A new neuro-based state estimator

Author

Menhaj, Mohammad B. ; Rajaei, Farzad

Author_Institution

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

Volume

3

fYear

2002

fDate

4-7 Aug. 2002

Abstract

This paper presents a new neuro-computing approach to the problem of state estimation. The neuro-estimator is established by a hybrid combination of a Hopfield network and a feedforward multilayer neural net. The former is very useful for optimization and the latter is well known for function approximation. This neuro-estimator is very appropriate for real-time implementation of nonlinear state estimators especially when modeling uncertainty is included.

Keywords

Hopfield neural nets; convergence of numerical methods; feedforward neural nets; function approximation; nonlinear dynamical systems; optimisation; state estimation; Hopfield network; convergence rate; feedforward multilayer neural net; function approximation; hybrid nonlinear state estimators; modeling uncertainty; neuro-based state estimator; neuro-computing; optimization; real-time estimator implementation; state estimation; Feedforward neural networks; Filtering; Function approximation; Hopfield neural networks; Multi-layer neural network; Neural networks; State estimation; State-space methods; Uncertainty; Working environment noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 2002. MWSCAS-2002. The 2002 45th Midwest Symposium on

Print_ISBN

0-7803-7523-8

Type

conf

DOI

10.1109/MWSCAS.2002.1186985

Filename

1186985