Uncertainty Quantification in Reacting Flow Modeling

Abstract-
Uncertainty quantification (UQ) in the computational modeling of physical systems is important for both engineering design and model validation. We have developed spectral/pseudo-spectral stochastic UQ techniques in which model parameters and field variables are modeled as stochastic quantities, and are represented using polynomial chaos (PC) expansions. We present results analyzing uncertainty propagation in Hydrogen-air reacting flow under supercritical conditions, allowing for known uncertainties in reaction rate constants and enthalpies of formation. We use the results to evaluate confidence intervals in model predictions, identify dominant sources of uncertainty, and provide a general assessment of the robustness of the model for prediction of specific observables. We compare sensitivities extracted from the UQ computations with those available from conventional sensitivity analysis. We also discuss numerical issues pertaining to the accurate representation of uncertainty with truncated PC expansions, and ensuing stability of the time integration of chemical systems.

GALCIT Home Page

2003-2004 Fluids Seminar Page