Uncertainty quantification of discontinuous problems in computational fluid dynamics ( Jeroen A.S. Witteveen, CWI Amsterdam, Netherlands)

Computational Fluid Dynamics (CFD) results are highly sensitive to boundary conditions, geometrical parameters, and turbulence models, because of the nonlinearity of the governing Partial Differential Equations (PDEs). Even more so, these inputs are often not exactly known, but uncertain to some extent. Uncertainty Quantification (UQ) is therefore necessary to estimate the impact of these uncertainties and to build a rigorous confidence in CFD results. UQ in CFD is a major challenge because of the high required robustness for approximating the discontinuous solutions of the nonlinear equations. Robustness properties which are consistent with CFD discretization methods can be obtained using adaptive UQ methods. These latter methods discretize the probability space using a tessellation containing multiple elements, similar to spatial CFD discretizations in the physical space. Introducing subcell resolution in UQ methods does no longer require adaptivity to resolve discontinuities robustly, which can be expensive in higher stochastic dimensions. 
Jeroen Witteveen is a scientific staff member in the Scientific Computing group at the Center for Mathematics and Computer Science (CWI) in Amsterdam, The Netherlands. He obtained his PhD degree from the Faculty of Aerospace Engineering of Delft University of Technology, The Netherlands, and he was a postdoctoral research fellow at the Uncertainty Quantification Laboratory of the Center for Turbulence Research at Stanford University, USA.