Optimization of Reactive Semi-Batch Distillation under Uncertainty Parameters Using Bayesian Network

摘要

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