An identification algorithm for parallel Wiener-Hammerstein systems

Maarten Schoukens, Gerd Vandersteen, Yves Rolain

Research output: Unpublished contribution to conferencePoster

14 Citations (Scopus)

Abstract

The class of Wiener-Hammerstein systems can be generalized to parallel Wiener-Hammerstein systems. Parallel Wiener-Hammerstein systems consist of a parallel connection of different Wiener-Hammerstein systems. An identification method for parallel Wiener-Hammerstein systems is presented. First, the overall dynamics are estimated in least squares sense. Second, these dynamics are decomposed over the parallel branches, and partitioned over the front and the back LTI blocks using a brute force scan. Finally, the static nonlinearities are estimated using a linear least squares estimator. The proposed method provides good initial estimates, which are refined in a nonlinear optimization step.
Original languageEnglish
Publication statusPublished - 22 Sep 2013
EventERNSI 2013, Nancy, France, September 22-25, 2013 - Nancy, France
Duration: 22 Sep 201325 Sep 2013

Conference

ConferenceERNSI 2013, Nancy, France, September 22-25, 2013
Country/TerritoryFrance
CityNancy
Period22/09/1325/09/13

Keywords

  • identification
  • Wiener-Hammerstein systems

Cite this