Learning two-phase microstructure evolution using neural operators and autoencoder architectures

TL;DR

This paper introduces a neural operator and autoencoder-based framework to efficiently predict two-phase microstructure evolution, significantly reducing computational costs compared to traditional phase-field modeling.

Contribution

The authors develop a novel combined autoencoder and DeepONet architecture to learn and accelerate microstructure evolution predictions from phase-field models.

Findings

Achieved accurate microstructure predictions with reduced computational time.

Successfully reconstructed microstructures from low-dimensional latent representations.

Demonstrated potential for surrogate modeling in materials design and optimization.

Abstract

Phase-field modeling is an effective but computationally expensive method for capturing the mesoscale morphological and microstructure evolution in materials. Hence, fast and generalizable surrogate models are needed to alleviate the cost of computationally taxing processes such as in optimization and design of materials. The intrinsic discontinuous nature of the physical phenomena incurred by the presence of sharp phase boundaries makes the training of the surrogate model cumbersome. We develop a framework that integrates a convolutional autoencoder architecture with a deep neural operator (DeepONet) to learn the dynamic evolution of a two-phase mixture and accelerate time-to-solution in predicting the microstructure evolution. We utilize the convolutional autoencoder to provide a compact representation of the microstructure data in a low-dimensional latent space. DeepONet, which consists of two sub-networks, one for encoding the input function at a fixed number of sensors locations (branch net) and another for encoding the locations for the output functions (trunk net), learns the mesoscale dynamics of the microstructure evolution from the autoencoder latent space. The decoder part of the convolutional autoencoder then reconstructs the time-evolved microstructure from the DeepONet predictions. The trained DeepONet architecture can then be used to replace the high-fidelity phase-field numerical solver in interpolation tasks or to accelerate the numerical solver in extrapolation tasks.