A Parallel Combustion Solver within an Operator Splitting Context for Engine Simulations on Grids

  • Authors:
  • Laura Antonelli;Pasqua D'Ambra;Francesco Gregoretti;Gennaro Oliva;Paola Belardini

  • Affiliations:
  • Institute for High-Performance Computing and Networking (ICAR)-CNR, Naples, Italy I-80131;Institute for High-Performance Computing and Networking (ICAR)-CNR, Naples, Italy I-80131;Institute for High-Performance Computing and Networking (ICAR)-CNR, Naples, Italy I-80131;Institute for High-Performance Computing and Networking (ICAR)-CNR, Naples, Italy I-80131;Istituto Motori (IM)-CNR, Naples, Italy 8 I-80125

  • Venue:
  • Large-Scale Scientific Computing
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Multidimensional engine simulation is a very challenging field, since many thermofluid processes in complex geometrical configurations have to be considered. Typical mathematical models involve the complete system of unsteady Navier-Stokes equations for turbulent multi-component mixtures of ideal gases, coupled to equations for modeling vaporizing liquid fuel spray and combustion. Numerical solutions of the full system of equations are usually obtained by applying an operator splitting technique that decouples fluid flow phenomena from spray and combustion, leading to a solution strategy for which a sequence of three different sub-models have to be solved. In this context, the solution of the combustion model is often the most time consuming part of engine simulations. This work is devoted to obtain high-performance solution of combustion models in the overall procedure for simulation of engines in a distributed heterogeneous environment. First experiments of multi-computer simulations on realistic test cases are discussed.