蔗汁品质成分的傅里叶变换近红外分析数学模型  

Mathematic Models for Analysis of Quality Components in Sugarcane Juice with Fourier Transform Near Infrared Spectroscopy

作　　者：曹干[1] 谭中文[2] 梁计南[2] 龙永惠[1] 周学秋 

机构地区：[1]广东省农业科学院作物研究所,广州510640 [2]华南农业大学农学院,广州510642 [3]布鲁克光谱仪器公司,北京100081

出　　处：《中国农业科学》2003年第3期254-258,共5页Scientia Agricultura Sinica

基　　金：广东省农业厅糖蔗良种繁育基地资助项目;广东省农作物遗传改良重点实验室资助项目

摘　　要：应用傅里叶变换近红外光谱技术 ,采用透射测定方式和PLS算法建立蔗汁蔗糖定量分析数学模型。蔗汁蔗糖分和转光度预测数学模型的决定系数 (R2 )分别为 0 .9980和 0 .9979,均方根差 (RMSECV)分别为 0 .14 3%和 0 .15 5 % ,预测误差与常规分析方法的误差接近。内部交叉证实和外部验证结果表明近红外定量分析有很高的准确度。近红外光谱法完全可以替代甘蔗糖分常规分析方法。傅里叶变换近红外光谱法作为一种快速而准确的定量分析手段 ,在糖厂甘蔗品质管理。With the technique of fourier transform near infrared spectroscopy(FT NIRS),the predicted models for quantitative analysis of sucrose and polarization in sugarcane juice were established by using transmittance mode and calibrating with partial least square (PLS) algorithm. The determination coefficients (R 2) of the predicted models for sucrose and polarization were 0.9980 and 0.9979 respectively; the root mean square errors of cross validation (RMSECV) were 0.143% and 0.155% for sucrose and polarization in juice, respectively. The predictive errors measured by FT NIRS were close to those by routine laboratory methods. The method of FT NIRS had high accuracy in predicted tests of cross validation and independent validation and was available to replace the routine laboratory analysis. As a rapid and accurate measurement, the FT NIRS method has a potentially application in quality control of mill sugarcane, establishment of payment system based on sugarcane quality, and selection of clones in sugarcane breeding.

关 键 词：蔗汁 品质成分 傅里叶变换近红外光谱 定量分析 甘蔗 蔗糖 

分 类 号：TS241[轻工技术与工程—制糖工程]