摘要
以不同等级的浓香型白酒为研究对象,利用气相色谱-质谱联用(GC-MS)技术建立浓香型白酒微量成分的指纹图谱,采用稀疏主成分分析(SPCA)提取GC-MS图谱前7个稀疏主成分构建白酒成分特性的综合指标,进而采用回归分析建立浓香型白酒质量评价的客观模型。验证实验结果表明,建立的白酒质量评价模型与感官评价的评分绝对误差<3,在实现特级、优级、一级、二级四个等级的评价中,正确率达到94%。研究表明,不同等级的白酒的GC-MS图谱的稀疏主成分存在较明显差异,该研究建立的白酒质量评价模型能有效地实现白酒等级的评判,为白酒质量控制及等级鉴定提供了一种客观方法。
Using the strong-flavor Baijiu(Chinese liquor) with different quality levels as research objects, the fingerprint of trace components in strong-flavor Baijiu was established by GC-MS, the comprehensive indexes of Baijiu component characteristics were established by extracting first seven sparse principal components with sparse principal component analysis(SPCA), and the objective model of quality evaluation of strong-flavor Baijiu was established by regression analysis. The results of verification test showed that the absolute error between the established Baijiu quality evaluation model and sensory evaluation was less than 3, and the accuracy rate reached 94% in the realization of evaluation of super grade, excellent grade, first grade and second grade. The research showed that there were significant differences in the sparse principal components of GC-MS spectrums of different grades of Baijiu. The Baijiu quality evaluation model established in the study could effectively realize the evaluation of Baijiu grade and provide an objective method for quality control and grade identification of Baijiu.
作者
陈明举
周迪
王鸿
熊兴中
CHEN Mingju;ZHOU Di;WANG Hong;XIONG Xingzhong(Artificial Intelligence Key Laboratory of Sichuan Province,Sichuan University of Science&Engineering,Yibin 644000,China;School of Automation and Information Engineering,Sichuan University of Science&Engineering,Yibin 644000,China)
出处
《中国酿造》
CAS
北大核心
2021年第10期102-107,共6页
China Brewing
基金
五粮液集团-四川轻化工大学产学研合作项目(CXY2020ZR006,HX2020034)
四川省科技厅项目(2021YFS0339)
四川轻化工大学研究生创新基金(y2021079)。
