清香型白酒酒龄鉴别的方法研究

Mathematical Modeling for Identification of Fen-Flavor Liquor Aging Time

以清香型白酒的典型代表汾酒为研究对象,利用气相色谱-质谱法定性、气相色谱三内标定量法,对新产和贮存0.5～30年的65°白酒中31种微量成分随酒龄的变化行为进行系统研究,并对其pH值、电导率等进行分析测定。在此基础上,运用多元线性回归和偏最小二乘技术,建立一个白酒组成与其酒龄的相关模型,同时采用内部及外部双重验证的办法对所建模型稳定性进行分析和验证。多元线性分析结果显示:建模相关系数为0.9990,交互检验复相关系数为0.9980,外部检验相关系数为0.9984;偏最小二乘回归结果显示:建模相关系数为0.9949,交互检验复相关系数为0.9840。所建立的数学模型具有良好的稳定性和预测能力,可用于清香型白酒的酒龄预测。

Fenjiu, which was a reprehensive of fragrant liquor, was taken as the research object. The pH values, electrical conductivities and the contents of flavor components analyzed by GC-MS were systematically analyzed with the age of liquor. Based on this, with the help of multiple linear regression （MLR） and partial least square regression （PLSR）, a model of liquor property and its age was established. The estimation stability and generalization ability of the models were strictly analyzed by both internal and external validation. The correlation coefficient R of established models, leave-one-out （LOt） cross-validation （CV）, Qext of MLR were 0.9990, 0.9980, 0.9984; the result of PLSR was the following： R and Rt,o were 0.9949, 0.9840. The results showed that the model is of great robustness and good predictive abilities. The method may play an important role in identifying the Fen-flavor liquor age in the theoretical and application.