Discrimination of Red Wine Age Using Voltammetric Electronic Tongue Based on Multifrequency Large-Amplitude Voltammetry And Pattern Recognition Method

Three methods of multivariate data analysis (MVAD), principal component analysis (PCA), soft independent modeling of class analogy (SIMCA) and partial least squares discriminating analysis (PLS-DA), were used for processing data from a multifrequency large-amplitude pulse electronic tongue (MLAP-ET) in this paper. The dry red wine samples from the same company, produced by the same type of grape from the same vineyard, but with different vintages were studied using MLAP-ET. The results showed that these three methods were all effective for the data treatment of MLAP-ET to assess the vintage of red wine samples but differ in their discriminating ability. PLS-DA had the best classification property and was most suitable for processing the data from MLAP-ET.