cs.LGNov 3, 2026

Surrogate Models for Stochastic Al-Cu Phase-Field Solidification

Authors: Simon Savukoski, Tatu Pinomaa

Organizations: VTT Technical Research Centre of Finland Ltd Espoo, Finland

Abstract

Phase-field models resolve microstructural evolution during rapid alloy solidification, but their computational cost limits long trajectories, stochastic ensembles, and studies across process conditions. We present a variance-preserving denoising diffusion bridge model (VP-DDBM) for state-conditioned prediction of two-dimensional Al-Cu solidification. Given the current phase, concentration, and prescribed temperature fields, the model predicts the subsequent state and is applied autoregressively. Across two held-out thermal gradients, VP-DDBM produces lower long-horizon errors than deterministic U-Net and SI-ConvNeXt baselines in domain-mean phase, domain-mean Cu concentration, and leading-tip position. It also retains fine dendritic structure over long rollouts, whereas the deterministic U-Net progressively smooths the morphology. VP-DDBM predicts the phase and concentration evolution approximately 73×73\times faster than the phase-field solver, showing potential for faster process-condition screening and ensemble-based microstructure studies.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Adaptive Uncertainty-Guided Surrogates for Efficient phase field Modeling of Dendritic Solidification

    Feb 17, 2026Eider Garate-Perez, Kerman López de Calle-Etxabe, Oihana Garcia +3Surrogate ModelingComputational Materials Science

  2. Deep Learning for Accelerated Long-Horizon Forecasting of Multicomponent Multiphase Microstructure Evolution in High-Entropy Alloys

    Jul 30, 2026Hamidreza Razavi, Nele MoelansNeural Surrogate ModelingMaterials Science

  3. Efficient and Scalable Physics-Guided Fully Convolutional Spatiotemporal Learning for 3D Microstructure Evolution Prediction

    Sep 29, 2026Michael Trimboli, Wenxi Liu, Xianqi LiPDE Surrogate ModelingPhysics-Informed ML