Estudo numérico sobre o comportamento de vigas de aço com almas esbeltas enrijecidas longitudinalmente

• Ano de defesa: 2019
• Autor(a) principal: José Osvaldo Ferreira Filho
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA ESTRUTURAS; Programa de Pós-Graduação em Engenharia de Estruturas; UFMG
• Palavras-chave em Português: Estruturas de aço; Vigas esbeltas; Almas enrijecidas; Análise de flambagem elástica; Análise avançada; Engenharia de estruturas; Aço - Estruturas; Análise numérica; Flambagem (Mecânica)
• Link de acesso: http://hdl.handle.net/1843/30928

The use of slender steel girders is a favorable alternative to optimize the design of bridges and viaducts because it provides high resistance coupled with low material consumption. However, special attention is needed to the buckling problems related to these structures. As a solution, stiffeners are commonly applied along plate girder web panels, increasing the resistance related to the buckling. There are several possibilities to compute the elastic critical stress of stiffened plates and determine the beginning of the structural instability. Despite the existence of practical methodologies, most part of them is inaccurate and limited, such as EN 1993-1-5:2006 code and EBPlate computer program. They are also responsible for oversizing and unnecessary expenses. Therefore, it was aimed to comprehend the critical and post-critical behavior of longitudinally stiffened web panels of slender steel I-girders. For this, a consistent numerical model was proposed, ensuring the reliability in correspondence to the real behavior of the structure. In this sense, a linear buckling analysis was performed to calculate the global critical stresses of plates stiffened by one, two or three longitudinal stiffeners and subjected to bending, shear and interaction of both. Furthermore, the reduction factor and the contribution from the web to the design shear resistance were measured by a geometrically and materially nonlinear analysis with initial imperfections (GMNIA). Then, the numerical model was confronted with the aforementioned approaches. In general, they set an inconsistent and overestimated analysis. The deviation surpassed 40% to the safe side. Moreover, the code even disregarded the presence of the second stiffener, leading to lower critical global stresses than in cases of plates stiffened by just one stiffener. At the end, this work resulted in a numerical approach capable to contribute for an in-depth study about the structural behavior of longitudinally stiffened plates, covering a large range of geometries with diverse stiffener configurations.