Rapid Detection of Baijiu Alcohol Content Based on NIR and SNV-UVE-PLS

这段英语中没有明显的语法错误。它在语法上 是正确的。然而, 可能需要注意一些表达方式和句子结构, 以提高流畅性和可读性。下面是对原文稍作修改的版本:The main components of Chinese Baijiu are ethanol, water, and various flavor substances. The ethanol content of Chinese Baijiu is directly related to its yield and quality. Online nondestructive testing for alcohol accuracy is a developing trend in the Chinese Baijiu industry. In this study, we analyzed the near-infrared spectrogram obtained from detecting ethanol aqueous solutions. After applying five different preprocessing methods to reduce noise in the original dataset of near-infrared spectroscopy, we screened the feature bands using Uninformative Variable Elimination (UVE), successive projections algorithm (SPA), and Principal Component Analysis (PCA) to simplify the model. We then used Partial Least Squares (PLS) to build a rapid prediction model for the alcohol content of liquor, focusing on the optimal bands screened. The results indicate that the alcohol prediction model for liquor, pretreated with the Standard Normal Variate (SNV) and processed by the UVE wavelength screening algorithm, performs well. The short-infrared band of the screened liquor better represents the characteristics of liquor alcohol, and the accuracy and stability of the prediction model are high, and R² = 0.9928, mean absolute error = 0.0132, root mean square error of prediction = 0.0117. The number of spectral variables is reduced to 28. The prediction model is applied to two commercially available Baijiu samples, yielding good results.