Potential of predictive model-based dissolved oxygen control for intermittent fed-batch processes

The level of dissolved oxygen (DO) in a bioreactor is essential for the cultivation of microorganisms[a]. Traditional control strategies acting on the stirrer speed and aeration rate usually struggle with the dynamic nature of the system[b]. Their
purely reactive algorithms especially show their limitations when challenged with abrupt changes in nutrient additions. The resulting drops in DO can negatively influence the cells metabolism and physiological state[c].

Model predictive control (MPC) algorithms present promising alternatives for such intermittent feeding profiles. Applied to high-throughput small scale multireactor systems they can avoid oxygen limitations during the cultivation and
enable optimal process conditions.