Identification of a dynamical system with input nonlinearity

Parameter bounds for discrete-time hammerstein models with bounded output errors

“…1, where the nonlinear block maps the input signal u t into the unmeasurable inner variable x t through the following nonlinear function: In line with the work done by a number of authors, we assume that: 1) the linear system is asymptotically stable (see, e.g., [12], [18], and [28]- [30]); 2) nb j=0 bj 6 = 0, that is, the steady-state gain is not zero (see, e.g. [28]- [30]); and 3) the only a priori information needed is an estimate of the process settling-time (see, e.g., [31]). Let y t be the noise-corrupted output yt = wt + t:…”

Identification of Nonlinear Cascade Systems with Time-Varying Backlash

“…Unfortunately, there are only few contributions in the literature on the identification of systems with hard nonlinearities [1,30,34] and even fewer on backlash identification [3,12,13,29,38]. To the authors best knowledge, no contribution can be found which addresses the identification of cascade systems with time-varying input backlash.…”

Identification of cascade systems with backlash

“…Unfortunately, there are only few contributions in the literature on the identification of systems with hard nonlinearities including input backlash (Tao and Canudas de Wit 1997;Vo¨ro¨s 1997;Sun, Liu, and Sano 1999;Bai 2002;Cerone and Regruto 2007;Giri, Rochdi, Chaoui, and Brouri 2008). One often used approach how to overcome the problems with the backlash in control systems is based on its replacing by the deadzone (Aguilar, Orlov, Cadiou, and Merzouki 2007;Djeziri, Merzouki, Ould-Bouamama, and Dauphin-Tanguy 2007;Merzouki, Davila Montoya, Fridman, and Cadiou 2007a;Merzouki, Medjaher, Djeziri, and Ould-Bouamama 2007b).…”