Identification of nonlinear systems with time-varying parameters using a sliding-neural network observer

Authors:
Tarek Ahmed-Ali;Godpromesse Kenné;Françoise Lamnabhi-Lagarrigue
Affiliations:
Ecole Nationale Supérieure des Ingénieurs des Etudes et Techniques d'Armement (ENSIETA), 2 Rue François Verny, 29806 Brest Cedex 9, France;Laboratoire d'Automatique et d'Informatique Appliquée (LAIA), Département de Génie ílectrique, IUT FOTSO Victor, Université de Dschang, B.P. 134 Bandjoun, Cameroun;Laboratoire des Signaux et Systémes (L2S), C.N.R.S-SUPELEC, Université Paris XI, 3 Rue Joliot Curie, 91192 Gif-sur-Yvette, France
Venue:
Neurocomputing
Year:
2009

Citing 3
Cited 0

Algorithms for clustering data

Algorithms for clustering data
Robust exact differentiation via sliding mode technique

Automatica (Journal of IFAC)
Approximation of nonlinear systems with radial basis function neural networks

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.01

Visualization

Abstract

In this paper, a new method for the identification of nonlinear systems with time-varying parameters using a sliding-neural network observer is investigated. The proof of the finite-time convergence of the estimates to their true values is achieved using Lyapunov arguments and sliding mode theories. An application example illustrated the effectiveness of the approach and the obtained results show high convergence rate and very satisfactory parameter estimation accuracy. The computing results under noisy condition also demonstrate that good state and parameter estimation can be achieved despite the disturbance (noise) in the system. The reduced number of hidden units and the small transient period demonstrate that the proposed method can be easily implementable in real-time.