Project: CryoFARE - Cryogenic Flame Acoustic Response

Prace Call: 17th
ID: 2018184445, Leader: Thierry POINSOT
Affiliation: CERFACS, FR
Research Field: Engineering
Collaborators: Gabriel Staffellbach CERFACS FR , Charlelie Laurent CERFACS FR , Simon Blanchard CERFACS FR
Resource Awarded: 29.2 Mil. core hours on Joliot Curie - SKL

Abstract

Transcritical conditions correspond to the thermodynamic state of a fluid subjected to a pressure exceeding its own critical pressure, and a temperature below its critical value. These extreme thermodynamic conditions are found for instance in high performance liquid rocket engines (LREs), where oxidizer and fuel are injected as dense cryogenic fluids through coaxial injectors. These real-gas combustion regimes are expected to become more and more prevalent in future LREs, due to a growing need for reusability and increased operability. Both SPACE X and ARIANEGROUP are working on methane/oxygen engines which will operate in these transcritical conditions. However, developments of innovative LREs are historically known to be plagued by thermoacoustic instabilities, that is highly unsteady fluctuations in the combustion chamber resulting from a two-way coupling between the acoustic field and the flame dynamics. The most dramatic outcome of a thermoacoustic instability is the destruction of the combustion chamber and the loss of the whole space launcher. The current state-of-the-art knowledge concerning transcritical flame dynamics is rather limited, due to the inherent difficulty in reproducing such extreme thermodynamic conditions both experimentally and numerically. The proposed work therefore aims at improving our fundamental understanding of transcritical flame dynamics under acoustic perturbations. A doubly-transcritical LO2/LCH4 flame in the geometry of the academic laboratory test rig Mascotte (operated by ONERA, France) will be computed by means of Large Eddy Simulation (LES), with non-ideal gas effects modelled by the Soave-Redlich-Kwong (SRK) equation of state. In a first stage, the structure of the stable flame will be studied. As the flame root is expected to have first order effects on the overall flame acoustic response, the study will focus on the mechanisms responsible for the flame stabilization at the coaxial injector lip. In particular, flame-wall interactions will be accounted for by using a realistic isothermal boundary condition extracted from a steady-state temperature field in the coaxial injector lip, as well as a detailed kinetic scheme valid for CH4/O2 combustion in high pressure conditions. Then, the stable LO2/LCH4 flame will be subjected to acoustic forcing over a wide range of frequencies, in order to partially build the classical Flame Transfer Function (FTF), which is the backbone for all studies of thermoacoustic instability. Most importantly, the proposed work will endeavour to compare results obtained with existing theoretical models of non-premixed flame FTF, and to extend these models to more realistic real-gas conditions by providing a fundamental understanding of doubly-transcritical flame response to acoustic perturbations. The size of the simulation to perform, the inclusion of complex chemistry and the coupling with heat transfer lead to a very large CPU effort which only a PRACE allocation can allow.