Large-Eddy Simulation of the PRECCINSTA burner with finite-rate chemistry and wall heat loss

In order to reduce nitrogen oxide emissions, aircraft gas turbine manufacturers try to lower the maximum temperature level in the combustion chamber. Lean premixed combustion is a simple way to achieve this. However, this type of flame is less robust than non-premixed flames, and therefore leads to a reduction of the operating margins. Engine manufacturers and researchers worked for several years to model and understand these flames.

Several studies have been carried out since 2009 at CORIA using the YALES2 code. Highly resolved simulation databases of the PRECCINSTA burner have been carried out in a perfectly premixed regime [1]. The mesh resolution was in the order of 100 microns with 2.6 billion tetrahedra in the perfectly premixed regime and 60 microns with 12 billion tetrahedra in the partially premixed regime. In the context of tabulated chemistry models, these resolutions make it possible to solve all the flame / turbulence interactions. These unique databases enabled to develop new combustion models [1,2] and they served as a priori and a posteriori validation for existing models such as the dynamic thickened flame model [3,4].
Since 2016, new simulations of the PRECCINSTA burner have been carried out with a detailed transport of the species, which therefore no longer assumes a flamelet hypothesis. These simulations used the LU17 kinetic scheme [Sankaran et al., PCI 2007], which comprises 17 species and 72 reactions, coupled to the dynamic thickened flame model (DTFLES). This mechanism enables to take into account the flame speed variation over a wide range of equivalence ratio as well as the production / destruction of intermediate species such as OH. By taking into account the thermal losses at the walls and by pushing the resolution to 150 microns in the flame front for a mesh of 878 million tetrahedra, it was possible to reproduce the flame topologies observed experimentally [5].
More recently, these simulations have been reproduced with dynamic mesh adaptation. With iso-resolution of mesh in the flame front and iso-quality of simulation, a significant gain in computation time is observed [6].

References

[1] MOUREAU, V., DOMINGO, P. & VERVISCH, L. (2011) From large-eddy simulation to direct numerical simulation of a lean premixed swirl flame : Filtered laminar flame-pdf modelling. Comb. and Flame 158, 1340–1357.

[2] MERCIER, R., MEHL, C., FIORINA, B. & MOUREAU, V. (2019) Filtered wrinkled flamelets model for large-eddy simulation of turbulent premixed combustion. Comb. and Flame 205, 93–108.

[3] VEYNANTE, D. & MOUREAU, V. (2015) Analysis of dynamic models for large eddy simulations of turbulent premixed combustion. Comb. and Flame162(12), 4622 – 4642.

[4] MERCIER, R., MOUREAU, V., VEYNANTE, D. & FIORINA, B. (2015) LES of turbulent combustion: on the consistency between flame and flow filter scales. Proc. Combust. Inst.35(2), 1359–1366.

[5] BENARD, P., LARTIGUE, G., MOUREAU, V. & MERCIER, R. (2019) Large-Eddy Simulation of the lean-premixed PRECCINSTA burner with wall heat loss. Proceedings of the Combustion Institute,1–11.

[6] MOUREAU,  V.,  BENARD,  P.,  LARTIGUE,  G.  &  MERCIER,  R.  (2019)  Dynamic  adaptation of tetrahedral-based meshes for the simulation of turbulent premixed flames. 17th International Conference on Numerical Combustion. Aachen, Germany.