RANS modeling of turbulent mixing for a jet in superaonic cross flow: model evaluation and uncertainty quantification

摘要

In this paper a framework for epistemic uncertainty quantification (EUQ) of turbulent mixing models used in Reynolds-averaged Navier-Stokes (RANS) simulations is presented. The method consists of introducing perturbations in the values for the Reynolds stresses and scalar fluxes as obtained from their respective models. The range of perturbations is determined based on a comparison with large-eddy simulations (LES) data and the uncertainty is quantified by performing an optimization which varies the parameters in the perturbation functions to determine the minimum and maximum value of a quantity of interest that characterizes the downstream mixing.