Intelligent Dynamic Simulation of a Fed-Batch Enzyme Fermentation Process

This paper describes building of a dynamic simulation model for prediction of bioprocess variables. The simulator consists of three interacting dynamic model based on the method of linguistic equations. Each model has three versions, i.e. an own version for each phase of the fed-batch fermentation process. Steady state methods with dynamic structures were used in developing these linguistic equation models. With this simulator, it is possible to predict values of dissolved oxygen concentration, oxygen transfer rate and concentration of carbon dioxide in the exhaust gas through the whole process, using only the values of the control variables as inputs. Extension to fuzzy LE models provides useful information about uncertainties of the forecasted results. The complexity of the models is increased only slightly with the new system based on the extension principle and fuzzy interval analysis.