Gust-Induced Amplitude Effects in Transonic Flow

Kurzfassung

The design of a transport aircraft configuration requires a variety of load computations for optimal structural sizing. Certification specifications demand, e.g., the computation of gust encounters with different amplitudes at transonic cruise speeds. Due to the large number of necessary computations, frequency-based methods are state-of-the-art in an industrial context. However, these time-linearized methods are strictly valid only for excitations with very small amplitudes. In order to enable an accurate load prediction also for large-amplitude excitations, computed results are partially corrected, e.g., on the basis of wind tunnel data. For optimizing an aircraft's sizing even further and more automated, a more accurate, i.e., unsteady nonlinear, load prediction for excitations with large amplitudes is required. The present work analyzes the influence of an unsteady nonlinear aerodynamic modeling in comparison to a time-linearized aerodynamic modeling on the computation of transonic gust responses. The time-domain simulations are based on the unsteady Reynolds-averaged Navier-Stokes equations (URANS) and are carried out for an airfoil configuration as well as for a transport aircraft configuration. Sinusoidal, as well as "1-cos" gusts, are used for the excitation. Gust lengths and gust amplitudes are varied. Gust encounters take place under different transonic steady conditions, in order to assess the influence of the steady base flow field. Finally, the impact of the observed aerodynamic nonlinearities is analyzed also for a flexible model of the transport aircraft configuration. Based on these investigations, three types of nonlinear lift responses are identified: Nonlinear lift responses of type A lead to a lower peak lift coefficient prediction than time-linearized computations. Shock-induced flow separation dominates the unsteady flow here. Nonlinear lift responses of type B correspond to higher peak lift values than the prediction by the time-linearized method. These responses involve a significant topological change in the flow field without significant separation. Nonlinear lift responses of type C mark the transition between types A and B and correspond to comparable peak lift values for time-linearized and nonlinear computations, even though the underlying unsteady flows differ fundamentally. The results indicate that the steady lift polar, or more specifically, the distinct trend of the lift curve slope, might serve as an indicator for the occurrence of the respective response type: If the steady angle of attack is lower than the angle of attack at which the maximum of the lift curve slope is expected, the time-linearized method might underestimate or overestimate the actually occurring, nonlinear maximum value. If the steady angle of attack is larger, only conservative loads can be expected using the time-linearized method.