Extending DFDGM to Motion by Weighted Mean Curvature
Matt Elsey
University of Michigan
Department of Mathematics
melsey@umich.edu
University of Michigan
Department of Mathematics
melsey@umich.edu
Abstract: The distance function-based diffusion-generated
motion (DFDGM) algorithm for multiphase motion by curvature has been
demonstrated to be a powerful tool for very large-scale simulations of
isotropic grain growth and recrystallization. Here, we discuss an extension of
this algorithm for generating weighted motion by curvature while enforcing the
natural Herring angle conditions. This extension allows for simulations under
a simplified model for anisotropic grain growth. Some numerical convergence
results and preliminary simulation results will be presented.