Abstract
Advanced control strategies such as Model Predictive Control (MPC) offer a lot of benefits when applied to suitable processes. However, their application has been limited due to a number of reasons including the cost of implementation and unavailability in off-the-shelf control equipment. This paper presents a method of designing MPC controllers from historic plant data. Motivated by the concept of closed-loop identification, the developed method reduces the amount of engineering effort and cost associated with the tuning of MPC to replace an existing Proportional Integral/Proportional Integral Derivative (PI/PID) Controller. Simulation studies were carried out and the MPC parameters obtained were close to the true values. The resulting MPC controllers also had a performances similar to the existing PI controllers. These performances were easily improved using only a single tuning parameter. (C) 2019 Elsevier Ltd. All rights reserved.