Actions

Download Analytics Citations

Export to: EndNote  |  Zotero  |  Mendeley

Collections

This file is not currently in any collections.

Bayesian optimisation of resolved interlaminar stresses in composite laminates using high- aspect-ratio DG finite elements [dataset] Open Access

Direct optimisation of local interlaminar stresses in composite laminates is difficult because the relevant response is three-dimensional, localised and expensive to evaluate, while conventional laminate-design workflows usually rely on reduced models, design rules or post-design checks. This paper presents a coupled discontinuous Galerkin (DG) finite-element and Bayesian op- timisation (BO) framework for treating local interlaminar-stress severity as the optimisation objective itself. Candidate laminates are chosen from a finite material-and-fibre-angle library and evaluated with a full 3D DG finite-element solver that uses high-aspect-ratio meshes to concentrate refinement near monitored stress regions. Scalar patch objectives are constructed from the resolved interlaminar normal and shear stresses, including hardmax, smoothmax and monotonic power/log transforms, to test how objective representation affects BO performance without changing the underlying candidate laminate set. The framework is assessed on two local-stress benchmarks: a straight free-edge laminate and a rectangular-cutout laminate with a geometry-induced corner stress concentration. Against a matched genetic-algorithm baseline, BO consistently achieves higher exact-optimum reliability and better final ranks. In the free-edge benchmark, the best-performing transformed objectives allow BO to identify the global optimum in every run across the tested budgets, while GA needs about seven times more evaluations on average to identify the same optimum. In the cutout benchmark, the best-performing transformed objective identifies the global optimum in every BO run after evaluating only about 1.2% of the feasible design space; GA needs about eight times more evaluations. The results show that DG– BO can make resolved local stress objectives searchable within limited high-fidelity evaluation budgets.

Descriptions

Resource type

Dataset

Contributors

Creator: Navvabi, Soheil ¹
Creator: Giani, Stefano ¹
Creator: Coombs, William ¹

¹ Durham University, UK

Funder

Engineering and Physical Sciences Research Council

Research methods

Other description

Keyword

Composite laminates
Stacking-sequence optimisation
Interlaminar stress
Local stress optimisation
Bayesian optimisation
High-fidelity optimisation
Discontinuous Galerkin

Subject

Laminated plastics
Galerkin methods

Location

Language

Cited in

Identifier

ark:/32150/r2g445cd22w
doi:10.15128/r2g445cd22w

Rights

Creative Commons Attribution 4.0 International (CC BY)

Publisher

Durham University

Date Created

File Details

Depositor

S. Navvabi

Date Uploaded

7 June 2026, 14:06:00

Date Modified

8 June 2026, 09:06:26

Audit Status

Audits have not yet been run on this file.

Characterization

File format: zip (ZIP Format)

Mime type: application/zip

File size: 79236

Last modified: 2026:06:07 15:36:45+01:00

Filename: dg_bo_laminate_optimisation_data_availability.zip

Original checksum: ca67f1cba9601e5fb86fbd52e253c25c