Title :
An Optimized Generator of Finite Element Meshes Based on a Neural Network
Author :
Alfonzetti, Salvatore ; Dilettoso, Emanuele ; Salerno, Nunzio
Author_Institution :
DIEES, Univ. di Catania, Catania
fDate :
6/1/2008 12:00:00 AM
Abstract :
An automatic mesh generator of triangular finite elements, based on an artificial neural network, is optimized by means of Design of Experiments techniques and Genetic Algorithms. The optimization concerns the various parameters which are defined in the mesh generator, whereas the objective function is related to the mesh quality. By means of this optimization the generator allows us to obtain very high-quality meshes.
Keywords :
electromagnetic fields; genetic algorithms; mesh generation; neural nets; physics computing; artificial neural network; automatic mesh generator; design of experiment techniques; electromagnetic field problems; finite element meshes; genetic algorithms; optimized generator; triangular finite elements; Automatic mesh generation; finite element method; neural networks; stochastic optimization;
Journal_Title :
Magnetics, IEEE Transactions on
DOI :
10.1109/TMAG.2007.916035