Assessing the influence of material and geometrical uncertainty on the mechanical behavior of functionally graded material plates

• Autor(a) principal: Carvalho, Alda
• Data de Publicação: 2016
• Outros Autores: Silva, Tiago, Loja, Amélia, Damásio, Fábio Raimundo
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.21/8948
• topic: Finite element analysis; Functionally graded particulate composites; Multiple linear regression; Parametric uncertainty characterization; Structural analysis

Composite materials possessing a functional gradient are becoming strong candidates to enhance the performance of structures when severe operating conditions are a reality. These types of conditions may, for example, range from situations where a high thermal gradient is present to others where it is imperative to minimize abrupt stresses transitions between material interfaces. The manufacturing achievement of the gradients determined for a specific application may in practice face some limitations, which can be due, among other factors, to technological process constraints, eventual operating condition deterioration of production stages, or to nonconforming raw materials. Regardless of the origin of such limitations, the reality is that the uncertainty is always present to some extent; this is clearly reflected in the scattering of material and geometrical properties of these composites. The understanding that deterministic analyses are not enough to provide a complete prediction of the composite structures’ behavior emphasizes the crucial need to identify the effects that the variability in material and geometrical parameters will produce in the structural response.With the presentwork, one intends to study the influence of this variability in the static and free vibrations behavior of functionally graded plates. It is also an objective of this study to use regression models to predict these responses and to characterize the contribution of each model parameter to the explanation of the response variability. To this purpose, a set of numerical results is presented and discussed.