Introduction

Abstract

The need for high-performance control systems has accelerated over the past decade. Economic pressures, increased environmental and safety concerns, and a tighter integration of process units have all contributed towards this demand. At the same time, because of the increasing efficiency of process computers, the demand for robust and accurate control of non-linear industrial processes can now be realized by advanced control strategies. Among these strategies, non-linear model-based control (NMBC), a kind of multivariable control technique, has drawn considerable attention in recent years. Lee and Sullivan (1988) presented a framework for process controllers that allow for the inclusion of a non-linear mathematical model to approximate the plant behaviour. The so-called generic model control (GMC) formulation of the control law allows process models, linear and non-linear, to be embedded directly into the control structure, providing the potential for better process control due to more accurate process models being incorporated into the control structure.