Research Group Computational PDEs
Finite element methods for flow problems and problems on moving and complex geometries — University of Göttingen.
Many physical processes are modelled by partial differential equations (PDEs), yet only rarely can these be solved analytically. We develop and analyse modern numerical methods — primarily finite element methods — that deliver reliable approximations efficiently, stably and in a structure-preserving way.
News & Events
All news →
PFUZ 
UZ 
Research
All topics →
Geometrically unfitted finite element methods
High-order accurate, robust finite element methods for PDEs on complex and moving geometries that are not resolved by the computational mesh.
Space-time finite element methods
Higher-order space-time discretizations for PDEs on moving domains — provably high-order accurate in space and time.
Trefftz and embedded Trefftz DG methods
A flexible, efficient route to Trefftz discontinuous Galerkin methods that drastically reduces the cost of DG while keeping its approximation power.
Structure-preserving discretizations for flows
Pressure-robust, exactly divergence-free H(div)-conforming finite element methods for incompressible flows.
Incompressible flows on curved surfaces
Structure-preserving higher-order discretizations for incompressible viscous and inviscid flows posed on general curved surfaces.
Hybrid Discontinuous Galerkin methods
Reducing the cost of discontinuous Galerkin methods through hybridization, enabling efficient high-order and structure-preserving discretizations.
Wave propagation, Galbrun's equation & helioseismology
Stable and efficient discretizations for time-harmonic wave propagation in stratified media, driven by the modelling of oscillations of the Sun.
Further topics
Further threads of the group — analog computing, reduced-order modeling, and open science / reproducible research (LiveDocs).