Saved in:
Bibliographic Details
Main Authors: Zhu, Bokai, Zhang, Qinghui, Rabczuk, Timon
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.24651
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911713516847104
author Zhu, Bokai
Zhang, Qinghui
Rabczuk, Timon
author_facet Zhu, Bokai
Zhang, Qinghui
Rabczuk, Timon
contents We propose a Weak-form Physics-Informed Neural Operator (WINO), a data-free framework that combines the efficiency of neural operators with the geometric flexibility of the $φ$-finite element method ($φ$-FEM). $φ$-FEM is an unfitted method that accommodates geometric variations without body-fitted meshes, where the domain geometry is represented by the level-set function $φ$. To impose the boundary conditions, Dirichlet problems adopt the $φ$-FEM lifting so only the homogeneous displacement contribution is learned, whereas traction-driven Neumann problems additionally predict the auxiliary fields necessary for the unfitted weak formulation. Parameters are trained by minimizing squared weak-form residuals aligned with $φ$-FEM together with squared penalties on the cut-cell auxiliary equations, which removes the need for large paired datasets of converged reference solutions. After training, WINO outputs can seed the nonlinear $φ$-FEM solvers as neural operator warm starts (NOWS), which reduce iteration counts relative to traditional cold-started solvers. Numerical benchmarks show that WINO achieves high accuracy below 0.04 across all benchmarks, while reducing total computational time by 50--80\% compared with purely data-driven methods.
format Preprint
id arxiv_https___arxiv_org_abs_2605_24651
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle WINO: A Weak-Form Physics Informed Neural Operator for Hyperelasticity on Variable Domains
Zhu, Bokai
Zhang, Qinghui
Rabczuk, Timon
Numerical Analysis
Machine Learning
We propose a Weak-form Physics-Informed Neural Operator (WINO), a data-free framework that combines the efficiency of neural operators with the geometric flexibility of the $φ$-finite element method ($φ$-FEM). $φ$-FEM is an unfitted method that accommodates geometric variations without body-fitted meshes, where the domain geometry is represented by the level-set function $φ$. To impose the boundary conditions, Dirichlet problems adopt the $φ$-FEM lifting so only the homogeneous displacement contribution is learned, whereas traction-driven Neumann problems additionally predict the auxiliary fields necessary for the unfitted weak formulation. Parameters are trained by minimizing squared weak-form residuals aligned with $φ$-FEM together with squared penalties on the cut-cell auxiliary equations, which removes the need for large paired datasets of converged reference solutions. After training, WINO outputs can seed the nonlinear $φ$-FEM solvers as neural operator warm starts (NOWS), which reduce iteration counts relative to traditional cold-started solvers. Numerical benchmarks show that WINO achieves high accuracy below 0.04 across all benchmarks, while reducing total computational time by 50--80\% compared with purely data-driven methods.
title WINO: A Weak-Form Physics Informed Neural Operator for Hyperelasticity on Variable Domains
topic Numerical Analysis
Machine Learning
url https://arxiv.org/abs/2605.24651