Detail publikace

On-line Identification Based on Neural Networks using of Levenberg-Marquardt Method and Back-propagation Algorithm

PIVOŇKA, P., DOHNAL, J.

Originální název

On-line Identification Based on Neural Networks using of Levenberg-Marquardt Method and Back-propagation Algorithm

Typ

článek v časopise - ostatní, Jost

Jazyk

angličtina

Originální abstrakt

In order to be able to control a system by means of an automatically adaptable controller, we have to adjust its parameters to the changes taking place in the system. For this purpose a number of methods have been developed. A frequently used identification method is the least-squares method, but we are limited by the choice of the suitable sampling period. The neural network seems to be a desirable solution because of its adaptation characteristics. Among the most used learning algorithms belong the Levenberg-Marquardt and back-propagation.

Klíčová slova

Levenberg-Marquardt, Back-propagation, Neural network, Training set

Autoři

PIVOŇKA, P., DOHNAL, J.

Rok RIV

2004

Vydáno

25. 3. 2004

ISSN

1109-2777

Periodikum

WSEAS Transactions on Systems

Ročník

3

Číslo

2

Stát

Spojené státy americké

Strany od

381

Strany do

355

Strany počet

5

BibTex

@article{BUT41839,
  author="Petr {Pivoňka} and Jiří {Dohnal}",
  title="On-line Identification Based on Neural Networks using of Levenberg-Marquardt Method and Back-propagation Algorithm",
  journal="WSEAS Transactions on Systems",
  year="2004",
  volume="3",
  number="2",
  pages="381--355",
  issn="1109-2777"
}