No abstract

Results of numerical simulations of a forced shear flow in an annular geometry are presented. The particular geometry used in this work reduces the effects of centrifugal and Coriolis forces. However, there are still a large number of system parameters (shear width, shear profile, radius of curvature, initial conditions, etc.) to characterize. This set of variables is limited after the code has been validated with experimental results (Rabaud & Couder 1983; Chomaz et al. 1988) and with the associated linear stability analysis. As part of the linear stability characterization, the pseudo-spectrum for the associated Orr–Sommerfeld operator for plane, circular Couette flow is calculated, and it is found to be insensitive to perturbations.The numerical simulation code is a highly accurate de-aliased spectral method which utilizes banded operators to increase the computational efficiency. Viscous dissipation terms enter the code directly from the equations of motion. The results from the simulation code at low Reynolds numbers are compared with the results from linear stability analysis and are used to give predictions for the coefficients of the Landau equation describing the saturation behaviour near the critical Reynolds number. Numerical results at higher Reynolds numbers demonstrate the effects of spin-up and spin-down, including the experimentally observed hysteresis. The properties of two- dimensional shears at high Reynolds numbers, at which temporal states are formed, are also addressed.

No abstract

As the capabilities of computational fluid dynamics (CFD) to model full aircraft configurations improve, and the speeds of massively parallel machines increase, it is expected that CFD simulations will be used more and more to steer or in some cases even replace traditional flight test analyses. The mission of the US Air Force SEEK EAGLE office is to clear any new weapon configurations and loadings for operational use. As more complex weapons are developed and highly asymmetric loadings are requested, the SEEK EAGLE office is tasked with providing operational clearances for literally thousands of different flight configurations. High-fidelity CFD simulations employing the turbulent Navier-Stokes equations are in a prime position to help reduce some of the required wind-tunnel and/or flight test workload. However, these types of CFD simulations are still too time consuming to populate a full stability and control parameter database in a bruteforce manner. This article reviews results previously published by the authors, which validate the ability of high-fidelity CFD techniques to compute static force and moment characteristics of aircraft configurations. A methodology to generate efficient but non-linear reduced-order aerodynamic loads models from dynamic CFD solutions, which in-turn may be used to quickly analyse various stability and control characteristics at a particular flight condition, is introduced, and the results based on the US Air Force F-16C fighter aircraft that exemplify the process are discussed.

With many modern fighter aircraft experiencing unpredicted flight dynamics during flight tests, recent research has focused on developing methodologies for incorporating computational fluid dynamics into the aircraft development process. The goal of this ap proach is to identify configurations susceptible to stability and control issues early in the design process. Previous research has primarily focused on full aircraft configurations, how ever, to increase the rate of development the current study focused on a two-dimensional NACA0012 airfoil. The two-dimensional NACA0012 airfoil has the advantage of reducing the computational cost by orders of magnitude compared to full scale aircraft simulations, while still providing complicated aerodynamics at high angles of attack. Computationally predicted lift coefficients from a number of newly developed training maneuvers were used to generate reduced order aerodynamic loads models. For evaluation, these models were compared to generated static and dynamic validation data. Methods of improving both the computational training maneuver and the reduced order modeling approach are suggested. Nomenclature