The significance of turbulent inflow conditions in DNS*

Direct Numerical Simulation (DNS) is the most rigorous approach of spatially-developing turbulent boundary layers (SDTBL) due to the need to resolve all length/time scales of the fluid flow. Consequently, DNS requires substantial HPC resources to do so. The problem becomes harder if the idea is to evaluate external conditions such as Reynolds number dependency, compressibility, wall temperature and wall curvature effects.

For time-dependent inflow prescription, recycling techniques were employed in unsteady SDTBL (e.g., strong recycling, weak recycling, synthetic recycling, etc.)

Traditional weak recycling techniques have been widely utilized and upgraded in the last decades. The only major flaw of this method is the injection of artificial low frequencies in the flow (proportional to the inlet/recycling distance), which may affect the transport phenomena inside SDTBL. Therefore, this study aims to identify, asset and mitigate artificial periodicity of weak recycling approaches. To this end, the final research question to be addressed is:

Is this “synthetic spatial periodicity” really harmful to SDTBL?