A HYBRID RECURRENT NEURAL-NETWORK MODEL FOR YEAST PRODUCTION MONITORING AND CONTROL IN A WINE BASE MEDIUM

A dynamic model based on a recurrent neural network was established to follow the growth of yeast in a wine-base medium. It leads to the est imation and prediction of the yeast concentration in batch cultures, b ased on the on-line measurement of the volume of CO2 released and the initial yeast concentration. The mean error of the predicted value of the final yeast concentration is lower than 5%. A hybrid model combini ng this model with a measurement model (based on linear correlations r eflecting the reaction scheme) also leads to the estimation and predic tion of the sugar and ethanol concentrations in the culture medium wit h respective mean errors of 1.6 and 1 gl(-1). Moreover, this model was used in an open-loop control strategy in order to achieve a final con centration of yeast by setting the culture temperature. Adjusting cult ure temperature during growth was necessary for only 4% of the culture s, in order to remain within the range of measurement error (3 x 10(6) cells ml(-1)) of yeast concentration. The performance of the model an d of the control algorithm used could be assessed by controlling six s uccessive cultures. (C) 1997 Elsevier Science B.V.