Evaluation of non-linear behavior of timber–concrete composite structures using FE model

• Autor(a) principal: Lopes, S.
• Data de Publicação: 2011
• Outros Autores: Jorge, L., Cruz, H.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://repositorio.lnec.pt:8080/jspui/handle/123456789/1003425
• topic: Timber–concrete floors; Composite structures; Timber structures; Beams; Computational analysis

This article describes a numerical model that was developed for the analysis of composite timber–concrete beams. This model presents a simplified methodology for determining the effective bending stiffness of the timber–concrete composite structure. It is based on previous work done usually referred to in some non-normative literature by c-method. The implemented methodology assumes some simplifications, as for instance, linear elastic behavior of all components, constant stiffness of the connection and sinusoidal loading. For comparison purposes, the work benefits from an experimental program in which full-scale beams were tested in bending and timber–concrete connections were tested in shear. The FE model has shown the ability to overcome the simplifications of the Eurocode, namely the variation of shear force along the beam axis. The numerical model is capable of detecting and quantifying the influence of the non-linear behavior of the connections on the composite structure. Different parameters are analyzed and, for instance, the ductility behavior of the timber–concrete connection could be more important than the maximumstrength,which is an interesting result. By comparing theoretical predictions with test results, it is clear that the numerical model used in this work is a very interesting method when compared with the usual design models, such as that of Annex B of Eurocode 5 (EN 1995-1-1). The influence of the connections behavior on the ultimate load of the composite structure is very important and the described approach proved to give good predictions.