Surrogate-based Uncertainty Quantification of the High-Fidelity Coupled Aero-Structural Performance of a High-Aspect Ratio Transport Aircraft

Abstract

Uncertainties in operating and manufacturing conditions can cause significant alterations in aircraft performance. A step on the way to mitigating this off-design performance reduction is understanding how variations in certain inputs affect aircraft performance. Uncertainty Quantification and Propagation methods allow one to investigate the impact of several uncertainties on one or several quantities of interest. In this work, we quantify the effect of operational and manufacturing uncertainties on aircraft performance, represented by the estimated fuel mass necessary to complete a known flight of 2500 NM. A novel contribution in this work is propagating these uncertainties for a realistic configuration, with aero-structural characteristics computed using coupled high-fidelity solvers for the disciplines involved. As the high cost of these simulations makes conventional uncertainty quantification methods infeasible, we employ surrogate-based uncertainty quantification instead, leveraging Gaussian process models to alleviate the computational burden. In investigating the influence of operational uncertainties in Mach number, flight altitude, and payload mass, we find that Mach number and flight altitude have much larger first order effects on aircraft performance. In investigating the influence of structural and geometric uncertainties in structural thickness and spanwise twists, we find that spanwise twists have a much larger impact on aircraft performance.