High-Order Variable-Scaled Energetic Variational Neural Networks for Phase-Field Gradient Flows

DOI10.48550/arXiv.2609.25489MaRDI QIDQ7353486arXivFDO

Authors Xiaobo Jing, Leiyi Dong, Jia Zhao

Publication date 21 September 2026

Published in arXiv

Tldr We develop a high-order variable-scaled energetic variational neural network for phase-field gradient flows, where each stage's spatial inputs are normalized by the interface scale to adapt to evolving interfaces; this leads to lower solution errors compared to an unscaled baseline and higher-order schemes improve accuracy.

Full work available at URL https://doi.org/10.48550/arXiv.2609.25489

arXiv classification

math.NA

Cites work

• A Method for Representing Periodic Functions and Enforcing Exactly Periodic Boundary Conditions with Deep Neural Networks
• A new Lagrange multiplier approach for gradient flows
• A phase-field neural solver for moving contact line problems with dynamic boundary conditions
• An adaptive energy-based sequential method for training PINNs to solve gradient flow equations
• Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems
• Discrete Gradient Flow Approximations of High Dimensional Evolution Partial Differential Equations via Deep Neural Networks
• Energy Dissipation Preserving Physics Informed Neural Network for Allen-Cahn Equations
• Energy-dissipative evolutionary deep operator neural networks
• Free energy of a nonuniform system. I: Interfacial free energy
• Global Convergence and Error Propagation in Neural Gradient Flows: A Riemannian Optimization Framework
• Gradient flow based phase-field modeling using separable neural networks
• High order schemes for gradient flow with respect to a metric
• High order, semi-implicit, energy stable schemes for gradient flows
• Linear, first and second-order, unconditionally energy stable numerical schemes for the phase field model of homopolymer blends
• Maximum principle preserving exponential time differencing schemes for the nonlocal Allen-Cahn equation
• Numerical approximations of Allen-Cahn and Cahn-Hilliard equations
• PF-PINNs: physics-informed neural networks for solving coupled Allen-Cahn and Cahn-Hilliard phase field equations
• Phase-field DeepONet: physics-informed deep operator neural network for fast simulations of pattern formation governed by gradient flows of free-energy functionals
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Respecting causality for training physics-informed neural networks
• Solving Allen-Cahn and Cahn-Hilliard Equations using the Adaptive Physics Informed Neural Networks
• The deep minimizing movement scheme
• The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems
• The phase field method for geometric moving interfaces and their numerical approximations
• The scalar auxiliary variable (SAV) approach for gradient flows
• Variable linear transformation improved physics-informed neural networks to solve thin-layer flow problems
• Variational Extrapolation of Implicit Schemes for General Gradient Flows
• VS-PINN: A fast and efficient training of physics-informed neural networks using variable-scaling methods for solving PDEs with stiff behavior

This page was built for publication: High-Order Variable-Scaled Energetic Variational Neural Networks for Phase-Field Gradient Flows

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7353486)