Search

A term-by-term comparison has been conducted between the residuals of the continuous and discrete adjoint methods for quasi-1D flow equations. This comparison is done to identify the origin of numerical instabilities near boundaries of the discrete adjoint method. The results show that there is one-to-one correspondence between the terms of the two residuals. Furthermore, the adjoint boundary conditions are rigorously analyzed. It turns out that improvement can be achieved by replacing some of the terms of the discrete adjoint residual with the counterparts in the continuous adjoint method.

Unsteady turbulent flow simulations were performed based on the Reynolds-averaged Navier–Stokes (RANS) equations to investigate flow oscillation due to three-dimensional (3D) configuration of a Mach 1.5 supersonic open cavity flow with a length-to-depth ratio of 3. Two-dimensional (2D) and 3D unsteady simulation results were analyzed and compared with experimental data and Rossiter’s empirical prediction data. The three-dimensional cavity width-to-depth ratio (W/D) was 1, 3.8 and 7.6. Computational results indicated that pressure oscillation in the 2D flow was generated by a single-flow structure, whereas a multiple-flow structure generated multiple oscillation peaks in the 3D flow. The flow structure in the 3D cavity was investigated. For the 2D flow case, the cavity internal pressure wave was directly synchronized with the free shear layer. In the 3D flow case, an unstable spanwise flow due to the sidewall was observed. This spanwise fluctuation produced additional pressure oscillations coupled with the streamwise internal pressure wave. The numerical results indicate that the spanwise flow reduces the propagation speed of the internal pressure waves and the intensity of the corresponding pressure fluctuation.

Flapping wing micro-air vehicles are biologically inspired by nature flyers, specifically insects and birds. Specifically, insect wings generally consist of veins and membrane components. In this study, a structural analysis considering the vein/membrane components of an insect-like flapping wing is presented. Co-rotational (CR) finite elements are adopted in order to consider the complex wing configuration including both vein and membrane. The CR beam elements with warping degrees of freedom are employed for veins and CR shell elements for the wing membrane. The present structural analysis is verified against the analytical results obtained by an existing software, and it is validated by comparison to existing results from the literature. A fluid–structure interaction analysis is then performed. In the procedure, an aerodynamic analysis based on three-dimensional preconditioned Navier–Stokes equations is employed. Finally, a comparative study with respect to the structural characteristics is conducted. As a result, an efficiency of the present structural analysis is confirmed by comparing with the existing software. It is found that the present FSI results are in good agreement with the existing experimental and numerical results. Moreover, the passive wing twist may have a significant influence on the hover performance.

In this paper, we present a computational aeroelastic analysis of flexible flapping wings in the vicinity of solid walls. The wall effects change the aerodynamic forces and moments of the wings, and thus the aeroelastic behavior. The numerical simulation is carried out using a fluid–structure interaction framework by coupling the computational fluid dynamics and computational structural dynamics. A preconditioned Navier–Stokes solver based on a finite volume method is used for the aerodynamic analysis. The structural analysis is performed using a nonlinear structural model based on a geometrically exact beam formulation. The method is validated using previous numerical and experimental results. The aeroelastic characteristics of the flexible wings with and without the walls are computed and compared.

The RANS equations are widely used to analyze complex flows over aircraft. The equations require a turbulence model for turbulent flow analyses. A suitable turbulence must be selected for accurate predictions of aircraft aerodynamic characteristics. In this study, numerical analyses of three-dimensional aircraft are performed to compare the results of various turbulence models for the prediction of aircraft aerodynamic characteristics. A 3-D RANS solver, MSAPv, is used for the aerodynamic analysis. The four turbulence models compared are the Sparlart–Allmaras (SA) model, Coakley’s $$q-\omega $$ model, Huang and Coakley’s $$k-\varepsilon $$ model, and Menter’s $$k-\omega $$ SST model. Four aircrafts are considered: an ARA-M100, DLR-F6 wing–body, DLR-F6 wing–body–nacelle–pylon from the second drag prediction workshop, and a high wing aircraft with nacelles. The CFD results are compared with experimental data and other published computational results. The details of separation patterns, shock positions, and $$C_{p}$$ distributions are discussed to find the characteristics of the turbulence models.

Recent advance in flapping-wing MAVs has led to greater attention being paid to the interaction between the structural dynamics of the wing and its aerodynamics, both of which are closely related to the performance of a flapping wing. In this paper, an improved computational framework to simulate a flapping wing is developed. This framework is established by coupling a preconditioned Navier–Stokes solution and a co-rotational beam analysis with a restrained warping degree of freedom. Validation of the present framework is performed by a comparison with examples from either earlier analyses or experiments. Further, a numerical analysis of a wing under simultaneous pitching and plunging motion is examined. The results are compared with those obtained with a wing under pure plunging motion, in order to assess the additional motion effect within a spanwise flexible wing. The comparison shows different aerodynamic characteristics induced by the flexibility of the wing, which can be beneficial.

A numerical simulation of flows in an axisymmetric supersonic inlet with bleed regions is performed. An existing code which solves the Reynolds Averaged Navier-Stokes equations and the two-equation turbulence model equations is converted into an axisymmetric code. In addition, a bleed boundary condition model has been applied to the code. In this paper, the modified code is validated by comparing numerical results against experimental data and other computational results for flows on a bump and over an oblique shock with bleed region. Using the code, numerical simulation is performed for the flows in an inlet with multiple bleed regions.

A numerical procedure for aerodynamic load analysis of long span bridges is presented. The preconditioned Reynolds averaged Navier-Stokes equations are adopted to compute flows over the bridges. To capture the turbulent characteristics of the flows, two equation turbulence models, Coakley’s q − ω model and Menter’s k − ω SST model, are used to compute the turbulent viscosity. A dual time stepping method in conjunction with the AF-ADI method is used to advance the solution in time. A loosely coupled method of the preconditioned RANS equations with the turbulence model equations is employed for fast computation without losing numerical stability. The numerical method for the aerodynamic load analysis is verified against well-known benchmark problems. Aerodynamic loads of two real bridges are computed with the method to demonstrate the usefulness of the method.

A soft recovery system (SRS) is a device that stops a high speed projectile without damaging the projectile. The SRS is necessary to verify the shock resistant requirements of microelectronics and electro-optic sensors in smart munitions, where the projectiles experience over 20,000 g acceleration inside the barrel. In this study, a computer code for the performance evaluation of a SRS based on ballistic compression decelerator concept has been developed. It consists of a time accurate compressible one-dimensional Euler code with use of deforming grid and a projectile motion analysis code. The Euler code employs Roe’s approximate Riemann solver with a total variation diminishing (TVD) method. A fully implicit dual time stepping method is used to advance the solution in time. In addition, the geometric conservation law (GCL) is applied to predict the solutions accurately on the deforming mesh. The equation of motion for the projectile is solved with the four-stage Runge-Kutta time integration method. A small scale SRS to catch a 20 mm bullet fired at 500 m/s within 1,600 g-limit has been designed with the proposed method.