Numerical prediction of turbulent flow structure generated by a synthetic cross‐jet into a turbulent boundary layer

Abstract

SUMMARY A detailed numerical study using large‐eddy simulation (LES) and unsteady Reynolds‐averaged Navier–Stokes (URANS) was undertaken to investigate physical processes that are engendered in the injection of a circular synthetic (zero‐net mass flux) jet in a zero pressure gradient turbulent boundary layer. A complementary study was carried out and was verified by comparisons with the available experimental data that were obtained at corresponding conditions with the aim of achieving an improved understanding of fluid dynamics of the studied processes. The computations were conducted by OpenFOAM C++, and the physical realism of the incoming turbulent boundary layer was secured by employing random field generation algorithm. The cavity was computed with a sinusoidal transpiration boundary condition on its floor. The results from URANS computation and LES were compared and described qualitatively and quantitatively. There is a particular interest for acquiring the turbulent structures from the present numerical data. The numerical methods can capture vortical structures including a hairpin (primary) vortex and secondary structures. However, the present computations confirmed that URANS and LES are capable of predicting current flow field with a more detailed structure presented by LES data as expected. Copyright © 2011 John Wiley & Sons, Ltd.

Study of synthetic jets in boundary layers

“…They built a more complete physical mode for flow separation delay which involves a hierarchy of vortical structures showed in Figure 2.22b. Note that both the primary hairpin voritces and the pair of streamwise vortices near to the wall are similar to what are observed in studies of SJs in an attached TBL as discussed in previous section 2.3.2 (Dandois& Garner, 2006;Wu & Leschziner, 2009;Javadi & El-Askary, 2011) Zhang & Zhong (2011). suggested that the induced streamwise vortex pair, which are found locate closer to the wall and persist during a substantial portion of the actuation cycle, provides the main mechanism of flow separation delay by producing a downwash motion, which bring relatively more energetic fluids towards the wall.…”

“…For the hairpin vortices case, they revealed a hierarchy of coherent vortices consisting of primary hairpin vortices, secondary vortices and horseshoe vortex using Q-criterion (Hunt et al, 1988) in Figure 2.14a. For the tilted vortex ring case, the vortex ring was found accompanied by a pair of trailing vortices which further induced a pair of tertiary structures as shown in .16: (a)Vortices generated by single circular SJs in TBL, and (b) contours of phase-averaged streamwise velocity in a wall-parallel plane (Javadi & El-Askary, 2011).…”

“…circular SJs are commonly identified responsible for enhancing mixing in a TBL (Dandois & Garner, 2006;Wu & Leschziner, 2009;Javadi & El-Askary, 2011). In their work of examining grid resolution and turbulent models, Dandois & Garner (2006) Figure 2.16a.…”

“…To model µ t , Wall-Adapting Local Eddy-Viscoity (WALE) (Nicoud & Ducro, 1999) model is used in current LES simulation of single-SJ in TBL. Although Smagorinsky model (Smagorinsky, 1963) based on isotropy of the small-scale assumption has been broadly used because of its simplicity and accuracy, it does not have the versatility for various flow fields (Javadi & El-Askary, 2011;Ozawa et al, 2010). In the dynamic Smagorinsky-Lilly model (Germano et al, 1996), the Smagorinsky model coefficient is dynamically computed and varies in time and space over a fairly wide range.…”

“…15: Vortex structure induced by single circular SJs in TBL, (a)Iso-surface of Q = 0.2 colored with streamwise vorticity, and (b) contours of time-averaged streamwise vorticity in a spanwise wall-normal plane(Javadi & El-Askary, 2011).…”