Grid Refinement of Fifth High-Lift Prediction Workshop Cases Using Wall-Modeled Large-Eddy Simulation

This paper presents solution assessments and grid-convergence studies for the test cases outlined in the Fifth High-Lift Prediction Workshop (HLPW-5), focusing on the high-lift Common Research Models (CRM-HL). The study utilizes a wall-modeled large-eddy simulation (WMLES) methodology developed in the unstructured-grid, node-centered flow solver FUN3D. The second-order-accurate simulations conducted in this study employ a finite-volume spatial discretization and an implicit temporal scheme. Large-scale turbulent features are resolved away from the wall, with small-scale effects captured by the Vreman subgrid-scale model. An equilibrium wall function uses the first grid point off the wall, which serves as the critical interface between the wall model and the large-eddy simulation region, thus requiring careful placement in grid design. Grid-convergence studies are systematically conducted using a global uniform refinement approach, with mesh sizes ranging from several hundred million to 20 billion grid points at the finest resolution level for full high-lift landing configurations. WMLES solutions are assessed for HLPW-5 cases, including a clean wing–body configuration and geometry-buildup configurations corresponding to the ONERA 5.1% CRM-HL model. Moreover, simulation results and grid sensitivity are presented for the NASA 5.2% CRM-HL configuration at both moderate and flight-scale Reynolds numbers. Overall, the WMLES results are satisfactory and agree well with the available experimental data, especially on sufficiently fine grids.