Biotechnological processes are essential for production of biopharmaceuticals, e.g., medicine and vaccines. The project will develop software to support operation of biotechnological processes, which can increase production yield and ultimately increase availability of essential medicine.
In today’s practice, manufacturing of peptides (e.g., insulin, glucagon, amylin) and vaccines is conducted following recipes operating the processes in open-loop without feedback control and optimization.
These operating procedures may be improved substantially using novel techniques such as economic nonlinear model predictive control. In this project, we will develop economic nonlinear model predictive control algorithms for closed-loop operation of biotechnological processes such as fermentations operated in batch, fed-batch, or continuous mode as well as industrial chromatographic processes.
These processes have been selected for this PhD project, as their current operation offer significant improvement potential by development and application of novel advanced mathematical technology.
Additionally, we will apply high-performance Monte Carlo simulation to quantify closed-loop performance of stochastic systems. To achieve the goals of the project, we will develop high-performance optimization software for nonlinear model predictive control.
We will design the algorithms such that Monte Carlo simulations can be performed in parallel on computing clusters to utilize the full potential of modern multicore CPUs.
Main supervisor:
John Bagterp Jørgensen
Co- supervisor:
Allan Peter Engsig-Karup
Dimitri Boiroux