On Intelligent Extraction of an Internal Signal in a Wiener System Consisting of a Linear Block Followed by Hard-Nonlinearity

Article type: Research Article

Authors: Kazlauskas, Kazys | Pupeikis, Rimantas

Affiliations: Process Recognition Department, Institute of Mathematics and Informatics of Vilnius University, Akademijos 4, LT-08663 Vilnius, Lithuania, e-mail: [email protected], [email protected]

Abstract: The aim of the given paper is development of an approach based on reordering of observations to be processed for the extraction of an unmeasurable internal intermediate signal, that acts between linear dynamical and static nonlinear blocks of the Wiener system with hard-nonlinearity of the known structure. The technique based on the ordinary least squares (LS) and on data partition is used for the internal signal extraction. The results of numerical simulation and identification of a discrete-time Wiener system with five types of hard-nonlinearities, such as saturation, dead-zone, preload, backlash, and, discontinuous nonlinearity are given by computer.

Keywords: Wiener systems, hard-nonlinearity, internal signal, parametric identification, observations

Journal: Informatica, vol. 24, no. 1, pp. 35-58, 2013