白兰地酒龄模型的构建

Predictive Modeling of Brandy Age

为预测陈酿白兰地酒龄，实验采用顶空固相微萃取（HS-SPME）结合气相色谱-质谱联用（GC-MS）技术分析陈酿0～6a的42个张裕白兰地样品，并且用偏最小二乘回归（PLS）研究挥发性成分含量（自变量）与酒龄（因变量）之间的关系。结果表明：用28种最佳挥发性化合物结合PLS建模可以比较精确的预测白兰地酒龄，使用由13种乙基酯和3种饱和醇组成的小组群挥发性化合物也可以较好的预测白兰地酒龄。

Headspace solid-phase microextraction （HS-SPME） and gas chromatography coupled with mass spectrometry（GC-MS） were used to analyze Changyu brandy aged for 0 to 6 years （a total of 42 samples）, and the relationship betweenthe content of volatile components （X, independent variables） and brandy age （Y, dependent variable） was explored usingpartial least square regression （PLS） for predictive modeling. The results showed that the PLS model could predict brandyage through the optimal 28 volatiles as well as with a smaller subset consisting of ethyl esters and saturated alcohols.