In this paper non-intrusive uncertainty quantification (UQ) method is used to improve the accuracy of Smagorinsky Large eddy simulation (LES) model and a wall-modeled large eddy simulation (WMLES) model. Detailed UQ studies focusing on the closure coefficients of these two models are performed. A Polynomial Chaos (PC) surrogate model is used to evaluate the output sensitivities and uncertainties in the entire flow domain. The proposed UQ method allows for the investigation of specific flow features and phenomena within the domain. The results of the UQ analyses are then used to identify which closure coefficients in the models most influence the flow features of interest. Refinements are then made to the closure coefficients of interest to improve the accuracy of the two LES models. OpenFOAM is used as the flow solver and the UQ analyses are conducted with DAKOTA. The proposed UQ method is applied to the turbulent channel flow at various Reynolds numbers. The LES models with new closure coefficients show significant improvement in the prediction of the skin-friction coefficient on channel walls.

Document Type

Final Report

Author's School

School of Engineering and Applied Science

Author's Department

Mechanical Engineering and Materials Science

Class Name

Mechanical Engineering and Material Sciences Independent Study

Date of Submission