Probabilistic Deep Learning Framework for Phase Transformation Forecasting aided by In Situ High temperature Microscopy
Abstract
The mechanical performance of structural materials is governed by their microstructure, which is itself set by the thermal history imposed during processing. Predicting how that microstructure evolves along an arbitrary thermal trajectory, however, remains a challenging problem. Conventional continuous-cooling-transformation diagrams, for instance, provide only a static description, and they do not exist for all materials, especially emerging and novel ones. Deep-learning approaches address this issue, but they treat image-based characterization, temporal prediction, and microstructure forecasting as separate tasks. Here, we introduce a probabilistic deep-learning framework that unifies these tasks for in situ high-temperature confocal laser scanning microscopy data. The initial surface image is compressed with a beta-variational autoencoder into a low-dimensional latent representation, which, together with the cooling rate and the full temperature history, is passed to a temporal fusion transformer that produces a quantile forecast of the Widmanstatten ferrite ratio. In parallel, a Gaussian process predicts the end-state Widmanstatten ferrite ratio and corrects the forecast equilibrium level, yielding probabilistic prediction intervals. We demonstrate the efficacy of our method on Widmanstatten ferrite evolution in S235 steel across five cooling regimes and showcase significant accuracy. This verifies the relevance of our framework and paves the way for a streamlined and highly accelerated investigation of material evolution under thermal treatment.
Related
- Thermally Induced Structural Transformation in a Two-Dimensional Conductive Metal-Organic Framework
- fix uvt and fix pimd/uvt: A Unified LAMMPS Framework for Constant-Potential Constant-Temperature Molecular Dynamics
- Room temperature synthesis of novel azine-linked covalent organic framework: molecular docking, theoretical investigation studies, and antimicrobial activity
- Finite-temperature adiabatic-connection reduced density-matrix functional theory: a unified framework for i-DMFT and TAO-DFT
- Predicting Glass Transition Temperature of Polyimide Homopolymers and Copolymers Using Interpretable QSPR-Based Machine Learning: The MFKit Framework
BibTeX
@misc{kouroudis2026probabilistic,
title = {{Probabilistic Deep Learning Framework for Phase Transformation Forecasting aided by In Situ High temperature Microscopy}},
author = {Ioannis Kouroudis and Niki Balestrieri and Stefan Rotzsche and Niccolo Radice and Peter Mayr and Alessio Gagliardi},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.16940},
eprint = {2609.16940},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.16940},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Probabilistic Deep Learning Framework for Phase Transformation Forecasting aided by In Situ High temperature Microscopy
AU - Ioannis Kouroudis
AU - Niki Balestrieri
AU - Stefan Rotzsche
AU - Niccolo Radice
AU - Peter Mayr
AU - Alessio Gagliardi
PY - 2026
DA - 2026/09/15
PB - arXiv
DO - 10.48550/arxiv.2609.16940
UR - https://arxiv.org/abs/2609.16940
AB - The mechanical performance of structural materials is governed by their microstructure, which is itself set by the thermal history imposed during processing. Predicting how that microstructure evolves along an arbitrary thermal trajectory, however, remains a challenging problem. Conventional continuous-cooling-transformation diagrams, for instance, provide only a static description, and they do not exist for all materials, especially emerging and novel ones. Deep-learning approaches address this issue, but they treat image-based characterization, temporal prediction, and microstructure forecasting as separate tasks. Here, we introduce a probabilistic deep-learning framework that unifies these tasks for in situ high-temperature confocal laser scanning microscopy data. The initial surface image is compressed with a beta-variational autoencoder into a low-dimensional latent representation, which, together with the cooling rate and the full temperature history, is passed to a temporal fusion transformer that produces a quantile forecast of the Widmanstatten ferrite ratio. In parallel, a Gaussian process predicts the end-state Widmanstatten ferrite ratio and corrects the forecast equilibrium level, yielding probabilistic prediction intervals. We demonstrate the efficacy of our method on Widmanstatten ferrite evolution in S235 steel across five cooling regimes and showcase significant accuracy. This verifies the relevance of our framework and paves the way for a streamlined and highly accelerated investigation of material evolution under thermal treatment.
ER -