Advanced
search

To read this content please select one of the options below:

Approximation of the objective function: multiquadrics versus neural networks

Th. Ebner (Institut für Grundlagen und Theorie der Elektrotechnik, Graz University of Technology, Graz, Austria, and)
Ch. Magele (Institut für Grundlagen und Theorie der Elektrotechnik, Graz University of Technology, Graz, Austria, and)
B.R. Brandstätter (Institut für Grundlagen und Theorie der Elektrotechnik, Graz University of Technology, Graz, Austria, and)
M. Luschin (Institut für Grundlagen und Theorie der Elektrotechnik, Graz University of Technology, Graz, Austria)
P.G. Alotto (Dipartimento di Ingegneria Elettrica, Università di Genova, Genova, Italy)

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering

ISSN: 0332-1649

Article publication date: 1 September 1999

204

Abstract

Global optimization in electrical engineering using stochastic methods requires usually a large amount of CPU time to locate the optimum, if the objective function is calculated either with the finite element method (FEM) or the boundary element method (BEM). One approach to reduce the number of FEM or BEM calls using neural networks and another one using multiquadric functions have been introduced recently. This paper compares the efficiency of both methods, which are applied to a couple of test problems and the results are discussed.

Keywords

Citation

Ebner, T., Magele, C., Brandstätter, B.R., Luschin, M. and Alotto, P.G. (1999), "Approximation of the objective function: multiquadrics versus neural networks", COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, Vol. 18 No. 3, pp. 250-265. https://doi.org/10.1108/03321649910274766

Publisher

:

MCB UP Ltd

Copyright © 1999, MCB UP Limited

Related articles

Manage cookies

We’re listening — tell us what you think

Join us on our journey