无信息变量消除法变量筛选优化烟草中总氮和总糖的定量模型  

Model of Total Nitrogen and Total Sugar in Tobacco Optimizing after Uninformative Variable Elimination

作　　者：李倩倩[1] 田旷达[1] 李祖红[2] 郑波[2] 赖衍清[1] 唐果[1] 宋相中[1] 闵顺耕[1] 

机构地区：[1]中国农业大学理学院应用化学系,北京100193 [2]云南省烟草公司曲靖市公司,曲靖655000

出　　处：《分析化学》2013年第6期917-921,共5页Chinese Journal of Analytical Chemistry

基　　金：中国烟草公司项目(No.2010YN65)资助

摘　　要：应用近红外光谱技术对烟草常规化学成分中总氮和总糖进行了测定。无信息变量消除(UVE)剔除光谱矩阵中没有有效信息的数据点,并用偏最小二乘方法(PLS)建立总氮和总糖的定量分析模型,外部检验对模型效果进行了评价。总氮定量模型校正集的决定系数R2为93.35%,标准偏差SEC为0.10;外部检验集的决定系数R2为94.09%,标准偏差SEP为0.11,相对标准偏差RSD为6.12%;总糖的定量模型校正集的决定系数R2为98.20%,标准偏差SEC为0.95;外部检验集样品的决定系数R2为98.01%,标准偏差SEP为0.78,相对标准偏差RSD为2.93%。结果表明:采用UVE建立的总氮与总糖的模型优于用全谱建立的模型,UVE提高了PLS模型的预测能力。The total nitrogen and total sugar in tobacco were determined by near infrared spectroscopy.The uninformative wavelength was gotten rid in matrix after uninformative variable elimination（UVE） variable selecting.The quantitative model of total nitrogen and total sugar was established by partial least squares（PLS） regression and the robustness of the models were evaluated by independent validation samples.The determination coefficient R2 and the standard error of calibration set（SEC） of total nitrogen were 93.35% and 0.10 respectively.The determination coefficient R2 of estimated value and specified value,the standard error and the relative standard deviation（RSD） of the independent validation samples were 94.09%,0.11 and 6.12%,respectively.The determination coefficient R2 and the standard error of calibration set（SEC） of total suger were 98.20% and 0.95,respectively.The determination coefficient R2 of estimated value and specified value,the standard error and the relative standard deviation（RSD） of the independent validation samples were 98.01%,0.78 and 2.93%,respectively.The result demonstrates that the model developed by selecting informative variable is better than that of full spectrum.

关 键 词：近红外 烟草 无信息变量消除 偏最小二乘法 

分 类 号：TS47[农业科学—烟草工业] O657.33[理学—分析化学]