This investigation takes an analytical approach to the oscillations of a cantilevered plate immersed in a hypersonic flow with shock impingement. In particular, we derive a mathematical model keyed in on the fact that (1) the shock impingement point moves along the structure as it oscillates and (2) the local curvature of the structure changes the shock reflection angle and thus the compressible flow properties. For cantilever boundary conditions, large motion at the free end my render both effects significant. We show that the movement of the shock impingement point varies approximately with the cotangent of the oblique shock angle and this implicit relationship generates surprisingly strong nonlinear effects in the equation governing the first mode of vibration. A new, geometrically modified third-order Piston Theory is adapted to accurately capture structural curvature induced changes in flow properties as well as possible expansion wave interaction. Our model takes the form of a nonlinearly damped Duffing oscillator with quadratic and cubic nonlinearities stemming from the nonlinear aerodynamic generalized forces as well as geometric and inertial structural nonlinearities. Despite inviscid flow modeling, a perturbation solution to the governing equation shows good agreement with structural oscillations predicted by high-fidelity turbulent computational fluid dynamic simulations.

Observations of maneuvering humpback whales have revealed unique hydrodynamic performance hypothesized to be a result of tubercles on the leading-edge of the whales' pectoral flippers. Inspired by this biological observation, it is shown sinusoidal leading-edge wings prevent the dramatic loss of lift caused by stall and instead generate a gradual decrease in lift with as much as 25% higher lift in the poststall regime. Six different wing geometries, smooth and sinusoidal leading-edge models, swept and unswept configurations, were tested at angles of attack of −2 to 24 degrees at Reynolds numbers between 100,000 and 500,000. Oil surface flow visualization and CFD results reveal variations in flow phenomena between the smooth and sinusoidal leading-edge configurations.

An experimental and computational study was conducted to provide insight into the starting characteristics of blunt body models in the Mach 6 Ludwieg tube at the U.S. Air Force Academy. Experimental results showed that the tube had difficulty starting when large model sizes were present in the test section. Steady-state analysis with computational fluid dynamics was able to predict if the tube would start for a given blunt body model. This helps to increase the confidence that computational tools can be used to investigate the behavior of the tube before running an actual experiment. Time-accurate simulations were run to determine if suggested methods of getting the tunnel to start would actually succeed in starting the flow. The changes investigated involved moving the placement of the model in the test section, both upstream and off the tube centerline. While the computational results show the that flow would not start, the results do suggest that placing the model in a different location in the test section has beneficial qualities. More investigation would be needed to quantify the full effect of shifting the model in the test section.

This a combined experimental and computational study to evaluate the aerodynamic forces resulting from the interaction of small pin protuberances and the boundary layer of a flat plate in Mach 3.0 flow. Cylindrical pins with diameters of 2 mm and 6 mm were tested. The pin heights were 2 mm, 4 mm and 6 mm and they were tested on the flat plate with a turbulent boundary layer thickness of approximately 3 mm. The 2 mm diameter pin geometry was also tested in a group of four to assess mutual interference between neighboring pins. Oil flow, schlieren imagery, and pressure data were recorded on the nine pin configurations to capture a three dimensional understanding of the flow structures. The numerical results were calculated using a Reynolds-Averaged Navier Stokes code. The numerical results were compared to the experimental results for validation. Then they were used to predict the aerodynamic forces induced by a group of protuberances. This was a first step in assessing the possibility of using an array of independently deployable pin protuberances on a missile for flight control at supersonic speeds. Under this concept both the height and number of pins deployed could be controlled. The results showed that the scale of the upstream fluid structures is a function of the pin height to diameter ratio similar to previous studies. There were some fundamental changes in the flow downstream as the pin height was increased from within the boundary layer to outside the boundary layer. A periodic pressure of signal between 300 to 400 Hz was recordeded for all cases in the wake of the pin. The side force produced by the smaller pin ranges up to 1.2 N. The side force form the larger pin ranges up to 4.2 N. The drag forces for the smaller and larger pins ranged up to 0.87 N and 2.5 N respectively. To the first order the forces scaled to the product of the pin height and the combined width of the pins in each configuration.

