Fuzzy Model-Based Predictive Hybrid Control of Polymerization Processes

Abstract

This paper presents a fuzzy model-based predictive hybrid controller (FMPHC) for polymerization processes based on Takagi−Sugeno models and moving horizon methodology. Such processes are characterized by strongly nonlinear behaviors, which may either require significant effort to tune model-based controllers or render them ineffective. The proposed FMPHC is a promising integrated approach to handle nonlinearities and control issues. An industrial copolymerization process of ethylene and 1-butene is adopted to validate the proposed approach and to compare it to the most widespread advanced multivariable control.