Identification of a Class of Nonlinear Systems Using Correlation Analysis

Most control systems encountered in practice are nonlinear to some extent and although it may be possible to represent systems which are only mild 'mildly' nonlinear or which operate over a restricted range by a linear model, in general nonlinear processes can only be adequately characterised by a nonlinear model. Since the analysis of control systems is often dependent uopn finding a mathematical description which defines the relationship between the system input and output, system identification is of fundamental importance in control theory. However, whereas system identification methods are well established for linear systems, the identification of non-linear systems appears to have been largely neglected. This can of course be attributed to the inherent complexity which results in the analysis of non-linear systems and the absence of a general theory for such systems.