OTIMIZAÇÃO TOPOLÓGICA DE ESTRUTURAS TRIDIMENSIONAIS USANDO MMA

Resumo

In structural optimization, one of the most fascinating fields is Topological Optimization (TO). In this article, the evolutionary topological optimization methods Smoothing Evolutionary Structural Optimization (SESO) and Sequential Element Rejection and Admission (SERA) are used to minimize the growth of compliance using the Method of Moving Asymptotes (MMA). The effect of changing asymptotes is to control the generation of subproblems, which can both stabilize and accelerate the convergence of the overall process. The three-dimensional topological optimization of two large-scale problems: a) a cantilever beam and  b) an MBB beam is realized. Therefore, to solve computational problems such as cost and excess memory, which occur in these types of problems, the Preconditioned Conjugated Gradient Method (PCG) is used as an iterative solver. The results obtained are compared with the Solid Isotropic Material with Penalization (SIMP) that uses for convergence the Optimality Criteria (OC).