Modelling the production of soluble hydrogenase in Ralstonia eutropha by on-line optimal experimental design**This work was supported by the DFG in the framework of the cluster of excellence UniCat.

This paper presents a case study on integrating a new model extension describing the production of the enzyme soluble hydrogenase (SH) into an already existing model by means of on-line optimal experimental design (OED). These investigations were done on the autotrophic cultivation system and process model of Ralstonia eutropha, i.e., a cultivation where aeration by air is replaced by a gas consisting of H2/O2/CO2. Prior to the experiment, three different structures of the new model extension are postulated and coarsely identified. Off-line OED with the best model yields initial feeding trajectories. During the experiment, a repetitive model refinement is performed. This consists of identifying the parameters of the new model extension for all three models, selecting the best model and recalculating optimal feeding trajectories. It is shown that on-line OED in the early modelling stage followed by a manual modelling step results in an acceptable quality of the description of SH production.