Physics-Guided Fully Convolutional Spatiotemporal Learning Toward Digital-Twin-Enabled Microstructure Evolution Prediction

2026-06-18

AI Summary

This work addresses the physical inconsistency and degraded long-term accuracy of purely data-driven models in predicting microstructural evolution by proposing a physics-guided fully convolutional spatiotemporal learning framework. The method explicitly embeds thermodynamic and kinetic governing equations as residual regularization terms into a self-supervised training objective, balancing data efficiency with physical fidelity. Leveraging a fully convolutional architecture and a physics-constrained loss function, the model enables efficient multiscale spatiotemporal modeling. Experimental results demonstrate that the proposed approach significantly outperforms data-driven baselines on spinodal decomposition tasks, accurately reproducing microstructural morphologies, statistical characteristics, and evolutionary trends while exhibiting strong cross-resolution generalization and stable long-horizon predictions.

Abstract

Understanding and predicting microstructure evolution is central to materials design, yet purely data-driven spatiotemporal learning models often suffer from limited physical consistency and degraded long-term prediction accuracy. In this work, we introduce a physics-guided fully convolutional spatiotemporal learning framework for microstructure evolution prediction. Unlike prior self-supervised approaches, the proposed method explicitly incorporates governing physical equations into the training objective, thereby encouraging the learned dynamics to remain consistent with known thermodynamic and kinetic laws. This physics-guided formulation improves predictive accuracy, long-horizon stability, and robustness across spatial resolutions and temporal prediction settings. Extensive experiments for spinodal decomposition demonstrate that incorporating physics-guided residual regularization leads to more faithful reproduction of microstructural morphology, statistics, and evolution trends compared with purely data-driven baselines. The proposed framework preserves the scalability and computational efficiency of fully convolutional architectures while bridging the gap between high-fidelity physics-based simulations and data-driven surrogate modeling, offering a reliable and efficient surrogate-modeling step toward digital-twin-enabled microstructure evolution prediction.

Problem

• microstructure evolution
• physical consistency
• long-term prediction accuracy
• data-driven modeling
• digital twin

Innovation

• physics-guided learning
• fully convolutional network
• microstructure evolution
• digital twin
• spatiotemporal prediction