Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 8 min ago
Peer-reviewedCC BY-NC-NDNature communications

Generative design of high-fidelity microstructures using physics-aware machine learning.

  • Weijie Liao1
  • Kaidi Li1,2
  • Bin Tang1,2
  • Jiangkun Fan1
  • Jun Wang1
  • Xiangyi Xue1
  • Jinshan Li3
  • Ruihao Yuan4
4 affiliations
  1. 1State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi'an, China.
  2. 2Chongqing Innovation Center, Northwestern Polytechnical University, Chongqing, China.
  3. 3State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi'an, China. [email protected].
  4. 4State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi'an, China. [email protected].
Read full textPDFDOICite

Abstract

Generative models are increasingly being applied in materials science to enable materials design. However, generating out-of-distribution microstructures with high fidelity that satisfy multiple target properties remains a major challenge. Here, we present a physics-aware generative-optimization strategy for the inverse design of microstructures, guided by mutually exclusive properties. The approach integrates interfacial and crystallographic information directly into the loss function of surrogate models, together with introducing a latent-space sampling scheme that balances microstructural diversity with physical realizability. We demonstrate the framework using a small yet diverse alloy dataset comprising well-characterized microstructures and tensile stress-strain responses. By targeting strength-ductility synergy, the generated out-of-distribution microstructures exhibit substantially improved fidelity, particularly in their interfacial features. The produced microstructures are validated by crystal plasticity finite element simulations and experiments. We attribute the enhanced mechanical performance to the emergence of a bimodal grain-size morphology absent from the training data. Our results demonstrate a data-efficient strategy for high-fidelity microstructure design that leverages interfacial physics to achieve multiple properties synergy.

Related

  1. Accelerating Proton Affinity Prediction with Multi-Fidelity Machine Learning
    Debjyoti Bhattacharya, Yifan Liu, Valentino R. Cooper +1
    ChemRxivPreprintLicence not statedGeneral chemistryChemRxiv
  2. Multi-fidelity machine learning interatomic potential for the energetics and finite temperature structure of Pt surfaces and nanoparticles
    Noor Aldin Alzghoul, Wo Bin Bae, Jun Hee Jang +1
    ChemRxivPreprintCC BYMaterials & polymersChemRxiv
  3. A physics-driven machine learning framework for predicting XRD patterns, magnetization, and magnetocaloric response in rare-earth-doped double perovskite oxides
    Sana Meftah, Emna Ammar Elhadjamor, Souhir Bouzidi +4
    RSC AdvancesPeer-reviewedCC BY-NCMaterials & polymersHAL
  4. Machine Learning-Empowered Electromagnetic Wave Absorbing Materials: From Forward Prediction to Generative Inverse Design
    Tongbaihui Qi, Jie Zhou
    MoleculesPeer-reviewedCC BY6 citedMaterials & polymersOpenAlex
  5. Physics-inspired feature engineering and interpretable machine learning for thermoelectric properties of multicomponent materials
    Thi Huynh Ho, Nguyen-Hieu Hoang, Thuat Thanh Trinh
    Materials & DesignPeer-reviewedCC BY5 citedMaterials & polymersOpenAlex

Cite

BibTeXRIS
BibTeX
@article{liao2026generative,
  title = {{Generative design of high-fidelity microstructures using physics-aware machine learning.}},
  author = {Weijie Liao and Kaidi Li and Bin Tang and Jiangkun Fan and Jun Wang and Xiangyi Xue and Jinshan Li and Ruihao Yuan},
  journal = {Nature communications},
  year = {2026},
  month = sep,
  doi = {10.1038/s41467-026-78118-3},
  url = {https://europepmc.org/article/MED/42957730},
  note = {CC BY-NC-ND},
}
RIS
TY  - JOUR
TI  - Generative design of high-fidelity microstructures using physics-aware machine learning.
AU  - Weijie Liao
AU  - Kaidi Li
AU  - Bin Tang
AU  - Jiangkun Fan
AU  - Jun Wang
AU  - Xiangyi Xue
AU  - Jinshan Li
AU  - Ruihao Yuan
PY  - 2026
DA  - 2026/09/30
JO  - Nature communications
DO  - 10.1038/s41467-026-78118-3
UR  - https://europepmc.org/article/MED/42957730
AB  - Generative models are increasingly being applied in materials science to enable materials design. However, generating out-of-distribution microstructures with high fidelity that satisfy multiple target properties remains a major challenge. Here, we present a physics-aware generative-optimization strategy for the inverse design of microstructures, guided by mutually exclusive properties. The approach integrates interfacial and crystallographic information directly into the loss function of surrogate models, together with introducing a latent-space sampling scheme that balances microstructural diversity with physical realizability. We demonstrate the framework using a small yet diverse alloy dataset comprising well-characterized microstructures and tensile stress-strain responses. By targeting strength-ductility synergy, the generated out-of-distribution microstructures exhibit substantially improved fidelity, particularly in their interfacial features. The produced microstructures are validated by crystal plasticity finite element simulations and experiments. We attribute the enhanced mechanical performance to the emergence of a bimodal grain-size morphology absent from the training data. Our results demonstrate a data-efficient strategy for high-fidelity microstructure design that leverages interfacial physics to achieve multiple properties synergy.
SN  - 2041-1723
ER  -