Bridging the gaps for engineering design use of large-eddy simulations

Several decades of research in turbulent flows and modeling have established that detailed flow simulations (e.g,, direct numerical simulations, large eddy simulations) can offer accurate prediction for a number of complex flow phenomena, including separating boundary layers, heat transfer, or flow generated sound. The use of these simulations in engineering design has remained limited due to their relative computational cost and obstacles associated with their robustness and tractability in practical engineering configurations. In this presentation, I will explore several critical advancements that have enabled penetration into engineering design use, including: rapid and high quality mesh generation techniques; low dissipation, nonlinearly stable numerical schemes for unstructured grids; advances in wall modeling and subgrid scale modeling for complex flows; algorithmic advances to leverage massively parallel computers using accelerators (e.g., GPUs). The efficacy of these developments will be assessed in the context of several pacing simulation challenges from the aerospace sector with applications in both external aerodynamics and propulsion. It will be shown that salient quantities of engineering interest are predicted with sufficient accuracy for engineering design within a day utilizing modest resources outpacing earlier estimates.