Turbulent fuel-air mixing study of jet in crossflow at different velocity ratios using LES

In order to study the mixing mechanism of fuel and air in gas turbine, large eddy simulation has been used to investigate the methane jet-in-crossflow with the velocity ratio (R) of 1.5 and 4. This study aims to explore the formation mechanism of vortices such as the hairpin vortices, hovering vortices and horseshoe vortices, the relationship between the fuel–air mixing and flow characteristics at different velocity ratios. The numerical methods in the present work are firstly validated with the experimental data in terms of mean and root mean square values of velocity. For R = 4, the shear layer vortices, horseshoe vortices, counter-rotating vortices pairs (CVP) and wake vortices can be observed, while the jet shear layer cannot be observed for R = 1.5. The hairpin vortices originating from the vortice-ring are lifted and shed from the downstream of the jet-outlet due to Kutta-Joukowski lift. The hairpin vortices are similar to CVP. The horseshoe vortices in R = 1.5 and 4 are formed due to the blockage of the jet (CH₄) and the crossflow (air) respectively, and its evolution is associated with the hovering vortices which only exist for R = 1.5. The uniform index and pr-obability density function are used for quantitative analysis of the mixing performance. The uniform index at X/D = 0 (fuel-inlet) and at X/D = 25 (outlet) are 0.033 and 0.335 for R = 1.5 and 0.130 and 0.047 for R = 4. For R = 4, the jet penetration is higher and the deflection angle of jet is smaller than that in case of R = 1.5. Higher R will provide more region for mixing, therefore uniform index is higher and the mixing is more uniform in the downstream.     

A hybrid dynamic Smagorinsky model for large eddy simulation

A hybrid dynamic subgrid-scale model (HDSM) pertaining to Large-eddy simulation (LES) has been developed. The coefficient obtained by German dynamic Smagorinsky model (DSM) was integrated with a new dynamic coefficient, based on the dynamic subgrid characteristic length and controlled by the subgrid-scale (SGS) motions. In HDSM, the characteristic wave number determining the characteristic length of the dynamic subgrid is calculated from a new energy weighted mean method when the subgrid scale turbulent kinetic energy and the dissipation wave number are known. The dissipation wave-number is derived from the SGS turbulent kinetic energy spectrum equation. The total dissipation rate spectrum equation is based on the Pao energy spectrum and local equilibrium assumption. The dynamic subgrid characteristic length could take into account the rapidly fluctuating small scale behaviours and the spatial variation of turbulent characteristics. HDSM was used to simulate the fully developed channel and turbulent flow past a circular cylinder, and to determine the impact of the dam-break flow on downstream structure. The HDSM is robust in respect to anisotropic mesh and is less sensitive to grid resolution, and would accurately describe the energy transfer from large-scale to SGS fluctuations and capture more fluctuations of turbulence with same meshes compared to the DSM.     

Imposing virtual origins on the velocity components in direct numerical simulations

The relative wall-normal displacement of the origin perceived by different components of near-wall turbulence is known to produce a change in drag. This effect is produced for instance by drag-reducing surfaces of small texture-size like riblets and superhydrophobic surfaces. To facilitate the research on how these displacements alter near-wall turbulence, this paper studies different strategies to model such displacement effect through manipulated boundary conditions. Previous research has considered the effect of offsetting the virtual origins perceived by the tangential components of the velocity from the reference, boundary plane, where the wall-normal velocity was set to zero. These virtual origins are typically characterised by slip-length coefficients in Robin, slip-like boundary conditions. In this paper, we extend this idea and explore several techniques to define and implement virtual origins for all three velocity components on direct numerical simulations (DNSs) of channel flows, with special emphasis on the wall-normal velocity. The aim of this work is to provide a suitable foundation to extend the existing understanding on how these virtual origins affect the near-wall turbulence, and ultimately aid in the formulation of simplified models that capture the effect of complex surfaces on the overlying flow and on drag, without the need to resolve fully the turbulence and the surface texture. From the techniques tested, Robin boundary conditions for all three velocities are found to be the most satisfactory method to impose virtual origins, relating the velocity components to their respective wall-normal gradients linearly. Our results suggest that the effect of virtual origins on the flow, and hence the change in drag that they produce, can be reduced to an offset between the virtual origin perceived by the mean flow and that perceived by the overlying turbulence, and that turbulence remains otherwise smooth-wall-like, as proposed by Luchini (1996). The origin for turbulence, however, would not be set by the spanwise virtual origin alone, but by a combination of the spanwise and wall-normal origins. These observations suggest the need for an extension of Luchini’s virtual-origin theory to predict the change in drag, accounting for the wall-normal transpiration when its effect is not negligible.     

