Identification of a Packed Distillation Column for Control Via Artificial Neural Networks

Artificial Neural Networks (ANN) have been used by several researchers for the purpose of nonlinear system identification. In this study, we address some of the issues associated with developing dynamic models of nonlinear chemical processes using ANN from the viewpoint of eventually using the ANN model for model prdictive control. Specifically, we look at the sufficiency of the training set for modeling of a highly nonlinear MIMO process, a packed distillation column. An ANN model of the packed column gives comparable (and sometimes superior) performance to a simplified first principles model when used for model predictive control.