Retrainable physics-integrated neural differentiable modeling of sintering across material systems
Abstract
Sintering is widely used to manufacture ceramics, but coupled densification and grain growth, material-dependent kinetics, and sparse measurements complicate predictive modeling and process design. We present Sinter-PiNDiff, a retrainable physics-integrated neural differentiable framework for predicting density and grain-size evolution. Two neural networks learn densification and grain-growth coefficients within coupled rate equations, while a smooth saturation factor attenuates densification near theoretical density. The same governing structure, network architecture, and training procedure were fitted independently to published data for MgO, Al-doped ZnO, and CaO-doped ThO2. Tests at held-out temperatures and compositions yielded the lowest mean error in all twelve material-metric comparisons against multilayer perceptron and residual network baselines. For MgO, Al-doped ZnO, and CaO-doped ThO2, respectively, density normalized root-mean-square errors were 14.6%, 10.8%, and 14.4%, and grain-size errors using the same metric were 8.6%, 12.1%, and 19.3%. Removing evolving density from both neural-network inputs increased density and grain-size trajectory errors in all three systems and ten of twelve aggregate errors, supporting density-dependent kinetic feedback. Deep ensembles estimated model disagreement, but empirical coverage showed that the uncertainty bands were not calibrated and did not capture all model-data discrepancies. These results establish Sinter-PiNDiff as a retrainable framework for sparse-data prediction and uncertainty-informed selection of sintering conditions.
Related
- Fully differentiable framework for inverse identification of geometry and material parameters with application to determining stress-free configuration of soft tissues
- Generalizable data-driven turbulence closure modeling on unstructured grids with differentiable physics
- Efficient angle-aware differentiable Mask3D modeling for inverse lithography
- A Differentiable Ray-Wave Framework for Hybrid Refractive-Diffractive System Modeling and Optimization
- Finite Element Modeling of the Active Muscle: Combined Influence of Boundary Conditions and Material Properties
BibTeX
@misc{chen2026retrainable,
title = {{Retrainable physics-integrated neural differentiable modeling of sintering across material systems}},
author = {Zeping Chen and Ani Aprahamian and Khachatur V. Manukyan and Tengfei Luo},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.31518},
eprint = {2609.31518},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.31518},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Retrainable physics-integrated neural differentiable modeling of sintering across material systems
AU - Zeping Chen
AU - Ani Aprahamian
AU - Khachatur V. Manukyan
AU - Tengfei Luo
PY - 2026
DA - 2026/09/25
PB - arXiv
DO - 10.48550/arxiv.2609.31518
UR - https://arxiv.org/abs/2609.31518
AB - Sintering is widely used to manufacture ceramics, but coupled densification and grain growth, material-dependent kinetics, and sparse measurements complicate predictive modeling and process design. We present Sinter-PiNDiff, a retrainable physics-integrated neural differentiable framework for predicting density and grain-size evolution. Two neural networks learn densification and grain-growth coefficients within coupled rate equations, while a smooth saturation factor attenuates densification near theoretical density. The same governing structure, network architecture, and training procedure were fitted independently to published data for MgO, Al-doped ZnO, and CaO-doped ThO2. Tests at held-out temperatures and compositions yielded the lowest mean error in all twelve material-metric comparisons against multilayer perceptron and residual network baselines. For MgO, Al-doped ZnO, and CaO-doped ThO2, respectively, density normalized root-mean-square errors were 14.6%, 10.8%, and 14.4%, and grain-size errors using the same metric were 8.6%, 12.1%, and 19.3%. Removing evolving density from both neural-network inputs increased density and grain-size trajectory errors in all three systems and ten of twelve aggregate errors, supporting density-dependent kinetic feedback. Deep ensembles estimated model disagreement, but empirical coverage showed that the uncertainty bands were not calibrated and did not capture all model-data discrepancies. These results establish Sinter-PiNDiff as a retrainable framework for sparse-data prediction and uncertainty-informed selection of sintering conditions.
ER -