Predictive modelling of brewing fermentation: from knowledge-based to black-box models

Advanced monitoring, fault detection, automatic control and optimisation of the beer fermentation process require on-line prediction and off-line simu lation of key variables. Three dynamic models for the beer fermentation pro cess are proposed and validated in laboratory scale: a model based on biolo gical knowledge of the fermentation process, an empirical model based on th e shape of the experimental curves and a black-box model based on an artifi cial neural network. The models take into account the fermentation temperat ure, the top pressure and the initial yeast concentration, and predict the wort density, the residual sugar concentration, the ethanol concentration, and the released CO2. The models were compared in terms of prediction accur acy, robustness and generalisation ability (interpolation and extrapolation ), reliability of parameter identification and interpretation of the parame ter values. Not surprisingly, in the case of a relatively limited experimen tal data (10 experiments in various operating conditions), models that incl ude more process knowledge appear equally accurate but more reliable than t he neural network. The achieved prediction accuracy was 5% for the released CO2 volume, 10% for the density and the ethanol concentration and 20% for the residual sugar concentration. (C) 2001 IMACS. Published by Elsevier Sci ence B.V. All rights reserved.