Abstract: To rapidly and non-destructively detect beer quality, olfactory visualization technology was used to qualitatively distinguish among eight kinds of Tsingtao beer (namely Premium, 1903, Augerta, Weiss, Stout, Pilsner, IPA and Strong), and quantitatively determine their key quality indicators (namely alcohol content, original gravity and diacetyl concentration). A 4 × 4 colorimetric sensor array (CSA) was used to react with volatile compounds of beer, and the color difference images obtained were analyzed by chemometric methods. The results showed that the linear discriminant analysis (LDA) model was able to qualitatively distinguish among these eight kinds of beer with a 100% recognition rate for both the calibration and prediction sets, while the least square-support vector machine (LS-SVM) model was able to quantitatively predict alcohol, original gravity and diacetyl concentration with correlation coefficients over 0.98 for the calibration and prediction sets. Hence, olfactory visualization technology has good potential for application in the rapid and non-destructive detection of beer quality.

Key words: beer; olfactory visualization; alcohol; original wort; diacetyl