Overview

The finite element method turns equations for physical systems—such as solid deformation, fluid flow, and heat transfer—into numerical problems on a mesh of small elements. It uses weak or variational formulations, specialized meshes, and iterative solvers to handle complex geometries, material behavior, and boundary conditions. Current work also uses differentiable solvers, neural surrogates, and automated modeling tools to speed up simulations and inverse problems, with applications ranging from engineering design to soft robotics. A central challenge is gaining speed and automation without sacrificing accuracy, stability, and reliability on unfamiliar geometries or difficult nonlinear problems.

Latest papers

CardsList

35 papers2 papers in the last four weeks, down 33% on the four weeks before. 0.0% of all new papers.

  1. PDEAgent-Bench: A Multi-Metric, Multi-Library Benchmark for PDE Solver Generation

    May 10, 2026Zhen Hang, Yushan Yashengjiang, Junhui Li +21PDE SolvingBenchmark Design

  2. GRIFDIR: Graph Resolution-Invariant FEM Diffusion Models in Function Spaces over Irregular Domains

    May 5, 2026James Rowbottom, Elizabeth L. Baker, Nick Huang +3Graph Neural OperatorsFinite Element Method

  3. Applying Two-Grid Preconditioner for Subsurface Flow Simulation using Attention-enhanced Hybrid Network to Accelerate Multiscale Discretization in High-contrast Media

    Apr 15, 2026Peiqi Li, Jie Chen, Shubin FuPorous Media FlowFinite Element Method

  4. Physics-informed post-processing of stabilized finite element solutions for transient convection-dominated problems

    Mar 3, 2026Süleyman Cengizci, Ömür Uğur, Srinivasan NatesanPDE SolvingPhysics-Informed ML

  5. A hybrid FEM-PINN method for time-dependent partial differential equations

    Sep 4, 2024Xiaodong Feng, Haojiong Shangguan, Tao Tang +1PDE SolvingPhysics-Informed ML