Model predictive control of vaporizer in vinyl acetate monomer process

摘要

A Model Predictive Control (MPC) algorithm is developing on the Vinyl Acetate Monomer process provided by Luyben and Tyreus. Stress on the vaporizer, consideration only for the ‘Single Input Single Output’ (SISO) strategy with two control objective. Three closed loop have been applying with involve the vaporizer pressure and level. The MPC algorithm is implemented to the control loop by using Simulink/MATLAB 7.0. The limitation on the set-point using conversional controller would be overcome by using an advanced control method that is MPC. In direct comparison, we show that MPC algorithm would give better performance due to future prediction potential compare to conventional controller that is PI. The capabilities of MPC have been tested with three different inputs which represent the real mechanistic behaviors of chemical plant through the Simulink/MATLAB 7.0 simulations. The MPC algorithm is believed can capture the nonlinear and robust of Vinyl Acetate Monomer process that refers to real chemical industries.