Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-16T07:11:21.755Z Has data issue: false hasContentIssue false

Speed and structure of turbulent fronts in pipe flow

Published online by Cambridge University Press:  27 January 2017

Baofang Song*
Affiliation:
Center of Applied Space Technology and Microgravity, University of Bremen, 28359 Bremen, GermanyInstitute of Fluid Mechanics, Friedrich-Alexander-Universität Erlangen-Nürnberg, 91058 Erlangen, Germany
Dwight Barkley
Affiliation:
Mathematics Institute, University of Warwick, Coventry CV4 7AL, UK
Björn Hof
Affiliation:
Institute of Science and Technology Austria, 3400 Klosterneuburg, Austria
Marc Avila
Affiliation:
Center of Applied Space Technology and Microgravity, University of Bremen, 28359 Bremen, GermanyInstitute of Fluid Mechanics, Friedrich-Alexander-Universität Erlangen-Nürnberg, 91058 Erlangen, Germany
*
†Email address for correspondence: baofang.song@zarm.uni-bremen.de

Abstract

Using extensive direct numerical simulations, the dynamics of laminar–turbulent fronts in pipe flow is investigated for Reynolds numbers between $Re=2000$ and 5500. We here investigate the physical distinction between the fronts of weak and strong slugs both by analysing the turbulent kinetic energy budget and by comparing the downstream front motion to the advection speed of bulk turbulent structures. Our study shows that weak downstream fronts travel slower than turbulent structures in the bulk and correspond to decaying turbulence at the front. At $Re\simeq 2900$ the downstream front speed becomes faster than the advection speed, marking the onset of strong fronts. In contrast to weak fronts, turbulent eddies are generated at strong fronts by feeding on the downstream laminar flow. Our study also suggests that temporal fluctuations of production and dissipation at the downstream laminar–turbulent front drive the dynamical switches between the two types of front observed up to $Re\simeq 3200$ .

Information

Type
Papers
Copyright
© 2017 Cambridge University Press 
Figure 0

Figure 1.Temporal evolution of (a) a puff at $Re=2000$, (b) a slug at $Re=2600$, and (c) a slug at $Re=5000$. The flow is from left to right, and turbulence is visualised by the transverse turbulent fluctuations $q$ (defined in § 2.1) in a frame of reference co-moving at the average of the upstream and downstream front speeds. The length scale in the vertical (radial) direction is stretched by a factor of 2 for better visualisation. Dark areas correspond to small fluctuations and bright areas correspond to large fluctuations. Time evolves in the upward direction and panels are separated by $10D/U$ in (a,c) and by $100D/U$ in (b), where $U$ is the bulk velocity and $D$ the pipe diameter. (d) Front speeds as a function of Reynolds number taken from the literature as indicated.

Figure 1

Figure 2.Division of a slug into upstream front, bulk and downstream front. A slug at $Re=2800$ is plotted at three time instants in a reference frame co-moving with the bulk flow. The flow is from left to right. Red triangles mark the position of the upstream and downstream fronts given by the cutoff $q=0.02$. The bulk region is bounded between green circles, which are located 15$D$ to the inside relative to the front positions (red triangles). Blue squares mark the beginning of the fronts and are located 25$D$ to the outside relative to the front positions. With these choices the fronts span 40$D$ in length.

Figure 2

Figure 3.(a) The structure of turbulent–laminar fronts from DNS seen as trajectories in the $(u,q)$-phase plane. Upstream fronts for slugs are shown with solid lines while downstream fronts are shown as dashed lines. Filled circles correspond to the bulk region of the slug. The dashed arrow indicates the direction of increasing Reynolds number. The turbulent puff at $Re=2000$ does not have a bulk region and for this case we do not distinguish upstream and downstream fronts with different line types. Trajectories form counterclockwise loops in the phase plane as one travels in space from the upstream to the downstream along the pipe. Phase-space trajectories have been obtained by averaging over time and over multiple runs at each $Re$ (typically 20 runs with 150–500 snapshots per run, depending on $Re$). (b) For comparison, the structure of turbulent–laminar fronts from a model of turbulent pipe flow (Barkley 2016). The trajectories are taken directly from Barkley (2016) figure 20.

Figure 3

Figure 4.Switching between weak and strong front at $Re=3000$. (a) Space–time diagrams showing the colour map of $q$ for two typical slugs in a moving frame of reference with the speed of 1.06$U$ (the speed of the weak front at this Reynolds number). Grey areas correspond to small fluctuations and dark red areas correspond to large fluctuations. The bars show the slopes, i.e. the inverse of the speeds, of transient weak and strong fronts in this frame of reference. (b) Front speed as a function of Reynolds number from DNS (up-triangles) and from an asymptotic analysis of fronts in a model system (dotted lines), reproduced from Barkley et al. (2015). The circles show the speeds of the transient weak and strong fronts highlighted in (a). Black down-triangles show the mean turbulent advection speed in the bulk region of slugs computed using (4.2) and discussed in § 4.

Figure 4

Figure 5.Kinetic energy budget (integrated over pipe cross-section) in the frame of reference co-moving with the fronts. Production (red bold line), dissipation (blue dash-dotted line), energy flux (cyan thin line) and convection (violet dashed line) are normalised by the mean turbulence production rate in the bulk. (a,b) Upstream front (UF) and downstream front (DF) of a slug at $Re=2600$. (c,d) Upstream front and downstream front of a slug at $Re=5000$. In (a,c,d) the peak of the production term is positioned at $z=0$. At each Reynolds number, an ensemble of about 500 snapshots is used to perform the energy budget analysis.

Figure 5

Figure 6.Advection of vortices near weak and strong fronts. (a) Weak front at $Re=2600$, (b) strong front at $Re=5000$. Vortices, visualised by using the $\unicode[STIX]{x1D6EC}_{2}$ criterion, are shown in the frame of reference co-moving with the downstream front (time evolves in the upward direction). The bulk turbulent region is on the left and the downstream laminar flow is on the right in the pipe. In each case one vortex is highlighted by red colour and a circle and is tracked in time. In (a) the time separation between two consecutive panels is $1.5D/U$, whereas in (b) it is $0.625D/U$.

Figure 6

Figure 7.Comparison between the average turbulent advection speed and the average centreline velocity as a function of $Re$. (b) Difference between the advection speed and the centreline velocity.

Song et al. supplementary movie

Advection of vortices near a weak downstream front at Re = 2600. The flow is from left to right, therefore, the bulk of the slug is on the left and the laminar flow downstream is on the right. Vortices, visualised by using the ?_2 criterion, are shown in the frame of reference co-moving with the front.

Download Song et al. supplementary movie(Video)
Video 9.1 MB

Song et al. supplementary movie

Advection of vortices near a strong downstream front at Re = 5000. The flow is from left to right, therefore, the bulk of the slug is on the left and the laminar flow downstream is on the right. Vortices, visualised by using the ?_2 criterion, are shown in the frame of reference co-moving with the front.

Download Song et al. supplementary movie(Video)
Video 8.1 MB