A volume-penalization-based Entropy Conserving flux limiter on reducing near wall stochastic oscillations in wall-bounded flows.

• Propose analysis methodology for volume-penalization-induced errors of the wall-bounded flows. • Capture the elements with certain oscillation through the priori analysis of discretization errors. • Handle the state vector of inviscid term with an entropy conserving average function. • The proposed flux limiter can reduce numerical overshooting and undershooting in narrow regions. In the application of Discontinuous Galerkin method (DGM) to the Reynolds-averaged Navier-Stokes (RANS) equations, the discretization of the convection-diffusion operator are usually directly constructed on the same discontinuous function space. However, due to overshooting and undershooting generated by inviscid term in narrow regions along sharp layers, the compressibility constraint will be only satisfied in a weak sense for the test function from space of pressure functions, thus leading to stochastic oscillations in the simulation of wall-bounded flows with complex configuration. For this, the reckless addition of artificial viscosity may introduce spurious non-physical forces on the cross-element edges, and further cause the problem about approximation of curvature arises due to the discrete representation of the interface. In this work, we propose a Entropy Conserving flux limiter based on volume-penalization(VP) applied in the DGM-based k-ω -SST IDDES framework (in Appendix). The VP-induced errors will be first analyzed and discretization error of inviscid terms can be checked a priori so that we can capture the elements with certain oscillation errors; and then pass the eigenvalue of state vector of these elements to the entropy-conserving average function; the overshoot and undershoot will be processed by the flux limiter, and finally reduce the stochastic oscillations in wall- bounded flows. [ABSTRACT FROM AUTHOR]