carrier image

Spatially Accelerated Shape Embedding in Multimaterial Simulations

Weiss, Kenneth, George Zagaris, Robert Rieben, Andrew Cook

Research Note, 25th International Meshing Roundtable, Sandia National Laboratories, September 26-30 2016

INTERNATIONAL
MESHING
ROUNTABLE

25th International Meshing Roundtable
Washington DC, U.S.A.
September 26-30, 2016

Kenneth Weiss, Lawrence Livermore National Laboratory, US, kweiss@llnl.gov
George Zagaris, Lawrence Livermore National Laboratory, US, zagaris2@llnl.gov
Robert Rieben, Lawrence Livermore National Laboratory, US, rieben1@llnl.gov
Andrew Cook, Lawrence Livermore National Laboratory, US, awcook@llnl.gov

Research Note Abstract
Multimaterial numerical simulations typically embed the shape of their materials into the computational mesh by determining the volume fractions of each material within the mesh elements, which requires the means to numerically describe the material boundaries. In this paper, we present a mesh-agnostic technique for generating and querying an implicit function defining the containment field of a surface with complex geometric boundaries. Specifically, given a closed, oriented surface representing a material boundary, we construct an In/Out octree to accelerate point containment queries which can efficiently determine whether an arbitrary point in space is enclosed by the surface. We apply this technique to initialize material volume fractions in the elements of a multimaterial high-order finite element Arbitrary Lagrangian-Eulerian (ALE) hydrodynamics code.

Download Full Paper (PDF)


Contact author(s) or publisher for availability and copyright information on above referenced article