Otimização paramétrica de estruturas reticuladas do tipo pórticos tridimensionais utilizando algoritmo TLBO modificado

Abstract

With the intense techno-scientific development experienced in the last decades, currently there are countless techniques and methods to optimize structures. One of them is parametric optimization, which has, among other advantages, ease of implementation and adaptability to specific projects. Structural optimization is a tool that assists the designer in the confiability and guarantee of good project results, thus becoming an important ally of the engineer in the analysis and decision-making process. In the present work, the objective is to optimize the weight of tridimensional frames reticulated structures, which are widely used in structures of mechanical components, such as chassis, towers, cranes, etc. Another important point when it comes to structural optimization is regarding the computational routine used. In the present work, the Modified TLBO was used, a meta-heuristic algorithm that is based on the interaction between students and teachers for the transmission of knowledge and thus improving the aptitude of the entire project group. The computational implementation was performed using MatLab textsuperscript textregistered software, and its verification was done by contrasting the results obtained with those available in the literature. It is important to emphasize that the work, although it deals with reticulated structures - which are well disseminated in the literature, models them as a frame, and not a truss, as most of the time it is done. This opens up the possibility of opposing the distinctions of results, at times modeling a structure as one, now as another. The results obtained show that, because they consider flexion and torsion in the elements - efforts hitherto disregarded as a truss, frames are more prone to failure, therefore requiring a correct dimensioning and, in most cases, resulting in structures with greater weight. Finally, all the results of verification and authorship are shown and discussed, and the appropriate conclusions and considerations are made, proving the consistency not only of the results, but of the entire methodology adopted. In addition, suggestions for future work are made along the same lines.