便携式近红外谷物分析仪快速测定小麦蛋白质的研究  

Study on the determination of protein content in wheat by portable near-infrared grain analyser

作　　者：王旭[1,2] 张凤清[1] 林家永[2] 范维燕[2] 叶华俊 陈智锋 

机构地区：[1]长春工业大学,吉林长春130012 [2]国家粮食局科学研究院,北京100037 [3]聚光科技(杭州)股份有限公司,浙江杭州310052

出　　处：《粮食与饲料工业》2012年第1期13-17,共5页Cereal & Feed Industry

基　　金：科技部"十一五"重点科技攻关项目(2006BAD08B07-3)

摘　　要：收集中国小麦各个主产区不同品种样品104份,研究采用国产便携式近红外谷物分析仪快速测定小麦蛋白质含量,用化学计量方法建立了小麦蛋白质含量的近红外分析模型,并对模型进行了预测准确性评价。在建模过程中,分别探讨光谱散射和数学导数处理等优化对定标模型的影响。结果表明:采用偏最小二乘法(PLS)建立模型,光谱预处理最佳条件为:Savitzky-Golay平滑、Savitzky-Golay一阶导数、基线校正、均值中心化、主因子数为12。所建模型的定标标准偏差(SEC)和定标相关系数(RC)分别为0.177和0.988;外部验证的标准偏差(SEP)和相关系数(RP)分别为0.188和0.961。标准方法测定值与NIRS方法预测值之间的T检验结果为T=0.304(P<0.05),表明两种测定方法测定值之间无显著性差异,说明定标模型具有很好的预测准确性,可应用于优质小麦收购中蛋白质含量的快速测定。104 wheat samples collected from the major wheat growth areas in China were used to measure their protein content by portable near-infrared grain analyser. The calibration model of protein content in wheat was established by chemometrics method. The influence of the spectrum scattering and mathematics derivative processing on the calibration model was discussed. The resultes showed that： the optimum pre-treatment factors of the calibration model developed by the partial least square （PLS） regression were：Savitzky-Golay smooth, one rank derivative,baseline corrections,average center, and main factor number of 12;The standard error （SEC） and the coefficient of correlation （Rc） of the calibration equation were 0. 177 and 0.988, respectively; The square error of prediction （SEP） and the regression squared （Re） were. 0. 188 and 0. 961 respectively; T test value between the chemical standard methods and NIRS method was 0. 304 （P〈0.05）, which showed that the difference between the two methods wasn＇t significant, the calibration model had good predictive accuracy, and the method could be applied to fast predict the protein content in high quality wheat purchasing.

关 键 词：小麦 蛋白质含量 近红外光谱技术(NIRS) 定标模型 

分 类 号：S512.1[农业科学—作物学] TH744.4[机械工程—光学工程]