摘要
目的建立固态法五粮浓香型原酒、成品酒与掺假白酒的分类模型。方法采用电感耦合等离子体质谱(inductively coupled plasma mass spectrometry,ICP-MS)技术进行鉴别分析,分别对样品中Mg24、Mg25、Si28、Fe56、Fe57、Ni58、Ni60、Zn64、Zn66、Cu63、Cu65、As75、Pb206、Pb207、Pb208等元素的响应强度进行检测,并结合聚类分析、主成分分析、因子分析和线性判别分析等方法进行数据分析。结果采用聚类分析时,当样本的产酒年份差距为2年内时,能够对原酒和掺假白酒进行有效分类,当样本的产酒年份差距为10年时,则聚类分析方法并不适用;采用因子分析时,当因子数限定为3时,掺假白酒在第1个因子和第3个因子中的得分均是最低的,在第2个因子得分中,掺假白酒总体上高于原酒组,低于成品酒组;采用线性判别分析时,总共17个样本中,88.2%个交叉验证已分组观察值可以正确分类。结论建立了固态法五粮浓香型原酒、成品酒与掺假白酒的分类模型,具有一定的应用价值。
Objective To establish the classification models of Yibin-flavor original baijiu by solid fermentation,bottled baijiu and adulterate baijiu.Methods Used ICP-MS technology for analysis,detected the response intensity of Mg24,Mg25,Si28,Fe56,Fe57,Ni58,Ni60,Zn64,Zn66,Cu63,Cu65,As75,Pb206,Pb207,Pb208 and other elements in the sample,and combined with data analysis methods such as cluster analysis,principal component analysis,factor analysis and linear discriminant analysis to analyze the data.Results When cluster analysis was used,the original baijiu and adulterate baijiu could be effectively classified when the difference in the baijiu-making years of the sample was within 2 years.When the difference in the baijiu-making years of the sample was 10 years,the cluster analysis method was not applicable.When factor analysis was used,and the number of factors was limited to 3,the score of adulterate baijiu in the first factor and the third factor was the lowest,in the second factor score,adulterate baijiu was generally higher than the original baijiu group,lower than the bottled baijiu group.When using linear discriminant analysis,in a total of 17 samples,88.2%of cross-validated grouped observations could be correctly classified.Conclusion The classification models of Yibin-flavor original baijiu by solid fermentation,bottled baijiu and adulterate baijiu was successfully established,which has certain application value.
作者
程铁辕
夏于林
何开蓉
张莹
陈闯
CHENG Tie-Yuan;XIA Yu-Lin;HE Kai-Rong;ZHANG Ying;CHEN Chuang(The State Key Laboratory of Liquor Products Test,Chengdu Customs Technology Center Yibin Branch,Yibin 644000,China;Food Safety Detection Key Laboratory of Sichuan Province,Chengdu Customs Technology Center,Chengdu 610000,China)
出处
《食品安全质量检测学报》
CAS
2020年第13期4473-4481,共9页
Journal of Food Safety and Quality
基金
四川省重点研发计划(重大科技专项)项目(2018SZ0360)。
关键词
电感耦合等离子体质谱法
白酒
聚类分析
主成分分析
因子分析
线性判别分析
inductively coupled plasma mass spectrometry
baijiu
cluster analysis
principal component analysis
factor analysis
linear discriminant analysis