This a combined experimental and computational study to evaluate the aerodynamic forces resulting from the interaction of small cylindrical protuberances and the boundary layer of a cylindrical body in Mach 3.0 flow. It is a continuation of a study on the effects of the same protuberances on the flow over a flat plate model. Cylindrical pin protuberances with diameters of 2 mm and 6 mm were tested. Protuberance heights of 2 mm, 4 mm and 6 mm were tested on a 63.5 mm diameter cylinder with an ogive nosecone. The boundary layer thickness was approximately 2.6mm. The set 2 mm diameter protuberances were also tested in a group of four at all three heights. Oil flow visualization, schlieren photography, and pressure data were recorded with a variety of pin configurations to capture a three dimensional understanding of the flow structures. Both schlieren and pressure data were captured a 10 kHz to capture the unsteady flow behavior. The results were that the there was a decrease of between 5 and 9% in the side forces produced by the pins. The drag was increased in some cases but reduced by up to 31% for the largest pins. Flow visualization showed that the upstream separation point did not move due to the curvature but the lateral spread of the separation front was limited and the peak pressures decreased due to flow relief from the addition on curvature.

Mach numbers. The employed ROM methodology is based on using the Radial Basis Functions, RBF and set of maneuvers at dierent Mach numbers. The proposed methodology is shown to be able to predict accurately the strongly non-linear aerodynamic coecients for both steady and pitching airfoil by using a single ROM model.

As the practical flight regime extends farther into the supersonic and hypersonic regimes, the interaction between shock waves and the boundary layer that forms along a body will become more significant. These interactions lead to the potential for locally separated flows and high temperature regions on the surface. Because of the limitations of high Mach number wind tunnel testing, wind tunnel tests alone will not be able to provide the information needed to design these high Mach number aircraft. This study combines supersonic (M = 3) wind tunnel test of shock-wave boundary-layer interactions with computational modeling of the same. The particular interaction chosen was the impingement of a planar shock on the boundary layer on a von Karman ogive forebody with a finess ratio of 2.3 and cylindrical aft section with diameter of 5 cm. The shock was generated by a 10 beveled wedge mounted above the ogive-cylinder to provide a shock wave impacting on the cylindrical part of the body. Schlieren photography and oil flow visualization provided details of the flowfield around the model. Time-averaged surface pressure ports and a force balance were used to measure the interaction of the shock wave with the ogive-cylinder body. In general the computational results compared well with the experimental results though several discrepancies were observed with each of the measurements. The results confirmed the existence of a shock induced horseshoe vortex around the cylindrical body, and found that the primary measurable lateral force on the ogive-cylinder body was Cy = −0.081 in the direction normal to the wedge shock generator. This normal force was primarily due to the sharp pressure rise induced near the initial point of impact of the oblique shock wave upon the receiver body.

The goal of this study was to evaluate the time accuracy of five different Navier-Stokes flow solvers using five simple two-dimensional validation cases. The five flow solvers included Cobalt, USM3D, FUN3D, Beggar and NASCART-GT. The five validation cases included the laminar flow over a circular cylinder, the laminar shock tube, the inviscid convecting vortex, the 18%-thick circular-arc airfoil and the NACA 0015 pitching airfoil. This study was conducted in an effort to better understand the capabilities of each flow solver for time-accurate calculations. The eventual goal of the project was to use one or more of these flow solvers to analyze the unsteady transonic flow over a fighter aircraft with stores in captive carriage. Evaluating the capabilities of each flow solver with simple validation cases rather than a complicated fighter aircraft was desirable.

