“Numerical simulation of fluid flow on complex geometries using the Lattice-Boltzmann method and CUDA-enabled GPUs”

Conference:


Type(s):


Title:

    Numerical simulation of fluid flow on complex geometries using the Lattice-Boltzmann method and CUDA-enabled GPUs

Abstract:


    Within computational fluid dynamics (CFD) the Navier-Stokes (NS) equations are traditionally used to describe the physical properties of the fluid. An alternative approach to classical discretizations for the numerical solution of the Navier-Stokes equations, such as Finite-Difference and Finite-Volume schemes, is provided by the Lattice-Boltzmann equations [Benzi et al. 1992], [Chen and G. 1998]. The Lattice-Boltzmann method (LBM) uses a Cartesian grid for propagating and relaxing a discrete velocity distribution function on a lattice at discrete time steps. Usually a very large number of cells is necessary to obtain an accurate prediction of the macroscopic scales for pressure and velocity. However, due to the simple formulation of the underlying algorithm this method is well suited for parallelization and hardware acceleration using general purpose graphical processing units (GPGPU). LBM is used in engineering software for example to compute the aerodynamic drag of a car to improve its efficiency. Therefore LBM has a big practical importance. Improving the performance of a CFD simulation gives the engineers more time and better feedback during the engineering process leading to more efficient engineering processes and more efficient engineering products. Moreover a strong acceleration in simulation performance makes a new quality of physical simulation technology available for desktop computer software like entertainment and content creation software.

References:


    1. Benzi, R., Succi, S., and Vergassola, M. 1992. The lattice-boltzmann equation: theory and applications. Phys. Rep. 222, 145–197.
    2. Chen, S., and G. D. D. 1998. Lattice boltzmann method for fluid flows. Annu. Rev. Fluid Mech. 30, 329–364.
    3. Li, W., Wei, X., and Kaufman, A. 2003. Implementing lattice boltzmann computation on graphics hardware. The Visual Computer 19, 444–456.
    4. Zhao, Y. 2008. Lattice boltzmann based pde solver on the gpu. Visual Computing 24, 323–333.


ACM Digital Library Publication:



Overview Page: