Fig. 1: Schematic of the dominant balance identification procedure applied to a turbulent boundary layer.
From: Learning dominant physical processes with data-driven balance models
High-resolution direct numerical simulation results (a, visualized with a turbulent kinetic energy isosurface) are averaged to compute the terms in the Reynolds-averaged Navier–Stokes equations (b). The equation space representation of the field enables clustering and sparse approximation methods to extract the distinct geometrical structures in the six-dimensional space corresponding to dominant balance physics (c). Finally, the entire domain can be segmented according to these interpretable balance models, identifying distinct physical regimes (d). A curve fit to the wall-normal extent of the post-transition region of the identified inertial sublayer shows an approximate scaling of ℓ ~ x0.81, consistent with the theoretical prediction of x4/5 from boundary layer theory. The 99% free-stream velocity (U∞) contour is also shown for reference.
