Title :
Semi-parametric identification of Wiener systems using a single harmonic input and retrospective cost optimisation
Author :
D´Amato, Anthony M. ; Teixeira, B.O.S. ; Bernstein, D.S.
Author_Institution :
Dept. of Aerosp. Eng., Univ. of Michigan, Ann Arbor, MI, USA
Abstract :
The authors present a two-step method for identifying single-input, single-output (SISO) Wiener systems. First, using a single harmonic input, they estimate a non-parametric model of the static non-linearity, which is assumed to be only piecewise continuous. Second, using the identified non-parametric map, the authors use retrospective cost optimisation to identify a parametric model of the linear dynamic system. This method is demonstrated on several examples of increasing complexity.
Keywords :
control nonlinearities; cost optimal control; identification; linear systems; nonparametric statistics; stochastic processes; SISO Wiener system; linear dynamic system; nonparametric estimation; nonparametric map; nonparametric probabilistic method; parametric model; retrospective cost optimisation; semiparametric identification; single harmonic input; single-input single-output Wiener system; static nonlinearity;
Journal_Title :
Control Theory & Applications, IET
DOI :
10.1049/iet-cta.2010.0023
