Comparative Study of Turbulent Inflow Techniques for High-Fidelity Simulations

Turbulence generation has received considerable attention as high-fidelity simulations have become more tractable for a widening range of applications. Many inflow strategies have emerged to reduce the numerical expense of generating equilibrium turbulent profiles, but ambiguity exists in the appropriate turbulent-inflow condition for each use case. This work aids in properly selecting turbulent inflows by comparing two common techniques: a synthetic digital filtering method and a body-force-based trip. The inflows are rigorously compared to identify parameters sensitive to the turbulence generation. A supersonic case with a Mach number of 1.5 and a subsonic case with a Mach number of 0.2 are considered. For both flow regimes, the skin-friction coefficient using the trip recovers faster than the digital filter. Conversely, the shape factor predictions recover faster for the digital filter than the trip. The results indicate that selecting the optimal inflow turbulence strategy is a multifaceted problem with many interrelated effects of flow conditions and desired target parameters. The numerical framework in which the technique is embedded is equally important. Therefore, code-specific comparisons like those provided here are a crucial benchmark for informed selection and guidance of turbulent inflows.