CNN model for predicting the transverse properties of uni-drectional composites

---

The repository contains the supporting data for the work titled "A generalised deep learning-based surrogate model for homogenisation utilising material property encoding and physics-based bounds" authored by Rajesh Nakka, Dineshkumar Harursampath and Sathiskumar A Ponnusami.

https://doi.org/10.1038/s41598-023-34823-3

---

• Data sets can be accessed on Kaggle or Zeonodo

• This repository contains

  • PyTorch scripts for building and training the models
  • Trained models

Cite as

@article{Nakka2023,
  doi = {10.1038/s41598-023-34823-3},
  url = {https://doi.org/10.1038/s41598-023-34823-3},
  year = {2023},
  month = jun,
  publisher = {Springer Science and Business Media {LLC}},
  volume = {13},
  number = {1},
  author = {Rajesh Nakka and Dineshkumar Harursampath and Sathiskumar A Ponnusami},
  title = {A generalised deep learning-based surrogate model for homogenisation utilising material property encoding and physics-based bounds},
  journal = {Scientific Reports}
}