Abstract:As one kind important phase field equations, Cahn-Hilliard equations contain spatial high order derivatives, strong nonlinearities, and even singularities. When using the physics informed neural network (PINN) to simulate the long time evolution, it is necessary to decompose the time domain to capture the transition of solutions in different time. Moreover, the baseline PINN can't maintain the mass conservation property for the equations. We propose a mass-preserving spatio-temporal adaptive PINN. This method adaptively dividing the time domain according to the rate of energy decrease, and solves the Cahn-Hilliard equation in each time step using an independent neural network. To improve the prediction accuracy, spatial adaptive sampling is employed in the subdomain to select points with large residual value and add them to the training samples. Additionally, a mass constraint is added to the loss function to compensate the mass degradation problem of the PINN method in solving the Cahn-Hilliard equations. The mass-preserving spatio-temporal adaptive PINN is employed to solve a series of numerical examples. These include the Cahn-Hilliard equations with different bulk potentials, the three dimensional Cahn-Hilliard equation with singularities, and the set of Cahn-Hilliard equations. The numerical results demonstrate the effectiveness of the proposed algorithm.

What problem does this paper attempt to address?

This paper attempts to solve several key problems encountered in the long - time evolution simulation of the Cahn - Hilliard equation using the Physics - Informed Neural Network (PINN). Specifically, these problems include: 1. **Strong Non - linearity and Singularity**: The Cahn - Hilliard equation contains high - order spatial derivatives, strong non - linear terms, and possible singularities. These characteristics make it difficult for traditional numerical methods and standard PINN methods to accurately capture the behavior of the solution. 2. **Mass Conservation Problem**: The standard PINN method cannot guarantee the mass conservation property of the solution, which is an important physical property of the Cahn - Hilliard equation. Non - conservation of mass will lead to a discrepancy between the predicted results and the actual physical phenomena, thus affecting the accuracy of the model. 3. **Time - Domain Decomposition**: In order to capture the changes of the solution in different time periods, it is necessary to divide the time domain reasonably. Especially in the time period of rapid change, the standard uniform time step may lead to insufficient accuracy. To solve the above problems, the author proposes a **mass - preserving spatio - temporal adaptive PINN method**. The main improvements of this method include: - **Spatio - temporal Adaptive Sampling**: The time domain is adaptively divided according to the energy decay rate, and the neural network is independently trained in each sub - domain. In addition, an adaptive sampling strategy is adopted in space, and points with larger residual values are selected as training samples to improve the training effect. - **Mass Constraint**: A mass conservation constraint term \( L_m \) is added to the loss function to ensure the mass conservation of the predicted solution. The specific formula is as follows: \[ L_m(t, \theta) = \frac{1}{N_t} \sum_{i = 1}^{N_t} \left| m(\hat{u}_\theta, t_i^m) - m(u, 0) \right|^2 \] where \( m(\hat{u}_\theta, t_i^m) \) represents the mass at time \( t_i^m \), and the calculation method is: \[ m(\hat{u}_\theta, t_i^m) = \int_\Omega \hat{u}_\theta(x, t_i^m) \, dx \approx \sum_{j = 1}^{N_q} \omega_j \hat{u}_\theta(x_j^q, t_i^m) \] Through these improvements, this method can more accurately simulate the long - time evolution behavior of the Cahn - Hilliard equation while maintaining mass conservation, especially in dealing with singularities and complex boundary conditions.

Mass-preserving Spatio-temporal adaptive PINN for Cahn-Hilliard equations with strong nonlinearity and singularity

Enhancing PINNs for solving PDEs via adaptive collocation point movement and adaptive loss weighting

A Physics Informed Neural Network for Time-Dependent Nonlinear and Higher Order Partial Differential Equations

Physics-informed neural networks with parameter asymptotic strategy for learning singularly perturbed convection-dominated problem

Structure Preserving PINN for Solving Time Dependent PDEs with Periodic Boundary

Physics-informed neural networks based on adaptive weighted loss functions for Hamilton-Jacobi equations

NH-PINN: Neural homogenization based physics-informed neural network for multiscale problems

A Spatial-Temporally Adaptive PINN Framework for 3D Bi-Ventricular Electrophysiological Simulations and Parameter Inference

TCAS-PINN: Physics-informed neural networks with a novel temporal causality-based adaptive sampling method

Solution of conservative-form transport equations with physics-informed neural network

Pre-training strategy for solving evolution equations based on physics-informed neural networks

Improvement of Bayesian PINN Training Convergence in Solving Multi-scale PDEs with Noise

Pre-Training Physics-Informed Neural Network with Mixed Sampling and Its Application in High-Dimensional Systems

Gradient auxiliary physics-informed neural network for nonlinear biharmonic equation

A Priori Error Estimation of Physics-Informed Neural Networks Solving Allen--Cahn and Cahn--Hilliard Equations

The robust physics-informed neural networks for a typical fourth-order phase field model

Physical informed neural networks with soft and hard boundary constraints for solving advection-diffusion equations using Fourier expansions

Generalized conditional symmetry enhanced physics-informed neural network and application to the forward and inverse problems of nonlinear diffusion equations

Semi-analytic PINN methods for singularly perturbed boundary value problems

MRF-PINN: A Multi-Receptive-Field convolutional physics-informed neural network for solving partial differential equations