Non-linear mechanical field reconstruction coupling recurrent neural networks with physics-informed graph neural networks
arXiv:2606.10909v1 Announce Type: cross Abstract: Reconstructing local stress fields in heterogeneous microstructures under non-linear, history-dependent loading remains a major computational bottleneck in multi-scale simulations. We propose a coupled LSTM-GNN framework that links the temporal and spatial aspects of local stress field reconstruction. A Long Short-Term Memory network encodes macroscopic stress-strain sequences into a compact hidden state that captures the path-dependent constitutive response, while a physics-informed Graph Neural Network reconstructs the spatially-resolved stre
The convergence of advanced AI (LSTM, GNNs) with complex physics simulations represents a current frontier in artificial intelligence research, enabled by increased computational power.
This development proposes a method to significantly reduce computational bottlenecks in multi-scale material simulations, which is critical for advanced materials design, engineering, and manufacturing.
The ability to rapidly and accurately reconstruct stress fields will accelerate the design and testing of new materials, improving efficiency and potentially reducing costs in various engineering fields.
- · Materials Science and Engineering
- · Manufacturing sector
- · AI/ML researchers
- · High-performance computing
- · Traditional simulation software relying solely on classical finite element metho
Faster development cycles for new materials with tailored properties.
Reduced need for extensive physical prototyping and destructive testing in some applications.
Potential for entirely new classes of materials or designs previously computationally infeasible.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG