NONLINEAR RULE-BASED PROCESS CONTROL-ANALYSIS AND REDUCTION OF THE RULE SET BY NONLINEAR-THEORY

Abundant time-series dynamic data can be accumulated from a chemical p lant during long term operations. In our previous work, these plant da ta were directly implemented for the purpose of model predictive contr ol. In this work, fractal analysis is performed to reduce the size of a time-series data set for high quality nonlinear model predictive con trol. Results in this study indicate that on-line identification of no nlinear models is unnecessary if the disturbances to the process satis fy the fractal-equivalence condition. Simulation examples, including t he dual composition control of a high-purity distillation column demon strate that the nonlinear model predictive scheme is guile useful for those cases in which linear model predictive controller has failed.