Simulación estructural de espumas de aluminio a partir de imágenes 2D mediante la combinación de técnicas de homogeneización y machine learning

[EN] The use of resistant, rigid, low-weight materials with good both acoustic and thermal properties is very interesting intoday¿s industry. Among these materials, one can find aluminium foams, whose mechanical behaviour is necessaryfor their application. In order to obtain the geometry of an alumi...

ver descrição completa

Detalhes bibliográficos
Autores: Ferrándiz-Catalá, Borja, Tur Valiente, Manuel|||0000-0001-7683-4771, Nadal, Enrique|||0000-0002-2808-298X
Formato: artículo
Fecha de publicación:2018
País:España
Recursos:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:español
OAI Identifier:oai:riunet.upv.es:10251/123279
Acesso em linha:https://riunet.upv.es/handle/10251/123279
Access Level:acceso abierto
Palavra-chave:Homogeneización
Espuma de aluminio
Red neuronal
Machine learning
Homogenization
Aluminium foam
Neural network
INGENIERIA MECANICA
Descrição
Resumo:[EN] The use of resistant, rigid, low-weight materials with good both acoustic and thermal properties is very interesting intoday¿s industry. Among these materials, one can find aluminium foams, whose mechanical behaviour is necessaryfor their application. In order to obtain the geometry of an aluminium foam, several techniques can be applied, and allof them are based in the fact that information is initially obtained by a Computed Axial Tomography (CAT). One ofthese techniques, known as segmentation, involves a CAD being generated from an image in order to build the FiniteElement (FE) model. Another option is to use techniques such as CutFEM or cgFEM, in which a certain amount ofpixels, which define the properties of the material, are embedded in each element. Among the existing methods forevaluating the material properties matrix, this study proposes the use of homogenization techniques, sped up by the useof machine learning techniques. This method has been applied to real problems obtaining a high speed up, conservingprecision.