Optimal operation of baker's yeast fermentation in the presence of uncertainty

In this paper, we consider the problem of determining the optimal feeding policy for a fed-batch fermentation for the production of baker's yeast in the presence of uncertainty. Two algorithms are discussed for computing the optimal solution in the presence of uncertainty. The first method is a measurement-based closed-loop approach where process measurements are utilized to make on-line adjustments to the optimization routine. The second method is a simulator-based open-loop approach where simulations are run assuming various operating policies, and utilizing the simulator as if it were the real process. These two methods are implemented on a baker's yeast fermentation with uncertain parameters, and it is shown that both approaches lead to improved performance over the conventional open-loop operation. Copyright © 2006 Curtin University of Technology and John Wiley & Sons, Ltd.