Design of Sub-Optimal Robust Gain-Scheduled PI Controllers

Abstract

A methodology is proposed for the analysis and design of a robust gain-scheduled PI controller for nonlinear chemical processes. The stability and performance tests can be formulated as a finite set of linear matrix inequalities (LMI) and hence, the resulting problem is numerically tractable. Input saturation and model error are explicitly incorporated into the analysis. A simulation study of a nonlinear CSTR (continuous stirred tank reactor) process indicates that this approach can provide useful sub-optimal robust controllers.