Deep iterative convergence of Reynolds-Averaged Navier–Stokes methods for high-lift aircraft geometries

The fifth edition of the High-Lift Prediction Workshop (HLPW5) was organized to evaluate the computational fluid dynamics (CFD) community’s capability to accurately predict flows and aerodynamic loads in complex high-lift configurations. While the workshop led to significant progress, several questions remained unanswered, and recurring challenges persisted across multiple editions. The complex geometries introduced by high-lift devices, such as slats, flaps, and their associated brackets, lead to the formation of large separation regions in Reynolds-averaged Navier–Stokes (RANS) solutions that are not observed in experimental oil-flow visualizations. Furthermore, the emergence of these large separation areas compromises the ability of RANS solvers to achieve deep iterative convergence. Addressing and overcoming these challenges is a critical step toward ensuring consistency and alignment across different CFD RANS solvers. To tackle this issue, the present work focuses on one of the challenges raised by the HLPW5 RANS Technology Focus Group: achieving iteratively converged solutions for complex high-lift configurations. This is accomplished through the successful application of the selective frequency damping method, which enables RANS solvers to converge toward unstable equilibrium solutions. Additionally, this study presents observations regarding the pseudo-unsteady nature of the large separation regions identified in the RANS solutions.

Mots clés

• Reynolds Averaged Navier Stokes
• Aircrafts
• Aerodynamic Coefficients
• Computational Fluid Dynamics
• Aerodynamic Performance
• Numerical Algorithms
• High-Lift