基于低场核磁共振的白酒品牌判别分析

为探求一种白酒品牌判别的方法，基于低场核磁共振（LF-NMR）技术，综合运用主成分分析（PCA）、偏最小二乘法-判别分析（PLS-DA）、反向传播人工神经网络（BP-ANN）方法对6种浓香型白酒品牌共300个样本进行模式识别分析，解析了对不同品牌浓香型白酒进行判别的可行性。结果表明，运用PCA方法对样品进行识别，无法区分白酒品牌；运用PLS-DA的方法对白酒样品进行识别，训练集的识别率约为99.5%，预测集识别率约为96.7%；运用BP-ANN的方法对白酒样品进行识别，训练集识别率约为99.5%，预测集识别率约为98.9%。结果表明，PLS-DA方法和BP-ANN方法对浓香型白酒样品的区分成功，表示将低场核磁共振方法应用到浓香型白酒的品牌判别中是可行有效的。

Discriminant analysis of Baijiu brand based on low field nuclear magnetic resonance

In order to explore a method for distinguishing Baijiu (Chinese liquor) brand, based on low-field-nuclear magnetic resonance (LF-NMR) technology, pattern discrimination analysis of a total of 300 samples from 6 strong-flavor Baijiu brands were applied comprehensively by principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA) and back-propagation artificial neural network (BP-ANN), and the feasibility of discriminating strong-flavor Baijiu of different brands was analyzed. The results showed that the Baijiu samples were identified by PCA, Baijiu brands could not be distinguished. The Baijiu samples were identified by PLS-DA, the recognition rate of training set was about 99.5%, and the prediction set was about 96.7%; The Baijiu samples were identified by BP-ANN, the recognition rate of training set was about 99.5%, and the prediction set was about 98.9%. The results showed that strong-flavor Baijiu samples could be distinguished successfully by the method of PLS-DA and BP-ANN, indicating that the application of low field NMR method to brands discrimination of strong-flavor liquor was feasible and effective.