Bioprocess Monitoring using Resource Allocation Models: Simulation and State Estimation for a Methane Bioconversion Process

Promovendus/a: Kobe De Becker

Promotor(en): Prof. dr. Steffen Waldherr, prof. dr. ir. Kristel Bernaerts

The biotechnological industry belongs to the most profitable branches of industry in Belgium. With this work we aim to improve the performance of bioprocesses in general, and of one process in particular: a methane bioconversion process. To increase process efficiency, the process should be carefully controlled with respect to a predefined objective which requires (1) an accurate model representation of the bioprocess, and (2) adequate monitoring software. Monitoring software is essential since not all process variables can be reliably measured. Therefore, monitoring software is applied in combination with a model representation of the process to estimate then non-measured variables, based on the variables that are measured. What adds to the difficulty is that bioprocess models have increased significantly in complexity in the last decade due to the availability of more detailed and elaborate biological data sets, making these new models not compatible with classical monitoring tools. We have developed a new solution strategy for one particular class of bioprocess models that implicitly imply that micro-organisms aim to maximize their own growth . In addition, we developed monitoring software tools that are compatible with the new model formulation. Both the solution strategy and the monitoring software are applicable to multiple bioprocesses.

Both the new solution strategy and the monitoring software tools are applied to a methane bioconversion process. This process consumes the green-house gas methane to produce lactic acid, a material used for the production of a type of biodegradable plastic, and biomass from which useful components might be extracted. The results show that the new solution strategy is capable to accurately describe the growth behavior of the micro-organism with the exception of a growth phase during which the biomass amount increases linearly. The challenge in monitoring bioprocesses lies in the fact that we cannot measure amounts of components that are inside the cells. The developed monitoring software is capable to reliably estimate these non-measured amounts for this methane bioconversion process.