Prediction of some physico-chemical parameters in red wines from ultraviolet–visible spectra using a partial least-squares model in latent variables

Abstract

A method for the rapid prediction of some physico-chemical parameters usually determined by classical methods in red wines (total acidity, volatile acidity, pH, free sulfur dioxide, tannins and total anthocyans) from the UV/VIS spectrum is proposed. A partial least-squares (PLS) model in latent variables was developed to relate the chemical variables to the spectra. Prediction of composition was achieved with a low prediction error in young red wines, some of them ‘Grand Cru’ from the Bordeaux production area. A similar procedure was applied to commercial red wines, that constituted a more heterogeneous group, with acceptable results. Finally, the model was extended to any red wine, without loss of prediction quality. In the last case, prediction errors were: total acidity, 5.0%; volatile acidity, 9.9%; free sulfur dioxide, 29%; pH, 3.3%; tannins, 11% and total anthocyans, 22%.