AN APPLICATION OF MODEL PREDICTIVE CONTROL TO GRADE ALTERNATION OPERATION IN A CONTINUOUS POLYMERIZATION REACTOR

In a continuous polymerization reactor, product grade alternation is quite important from the viewpoint of lower production loss, uniform production quality and labor saving. The above alternation step should proceed as quickly and stably as possible for good process economic. A model predictive control system is designed for the satisfactory grade alternation step in a continuous polyethylene polymerization reactor. The ARX model was applied to a reactor dynamics approximation using the Recursive Maximum Likelihood method. Actual results showed that the control system so developed is sufficient in practice for product grade alternation in a continuous polyethylent polymerization reactor. © 1990, The Society of Chemical Engineers, Japan. All rights reserved.