Publications

Journal Articles


Toward signed distance function based metamaterial design: Neural operator transformer for forward prediction and diffusion model for inverse design

Published in Computer Methods in Applied Mechanics and Engineering, 2025

Liu, Qibang; Koric, Seid; Abueidda, Diab; Meidani, Hadi; Geubelle, Philippe (2025). "Toward signed distance function based metamaterial design: Neural operator transformer for forward prediction and diffusion model for inverse design." Computer Methods in Applied Mechanics and Engineering, 446:118316. DOI: 10.1016/j.cma.2025.118316.
Download Paper

Univariate conditional variational autoencoder for morphogenic pattern design in frontal polymerization-based manufacturing

Published in Computer Methods in Applied Mechanics and Engineering, 2025

Liu, Qibang; Cai, Pengfei; Abueidda, Diab; Vyas, Sagar; Koric, Seid; Gomez-Bombarelli, Rafael; Geubelle, Philippe (2025). "Univariate conditional variational autoencoder for morphogenic pattern design in frontal polymerization-based manufacturing." Computer Methods in Applied Mechanics and Engineering, 438:117848. DOI: 10.1016/j.cma.2025.117848.
Download Paper

Towards long rollout of neural operators with local attention and flow matching-inspired correction: An example in frontal polymerization PDEs

Published in Machine Learning and the Physical Sciences Workshop @ NeurIPS 2024, 2024

Cai, Pengfei; Liu, Sulin; Liu, Qibang; Geubelle, Philippe H.; Gomez-Bombarelli, Rafael. "Towards long rollout of neural operators with local attention and flow matching-inspired correction: An example in frontal polymerization PDEs." Machine Learning and the Physical Sciences Workshop @ NeurIPS 2024.
Download Paper

A non-linear material model for progressive damage analysis of woven composites using a conformal meshing framework

Published in Journal of Composite Materials, 2022

<!– ## Abstract A progressive damage model is presented to investigate the damage and failure behaviors of woven composites. The conformal finite element mesh for the woven composites, used for this analysis, is generated using fabric micro-geometry from DFCA (Digital Element Approach (DEA) Fabric and Composite Analyzer, Kansas State University). The mesh generation strategy is discussed briefly – it does not make any idealized assumptions about yarn geometry and thus can be used for any woven composite micro-geometry. The woven composite domain consists of homogeneous and transversely isotropic yarns, and the surrounding homogeneous isotropic matrix.

Mazumder, Agniprobho; Liu, Qibang; Wang, Youqi; Yen, Chian-Fong (2022). "A non-linear material model for progressive damage analysis of woven composites using a conformal meshing framework." Journal of Composite Materials. DOI: 10.1177/00219983221113615.
Download Paper

Conference Papers


Sequential Deep Operator Neural Networks for Thermomechanical Modeling of Steel Solidification with Multi-inputs

Published in International Symposium on Innovative and Interdisciplinary Applications of Advanced Technologies, 2025

This work introduces Sequential Deep Operator Neural Networks (DeepONets) for thermomechanical modeling of steel solidification, leveraging multi-inputs for accurate and efficient predictions.

Koric, Seid; Liu, Qibang; Abueidda, Diab (2025). "Sequential Deep Operator Neural Networks for Thermomechanical Modeling of Steel Solidification with Multi-inputs." International Symposium on Innovative and Interdisciplinary Applications of Advanced Technologies, Springer, pp. 561–573. DOI: 10.1007/978-3-032-05159-2_35.
Download Paper