Temperature Control of Distillation Column in Hydro-alkylation of Toluene: A Supervisory Predictive Approach

one of the most non-linear basic process units is distillation unit and has many uses in the main process units. In Hydro-alkylation unit of toluene, three distillation columns used in several ways that generally makes the process to be non-linear. On the other hand, Hydro-alkylation of toluene (HAD) unit has another six basic non-linear units. The importance of this unit is in the production of chemical Benzene. In this paper, first the mathematical model of the process has been extracted using system identification method. Using a model-based predictive control algorithm in the regulatory layer leads to more favorable responses in controlling of the unit. The comparison between two responses comes from the two-control approach. The first one is PID controller and the second one is a model-based predictive control in the regulatory layer, which is responsible for changing the reference signal of PID controller. Simulation of this process is done by Aspen HYSYS process simulator software (v7.2). The model-based predictive controller is located for monitoring the classic PID controller, which improves the optimal point tracking as well as optimization in the thermal load rate of boilers in this process. By prediction on the obtained model the effect of disturbances on the process decreased dramatically which is another important achievement of the proposed approach.