High-throughput screening of optimal process conditions using model predictive control.

Area of Science:

• Biotechnology and Bioprocess Engineering
• Control Systems Engineering
• Computational Biology

Background:

• Modern biotechnological labs utilize advanced parallel mini-bioreactor systems for sophisticated cultivations.
• These systems generate vast data, necessitating efficient experimental design and operational algorithms.
• Existing control algorithms require adaptation for complex biological systems and dynamic operating conditions.

Purpose of the Study:

• To implement a model predictive control (MPC) framework for parallel mini-bioreactor operations.
• To address challenges in mathematical modeling, estimation, and optimization for nonlinear biological systems.
• To demonstrate the value of MPC in high-throughput bioprocess development under constrained conditions.

Main Methods:

• Implementation of a model predictive control (MPC) framework.
• Development of mathematical models accounting for nonlinear and stiff dynamics.
• Adaptation of MPC for high-throughput bioprocess development and parallel operation.
• Optimization strategies to maximize biomass concentration under dissolved oxygen constraints.

Main Results:

• Demonstrated challenges in mathematical model structure, state, and parameter estimation for biological systems.
• Showcased necessary adaptations for applying MPC in high-throughput bioprocess development.
• Quantified the added value of MPC in maximizing biomass concentration within operational constraints.

Conclusions:

• MPC is a viable and valuable framework for optimizing parallel mini-bioreactor operations.
• Adaptations in modeling and control strategies are crucial for complex biological systems.
• MPC enables efficient high-throughput bioprocess development by managing constraints and maximizing productivity.