Accurate numerical prediction of flutter boundary for fighter aircraft is of great importance. Existing models are deterministic, and do not allow for inherent variations in the system parameters. These variations (e.g. structural dimensions, aerodynamic flow field, stores properties) propagate to uncertainty in the model predictions. In this paper we examine variations in structural dimensions of a "heavy" version of the Goland wing on the flutter boundaries. Initially, the large number of random quantities (component thicknesses and areas) are efficiently reduced by conducting a sensitivity analysis of the baseline wing. Next, an optimization study is carried out to provide a design of the wing that maximizes its first natural frequency while constraining the frequency of the remaining nine modes to no less than their baseline wing counterpart values. The sensitivity study enables selection of a random variable set of the wing components having significant impact on the wing natural frequencies. Monte Carlo simulation is used to propagate the variation in the dimensional properties of the selected set of random quantities of the designed wing. The effect of correlation between random variables is considered. A modal analysis of each realization is evaluated using MSC.Nastran. Flutter boundaries of the propagated sample are predicted based on linear aerodynamic theory (ZAERO R), resulting in a "banded" stability boundary. Results indicate the high sensitivity of the flutter speed to small changes in the structure with an apparent switching in the failure modes.

No abstract

The morphing of an air vehicle is to change its shape and size substantially during flight. Thus, the morphing vehicle is to achieve a broader range of operational modes, all of which will maximize the vehicle performance throughout its mission profile. The dream of human flight has been to mimic birds or insect flights in similar manner since the days of Leonardo da Vinci. Our current aeronautical technology brings us closer to such a feat by vehicle morphing. This is evidenced by the ongoing DARPA contracts on designs of a Sliding-skin concept (in-plane morph) and a Folding wing concept (out-of-plane morph). [1][2][3][4]. However, the R&D of its engineering design/analysis methodology appears to be lagging behind. One such important methodology is the computational capability to assess the flight dynamics and aeroelastic instability, or stability, of a morphing vehicle during the course of its morphing motion.

This paper develops and demonstrates a nonlinear aeroelastic scaling procedure. Previous work showed that matching scaled structural frequencies and mode shapes as well as a buckling eigenvalue produced a scaled model that did not have adequately consistent nonlinear behavior. A new scaling methodology is developed that uses direct matching of the nonlinear static response to account for geometric nonlinearities. The optimization is facilitated by an Equivalent Static Loads (ESL) approach. Natural frequencies and mode shapes are matched by optimizing nonstructural mass. A joined-wing beam model is used as a target for scaling a joined wingbox model. The models are representative of a full scale SensorCraft concept developed by AFRL and Boeing. The aeroelastic frequency and damping response and the nonlinear static response are matched successfully. IntroductionA multi-national team of researchers are working to aeroelastically scale, manufacture, and flight test a 1/9 th scale joined-wing remotely piloted aircraft. Scaling of nonlinear aeroelastic effects are crucial to the program goals. Previous attempts to develop a methodology for nonlinear scaling produced inadequate matching of the nonlinear response. This paper develops a methodology that successfully scales both the linear aeroelastic response and nonlinear stiffness of a joined wingbox model. The developed methodology is useful for any flight test or wind tunnel model needing to scale geometrically nonlinear behavior. Motivation