电子鼻对不同品牌食醋的识别

发布时间：2018-03-14 12:39

本文选题：电子鼻　切入点：食醋　出处：《食品工业科技》2017年09期 　论文类型：期刊论文

【摘要】：以东湖陈醋、水塔米醋、海天白米醋、恒顺香醋、水塔陈醋、紫林陈醋六种食醋为实验材料,利用电子鼻技术对其挥发性成分进行检测,采用径向基函数神经网络和Fisher判别建立其预测模型。结果表明:传感器S7、S3对六种样品的响应值差异显著(p0.05),即六种食醋的硫化氢、氨类物质的含量差异显著(p0.05)。同时线性判别分析比主成分分析能更好地区分不同品牌的食醋。Fisher判别和径向基函数神经网络模型均能较好地预测不同品牌的食醋,但Fisher判别的效果更好,对六种食醋的正确判别率均为100%。由此说明利用电子鼻技术能实现食醋种类的快速识别,以维护食醋品牌的名誉和规范市场。
[Abstract]:Six kinds of vinegar, such as Donghu vinegar, water tower vinegar, Haitian white rice vinegar, Hengshun vinegar, water tower vinegar and purple forest vinegar, were used as experimental materials, and their volatile components were detected by electronic nose technique. The prediction model was established by radial basis function (RBF) neural network and Fisher discriminant. The results showed that the response value of sensor S7 / S3 to six kinds of samples was significantly different (p 0.05), that is, hydrogen sulfide of six kinds of vinegar. The difference of ammonia content was significant (p 0.05). Meanwhile, linear discriminant analysis (LDA) was better than principal component analysis (PCA) in distinguishing vinegar. Fisher discriminant and radial basis function neural network model of different brands could predict vinegar of different brands better than that of principal component analysis (PCA), and linear discriminant analysis (LDA) was better than principal component analysis (PCA). But the effect of Fisher discrimination is better and the correct discriminant rate of six kinds of vinegar is 100. It shows that the electronic nose technology can be used to realize the rapid identification of vinegar types in order to maintain the reputation of vinegar brands and regulate the market.
【作者单位】：山西师范大学食品科学学院;
【分类号】：O657;TS264.22

【相似文献】