Contributions to Physics-Based Aeroservoelastic Uncertainty Analysis

Thesis (Ph.D.)--University of Washington, 2016-06

The thesis presents the development of a new fully-integrated, MATLAB based simulation capability for aeroservoelastic (ASE) uncertainty analysis that accounts for uncertainties in all disciplines as well as discipline interactions. This new capability allows probabilistic studies of complex configuration at a scope and with depth not known before. Several statistical tools and methods have been integrated into the capability to guide the tasks such as parameter prioritization, uncertainty reduction, and risk mitigation. The first task of the thesis focuses on aeroservoelastic uncertainty assessment considering control component uncertainty. The simulation has shown that attention has to be paid, if notch filters are used in the aeroservoelastic loop, to the variability and uncertainties of the aircraft involved. The second task introduces two innovative methodologies to characterize the unsteady aerodynamic uncertainty. One is a physically based aerodynamic influence coefficients element by element correction uncertainty scheme and the other is an alternative approach focusing on rational function approximation matrix uncertainties to evaluate the relative impact of uncertainty in aerodynamic stiffness, damping, inertia, or lag terms. Finally, the capability has been applied to obtain the gust load response statistics accounting for uncertainties in both aircraft and gust profiles. The gust response analysis shows a significant increase of the critical loads when system’s uncertainties are included. The work is expected to make a contribution to the understanding of the propagation of uncertainty and the resulting reliability of realistic integrated aeroservoelastic system inherent in real modern aircraft. This work will motivate more research in the aeroservoelastic uncertainty area regarding both methods development in general and reliability studies of different configurations of interest. The mathematical derivations which serve as the foundation of the capability are also well documented.