Physics-informed operator learning for transferable energy-dissipative microstructure dynamics

TL;DR

This paper introduces PFNet, a physics-informed neural operator that efficiently predicts microstructure evolution in phase-field simulations, enabling transferable surrogates for complex energy-dissipative systems.

Contribution

PFNet combines a diffusion-inspired U-Net with thermodynamic modulation to accurately model microstructure dynamics across various parameters without retraining.

Findings

PFNet achieves accurate one-step predictions for Cahn-Hilliard coarsening.

PFNet maintains stability in autoregressive rollouts across different morphologies.

The framework extends effectively to martensitic-transformation benchmarks.

Abstract

Phase-field simulations provide mechanistic descriptions of microstructure evolution, but repeated high-fidelity integration over long horizons and broad parameter spaces remains computationally expensive. We present PFNet, a physics-informed neural operator framework that advances microstructural states by learning conditional evolution operators rather than direct correlations. PFNet combines a diffusion-inspired U-Net with periodic padding, entropy-based state conditioning and thermodynamic-parameter modulation to encode boundary consistency, instantaneous ordering state and changes in the free-energy landscape. For Cahn-Hilliard coarsening, PFNet achieves accurate one-step prediction and stable autoregressive rollouts across composition, gradient-energy coefficient, coarsening stage and morphology class, with errors concentrated near diffuse interfaces and topology-changing regions. The same framework extends to a four-channel martensitic-transformation benchmark without martensite-specific redesign. These results indicate that physics-informed operator learning can provide transferable surrogates for phase-field dynamics and broader energy-dissipative dynamical systems.