Otimização topológica evolucionária de estruturas submetidas à multiplos casos de carregamento

Abstract

Structural optimization has received ever increasing attention in all engineering areas and it has been identified as the most challenging task in structural design. As a result of high speed computers, several topology optimization methods have been developed, including those that apply heuristic procedures in the analyses and act in an evolutionary manner. In this context, it is intended to develop, implement and validate a methodology for Evolutionary Topology Optimization applied for structural components by using efficient computational techniques. This methodology represents an evolutional procedure in which the finite elements with low levels of stress are progressively removed from the structure. This paper will propose some optimization examples applied to structures with elastoplastic behavior under dynamic conditions in order to evaluate the influence of the damping ratio variation in the oscillatory motion and consequently in the outcome of the optimal solution. The implemented algorithm is also able to perform optimization studies for structures under multiple load cases, where by using the proposed method it will be possible to combine different types of analyses, which may be considered either linear or non-linear analyses under either static or dynamic conditions. The implementation was accomplished through an algorithm in Python programming language, due to its interface with the analysis software for multiphysics problems Abaqus® . The effectiveness of this new approach by evolutionary procedure optimization will be demonstrated through the comparison with classical examples from literature. Some numerical examples not contained in the literature have also been added and validated by commercial softwares.