Quantification of Parametric Uncertainty via an Interval Model

The quantification of model uncertainty becomes increasingly important as robust control is an important tool for control system design and analysis. This paper presents an algorithm to characterize the model uncertainty in terms of parametric and nonparametric uncertainties directly from inputloutput data. We focus on the quantification of parametric uncertainty, which is represented as an interval system of the transfer function. Using this family of transfer functions (interval system), we give complete analysis of the system. A numerical example is used to demonstrate and verify the developed algorithm. The example illustrates the application of recently developed interval system techniques to the identified interval models.