摘要
结合主成分分析法(PCA)、线性判别分析法(LDA)和方差分析法(ANOVA),通过单因素试验确定了样品测试时的较佳酒精度、装样体积和装样瓶顶空生成时间分别为1.9%vol、15 mL和30 min。并在优化参数条件下,对哈达山粮食酒和酒精酒进行鉴别。结果表明,采用主成分分析法及线性判别分析法,电子鼻技术可鉴别哈达山粮食酒与食用酒精含量分别为0.38%vol、0.76%vol、0.95%vol、1.14%vol、1.52%vol、1.90%vol的酒精酒。两种分析方法相比,线性判别分析法比主成分分析法鉴别效果更好。
Combined with principal component analysis(PCA),linear discriminant analysis(LDA)and variance analysis method(ANOVA),the optimal alcohol content,sample loading volume,and headspace generation time of packing bottles during sample testing were determined as 1.9%vol,15 ml and 30 min,respectively by single factor tests.Under the optimal conditions,Hadashan Baijiu(Chinese liquor)made from grain and alcoholic drink blended by alcohol were identified.The results showed that using principal component analysis and linear discriminant analysis method,the electronic nose technology could identify the alcohol content of Hadashan Baijiu made from grain and alcoholic drink blended by edible alcohol of 0.38%vol,0.76%vol,0.95%vol,1.14%vol,1.52%vol,and 1.90%vol,respectively.Compared with two kinds of analysis methods,the linear discriminant analysis method had good identification effects than the principal component analysis method.
作者
董画
何雨
薛桂新
DONG Hua;HE Yu;XUE Guixin(College of Agriculture,Yanbian University,Yanji 133002,China)
出处
《中国酿造》
CAS
北大核心
2018年第9期43-47,共5页
China Brewing
基金
校企合作横向项目(2018222409000036)
关键词
粮食酒
酒精酒
掺假
电子鼻技术
Baijiu made from grain
alcoholic drink blended by alcohol
adulterate
electronic nose technology
