320 Advances in Coupled ALE-AMR Methods

Jacob Waltz, Los Alamos National Laboratory
 
During the last few years a surge in activity has ocurred in numerical methods that combine moving meshes, either Lagrangian or arbitrary Lagrangian-Eulerian (ALE), with adaptive mesh refinement (AMR). These coupled ALE-AMR schemes  represent a significant advance in simulation capability due to their ability to tailor the relative amounts of mesh motion and mesh adaption. Example applications include high speed compressible flows, multi-material shock hydrodynamics, free surface flows, and unsteady aerodynamics. Coupled ALE-AMR methods also create unique challenges having to do with mathematically consistent coupling between ALE and AMR, error estimation, effective use of supercomputing hardware, and practical application.

The goal of this minisymposium is to bring together researchers who develop and/or apply coupled ALE-AMR methods and thereby advance the state-of-the-art in this area. Abstracts are solicited that address some or all of the following topics:

- Mathematical theory and methods for the coupling of ALE and AMR

- Implementation of coupled ALE-AMR methods in parallel computing environments, and in particular approaches to dynamic load balancing

- Error estimation techniques that account for both local flow features and variations in mesh size due to mesh motion

- Applications of coupled ALE-AMR methods that go beyond what can be achieved with ALE or AMR alone

Although coupled ALE-AMR methods have been developed primarily in the area of unstructured grids, this minisymposium is not intended to be limited to certain types of grids or discretization schemes. Contributions from a range of discretization techniques (finite difference, finite volume, finite element) as well as application areas (compressible flows, incompressible flows, solid dynamics, high-energy density physics, etc) are therefore encouraged.

Keywords: arbitrary Lagrangian-Eulerian; adaptive mesh refinement; error estimation; dynamic load balancing