Turbulent transport dissimilarity with modulated turbulence structure in channel flow of viscoelastic fluid

This experimental study investigated the turbulent transport dissimilarity with a modulated turbulence structure in a channel flow of a viscoelastic fluid using simultaneous particle image velocimetry and planar laser-induced fluorescence measurements. An instantaneous dye concentration field with fluctuating velocity vectors showed that mass was transferred by hierarchically large-scale wavy motions with inclination. A co-spectral analysis showed that the spatial phase modulation of the streamwise velocity and dye concentration fluctuations for the wall-normal velocity fluctuation corresponded to the relaxation time. The occurrence of intense dye concentration fluctuation and small streamwise velocity fluctuation in a thin boundary layer caused dissimilar turbulent transport because of the non-zero negative correlation of the streamwise velocity and dye concentration fluctuations for the wall-normal velocity fluctuation only on large scales. This explains the turbulent transport dissimilarity which leads to the zero averaged Reynolds shear stress and non-zero wall-normal turbulent mass flux.     

Delay in response of turbulent heat transfer against acceleration or deceleration of flow in a pipe

To predict the heat transfer enhancements that result from the application of a pulsating flow in a pipe, we experimentally investigated the turbulent heat transfer variations produced in response to sudden accelerations or decelerations to flows within a pipe. To accomplish this, the Reynolds numbers with the valve open (Re₁) and close (Re₀) were systematically varied in the range of 8,000 ≤ Re₁ ≤ 34,000 and 700 ≤ Re₀ ≤ 23,000, respectively, and in-pipe spatiotemporal heat transfer variations were measured using infrared thermography simultaneously with temporal variations to the in-pipe flow properties. Based on the experimental results, it was found that the heat transfer delays that occur in response to accelerations or decelerations can be characterized using the corresponding time lag Δt and first-order time constant τ. The values of Δt and τ can be expressed as non-dimensional forms of Δt/(ν/u_τ²) and τ/(R/u_τ), respectively, where u_τ is the pipe wall friction velocity, ν is the kinematic viscosity of the fluid, and R is the pipe radius.     

Surface pressure on a cubic building exerted by conical vortices

Experiments were performed to study surface pressure on a cubic building underlying conical vortices, which are known to cause severe structural damage and failure. The focus is on the effects of turbulence in the incident flow. Three turbulent boundary layers were created in a boundary layer wind tunnel. A wall-mounted cube, i.e. a cube situated on the horizontal ground floor surface of the wind-tunnel test section, was used as an experimental model. The cube was subjected to the incidence flow at 40°. Steady and unsteady pressure measurements were performed on the cube surface. The analysis suggests that conical vortices developed above the top surface of the wall-mounted cube. A larger mean suction was observed on the top cube surface in the less turbulent boundary layer. With an increase in turbulence in the incoming flow, the strong suction zones decreased in size. The fluctuating pressure coefficient profiles retained their shape when the turbulence in the upstream flow of the cube increased. The fluctuating pressure coefficient was observed to be larger in more turbulent flows. The pressure fluctuations were larger on the cube surface underlying outer boundaries of the conical vortex. The fluctuating pressure coefficient under the conical vortex was three to four times larger than in the weak suction zone on the central area of the top cube surface. Close to the leading cube corner, the pressure spectra were dominated by a single low frequency peak. As the conical vortex developed, this primary peak weakened and a secondary peak emerged at a higher reduced frequency. There is a general trend of shifting the pressure spectra towards higher reduced frequencies when the turbulence in the undisturbed incident flow increases.