荧光光谱结合化学计量学原酒鉴别方法的建立及其在小曲清香酒中的应用

Original liquor quality control and grade identification is important for the quality improvement of finished liquor. In this study, a total of 209 samples of original Chinese mild-flavor liquors with different grades were carefully collected. The grading of these samples was derived from sensory tasting, which is currently the common grading method in major liquor factory. These samples were scanned by three-dimensional fluorescence to obtain the excitation emission matrices (EEM). Principal component analysis (PCA), discrete cosine transform (DCT) and fast independent component analysis (Fast-ICA) are used to reduce the dimension of the spectrum, and linear discriminant analysis (LDA) and support vector machine (SVM) are used for classification. In general, it was found that a fast and man-saving model for quality identification of original liquor was obtained, theclassification accuracy of the original liquor obtained by the SVM method after PCA dimensionality reduction was the highest, the training set accuracy was 89.81%, and the test set accuracy was 84.62%, which can well meet the need for rapid grading of original liquor. The model has undergone five-fold cross-validation (CV), and the CV accuracy is 81.53%, which proves that the model is sufficiently stable. At the same time, the method of spectrum combined with chemometrics analysis also provides an idea for the renewal of many traditional detection methods.