Session: 34-20 Solver methods 4

Paper Number: 122355

122355 - High-Fidelity Simulation of the Aerothermal Performances of a Turbofan Thrust Reverser 

Thrust reversers (TR) are one of the primary braking systems in current aircraft. They rely on a deflection of the airflow exiting the engine to generate a counter-thrust. TR can be found in mainly two forms, depending on the size of the engine. For low bypass ratios, the TARGET configuration is commonly used. It consists of a series of doors integrated into the engine’s nacelle, which are opened by a set of hydraulic actuators. A hot and cold air mixture coming from the engine then impinges on these doors and is redirected forward. On a wet runway, the use of reversers reduces the braking distance by at least a third. Under icy conditions, often encountered in winter, their use is compulsory to ensure the safe stopping of the aircraft.

Steady-state approaches, such as RANS (Reynolds Averaged Navier-Stokes), fail to correctly predict the aerothermal performances of impinging configurations. Indeed, the flow is characterized by strong mixing and turbulence. Under these conditions, classical closure models tend to overestimate the heat-transfer distribution. Large-Eddy Simulation (LES) is a numerical method based on the explicit resolution of large-scale structures and the modeling of the influence of the smallest ones by a subgrid-scale (SGS) model. For the past decades, LES has proven to be a valuable tool for the prediction of unsteady turbulent flows, especially for impinging configurations. Nevertheless, a significant amount of time is often spent by the user designing a suitable grid. The resolution of the grid is of utmost importance, as it implicitly imposes the cut-off frequency between the large resolved scales and the modeled ones. Static and dynamic feature-based mesh adaptation, and more generally h-refinement methods, have gained momentum in the last decade due to the great advances in the robustness, efficiency and quality of mesh adaptation libraries. These methods enable to improve LES results by refining areas to capture specific flow features. The present paper proposes a framework for the generation of a suitable grid used to evaluate the aerothermal performances of a reduced-scaled TARGET Thrust Reverser (TR) model under test-bed conditions.  

The methodology relies on a two-step workflow. First, a Low-Mach number variable density simulation is performed. Despite not taking into account compressible effects, the accumulated statistics are used as a first guess to adapt the mesh. A dedicated procedure based on the homogenization of the dissipation of the turbulent kinetic energy allows to generate a grid with a controlled number of cells. The grid is refined in regions of prime interest for the study of this configuration, i.e. in the mixing layer and the impinging region, while it is coarsened in the far field. Then, the second step relies on the use of an explicit compressible solver to solve for the flow dynamics. Statistics are converged on this grid for later post-processing.  

The whole methodology is applied to a reduced-scale TARGET-type Thrust Reverser under realistic test-bed conditions.  Flow field quantities, including counter-thrust, upstream core and bypass pressures and mass flow rates, as well as impinging near-wall pressure and temperature, are in good agreement with experimental data. The proposed methodology highlights the ability of LES to predict this kind of complex flow. Overall, results are improved with respect to a RANS resolution.

Presenting Author: Adrien Grenouilloux CORIA - UMR6614 - CNRS

Presenting Author Biography: Post-doctoral researcher at CNRS.

Authors:

Adrien Grenouilloux CORIA - UMR6614 - CNRS
Yacine Bechane CORIA - UMR 6614 - CNRS
Julien Carmona CORIA - UMR 6614 - CNRS
Pierre Bénard INSA de Rouen
Ghislain Lartigue CORIA - UMR 6614 - CNRS
Vincent Moureau CORIA - UMR 6614 - CNRS
Paul Ferrey Safran Nacelles