短波近红外光谱-偏最小二乘法对食醋中总酸含量的无损定量分析

将短波近红外光谱技术与偏最小二乘法（PLS）结合，对食醋中的总酸含量进行快速、无损定量分析，建立了最佳数学校正模型．讨论了光谱预处理方法和主成分数对PLS模型预报精度的影响．研究表明，采用一阶导数预处理光谱建立的数学校正模型能够得到最佳的预报结果，在对预测集8个样本中总酸的含量进行预报时，所得的集相对标准偏差为1．386％．实验结果表明，该方法方便快捷，并且具有较高的预报精度，可以用于食醋中总酸含量的测定．

Nondestructive Quantitative Analysis of the Concentrations of Total Acid in Vinegar by Short-wave Near Infrared Spectroscopy with PLS

In this paper, Near Infrared spectroscopy combined with partial least squares （PLS） method was applied to nondestructively analyzing the concentrations of total acid in vinegar, and the optimal mathematic calibration models was established. Furthermore, the influences of spectra pretreatment methods and the numbers of principal components on the prediction ability of PLS were discussed. Study showed that, the calibration model established with first-derivative preprocessing spectra was the best, 8 samples of test set were predicted, the relative standard error was 1.386%. Experimental results demonstrate that this method is fast and convenient, and it has a high ability of prediction, thus it is promising for the non-destructive quality control of vinegar.