Modelling and uncertainties characterization for robust control

Abstract

In this work, multi-input multi-output (MIMO) process identification is studied, where the model identification is dedicated to the control design goal. An ad hoc identification procedure is presented which allows estimating not only a nominal parametric process model, but also a bound of the model uncertainty (i.e. modelling errors). The model structure is defined in a way that the identified nominal model and the uncertainties can readily be used for the analysis and design of a robust control system by means of many of the techniques available in the literature. Simulation examples are given to illustrate the method.