶的芳香物质等特征气味并对其进行区分和归类,可在茶叶陈化年份识别中推广应用.

Application of electronic nose in identification for aging years of Liupao tea

[Objective]The application of electronic nose(EN) in identification for aging years of Liupao tea was re-searched to provide reference for establishing new quality evaluation technology system of Liupao tea. [Method]Response characteristic values of dry tea, tea infusion and wet tea leaf at different aging years from various producers were obtained by PEN3 portable EN smell analysis system. Principle component analysis(PCA) and linear discriminant analysis(LDA) were used to analyze their aging years and prediction model, and mahalanobis distance(MA),euclidean metric(EU) and discriminant function analysis(DFA) were applied in verification of accuracy of the prediction model. [Result]Liupao tea of different aging years had different response signals on EN. The EN response values of 171-175,166-170 and 155-160 s were selected as the characteristic values for dry tea,tea infusion and wet tea leaf to establish the identification model. LDA was better than PCA in distinguishing Liupao tea of different aging years. Samples of different aging years distributed in di-fferent regions. The cumulative contribution rates of discriminant LD1 and LD2 for dry tea,tea infusion and wet tea leaf were 93.00%,88.31% and 80.52% respectively. The verification accuracy of identification model for dry tea with EU,MA and DFA reached 88.23%,88.23% and 94.12%,identification model for tea infusion all reached 82.35% and for wet tea leaf reached 70.58%,82.35% and 94.12%. [Conclusion]EN can identify the smell characteristics such as aromatic substance in Liupao tea of different aging years, and realize distinguishing and classification. It can be used and promoted in the identification of tea aging years.