Verification and Validation of a Solver in OpenFOAM

Project Abstract

A framework for the estimation of predictive uncertainty is applied to an application area. The framework treats the aleatory and epistemic inputs separately and the numerical and model form uncertainty is also estimated to account for the total uncertainty. The benefits of this framework in taking risk-informed decisions are discussed. The framework includes code verification, solution verification, predictive model form uncertainty and segregated uncertainty propagation. It was observed that the main source of uncertainty is from numerical error and this is due to the use of coarser grid for the parametric study. The under estimation of uncertainty is also discussed if the epistemic uncertainty is considered as an aleatory uncertainty.