Mechanical performance and prediction of a novel reinforced octagonal honeycomb
•A novel honeycomb (ROHC) with two plateau stress regions is proposed.•Parameter analysis proves ROHC's adjustability of two-stage mechanical performance.•The performance prediction method for ROHC is established by neural network.•The reverse design of ROHC is realized with small error. Honeyc...
Gespeichert in:
Veröffentlicht in: | International journal of mechanical sciences 2024-12, Vol.284, p.109758, Article 109758 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •A novel honeycomb (ROHC) with two plateau stress regions is proposed.•Parameter analysis proves ROHC's adjustability of two-stage mechanical performance.•The performance prediction method for ROHC is established by neural network.•The reverse design of ROHC is realized with small error.
Honeycomb structures are widely used in various engineering applications due to their lightweight and excellent energy absorption capabilities. Materials with two plateau stress regions exhibit unique advantages in multi-stage energy dissipation and multi-task applications. This paper presents a reinforced octagonal honeycomb structure (ROHC) inspired by the topology of an octagonal tensegrity structures. The paper demonstrates the phenomenon of two plateau deformation stages through the 3D printing of ROHC specimens and finite element simulation. By adjusting three geometric parameters of ROHC (angle α, length ratio r, and thickness t of internal reinforcement), the paper obtains the influence laws on the first and second plateau stress and strain, and proves the controllability of two-stage mechanical performance of ROHC. Based on deep learning technology, a performance prediction model for ROHC's two-stage mechanical performance is proposed, with the MSE and R values confirming the accuracy of the prediction model. Based on the prediction model, a rapid reverse design method is proposed, capable of designing structures with expected two-stage mechanical performance, with errors |
---|---|
ISSN: | 0020-7403 |
DOI: | 10.1016/j.ijmecsci.2024.109758 |