Optimizing truss structures with natural frequency constraints using the cross-entropy method - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Optimizing truss structures with natural frequency constraints using the cross-entropy method

Résumé

This work addresses the challenging problem of structural optimization, specifically focusing on minimizing the mass of a structure while satisfying constraints related to natural frequencies. Traditional optimization methods that rely on gradient information are not suitable for such complex problems. To overcome this limitation, metaheuristic methods have emerged as effective alternatives. In this study, we propose a novel framework that employs the Cross-Entropy (CE) method, a powerful Monte Carlo technique for handling nonconvex optimization problems. By applying the CE method to optimize structural trusses as benchmark tests, we compare its performance with the genetic algorithm (GA). Through numerical experiments, we demonstrate that the CE method provides accurate and computationally efficient solutions, outperforming other metaheuristic optimization methods. This research highlights the efficacy of the CE method as a valuable tool for addressing non-convex structural optimization problems, enabling efficient solutions with natural frequency constraints.
Fichier principal
Vignette du fichier
ConferencePaper_2023_COBEM2023_2.pdf (381.64 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04373361 , version 1 (05-01-2024)

Licence

Copyright (Tous droits réservés)

Identifiants

  • HAL Id : hal-04373361 , version 1

Citer

Marcos Vinícius dos Santos Issa, Americo Cunha Jr, Francisco Soeiro, Anderson Pereira. Optimizing truss structures with natural frequency constraints using the cross-entropy method. 27th International Congress on Mechanical Engineering (COBEM 2023), Dec 2023, Florianópolis, Brazil. ⟨hal-04373361⟩

Collections

TDS-MACS
13 Consultations
6 Téléchargements

Partager

Gmail Facebook X LinkedIn More