Mechanics-Informed Machine Learning Using Classical Lamination Theory-Derived Data for Screening and Ranking Carbon/Basalt/Glass Hybrid Laminates

Hybrid composite laminates provide broad design flexibility, but their mechanical response depends on coupled effects of ply material, stacking sequence, ply number, ply position, and fiber volume fraction. This study presents a mechanics-informed machine learning framework for the preliminary screening and ranking of symmetric carbon/basalt/glass epoxy hybrid laminates. Symmetric 4-, 6-, and 8-ply configurations were generated using unidirectional carbon, basalt, and glass plies. Classical Lamination Theory was used to calculate tensile modulus, flexural modulus, a lamina-level longitudinal tensile strength indicator, nominal first-failure stress, and tensile resilience at selected fiber volume fractions. The resulting analytical dataset was used to train Support Vector Regression, Multi-Layer Perceptron, and Gaussian Process Regression models after Bayesian hyperparameter optimization. On the held-out analytical test set, Gaussian Process Regression reproduced the CLT-derived targets most closely, with an overall R² of 0.991. The trained model was then used for laminate ranking, feature-importance analysis, and Monte Carlo-based design-space exploration. Carbon-rich laminates showed the highest overall performance, while selected carbon–glass hybrids offered competitive balanced responses. The framework provides a physically interpretable tool for screening hybrid laminate designs before experimental validation.

Authors

Institutions

Publication Details

Journal
Materials Research Express
Published
2026-11-20
DOI
https://doi.org/10.1088/2053-1591/aec544
Primary Topic
Mechanical Behavior of Composites
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Mechanics-Informed Machine Learning Using Classical Lamination Theory-Derived Data for Screening and Ranking Carbon/Basalt/Glass Hybrid Laminates

İhsan Uluocak, Ahmet Çetin, Sefa Yıldırım
4 links
Materials Research Express
Mechanical Behavior of Composites
article

Mechanics-Informed Machine Learning Using Classical Lamination Theory-Derived Data for Screening and Ranking Carbon/Basalt/Glass Hybrid Laminates

İhsan Uluocak, Ahmet Çetin, Sefa Yıldırım
article en

Abstract

Hybrid composite laminates provide broad design flexibility, but their mechanical response depends on coupled effects of ply material, stacking sequence, ply number, ply position, and fiber volume fraction. This study presents a mechanics-informed machine learning framework for the preliminary screening and ranking of symmetric carbon/basalt/glass epoxy hybrid laminates. Symmetric 4-, 6-, and 8-ply configurations were generated using unidirectional carbon, basalt, and glass plies. Classical Lamination Theory was used to calculate tensile modulus, flexural modulus, a lamina-level longitudinal tensile strength indicator, nominal first-failure stress, and tensile resilience at selected fiber volume fractions. The resulting analytical dataset was used to train Support Vector Regression, Multi-Layer Perceptron, and Gaussian Process Regression models after Bayesian hyperparameter optimization. On the held-out analytical test set, Gaussian Process Regression reproduced the CLT-derived targets most closely, with an overall R² of 0.991. The trained model was then used for laminate ranking, feature-importance analysis, and Monte Carlo-based design-space exploration. Carbon-rich laminates showed the highest overall performance, while selected carbon–glass hybrids offered competitive balanced responses. The framework provides a physically interpretable tool for screening hybrid laminate designs before experimental validation.

Materials Research Express
Alanya Alaaddin Keykubat Üniversitesi (TR), Cukurova University (TR)
Mechanical Behavior of Composites
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon