Uncertainty Quantification for Airfoil Icing

Abstract

Ensuring the safety of airplane flight in icing conditions is an important and active arena of research in the aerospace community. Notwithstanding the research, development, and legislation aimed at certifying airplanes for safe operation, an analysis of the effects of icing uncertainties on certification quantities of interest is generally lacking. The central objective of this thesis is to examine and analyze problems in airfoil ice accretion from the standpoint of uncertainty quantification.

We focus on three distinct areas: user-informed, data-driven, and computational uncertainty quantification. In the user-informed approach to uncertainty quantification, we discuss important canonical icing classifications and show how these categories can be modeled using a few shape parameters. We then investigate the statistical effects of these parameters. In the data-driven approach, we build statistical models of airfoil ice shapes from databases of actual ice shapes, and quantify the effects of these parameters. Finally, in the computational approach, we investigate the effects of uncertainty in the physics of the ice accretion process, by perturbing the input to an in-house numerical ice accretion code that we develop in this thesis.