Js. Lin et Ss. Jang, NONLINEAR RULE-BASED PROCESS CONTROL-ANALYSIS AND REDUCTION OF THE RULE SET BY NONLINEAR-THEORY, Computers & chemical engineering, 20, 1996, pp. 877-882
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.