Development of topology optimization methodologies for a modular simulation platform

• Autor(a) principal: Gonçalves, Mafalda Pereira
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/31314
• topic: Topology optimization; Design-dependent loads; Thermoelasticity; Software development; MATLAB

Structural optimization has gained popularity since its first studies in the late 19th century. Over the years, due to the improvement of technology, several works have been focused on its computational implementation. Among the most popular applications, topology optimization deals with the non-homogeneous material distribution in a structure in order to optimize a given structural objective. A compliance approach is usually carried out to evaluate a topology optimization problem. Moreover, it is also considered the Optimality Criterion and SIMP as the optimization method and the material interpolation scheme, respectively. In this work, several topology optimization problems are carried out and evaluated, from a multi-objective approach, where thermal and mechanical analyses are simultaneously considered, to thermoelastic phenomena. These problems are recognized for incorporating loads that depend on the solution (in this case, thermal loads). Also known as design-dependent loads, they depend on the material layout inside the structure and their magnitude has a direct impact on the optimization process. Therefore, in the resultant topologies, the instabilities associated with this type of loading become evident. The main focus of this work consists in introducing alternative ways to prevent these issues and deal with the problems’ instability. Therefore, an alternative procedure is proposed to control the problems that arise from the mentioned analysis. An adaptation of the Evolutionary Structural Optimization (ESO) method, also known as Bi-directional ESO, is implemented and the obtained results are compared with the conventional ones. The development of a computational tool consists in an additional outcome of this work and, therefore, the mentioned methodologies are implemented considering a numerical simulation software, based on the Finite Element Method (FEM), as background. Besides the FEM analysis, the computational tool becomes capable of solving different types of topology optimization problems