System identification method for Hammerstein processes

Research output: Contribution to journalArticlepeer-review

72 Scopus citations

Abstract

A new strategy is proposed to identify Hammerstein-type nonlinear processes composed of a nonlinear static function and a linear dynamic subsystem. It completely separates the identification problem of the linear dynamic subsystem from that of the nonlinear static function using a special test signal. Then, quite useful advantages over previous approaches are guaranteed: we can use existing well-established linear system identification methods and their asymptotic properties of parameter estimates as well as excellent techniques for minimal parametrization to identify the linear dynamic subsystem of the Hammerstein process. Also, we can identify the nonlinear static function of the Hammerstein process analytically without any iterative optimization.

Original languageEnglish
Pages (from-to)4295-4302
Number of pages8
JournalIndustrial and Engineering Chemistry Research
Volume41
Issue number17
DOIs
StatePublished - 21 Aug 2002

Fingerprint

Dive into the research topics of 'System identification method for Hammerstein processes'. Together they form a unique fingerprint.

Cite this