Metodologia para simulação numérica de estruturas soldadas via elementos finitos com Software Sysweld

• Autor(a) principal: Vieira, Leandro Coutinho
• Data de Publicação: 2008
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/14875
• topic: Elementos finitos; Tensões residuais; Deformações; Soldagem; Método dos elementos finitos; Deformações e tensões; Welding; Finite elements; Residual stress; Distortion; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA

The establishment of manufacturing process with high productivity and quality is a constant requirement of the industrial sector. A possible methodology for carrying on this need of process optimization can be achieved by numerical-computational simulation. Specifically for welding processes, recently developed commercial software has called the attention in both academic and industrial sectors: the Sysweld®. This software has the capacity of simulating different welding processes, including heat source movement, material deposition, transient material properties as a function of temperature, phase transformation and also thermal and superficial treatments. As result, Syweld® is capable of predicting temperature field, residual stress, distortion and phase proportion (microconstituents). However, despite the great possibility of the software usage, it demands an experimental calibration to assure confidentiality on the results. Therefore, within this context, a case study was performed in an automotive component nationally fabricated to delimitate a methodology for carrying out the numerical-computational simulation for the welding process using the commercial software Sysweld®. Simulations were done for the welds made in a part of the rear axle of a hatch car and the results are shown for the temperature field, generated distortion in the part, residual stresses and phase proportion. The thermal results and displacements were experimentally compared to measurements done by thermocouples and three-dimensional table in the partner industry. This comparison led to a poor correlation between numerical and experimental data. In a second stage, a numerical-statistical study was carried out for assessing the parameters influence (gap, thickness, heat source level and clamping system) on the residual stress, maximum displacement and cooling rate between 800°C e 500°C (ΔT8/5). The obtained results are in agreement with technical literature.