Application of Fuzzy Modeling and Optimization in Enzymatic Esterification Process

In this paper, an application of a neuro-fuzzy modeling approach is presented in order to characterize the essential behavior of enzymatic esterification processes. The accuracy of the developed model was validated by comparing the response of the model and actual experimental data. The simulation results showed good generalization of the proposed model and its ability to predict the reaction yield, where the error of prediction for training data was less than 3%, and for validating and testing data less than 3 and 1.5%, respectively. A model-based optimization was performed to obtain the best operating conditions by using genetic algorithm. A fair comparison between the optimization results obtained from simulation experiments and laboratory data indicated the accuracy and feasibility of the proposed approach for estimating the optimal profiles in biotechnological processes. This can further facilitate up-scaling of the process by selecting the appropriate combinations of potential manufacturing parameters.