Hybrid dynamic modeling of 4-CBA hydrogenation fixed-bed catalytic reactor of PTA production plant

Abstract

In this paper, a dynamic hybrid model of 4-carboxybenzaldehyde (4-CBA) hydrogenation fixed-bed catalytic reactor of Purified Terephthalic Acid (PTA) production plant has been developed. At the first step, the deactivation model of the reaction system has been predicted by using artificial neural network (ANN) technique. Then, the deactivation model estimated by ANN as black box incorporated to first principle model (FPM) as white box. Therefore, the established hybrid model has been utilized to model the hydrogenation reactor of PTA plant. Finally, the simulated results have been compared with the industrial reactor data. It has been concluded that hybrid model is more precise than a FPM model.