R&D Program 

All developments and applications of SGEmac make use of the AVBP LES code co-developed and co-owned by IFP and CERFACS. AVBP solves the unsteady, compressible, reactive, multi-species Navier-Stokes equations on unstructured moving meshes. Time advancement is explicit, and convective schemes are centered second order FV and third order FE schemes.

 

AVBP has demonstrated its unique linear efficiency on all parallel computers on the market and for up to 8000 processors.
Moving meshes are handled via a Conditional Temporal Interpolated (CTI) combined with an ALE method.

 

Available LES models include turbulence models (constant coefficient and dynamic Smagorinsky, WALE, filtered structure function, subgrid kinetic energy), combustion models (TF-LES, ECFM-LES, PCM-FPI) and liquid injection models (Eumlerian and Lagrangian).

 

Boundary conditions are handled via the NSCBC approach particularly suited for accurate treatment of acoustics.

 

Project work is structured based on 5 Tasks:

  • Task 1 is the starting point and concerns the acquisition, based on advanced optical diagnostics, of an experimental database on a single cylinder, PFI and DI spark ignition engine, comprising stable and unstable operating points;
  • Task 2 aims at developing complementary LES methods allowing to simulate the behaviour of the combustion chamber, but also of the whole intake and exhaust system, as they may be of importance in the appearance of cyclic variabilities. These developments will be validated on stable operating points acquired in Task 1;
  • Task 3 consists of having the industrial partners use the methods from Task 2 to define methodologies for the  prediction of cyclic variabilities in the case of a PFI SI engine, which will be validated using unstable operating points from Task 1;
  • Task 4 is dedicated to exploring the possibility of extending the PFI LES methodologies to predicting cyclic variability in DI engines;
  • Finally Task 5 aims at synthesizing all the realised work, by comparing the obtained predictions in order to identify  their respective strong and weak points, as well as by identifying elements that will have to be addressed in future research  and in particular for DI studies.
 |  Legal information © IFP Energies nouvelles / IFP 2011