| Summary: | Two optimisation techniques for the fed-batch cultivation of high cell density Escherichia coli producing recombinant proteins were compared. An unstructured model for the growth, based on the General State Space Dynamical Model [1] was used to represent the four major metabolic pathways: oxidative growth on glucose, fermentative growth on glucose, oxidative growth on acetate, and maintenance. The dilution rate (dependent on the substrate feed rate) was chosen as the input variable. Recombinant protein production is known to be proportional, in our system, to the biomass concentration. Thus, biomass productivity was chosen as the criterion to be maximized. The two methods compared were a first order gradient method based on Pontryagin’s minimum principle and a stochastic method based on the biological principle of natural evolution, using a genetic algorithm. The former method revealed less efficient concerning to the computed maximum, and dependence on good initial values.
|
|---|