(371e) Optimization of Reactive Semi-Batch Distillation Under Uncertainty Parameters Using Bayesian Network

This work deals with modeling and optimization of Reactive Semi-Batch Distillation using Bayesian Network and utilizing uncertainty parameters. In a distillation column, heterogeneously catalyzed esterification of Methanol and Acetic Acid takes place. Considering that the reaction is azeotropic and in equilibrium, using reactive distillation, we can overcome the azeotrops and reach to high level of conversion. In this dynamic process, minimum batch-time and maximum productivity are the aims of optimization. Feed flow of Acetic acid that enters in the middle of column, reflux of distillation, kinetic parameters, hold up of each tray, trays' efficiency and pressure are uncertainty parameters that are confronted in this research. Hence, by using Bayesian Network and utilizing uncertainty parameters, we optimize batch-time and purity of the production, Methyl acetate.