摘要
本文提出了一种以拉曼光谱为输入,采用层次聚类分析为手段的葡萄酒品质控制新方法。实验以张裕干红葡萄酒为主要研究对象,分别采集了样品在激光积分时间1～5s的拉曼光谱信息,运用层次聚类法,定量分析了同品牌不同系列(张裕干红葡萄酒、张裕酿酒师赤霞珠干红葡萄酒、张裕干红葡萄酒(佐餐级)),以及不同品牌(张裕干红葡萄酒、王朝干红葡萄酒、长城干红葡萄酒)间品质差异,并进行了实验参数优化。结果显示,张裕干红葡萄酒在欧氏距离结合平均距离法,激光积分时间4s表征条件下,可实现其与同品牌不同系列,以及不同品牌葡萄酒间的高效区分。本文所提快速检测方法具有光谱信号获取速度快,算法自动化分析程度高等优势。
In this work,a method of controlling wine quality with Raman spectra as inputs and hierarchical cluster analysis were proposed.Zhang Yu dry red wine was used as the main object in this work,Raman spectra of laser integration time from one to five seconds of the samples were collected,then,the quality difference among the different series of the same brand(Zhang Yu dry red wine,Zhang Yu Cabernet Sauvignon dry red wine and Zhang Yu dry red wine(table level)),and different brands(Zhang Yu dry red wine,Dynasty dry red wine and Great Wall dry red wine)were quantitatively analyzed by using the hierarchical clustering method.The results show that under the conditions of Euclidean distance plus mean distance method and laser integration time(four seconds),Zhang Yu dry red wine can be distinguished effectively from different series of the same brand and different brands.The rapid detection method proposed in this work has the advantages of fast acquisition of spectral signals and high degree of automatic analysis of the algorithm.
作者
张正勇
李晓烨
龙宸
姜逸雪
沙敏
王海燕
刘军
ZHANG Zheng-yong;LI Xiao-ye;LONG Chen;JIANG Yi-xue;SHA Min;WANG Hai-yan;LIU Jun(School of Management Science and Engineering,Nanjing University of Finance and Economics,Nanjing210023;State Key Laboratory of Chemo/Biosensing and Chemometrics,Hunan University,Changsha 410082;School of Management Engineering and Electronic Com merce,Zhejiang Gongshang University,Hangzhou 310018)
出处
《分析科学学报》
CAS
CSCD
北大核心
2019年第1期8-12,共5页
Journal of Analytical Science
基金
国家自然科学基金(No.61602217
61373058
71433006
91746202)
江苏省高校自然科学研究项目(No.16KJB150015)
湖南大学化学生物传感与计量学国家重点实验室开放课题(No.2017017)
关键词
拉曼光谱
层次聚类分析
相似度
葡萄酒
品质控制
Raman spectroscopy
Hierarchical cluster analysis
Similarity
Wine
Quality control
