Determination of the critical length of elastic buckling by distortion in Z-profiles stiffened under centered compression using artificial neural networks

Autores

  • Tailanne S. M. Silva
  • Leonardo P. Silva
  • Patrícia dos S. Andrade
  • Anderson de S. M. Gadéa
  • Koji de J. Nagahama

Palavras-chave:

Artificial Neural Networks, cold formed profiles, critical length

Resumo

NBR 14762 (ABNT, 2010) does not have a defined method to determine the critical distortional
buckling stress, making it difficult for designers to verify the PFF strength. Because of this, researches were
developed in order to solve this problem. However, there are difficulties in the methods developed in the literature.
An alternative to overcome such difficulties is the use of the critical length associated with the distortional mode
in order to help in the determination of the distortionary critical elastic stress. In this sense, this article aimed to
determine an equation for the critical length associated with the distortional mode of cold-formed profiles with Z
section stiffened under centered compression and with bi-articulated support with free warping. For this, Artificial
Neural Networks (ANNs) were trained. Such ANNs were validated using experimental and numerical results
available in the literature and some equations were generated. The equations obtained through the adopted ANN
obtained good correlations with the literature.

Downloads

Publicado

2024-05-30

Edição

Seção

Artigos