| Research |
The Applied Numerical Algorithms Group develops numerical methods for the solution of partial differential equations in complex geometries. The equations of particular interest are those that arise from compressible and incompressible fluid dynamics. The representation of complex geometries is based on an embedded boundary model, in which the boundary of a complex domain cuts away at a regular Cartesian grid. This allows the use of highly efficient and well understood methods in the regions of the grid away from the boundary, and confines the use of irregular grid methods to the small region of the domain near the boundary.
We are developing 2D and 3D methods for the solution of Poisson's equation on complex geometries using Cartesian meshes with embedded boundaries and adaptive mesh refinement. The solution procedure is based on the multigrid-like AMRPoisson algorithm.
The embedded boundary Poisson solver is incorporated into a cell-centered projection method to simulate incompressible fluid flow. The advective parts of the method are similar to the embedded boundary gas dynamics method. This is a joint project with Gerry Puckett and Mark Sussman at UC Davis.
We are developing a projection method for low speed compressible flow, based on Karen Pao's all-speed method.
| ANAG | Research | Staff |
| Publications | Software | Links |