FRACTAL ANALYSIS OF TIME-SERIES RULE-BASED MODELS AND NONLINEAR MODEL-PREDICTIVE CONTROL

Abundant time-series dynamic data can be accumulated from a chemical p lant during longterm operations. In our previous work, these plant dat a were directly implemented for the purpose of model predictive contro l. However, a large amount of time-series data is required to perform high-quality nonlinear model predictive control. In this work, fractal analysis is performed to reduce the size of a time-series data set fo r high-quality nonlinear model predictive control. Results in this stu dy indicate that on-line identification of nonlinear models is unneces sary if the disturbances to the process satisfy the fractal-equivalenc e condition. Simulation examples, including the dual composition contr ol of a high-purity distillation column, demonstrate that the nonlinea r model predictive scheme is quite useful for those cases in which the linear model predictive controller has failed.