The Air Vehicles Directorate of the Air Force Research Laboratory (AFRL/RB) has been investigating reusable launch vehicle (RLV) re-entry trajectory approaches in order to mitigate vehicle heating and improve operability. One approach that has been considered is to fly the re-entry trajectory at extremely high angles of attack (35°-75°), which has been shown to have the potential to reduce peak stagnation and integrated airframe heating. Although wind tunnel tests have been performed to characterize aerodynamic and aerothermal properties at these high angles of attack in hypersonic flows, such testing is very expensive and it is therefore highly desirable to be able perform computational predictions with a high level of accuracy. A study of existing high fidelity computational fluid dynamics codes was undertaken to assess their ability to accurately predict aerodynamic force and moment and aeroheating characteristics of RLVs at high angles of attack in hypersonic flows. Several different organizations using a number of different, but primarily unstructured, codes participated in the study. An initial set of blind simulations was performed by all participants, with computational predictions compared to experimental data once all simulations had been completed. Following the blind simulations and comparison with experimental data, a second set of simulations was performed to improve the results and identify guidelines and lessons learned. For all participants, computational predictions and experimental values for force and moment data showed excellent agreement at all angles of attack. Aeroheating predictions and experimental values showed good agreement, with simulation accuracy highly dependent on grid parameters and operator experience. Nevertheless, the study showed that at these high angles of attack, hypersonic flows can be modeled computationally with very good accuracy for the particular class of vehicle under consideration.

The development of a flow-state database, including the optical aberrations caused by density variations, of the separated shear layer behind a backward-facing step is discussed in this paper. Delayed Detached-Eddy Simulations (DDES) using Cobalt are performed to investigate the flow field, both in its natural, unforced state and when periodic forcing is applied. The results indicate that the large, coherent structures in the shear layer cause the most severe optical aberrations, while the smaller scales do not significantly affect the beam. The computational results are part of a flow state database that will serve as the basis for the development of feedback control strategies to mitigate the optical aberrations caused by large coherent structures in the flow.

Grid adaptation using general elements was applied to a hypersonic high angle of attack re-entry vehicle to determine if there could be improvements made in the prediction of heat transfer starting from a baseline unstructured grid. Grid adaptation was able to greatly reduce the errors in heat transfer prediction that are common when computing on unstructured grids. The reduction in error is due to a shrinking of average cell size near the bow shock. The smaller cells cause a less drastic jump in the solution of the Riemann problem at the shock. Additionally, a hybrid numerical scheme was used to add numerical dissipation near the bow shock region to help further reduce the jump in the solution of the Riemann problem at the shock. Average errors in heat transfer near the stagnation region were reduced by approximately 50% due to the smaller average cell size and hybrid numerical scheme. Overall the approach of combining grid adaptation and a hybrid numerical scheme appears to provide a robust and reliable method of computing heat transfer rates for hypersonic re-entry vehicles using unstructured grids.

Simulations of gas lasers are performed to examine high power gas laser components. These simulations are performed in a manner that illustrates the application of high performance computing to achieve an increased understanding of the physics underlying gas lasers, and improve their operation. Computational fluid dynamic (CFD) simulations coupled to a geometric optics resonator model for the chemical oxygen-iodine laser (COIL) are executed to provide this information.

This paper documents a comparison of overset grid and grid deformation schemes ap­ plied to flapped and non-flapped NACA airfoil configurations in order to determine the relative accuracy and computational efficiency of each method. This study is part of a larger effort to use computational fluid dynamics to perform moving control surface cal­ culations. Three different cases, using both overset and deformed grids, are considered, including: a) lift and moment comparison of a quasi-steady, non-flapped 0012 airfoil, b) lift and moment comparison of a dynamically pitching non-flapped 0012 airfoil, and c) lift comparison of a dynamic pitching and oscillating flapped 0012 airfoil. These results are compared to experimental data from various sources. Two flow solvers of common lineage were used for the computations: Cobalt for overset and rigid mesh motion and AVUS for the deformable mesh motion. All of the methods produced nominally similar results. As expected, the rigid mesh technique required the least amount of computational resources, while the deformable mesh technique required the greatest amount of computational re­ sources due to its serial implementation. However, in the end, it is difficult to recommend one method over another as the application of each method may be dependent on the project being solved.