Robust model predictive control of an industrial solid phase polymerizer

This study presents the application of model predictive control (MPC) to regulate a continuous, solid phase, industrial polymerization reactor. Polymerization is an extremely complex process which is a function of operating conditions such as, temperature, humidity, and pressure. This solid phase polymerizer is used for re-melting and re-polymerization of the solid polymer flake. PID algorithms, although effective, seldom lead to optimal utilization of resources and raw materials, hence the choice of an advanced supervisory control strategy to regulate the process. The paper presents the implementation of one particular MPC strategy, called robust model predictive control technology (RMPCT), developed by Honeywell. The identification of the process models for the various controlled and manipulated variable pairs is discussed and analyzed. Online performance results using RMPCT to control the molecular weight distribution are presented.