【摘要】 探索改善油菜籽芥酸近红外预测模型准确度与精密度的方法,利用无效变量消除法(UVE),对135个油菜籽样品近红外光谱信号进行筛选,并利用筛选后的光谱对油菜籽芥酸含量进行偏最小二乘法交叉验证。结果表明,UVE法筛选变量后建立的芥酸校正模型对未知样品预测结果的准确度和速度显著优于全波长参与建立的芥酸校正模型。散射校正加一阶导数对光谱预处理,UVE法筛选变量,偏最小二乘法交叉验证建立的校正模型效果最好,其预测值与标准值的相关系数R达到0.92,交叉验证预测均方差为2.2。因此,用UVE进行波长选择后建立的近红外模型,能准确快速地对油菜籽芥酸含量进行定量分析。更多还原

【Abstract】 Study was conducted to improve the accuracy and preciseness of the prediction model of near-infrared reflectance spectroscopy(NIRS) for erucic acid in rapeseed.Based on uninformative variables elimination(UVE) method,the NIRS signals of 135 intact rapeseed samples were selected and some PLS-CV(partial least squared-cross validation) operations were proposed for the prediction of rapeseed erucic acid content with the selected spectra.The result showed that the model with selected variables by UVE was more accurate than the model optimized with the whole spectra data in predicting unknown rapeseed sample.The model with multiplicative scattering correction(MSC) and first derivative had the best predicting accuracy.The correlation coefficient of the predictive value and the standard value reached 0.92 and root mean square error of cross validation(RMSECV) was 2.2.It was concluded that the model built with UVE in selecting the wavelength variables could be used for rapid,accurate quantitative analysis of erucic acid in rapeseeds.更多还原