The Relationship between DNS and RANS
DNS provides all the information on a flow, with limitations on geometry and Reynolds numbers. Datasets have been available since the 1980’s for homogeneous and free shear flows, as well as wall-bounded flows. The Reynolds number Re_tau in channel flow has risen from 180 to 8,000. The impact of DNS on the most-used eddy-viscosity models, however, has been extremely small. The impact on Reynolds-stress models is also modest. There are substantial structural difficulties when trying to match RANS to DNS, including conflicts over whether the Reynolds stresses obey a law of the wall, and over whether the stresses are uniform in the log layer. Another factor is that all models accept cancellations of errors between terms. In recent work, DNS fields have been used to define an effective eddy viscosity, to provide a target for simple models. Examples are given, which led to limited success. Machine Learning may be a key to bridging DNS and RANS. The core problem is that data in itself does not create (good) ideas.
Philippe Spalart studied Mathematics and Engineering in Paris, and obtained an Aerospace PhD at Stanford/NASA-Ames in 1982. Still at Ames, he conducted extensive Direct Numerical Simulations of transitional and turbulent boundary layers. Moving to Boeing in 1990, he created the Spalart-Allmaras one-equation Reynolds-Averaged Navier-Stokes turbulence model. He wrote a review and co-holds a patent on airplane trailing vortices. In 1997 he proposed the Detached-Eddy Simulation approach, blending RANS and Large-Eddy Simulation to address separated flows at high Reynolds numbers with a manageable cost. Recent work includes refinements to the SA model, computational aeroacoustics, and theories for aerodynamics and turbulence.