No abstract

An unstructured and massively parallel Reynolds-Averaged Navier-Stokes (RANS) code is used to simulate 3-D, turbulent, non-reacting, and confined swirling flow field associated with a single-element and a nine-element Lean Direct Injection (LDI) combustor. In addition, the computed results are compared with the Large Eddy Simulation (LES) results and are also validated against the experimental data. The LDI combustors are a new generation of liquid fuel combustors developed to reduce aircraft NOx emission to 70% below the 1996 International Civil Aviation Organization (ICAO) standards and to maintain carbon monoxide and unburned hydrocarbons at their current low levels at low power conditions. The concern in the stratosphere is that NOx would react with the ozone and deplete the ozone layer. This paper investigates the non-reacting aerodynamics characteristics of the flow associated with these new combustors using a RANS computational method. For the single-element LDI combustor, the experimental model consists of a cylindrical air passage with air swirlers and a converging-diverging venturi section, extending to a confined 50.8-mm square flame tube. The air swirlers have helical, axial vanes with vane angles of 60 degree. The air is highly swirled as it passes through the 60 degree swirlers and enters the flame tube. The nine-element LDI combustor is comprised of 9 elements that are designed to fit within a 76 mm 76 mm flametube combustor. In the experimental work, the jet-A liquid fuel is supplied through a small diameter fuel injector tube and is atomized as it exits the tip and enters the flame tube. The swirling and mixing of the fuel and air induces recirculation zone that anchors the combustion process, which is maintained as long as a flammable mixture of fuel and air is supplied. It should be noted that in the numerical simulation reported in this paper, only the non-reacting flow is considered. The numerical model encompasses the whole experimental flow passage, including the flow development sections for the air swirlers, and the flame tube. A low Reynolds number K-e turbulence model is used to model turbulence. Several RANS calculations are performed to determine the effects of the grid resolution on the flow field. The grid is refined several times until no noticeable change in the computed flow field occurred; the final refined grid is used for the detailed computations. The results presented are for the final refined grid. The final grids are all hexahedron grids containing approximately 861,823 cells for the single-element and 1,567,296 cells for the nine-element configuration. Fine details of the complex flow structure such as helical-ring vortices, re-circulation zones and vortex cores are well captured by the simulation. Consistent with the non-reacting experimental results, the computation model predicts a major re-circulation zone in the central region, immediately downstream of the fuel nozzle, and a second, recirculation zone in the upstream corner of the combustion chamber. Further, the computed results predict the experimental data with reasonable accuracy.

No abstract

No abstract

This paper describes an emerging computational capability for physics-based flow simulation and maneuvering predictions for appended submarine/propulsor geometries. The solution methodology for the unsteady Reynolds-averaged Navier-Stokes equations is summarized, including the transition of this capability from single-processor to scalable parallel computing. The current status of validation efforts for this methodology is discussed, including comparisons for appended-hull force and moment coefficients andpropulsor thrust and torque coefficients. Results are given from several simulations related to maneuvering of appended submarines with rotating propulsors. This capability will enable new complex simulations in computational naval hydrodynamics that can support the submarine design process as well as provide understanding leading to improved safety margins in submarines undergoing complicated maneuvers. To illustrate the impact of the scalable parallel code, a submarine maneuver requiring 3 million grid points and covering a distance of five hull lengths can be run in less than two days on 80 T3Eproc-essors, as compared with over four months on a single processor.

No abstract

Behavior of unconfined transverse jets has been studied extensively, but little work is reported on the flow characteristics of confined transverse jets. The latter has been numerically investigated using a number of RANS codes. The computational results obtained from these codes have been evaluated against the existing experimental data, and the results of a Large-Eddy Simulations (LES) code reported in the literature. Furthermore, an extensive validation effort has been conducted to characterize the performance of the codes for predicting the flow within a propulsion-related mixing configuration. The validation case involves eight circumferentially spaced transverse jets issuing radially into an axisymmetric main flow, a configuration relevant for gas turbine burners and new liquid rocket engine preburners. The main flow Reynolds number was 1.7 x 10 5 and the jet-to-main flow momentum flux ratio was sixteen. The momentum and scalar mixing was investigated through the solution of the Reynolds-Averaged Navier Stokes (RANS) equations. The solutions of three commercial RANS solvers, Fluent, STAR-CCM+, and CFD++, are compared to experimental data and large-eddy simulation (LES) results available in literature. Due to demonstrated periodicity, only a one-eighth pie-shaped section of the geometry was considered. The different commercial codes used the same geometry, grid, boundary conditions, and variations of the k-ϵ turbulence model. The LES results obtained from literature used a different grid, but the same geometry. All numerical simulations using the above mentioned codes capture salient flow structures such as the counter-rotating vortex pair (CRVP). Experimental data used for validation of the codes include mean axial velocity and jet fluid mixture fraction profiles (at three distinct axial locations), jet trajectory, turbulent kinetic energy distributions, and velocity and mixture fraction cross-plane distributions. All CFD results except CFD++, exhibit symmetrical solutions about the center plane. The current investigation shows that although all codes considered predict the experimental data with various degrees of accuracy, Fluent using the standard k-ϵ turbulence model with the standard wall function, and LES results compare exceptionally well with the experimental data for this flow regime and configuration. NomenclatureA, b, c = empirical constants C = local mean mass fraction of the jet C = average mass fraction of the jet over the cross sectional area A D = diameter J = momentum flux ratio m  = mass flow rate n = number of jets Q = volume flow rate r = velocity ratio x = streamwise coordinate y = cross stream coordinate j = subscript for jet flow Us = Unmixedness

Transonic strong blade-vortex interaction is numerically analyzed by solving the unsteady 2-D Navier–Stokes equations using an iterative implicit second order scheme. The dominant processes during the interaction are the development of large transverse pressure gradients in the upper leading edge region and the development of disturbances at the root of the lower surface shock wave. As a result of this interaction, high pressure pulses are emitted from the leading edge, and acoustic waves are radiated from the lower surface in a region originally occupied by a supersonic pocket. In addition, severe load variations occur when the vortex is within one chord length of the blade.