Simulation of masonry arch experimentation using 3D finite element analysis
In the design world, there exists two spaces. One space is the physical domain, and the other, the numerical domain. Researchers, engineers and mathematicians alike have all been trying to bridge the gap between these two domains. In the case of the engineering field, this motivation is d riven by t...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2012 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/121896 |
| Acceso en línea: | https://hdl.handle.net/2117/121896 |
| Access Level: | acceso abierto |
| Palabra clave: | Finite element method Structural design Masonry bridges Elements finits, Mètode dels -- Anàlisi numèrica Disseny d'estructures Ponts de maçoneria Àrees temàtiques de la UPC::Enginyeria civil::Materials i estructures |
| Sumario: | In the design world, there exists two spaces. One space is the physical domain, and the other, the numerical domain. Researchers, engineers and mathematicians alike have all been trying to bridge the gap between these two domains. In the case of the engineering field, this motivation is d riven by the requirement to understand the exact responses of a structure so they can be confident their designs will succeed. The Finite Element Method is a huge leap in goal of bridging these two domains together. Since its implementation in the study of structural responses, electromagnetic fields and flow theory, several analysts and researchers have come up with new innovations in order to improve the method and have the numerical results approach the realistic responses. This dissertation is a study to determine how close an experiment on a scale model masonry bridge can be simulated. A scale model segmental arch bridge (3.2m span, 1 m thick) was loaded until failure in the structural laboratory at the Universitat de Politecnica de Catalunya in 2001. The results of the experiment were recorded and were used as a source of comparison for models generated with finite element software. Originally this dissertation was too being a continuation of a previous dissertation done on the same experiment. However, the models were not able to be verified so it became necessary to generate new models. For generation of the new models, an innovative interface insertion technique was applied to ensure the correct implementation of the interface elements. The 3D model was completed in stages. At each stage, the results were compared with both those of the previous dissertation and the experimental data. Once the results were verified, the next model was generated. The first model was a plane stress model fully supported with only one layer of soil. The second model incorporated five layers of soil and a steel wall and tie lateral constraint system. The third model was a 3D model of half of the actual bridge (one symmetrical half).The first two models yielded acceptable results when compared to the previous models and the experimental data. The third model, however, did not possess the expected structural responses. This is attributed to the complex nature of modeling 3D structures due to the multitude of variables involved. Due to obtaining acceptable results from the predecessor models, it is possible that with continued work on the 3D model, it will yield acceptable results as well. It will be a matter of manipulating variables such as the nonlinear properties of the masonry and of the interfaces. |
|---|