STRUCTURE IDENTIFICATION OF NONLINEAR DYNAMIC-SYSTEMS - A SURVEY ON INPUT OUTPUT APPROACHES

In the past several methods have been elaborated for the identification of nonlinear dynamic systems. Most of the methods assume that the structure of the system is given a priori. Therefore they are in reality parameter estimation algorithms and structure identification is thus usually performed by repeated parameter estimation. However in nonlinear system theory several methods are known to determine the structure of a system. In this paper structure identification of block-oriented (especially cascade) models, of semi-linear dynamic models with signal-dependent parameters and of nonlinear dynamic models being linear in the parameters will be considered. Different structure selection methods are summarized based on step and impulse tests, frequency response measurements, correlation analysis, repeated reproducible tests and normal operating data. © 1990.