基于味觉分析系统鉴别干邑白兰地真假的 应用初探

Application of Cognac brandy identification based on taste analysis system

中文关键词:  味觉分析系统(电子舌)  干邑白兰地  多元统计分析  真假鉴别

英文关键词:taste analysis system (electronic tongue)  Cognac brandy  multivariate statistical analysis  identification of true or false

中文摘要:

      目的 运用味觉分析系统分析评价干邑白兰地滋味指标, 结合多元统计方法, 研究该系统在鉴别真假干邑白兰地方面的应用。方法 选择具有市场流通代表性的轩尼诗XO干邑白兰地作为研究样品, 将样品用水按1:2(V:V)稀释后, 采用味觉分析系统SA402B进行检测, 运用SPSS 22.0, 对酸味、苦味、涩味、苦回味、涩回味、丰富性、咸味和鲜味8种味觉信号进行主成分分析(principal component analysis, PCA)和判别分析, 并模拟识别未知样品, 考察方法对干邑白兰地真假鉴别的应用效果。结果 PCA图显示真假样品能够明显区分, 主成分1(PCA1)贡献率为68.187%, 主成分2(PCA2)贡献率为26.189%, 两大主成分累积贡献率达到94.377%, 能够较全面反应样品滋味特征; 依据判别分析模拟识别6批次未知样品, 所有未知样品被正确识别。结论 利用味觉分析系统能够简单方便分析轩尼诗XO干邑白兰地滋味信息, 结合多元统计分析样品味觉指标, 能够较好区分真假样品, 可以为干邑白兰地的真假鉴别提供一种新途径。

英文摘要:

      Objective To evaluate the taste index of Cognac by using the taste analysis system, and to study the application of the system in the identification of true and false Cognac by combining the multivariate statistical method. Methods Hennessy XO. Cognac samples with the representative of the market circulation were chose as the research samples, The samples were diluted with water 1:2 (V:V) and detected by the taste analysis system SA402B. Principal component analysis (PCA) and discriminant analysis were performed on 8 kinds of taste signals of sourness, bitterness, astringency, bitterness, aftertaste, richness, salty taste and umami taste by using SPSS 22.0, and simulated to identify the unknown samples, inspection method of Cognac and application effect of true and false identification. Results The PCA diagram showed that the true and false samples could be clearly distinguished. The contribution rates of PCA1 and PCA2 were 68.187% and 26.189%, respectively. Totally 6 batches of unknown samples were identified by discriminant analysis simulation, and all unknown samples were correctly identified. Conclusion The use of the taste analysis system can be simple and convenient to analyze Hennessy XO Cognac taste information, and can better distinguish the true and false samples combining with the multivariate statistical analysis of sample taste index, which can provide a new way for the true and false Cognac identification.