Learning coupled Allen-Cahn and Cahn-Hilliard phase-field equations using Physics-informed neural operator(PINO)

TL;DR

This paper introduces a Physics-informed Neural Operator (PINO) approach to efficiently predict microstructural evolution in materials by solving coupled Allen-Cahn and Cahn-Hilliard equations, outperforming traditional numerical methods.

Contribution

The study demonstrates the use of PINO with Fourier derivatives to accurately and efficiently model coupled phase-field equations, reducing computational cost and complexity.

Findings

Fourier derivatives significantly improve loss accuracy for Cahn-Hilliard equations.

PINO can learn and solve coupled Allen-Cahn and Cahn-Hilliard equations simultaneously.

Fourier methods enable easy computation of higher derivatives in PINO.

Abstract

Phase-field equations, mostly solved numerically, are known for capturing the mesoscale microstructural evolution of a material. However, such numerical solvers are computationally expensive as it needs to generate fine mesh systems to solve the complex Partial Differential Equations(PDEs) with good accuracy. Therefore, we propose an alternative approach of predicting the microstructural evolution subjected to periodic boundary conditions using Physics informed Neural Operators (PINOs). In this study, we have demonstrated the capability of PINO to predict the growth of $\theta^{\prime}$ precipitates in Al-Cu alloys by learning the operator as well as by solving three coupled physics equations simultaneously. The coupling is of two second-order Allen-Cahn equation and one fourth-order Cahn-Hilliard equation. We also found that using Fourier derivatives(pseudo-spectral method and Fourier extension) instead of Finite Difference Method improved the Cahn-Hilliard equation loss by twelve orders of magnitude. Moreover, since differentiation is equivalent to multiplication in the Fourier domain, unlike Physics informed Neural Networks(PINNs), we can easily compute the fourth derivative of Cahn-Hilliard equation without converting it to coupled second order derivative.