Partition of Unity Data-driven Finite Element Method for Anisotropic Diffusion in Magnetized Plasmas

Numerical simulation of transport processes in magnetized plasmas is characterized by extreme anisotropy, where diffusion along magnetic field lines can be many orders of magnitude larger than in the perpendicular direction. This anisotropy leads to thin, steep boundary layers and numerical pollution that standard finite element methods struggle to resolve without prohibitive mesh refinement. In this paper, we introduce the Partition of Unity Data-Driven Finite Element Method (PU-DDFEM), a data-driven local model reduction approach, for anisotropic diffusion. The method utilizes a bottom-up construction to enable a complete offline-online splitting. We parametrize the local magnetic field and boundary conditions to generate a snapshot space that captures the essential physics of plasma confinement. Dominant modes are extracted via a local spectral problem to identify representative features. These modes are then mapped to the global domain using a partition of unity to form a conforming approximation space. This ensures that the basis functions are fully reusable across different global magnetic configurations, significantly reducing the online computational overhead for large-scale plasma simulations.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-12
DOI
https://doi.org/10.48550/arxiv.2611.17086
Primary Topic
Model Reduction and Neural Networks
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Partition of Unity Data-driven Finite Element Method for Anisotropic Diffusion in Magnetized Plasmas

Youngsoo Choi, Siu Wun Cheung
arXiv (Cornell University)
Model Reduction and Neural Networks
preprint

Partition of Unity Data-driven Finite Element Method for Anisotropic Diffusion in Magnetized Plasmas

Youngsoo Choi, Siu Wun Cheung
preprint en

Abstract

Numerical simulation of transport processes in magnetized plasmas is characterized by extreme anisotropy, where diffusion along magnetic field lines can be many orders of magnitude larger than in the perpendicular direction. This anisotropy leads to thin, steep boundary layers and numerical pollution that standard finite element methods struggle to resolve without prohibitive mesh refinement. In this paper, we introduce the Partition of Unity Data-Driven Finite Element Method (PU-DDFEM), a data-driven local model reduction approach, for anisotropic diffusion. The method utilizes a bottom-up construction to enable a complete offline-online splitting. We parametrize the local magnetic field and boundary conditions to generate a snapshot space that captures the essential physics of plasma confinement. Dominant modes are extracted via a local spectral problem to identify representative features. These modes are then mapped to the global domain using a partition of unity to form a conforming approximation space. This ensures that the basis functions are fully reusable across different global magnetic configurations, significantly reducing the online computational overhead for large-scale plasma simulations.

arXiv (Cornell University)
Model Reduction and Neural Networks
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon