Comparative analysis of the boundary transfer method with other near-wall treatments based on the k-ε turbulence model

Accurate description of wall-bounded turbulent flows requires a fine grid near walls to fully resolve the boundary layers. We consider a locally simplified transport model using an assumed near-wall viscosity profile to project the wall boundary conditions using the boundary transfer method. Related coefficients are obtained numerically. By choosing a near-wall viscosity profile, we derive an analytic wall function, which significantly reduces the CPU costs. The performance of this wall function is compared to other near-wall treatments proposed in the literature for two frequently used benchmark cases: near-equilibrium channel flow and flow over a backward-facing step with separation and reattachment. The results show that the suggested wall function is computationally efficient. Furthermore, the performance of this wall model is not very sensitive to the size of the wall-adjacent cell.