EFFECTS OF THE SAMPLING TIME ON THE DYNAMICS AND IDENTIFICATION OF NONLINEAR MODELS

Abstract

This paper investigates how the choice of the sampling time affects the identification of nonlinear models. It is shown how the choice of the sampling interval can affect structure selection and parameter estimation in different ways so that a compromise is required to properly select an appropriate sampling time. The use of a nonlinear correlation function to aid in the choice of the sampling rate is investigated. Several numerical examples are included to illustrate the main points of the paper.