Design and structural optimization of a flying wing of low aspect ratio based on flight loads

Abstract

The design process for new aircraft configurations is complex, very costly and many disciplines are involved, like aerodynamics, structure, loads analysis, aeroelasticity, flight mechanics and weights. Their task is to substantiate the selected design, based on physically meaningful simulations and analyses. Modifications are much more costly at a later stage of the design process. Thus, the preliminary design should be as good as possible to avoid any “surprises” at a later stage. Therefore, it is very useful to include load requirements from the certification specifications already in the preliminary design. In addition, flying wings have some unique characteristics that need to be considered. These are a differentiating factor with respect to classical, wing-fuselage-empennage configurations. The aim of this thesis is to include these requirements as good and as early as possible. This is a trade-off, because the corresponding analyses require a detailed knowledge and models, which become available only later during the design process. New methodologies in the form of a comprehensive, algorithmic design process and a parametric aeroelastic modeling are developed.

The first aspect of this work concentrates on the gust encounter of flying wings. The open loop gust encounter is studied at the example of two flying wing configurations and both show a pronounced tendency of pitch up when encountering a positive gust. This has an increasing effect on section loads and should be included in every gust analysis. At the wing root for example, the peaks of the section loads are reduced by the unsteady aerodynamic influence but occur earlier in time, compared to quasi-steady aerodynamics. The closed loop gust encounter includes a controller for the pitching motion and significantly reduces the minimum and maximum pitch angles during a gust encounter. It was found the performance of the controller is limited by the maximal control surface rate, especially for short gust gradients. Concerning section load, two effects need to be considered. Because the controller reduces the pitch up tendency, section loads are decreased. Then, from the deflection of the control surfaces, loads along the trailing edge are added. These two effects were found to balance out with respect to the shear force and the bending moment at the wing root but the torsional moment was increased. Obviously, actuation of the control surfaces causes much higher hinge moments and attachment loads. The operation range of the aircraft is extended to unstable conditions by allowing mass configurations where the payload is positioned further rearwards. The rigid body motion is increased compared to the naturally stable closed loop configuration. An increase in section load is observed for most monitoring stations as well.

The second focus is the comparison of low fidelity panel methods with higher fidelity aerodynamics. Similarities and differences between VLM and CFD based maneuver loads are shown. For a horizontal level flight at low speed, CFD and VLM converge and deliver similar results in terms of trim conditions with small difference in pressure distribution. Then, all maneuver load cases are calculated using high fidelity aerodynamics within the preliminary design process. Comparison of the CFD to the VLM based maneuver loads shows load envelopes at the wing root that are similar in size and shape but have an offset. At the outer wing and at the control surfaces, the envelopes take different shapes and new dimensioning load cases are identified. Application of parametric modeling and an algorithmic design process result in a final aeroelastic model, optimized for minimum structural weight. The new structural mass is approximately 200kg (~13%) heavier than the reference.