Tools for Analyzing Flutter in the Presence of Aeroelastic Uncertainty

Because flutter can lead to catastrophic failure of the aircraft structure, reliable prediction of the aeroelastic behavior of the aircraft is vitally important to provide information about the boundaries of the flight envelope outside which the aircraft is susceptible to flutter. In spite of major progress in computational methods in recent years, particularly in computational fluid dynamic (CFD), there is still little reliance on using computational models to certify aeroelastic stability. One of the key difficulties is that the aeroelastic models are normally represented in a deterministic way, while in real-world fluid structure interactions, many uncertainties exist. Assessing the effects of uncertainties using traditional approaches such as Monte-Carlo analysis is impractical because of the computational complexity of high-fidelity aeroelastic models. The highly nonlinear nature of these models creates problems for other typical approaches such as linearization and worst-on-worst testing. The goal of the proposed research is to develop uncertainty analysis tools based on a generalized polynomial chaos expansion approach that can be effectively applied in the context of aeroelastic modeling. The tools will provide insight into the sensitivity of the flutter boundary with respect to various uncertainties, and will provide confidence interval estimates for the flutter boundary.