Abonnement à la biblothèque: Guest
TSFP DL Home Archives Comité de direction


Catherine Gorle
Center for Turbulence Research, Stanford University, Stanford, CA 94305, USA; EMAT, University of Antwerp, Antwerp, Belgium von Karman Institute for Fluid Dynamics, Sint-Genesius-Rode, Belgium

Gianluca Iaccarino
Department of Mechanical Engineering Institute for Computational Mathematical Engineering Stanford University Bldg 500, RM 500-I, Stanford CA 94305 - USA


The goal of the present study is to establish a method that can capture the uncertainty in the Reynolds-averaged Navier-Stokes prediction for dispersion from a point source in the flow over a wavy wall. The methodology is based on 1. perturbing the modeled Reynolds stresses in the momentum equations, thereby establishing a method that is completely independent of the initial model form and 2. introducing uncertainty in the turbulent scalar flux vector in the transport equation for the scalar by using the perturbed Reynolds stresses in the generalized gradient diffusion model. The Reynolds stress perturbations are defined in terms of a decomposition of the Reynolds stress tensor, i.e. based on the tensor magnitude and the eigenvalues and eigenvectors of the normalized anisotropy tensor. Results of a previous study are further analyzed and show that a realistic representation of the uncertainty in the velocity field is obtained. In addition, an a priori analysis of the scalar flux model alignment indicates that, provided sufficient uncertainty is introduced in the Reynolds stresses, an adequate representation of the uncertainty in the scalar flux vector can be obtained. Based on these results, a UQ study will be formulated to quantify the uncertainty in the scalar dispersion.