摘要
本研究采用近红外光谱技术快速无损鉴别技术,研究了不同产地,不同变型的82个天麻样品;通过采用Thermo scientific omnic TQ Analyst9分析软件的判别分析方法建立鉴别模型,同时选择乌天麻鉴别的3个敏感谱段:4007.02cm^(-1)~7254.89cm^(-1),4080.84cm^(-1)~7046.62cm^(-1),7594.57cm^(-1)~8651.10cm^(-1),通过马氏距离计算未知材料到每个类别中心的距离,来判断未知样品与标准样品中的相似性。结果表明:模型性能指数(Performance Index)为92.2,说明模型预测准确。该数学模型可靠,可为乌天麻的鉴别提供一种准确、快速、简便、无损的新方法。
In the research the rapid and non-destructive identification method Near Infrared Spectroscopy(NIRS)is adopted,eighty-two gastrodia elata samples from different producing areas and different varieties are selected as research subject.By using the analytic method of analysis software Thermo scientific omnic TQ Analyst 9(Thermo Fisher Scientific),the identification model is built,three sensitive spectrum ranges are selected to identify Gastrodia elata Bl.,respectively 4007.02cm-1~7254.89cm-1,4080.84cm-1~7046.62cm-1,7594.57 cm-1~8651.10cm-1.Through Mahalanobis distance,the distance between unknown material to the center of each category is calculated to determine the similarity of unknown samples and standard samples.The results show that the Model Performance Index is 92.2,which indicates that the model is accurate and reliable,and provides a new accurate,fast and simple,non-destructive method for the identification of Gastrodia elata Bl.
作者
彭佳华
余显伦
杨宝钦
吴浩玮
刘祥义
Peng Jiahua;Yu Xianlun;Yang Baoqin;Wu Haowei;Liu Xiangyi(Gastrodia elata Research Institute of Zhaotong,Zhaotong 657000,China;Southwest Forestry University,Kunming 650224,China)
出处
《云南化工》
CAS
2018年第6期66-67,共2页
Yunnan Chemical Technology
关键词
乌天麻
近红外光谱
鉴别
模型
Gastrodia elata Bl.
Near Infrared Spectroscopy
Identification
Model
