CFD-Based Determination of Optimal Design and Operating Conditions of a Fermentation Reactor Using Bayesian Optimization.

The efficiency of fermentation reactors is significantly impacted by gas dispersion and concentration distribution, which are influenced by the reactor's design and operating conditions. As the process scales up, optimizing these parameters becomes crucial due to the pronounced concentration gradients that can arise. This study integrates the kinetics of the fermentation process with hydrodynamic analysis using Bayesian optimization to efficiently determine the optimal reactor design and operating conditions. By utilizing computational fluid dynamics (CFD) simulations, the study provides a comprehensive assessment of distributions ranging from gas supply to cell growth. The results demonstrate that a combination of wide baffle width, narrow impeller gap, slow gas flow rate, and high agitation speed significantly enhances reactor performance by improving gas distribution and minimizing stagnant zones. These findings underscore the importance of considering both kinetic and hydrodynamic factors to achieve more precise and scalable fermentation processes, offering valuable insights for industrial applications.