State Space Model Predictive Control of a Reactive Distillation Process

This work has been carried out to demonstrate the performances of four different state space models in the model predictive control system of a reactive distillation process that was used for the production of ethyl acetate. The state space models of the reactive distillation process were developed with the aid of System Identification Toolbox of MATLAB using the data acquired from the reactive distillation column set up. The control algorithms were developed and simulated in MATLAB environment with the aid of Model Predictive Control Toolbox. The top segment, the reaction segment and the bottom segment temperatures were selected as the controlled variables while the reflux ratio, the feed ratio and the reboiler duty were respectively chosen as the manipulated variables. Compared to the other state space model predictive controllers investigated, the best closed-loop dynamic responses with the smallest number of oscillations, fastest rise time and fastest response time obtained from n4sid state space model predictive controller showed that it had the best performance. Further simulations of the n4sid state space model predictive controller for mısmatch revealed that it was very robust as the mismatch only affected the performance of the controller very slightly.