On-Line Simulation Techniques for Bioreactor Control Development

Abstract

During the last decade, bioprocess development automation tasks have become more significant for overall process performance. Quality requirements on the one hand and growing process knowledge on the other have resulted in applied control strategies of increasing complexity. At the same time modern information technology facilitates higher flexibility of the technical systems used for practical realization of control tasks [1, 2]. Nevertheless, one of the most important goals in bioprocess automation is the manipulation of the process to meet desired performance criteria. Examples are the control of p0₂ to a prescribed limit or the realization of an exponential substrate feeding strategy. In practice, the process and the controller form a closed loop, often realized in a feedback fashion (Fig.6.1a). For design and stability investigations of these control-loops both process and controller require treatment using a theoretical mathematical model (Fig.6.1b). Obviously the time-dependent changes of the relevant process variables are of major interest for these applications. Therefore the underlying model is highly dynamic in nature and typically a set of differential equations based on mass and energy balances is used for simulation of the process behavior.