A nonlinear observer design via differential neural networks for coupled tanks

Author

Sharifi, Farid ; Menhaj, Mohammad Bagher

Author_Institution

Electr. Eng. Dept., Amirkabir Univ. of Technol., Tehran

fYear

2008

fDate

4-7 May 2008

Abstract

A nonlinear observer for a general class of dynamical nonlinear systems is proposed based on a Differential Neural Networks (DNN). A learning rule to adjust the parameters of the neuro-observer is designed. By a Lyapunov-like analysis, the upper bounds for the weights and the averaged estimation error are derived. To illustrate the applicability of the DNN observer, we consider a coupled tank as a case study. The results of simulation are very promising.

Keywords

Lyapunov methods; control system synthesis; flow control; level control; neurocontrollers; nonlinear control systems; nonlinear dynamical systems; observers; stability; tanks (containers); Lyapunov method; coupled tank; differential neural network; dynamical nonlinear system; nonlinear observer design; stability; Artificial neural networks; Couplings; Estimation error; Lyapunov method; Neural networks; Nonlinear dynamical systems; Nonlinear systems; Observers; State estimation; Uncertainty; Coupled Tanks; Differential Neural Networks; Dynamic Neural Networks; Nonlinear observer; State Estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical and Computer Engineering, 2008. CCECE 2008. Canadian Conference on

Conference_Location

Niagara Falls, ON

ISSN

0840-7789

Print_ISBN

978-1-4244-1642-4

Electronic_ISBN

0840-7789

Type

conf

DOI

10.1109/CCECE.2008.4564783

Filename

4564783