Wall thermal effects in wall-modeled and wall-resolved large-eddy simulation of compressible flows

High-fidelity numerical simulations are performed to investigate the impact of wall temperature on shock wave-induced flow separation. The numerical methodology employs an unstructured-mesh, finite-volume, second-order shock-capturing LES flow solver. We conduct wall-resolved large-eddy simulations (WRLES) and wall-modeled large-eddy simulations (WMLES), comparing both to a direct numerical simulation (DNS) study by Bernardini et al. (2016) of a Mach 2.28 oblique shock-turbulent boundary layer interaction (STBLI) over a wall at cooled, adiabatic, and heated thermal conditions. An experimental study by Debiève et al. (1997) investigating Mach 2.3 flow over a flat plate subjected to changes in wall temperature provides an additional validation case for thermal effects in our WMLES simulations. We investigate the impact of several modeling parameters, such as the turbulent Prandtl number of the subgrid-scale and wall models, on the simulation results and evaluate the adequacy of various mean velocity profile transformations in the compressible flow regime. Results from WRLES and WMLES are compared, assessing how each simulation method performs against the reference data in the prediction of thermal transport, boundary layer separation, and unsteady low-frequency STBLI motions.