In this study, the methods for rapid detection of the alcohol content, total acidity and amino acid nitrogen of Chinese rice wine by near-infrared (NIR) spectroscopy were established. By comparison of different frequency regions and ranks, the optimal pretreatment methods were obtained and the three kinds of predictive models were established. R2 (coefficient of determination) and RMSECV (cross-validation mean square) were used to evaluate the prediction accuracy and stability of the three models, respectively. The R2 values were 97.68, 95.68 and 94.13, respectively, and the RMSECV values were 0.2030,0.1260 and 0.0232, respectively. There was no significant difference (p> 0.05) between predicted values and measured values from ten randomly selected samples of Chinese rice wine. The results demonstrated that near-infrared spectroscopy was feasible in rapid detection of the alcohol content, total acidity and amino acid nitrogen of Chinese rice wine.%采用傅立叶变换近红外透射光谱技术建立黄酒3项理化指标(酒精度、总酸和氨基酸态氮)的快速检测方法.通过比较不同的频率范围和维数,选择最佳光谱预处理方法,建立了3项指标的预测模型.用R2(决定系数)和RMSECV(交叉验证均方差)衡量3个模型的预测精度和稳定性,R2值分别为97.68,95.68、94.13,RMSECV值分别为0.2030,0.1260,0.0232.对10个随机抽取的黄酒样品的预测结果显示,样品的预测值和实测值间的之间没有显著性差异(p＞0.05).研究结果表明,近红外光谱法可用于黄酒酒精度、总酸和氨基酸态氮的快速检测.
展开▼
