Skip to main content

An Adaptive Embedded Boundary Method for the Compressible Navier-Stokes Equations

Speaker: Daniel Graves, Lawrence Berkeley National Laboratory

Presentation

Seminar Slide Deck (PDF-1.4MB)

Abstract

Embedded boundary methods generate grids by cutting a Cartesian grid with the surfaces of the complex geometry. These methods have substantial advantages but also present some interesting algorithmic challenges. To illustrate both, this talk presents a new algorithm to solve the time-dependent compressible Navier-Stokes equations in the presence of complex geometry using an embedded boundary method. The method is strongly conservative and second order accurate in space and time. This algorithm will be used to highlight the algorithmic choices that embedded boundary methods involve. The talk will also present some of the rich set of tools for embedded boundary calculations that the Chombo numerical infrastructure provides. This set of tools includes very robust, accurate, and fast grid generation as well as high level tools to facilitate adaptivity.

Biography

Graves received his B. S. in Mechanical Engineering (1989) and his PhD degree from U. C. Berkeley (1996). He has been working for the Applied Numerical Algorithms Group at Lawrence Berkeley National Laboratory for 18 years as a research scientist. His research is in the area of numerical methods for partial differential equations, with contributions in the areas of adaptive mesh refinement, Cartesian grid embedded boundary methods, massively parallel computation and library design for scientific computing. He has contributed in algorithms for incompressible flow, shock physics, viscoelastic flows and elliptic solvers for magneto-hydrodynamics.