Development of a Multi-physics Code with Adaptive Mesh Refinement

Date and Time: 
2015 April 13 @ 3:30pm
FL2-1022 Large Auditorium
William Dai

We have been working on the Roxane project for a few years. The Roxane project covers a variety of physics in standard geometries of one, two, and three dimensions, including hydrodynamics, volume fraction material advection, material mixing, elastic-plastic strength models, magnetohydrodynamics, 3-T radiation diffusion, detonation shock dynamics, high explosive burn models, etc. Although there are other kinds of AMR used in calculations on structured meshes, we use cell-based AMR.

In this presentation, in addition to AMR, parallel strategy, and IO package, we will particularly describe a numerical approach for solving nonlinear 3-T radiation diffusion equations. For systems of multi-materials with dramatically different material properties, the correct treatment for the discontinuity of material properties is important. We use the governing physics to obtain the effective diffusion coefficient across a material interface for flux calculations on polyhedral meshes. Formulations good for steady states are important even for time-dependent problems for the systems. This often challenges the second order accuracy (in time) of numerical schemes. We applied a rare time-stepping technique to have the both properties. Another important aspect in numerical simulations for 3-T radiation equations is the numerical treatment for interaction between radiation and material. The 3-T radiation diffusion equations are often solved through operator splitting. In our approach radiation and material are fully coupled, and three temperatures are updated simultaneously.

Speaker Description: 

William Dai received his Ph.D. degree in physics in University of Minnesota in 1993. After that he joined Laboratory for Computational Science and Engineering as a research scientist in the University, focusing on numerical methods for hydrodynamics, magnetohydrodynamics, radiation, and diffusion. William joined the Los Alamos National Laboratory in 2001 as a staff member in High Performance Computing Division. In 2002 he became a team leader and project leader responsible for software development and their integration to several multi-physics codes. Currently, William is a scientist in Computer, Computational, and Statistical Sciences Division, and he is one of the key developers of a large-scale multi-physics code, responsible for new physics capabilities, numerical solvers, and modernization of the code on future computer platforms.

PDF icon SEA2015_Dai.pdf18.86 MB

Event Category: