Improved control strategies for the environment within cell culture bioreactors

This paper describes the development of improved control strategies for the standard environmental conditions in a fed-batch bioreactor used for monoclonal antibody cell culture. The consequences of relying on fixed parameter PID based controllers are considered and poor performance is demonstrated as a consequence of non-linearity and loop interactions. The benefits from adopting a more sophisticated control strategy are considered. Model Predictive Control (MPC) relies on a process model that can be identified from small system perturbations. It considers the predicted longer-term response and consequently can deliver improved control and satisfy user defined constraints. Results from experimental trials demonstrate the capability of MPC and the merits are discussed with regards to industrial application.