Control Engineering Practice 11 (2003) 665–674 Optimal operation of fed-batch fermentations via adaptive control of overflow metabolite S. Valentinotti a, * ,1 , B. Srinivasan a , U. Holmberg c , D. Bonvin a,2 , C. Cannizzaro b , M. Rhiel b , U. von Stockar b a Laboratoire d’Automatique, EPFL, 1015 Lausanne, Switzerland b Laboratoire de G ! enie Chimique et Biologie, EPFL, 1015, Lausanne, Switzerland c School of Information Science, Computer and Electrical Engineering, Halmstad University, Halmstad, Sweden Received 23 July 2002; accepted 30 July 2002 Abstract The maximization of biomass productivity in the fed-batch fermentation of Saccharomyces cerevisiae is analyzed. Due to metabolic bottleneck, often attributed to limited oxygen capacity, ethanol is formed when the substrate concentration is above a critical value, which results in a decrease in biomass productivity. Thus, to maximize the production of biomass, the substrate concentration should be kept at the critical value. However, this value is unknown a priori and may change from experiment to experiment. A way to overcome this lack of knowledge is to allow the cells to produce a very small amount of ethanol. This way, the problem of maximizing the production of biomass is converted into that of regulating the concentration of ethanol, for which cell growth can be viewed as a perturbation. A novel adaptive control methodology based on the internal model principle is used to maintain the desired ethanol setpoint and reject the perturbation. Only a single parameter needs to be estimated on-line. Experimental results demonstrate the effectiveness of the proposed control methodology. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Fed-batch fermentation; Baker’s yeast; Optimization; Disturbance rejection; Adaptive control; Internal model principle 1. Introduction For centuries, man has employed biochemical reac- tions to his benefit. He has used microorganisms to obtain products such as bread, cheese, wine, beer, and yogurt, to mention only a few. Due to the importance of these products, scientists have ventured into under- standing how these reactions take place inside the microorganisms and have exploited their capabilities to perform more complex and more useful transforma- tions (Bailey & Ollis, 1986). Although the results in this paper are generally applicable to fermentation systems with microorganisms that present an overflow metabolism, the study con- centrates on the yeast Saccharomyces cerevisiae that is commonly used for making bread, alcohol and, recently, recombinant proteins. Numerous models have been proposed to describe the behavior of S. cerevisiae under different growth conditions (Bailey & Ollis, 1986; Lee, 1992). The model used in this work was proposed by Sonnleitner and K . appeli (1986). The contributions of this paper are twofold: * Mathematical analysis of the optimal operation of biofermenters: Nielsen and Villadsen (1994) argued that optimal productivity corresponds to operating at the critical substrate concentration. This will be by and large confirmed by the present analysis. However, it is difficult to maintain the substrate concentration at its critical value since the latter changes from experiment to experiment and from strain to strain (van Hoek, van Dijken, & Pronk, 1998). Different methods have been proposed in the literature to circumvent this problem, e.g., by *Corresponding author. Laboratoire de G! enie Chimique et Biologie, EPFL, Institut de Sciences de Bases, 1015 Lausanne, Switzerland. Tel.: +41216933836; fax: +41216932574. E-mail address: sergio.valentinotti@epfl.ch (S. Valentinotti). 1 The author thanks the Mexican Science and Technology Council, CONACYT, for financial support during part of this work. 2 Also for correspondence. 0967-0661/02/$-see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII:S0967-0661(02)00172-7