Predictive Design of a Multilayered Laminate Shell Based on AI and 1st-Order Classical Laminate Theory

Sofiene Helaili, Taysir Rezgui, Fehmi Najar, Moez Chafra

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The design of multilayered laminate shells and plates requires a very specific approach. This is a multistage approach that begins with a pre-design work to select the number of layers, the layer thickness, and the layer reinforcement type depending on the application and based on previous works. After that, the predesign is subject to verification, and the structural element made of multilayered laminates is checked. The loads, the load combinations, and the structural calculation make it possible to determine the loads to which the structural element would be subjected: the bending moments, the shear forces, and the concentrated forces. The shells are structural elements mainly subject to bending moments, but their dimensioning also depends on the nature of the supports: fixed, articulated, or a combination of them. In this paper, a pre-design approach using artificial intelligence has been developed. A database of laminated shells was generated based on input parameters. Criteria related to resistance and deformation made it possible to calculate an objective function to determine the optimal design. This same base was used to teach an artificial intelligence algorithm to predict the optimal design. The algorithm has been tested, and the results are predictive to some degree of accuracy.

Original languageEnglish
Title of host publicationAdvances in Mechanical Engineering and Mechanics III - Selected Papers from the 6th Tunisian Congress on Mechanics, CoTuMe 2023
EditorsTarak Bouraoui, Slah Mzali, Naoufel Ben Moussa, Farhat Zemzemi, Tarek Benameur, Nizar Aifaoui, Amna Znaidi, Ridha Ennetta, Fathi Djemal
PublisherSpringer Science and Business Media Deutschland GmbH
Pages147-153
Number of pages7
ISBN (Print)9783031704277
DOIs
StatePublished - 2024
Event6th Tunisian Congress on Mechanics, CoTuMe 2023 - Monastir, Tunisia
Duration: 17 Mar 202319 Mar 2023

Publication series

NameLecture Notes in Mechanical Engineering
ISSN (Print)2195-4356
ISSN (Electronic)2195-4364

Conference

Conference6th Tunisian Congress on Mechanics, CoTuMe 2023
Country/TerritoryTunisia
CityMonastir
Period17/03/2319/03/23

Keywords

  • AI
  • CLT theory
  • Multilayered laminate

Fingerprint

Dive into the research topics of 'Predictive Design of a Multilayered Laminate Shell Based on AI and 1st-Order Classical Laminate Theory'. Together they form a unique fingerprint.

Cite this