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

2026-09-28

AI Summary

This study addresses the prohibitive computational cost of high-fidelity phase-field simulations for three-dimensional microstructure evolution by proposing a physics-guided fully convolutional spatiotemporal prediction framework. Methodologically, we design a shared-encoding and decomposed latent translator architecture to directly generate complete 3D evolution sequences. Furthermore, a discrete physics regularization mechanism based on Cahn-Hilliard equation residuals is introduced to enforce kinetic constraints without incurring inference overhead. Experimental results demonstrate that the proposed framework achieves a 3D structural similarity exceeding 0.97 under nominal predictions and accelerates computation by more than 30 times compared to conventional spectral phase-field solvers, while significantly enhancing robustness in long-horizon forecasting.

Abstract

Accurate prediction of three-dimensional (3D) microstructure evolution remains computationally demanding because high-fidelity phase-field simulations require repeated numerical integration over large volumetric domains and long temporal horizons. This study develops an efficient and scalable physics-guided fully convolutional spatiotemporal framework for direct multi-frame prediction of complete 3D microstructure sequences. The model combines shared 3D spatial encoding and decoding with a factorized latent translator that integrates temporal, local 3D spatial, and channel interactions. A discrete Cahn--Hilliard (CH) residual is incorporated during training to regularize the learned evolution toward the governing dynamics without altering the inference pathway. The framework is evaluated on high-resolution 3D spinodal-decomposition trajectories under nominal, long-horizon, and reduced-temporal-context forecasting. Under full temporal context, the model accurately reproduces volumetric evolution, with average 3D structural similarity remaining above 0.97 over the nominal prediction horizon. Physics guidance becomes increasingly beneficial as temporal information is reduced, improving predictive robustness and preservation of interface-level morphology. The framework also achieves more than a 30-fold wall-clock speedup relative to the reference spectral phase-field solver, while physics guidance introduces no additional inference cost. These results establish direct multi-frame, physics-guided fully convolutional learning as a high-throughput surrogate strategy for dense 3D phase-field dynamics and repeated microstructure forecasting.

Problem

• 3D microstructure evolution
• phase-field simulation
• computational cost
• spatiotemporal prediction

Innovation

• Physics-guided learning
• Fully convolutional spatiotemporal framework
• 3D microstructure evolution
• Cahn-Hilliard residual
• Surrogate modeling