论文部分内容阅读
应用傅里叶变换近红外光谱技术,定量分析49种酿酒葡萄的可溶性固形物含量。采用偏最小二乘回归法建立校正模型和预测模型,评价模型的预测能力和实用性。在11 987.7~3 999.5 cm-1范围,通过KennardStone算法划分样品校正集与预测集,比较10种光谱预处理方法对模型检测结果的影响。结果表明,光谱经一阶导数法处理,在11 987.7~7 498.1 cm-1和6 101.8~4 597.6 cm-1范围建模,所得定量分析模型效果最佳,分析结果精度较高,其校正相关系数(Rca)、交互验证相关系数(Rcv)、主因子数、相对标准差(RSD)和相对分析误差(RPD)分别为0.972,0.951,7,2.76%和4.03。模型经预测集检验,预测相关系数Rp达到0.961,说明建立的模型可靠,预测效果好,能满足酿酒葡萄快速、无损检测的要求。
The Fourier transform near infrared spectroscopy was used to quantitatively analyze the soluble solids contents of 49 wine grape varieties. The partial least-squares regression method is used to establish a calibration model and a prediction model to evaluate the prediction ability and practicability of the model. In the range of 11987.7 ~ 999.5 cm-1, KennardStone algorithm was used to divide the sample calibration set and the prediction set. The effects of 10 spectral pretreatment methods on the model test results were compared. The results showed that the spectra were modeled by the first derivative method and modeled in the range of 11 987.7 ~ 7 498.1 cm-1 and 6 101.8 ~ 4 597.6 cm-1. The obtained quantitative analysis model was the best and the analysis results were more accurate. Rca, Rcv, main factor, relative standard deviation (RSD) and relative analytical error (RPD) were 0.972, 0.951, 7, 2.76% and 4.03, respectively. The model was tested by the predictive set, and the prediction correlation coefficient Rp reached 0.961, indicating that the established model is reliable and the forecasting effect is good, which can meet the requirements of rapid and non-destructive testing of wine grape.