Analysis of crack propagation in the wedge splitting test via digital image correlation and finite element analyses

• Ano de defesa: 2019
• Autor(a) principal: Maginador, Rafael Vargas
• Orientador(a): Canto, Rodrigo Bresciani
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência e Engenharia de Materiais - PPGCEM
• Palavras-chave em Português: Método da cunha; Correlação de imagens digitais; Atualização de modelo de elementos finitos; Elementos coesivos; Zona do processo de fratura
• Palavras-chave em Inglês: Wedge splitting test; Digital image correlation; Finite element model updating; Crack propagation; Cohesive elements; Fracture process zone
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA DE MATERIAIS E METALURGICA::MATERIAIS NAO METALICOS
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/11465

Castable refractories are applied in dangerous environments in which it is unpractical to avoid crack initiation. Therefore, their microstructure is usually complex in order to hinder crack propagation by toughening mechanisms. One way to quantify such mechanisms is calculating the work of fracture via the so-called Wedge Splitting Test (WST), which allows for stable crack growth. However, this method is experimentally laborious and the usual result is only one parameter per test. The stability of the propagating crack leads to opportunities for additional analyses to better understand the fracture process. The present work focuses on extracting more information from one single WST. Parameters relevant to linear elastic fracture mechanics (i.e., crack tip positions during the propagation, stress intensity factors, and T-stress) and insight into the Fracture Process Zone (FPZ) described by various cohesive zone models were sought as well. Three different numerical methodologies are compared on the same experiment, namely integrated digital image correlation, and two finite element model updating methods. The first one uses analytical fields accounting for the presence of a crack in an elastic medium. The second and third ones consist in updating finite element models using cost functions on the displacement fields and on the experimental force, respectively. Slightly different crack tip positions were found by the three methods. Stress intensity factors and T-stresses were also calculated. Using cohesive elements, it was possible to further analyze the FPZ length, which was found not to be fully developed in the test studied herein. This result proves that the examined material has the ability of bridging rather long cracks.