Modelling and predictive control of fed-batch yeast growth on industrial pilot plant scale

This paper presents an approach for modelling and control of fed-batch yeast growth which requires extensive calculation and a small number of measurements of component concentration. The process model is based on a simplified biochemical reaction description. An adapted predictive control algorithm is based on this model. It provides the future manipulated variable by analytical calculation, avoiding numerical optimisation methods. Experimental performance of this approach is shown for the control of ethanol production during fed-batch growth of Saccharomyces cerevisiae on an industrial pilot plant.