Model-based control to maximise biomass and PHB in the autotrophic cultivation of Ralstonia eutropha

This paper presents a realisation of a closed-loop controlled autotrophic fed-batch cultivation of Ralstonia eutropha H16 with the aim to maximise the amount of biomass and internal storage compound PHB. The special control unit is a model-based on-line trajectory planning with a sigma-point Kalman filter for state estimation and an additional gas phase controller. For the control, a medium-sized structured model describing the process is proposed and compared to cultivation data. The model considers the internal storage of PHB by R. eutropha and pays special attention to the production of the oxygen-tolerant membranebound hydrogenase. This enzyme is in the center of biological research as a component of a biological fuel cell. It is shown that on-line optimisation works for maximising the amounts of biomass and PHB.