Modelling and Identification for Robust Control of Ill-Conditioned Plants - a Distillation Case Study

The problem of obtaining linear models for effective control is discussed. First a good model structure is identified by comparing different models. It is demonstrated that the high-frequency behavior (initial time response) of the model is much more important for controller design than the steady-state characteristics. Finally, having determined a good model structure, the problem of obtaining models from experiments is discussed. It is shown that by combining experimental open-loop step responses with theoretically established structural properties one may obtain reasonably good models.