Performance of Linear Model Predictive Control (MPC) On the Control of a Nonlinear Continuous Stirred Tank Reactor (CSTR)

The purpose of this paper is to investigate and identify the control performance challenges of using a linear Model Predictive Control (MPC) for the control on a non-linear system which in this case is a Continuous Stirred Tank Reactor (CSTR). The goal is to develop the non-linear CSTR and an MPC using SIMULINK which can be found in MATLAB for the control of the non-linear CSTR and hence investigate the performance of the controller in various operating ranges for set-point tracking and disturbance rejection problems. The saponification process of ethyl acetate to form sodium acetate as the desired product is selected as a baseline for the implementation for the control of a non-linear process model for the CSTR. The objective is to control the sodium acetate concentration and temperature of the reactor by manipulating the cooling water and ethyl acetate flowrate despite the presence of the disturbances which is the flowrate of sodium hydroxide and cooling water temperature.