dc.contributor.author | Molemaker, Jeroen | en_US |
dc.contributor.author | Cohen, Jonathan M. | en_US |
dc.contributor.author | Patel, Sanjit | en_US |
dc.contributor.author | Noh, Jonyong | en_US |
dc.contributor.editor | Markus Gross and Doug James | en_US |
dc.date.accessioned | 2014-01-29T07:36:56Z | |
dc.date.available | 2014-01-29T07:36:56Z | |
dc.date.issued | 2008 | en_US |
dc.identifier.isbn | 978-3-905674-10-1 | en_US |
dc.identifier.issn | 1727-5288 | en_US |
dc.identifier.uri | http://dx.doi.org/10.2312/SCA/SCA08/009-018 | en_US |
dc.description.abstract | We present a combination of techniques to simulate turbulent fluid flows in 3D. Flow in a complex domain is modeled using a regular rectilinear grid with a finite-difference solution to the incompressible Navier-Stokes equations. We propose the use of the QUICK advection algorithm over a globally high resolution grid. To calculate pressure over the grid, we introduce the Iterated Orthogonal Projection (IOP) framework. In IOP a series of orthogonal projections ensures that multiple conditions such as non-divergence and boundary conditions arising through complex domains shapes or moving objects will be satisfied simultaneously to specified accuracy. This framework allows us to use a simple and highly efficient multigrid method to enforce non-divergence in combination with complex domain boundary conditions. IOP is amenable to GPU implementation, resulting in over an order of magnitude improvement over a CPU-based solver. We analyze the impact of these algorithms on the turbulent energy cascade in simulated fluid flows and the resulting visual quality | en_US |
dc.publisher | The Eurographics Association | en_US |
dc.title | Low Viscosity Flow Simulations for Animation | en_US |
dc.description.seriesinformation | Eurographics/SIGGRAPH Symposium on Computer Animation | en_US |