Wall turbulence manipulation by large-scale 
streamwise vortices

Iuso, Gaetano; Onorato, Michele; Spazzini, Pier Giorgio; Cicca, Gaetano Maria Di

doi:10.1017/s0022112002002574

Cited by 49 publications

(50 citation statements)

References 34 publications

(57 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, Willis, Hwang & Cossu (2009 have forced large-scale linearly optimal α = 0 vortices, computed with the same approach used in this study, in a DNS of the turbulent pipe flow. A large amplification of the forcing is observed and the induced large-scale finite-amplitude streaks modify the flow at leading order and induce a reduction of the turbulent drag consistently with the results of Schoppa & Hussain (1998) (plane channel flow, DNS) and Iuso et al (2002) (boundary layer, experimental results).…”

Section: Relevance Of the Harmonic Forcing Analysissupporting

confidence: 80%

Linear non-normal energy amplification of harmonic and stochastic forcing in the turbulent channel flow

2010

View full text Add to dashboard Cite

To cite this version:Yongyun Hwang, Carlo Cossu. Linear non-normal energy amplification of harmonic and stochastic forcing in the turbulent channel flow. Journal of Fluid Mechanics, Cambridge University Press (CUP), 2010, 664 (December), pp. 51-73. 10.1017/s0022112010003629. hal-01020668 J. Fluid Mech. (2010 The linear response to stochastic and optimal harmonic forcing of small coherent perturbations to the turbulent channel mean flow is computed for Reynolds numbers ranging from Re τ = 500 to 20 000. Even though the turbulent mean flow is linearly stable, it is nevertheless able to sustain large amplifications by the forcing. The most amplified structures consist of streamwise-elongated streaks that are optimally forced by streamwise-elongated vortices. For streamwise-elongated structures, the mean energy amplification of the stochastic forcing is found to be, to a first approximation, inversely proportional to the forced spanwise wavenumber while it is inversely proportional to its square for optimal harmonic forcing in an intermediate spanwise wavenumber range. This scaling can be explicitly derived from the linearized equations under the assumptions of geometric similarity of the coherent perturbations and of logarithmic base flow. Deviations from this approximate power-law regime are apparent in the pre-multiplied energy amplification curves that reveal a strong influence of two different peaks. The dominant peak scales in outer units with the most amplified spanwise wavelength of λ z ≈ 3.5h, while the secondary peak scales in wall units with the most amplified λ + z ≈ 80. The associated optimal perturbations are almost independent of the Reynolds number when, respectively, scaled in outer and inner units. In the intermediate wavenumber range, the optimal perturbations are approximatively geometrically similar. Furthermore, the shape of the optimal perturbations issued from the initial value, the harmonic forcing and the stochastic forcing analyses are almost indistinguishable. The optimal streaks corresponding to the large-scale peak strongly penetrate into the inner layer, where their amplitude is proportional to the mean-flow profile. At the wavenumbers corresponding to the large-scale peak, the optimal amplifications of harmonic forcing are at least two orders of magnitude larger than the amplifications of the variance of stochastic forcing and both increase with the Reynolds number. This confirms the potential of the artificial forcing of optimal large-scale streaks for the flow control of wall-bounded turbulent flows.

show abstract

Section: Relevance Of the Harmonic Forcing Analysissupporting

confidence: 80%

Linear non-normal energy amplification of harmonic and stochastic forcing in the turbulent channel flow

2010

View full text Add to dashboard Cite

show abstract

“…In their experiment using jets to create streamwise vortices in a channel flow, Iuso et al 22 observed a similar shift in the power spectra of the streamwise velocity signal towards the lower frequencies when drag reduction occurs. In studies on polymer additives for drag reduction, similar shifts in the power spectra are often observed together with drag reduction.…”

Section: Power Spectral Measurementsmentioning

confidence: 80%

“…Streamwise vorticity can be effective in introducing spanwise flow components near the wall. 21,22 Choi et al 19 and Schoppa and Hussain 21 have shown that it takes very small spanwise flow components to be introduced to result in significant skin friction drag reduction. Unlike riblets, the introduction of spanwise flow components can result in skin friction reductions of up to 40% under optimum flow conditions.…”

Section: Power Spectral Measurementsmentioning

confidence: 99%

“…This has been done by active means such as via spanwise wall movement 19,20 or using transverse jets to introduce streamwise vortices that result in spanwise flow components near the wall. 21,22 Whether the wall is moved spanwise below a streamwise moving fluid, or the fluid is moved spanwise over a stationary wall is not critical, the method only requires the fluid to have relative spanwise components near the wall over which it flows. Using such methods to introduce spanwise flow components near the wall, a much larger drag reduction can be achieved than with riblets.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Mechanics of drag reduction by shallow dimples in channel flow

2015

View full text Add to dashboard Cite

Arrays of shallow dimples with depth to diameter ratios of 1.5% and 5% are studied in a turbulent channel flow at Reynolds numbers between 5000 and 35 000. Pressure measurements show that drag reduction of up to 3% is possible. The mechanism of skin friction drag reduction with dimples is the same as that observed for flat surfaces using active methods such as spanwise wall motions or transverse wall jets. The three dimensional dimples introduce streamwise vorticity into the flow which results in spanwise flow components near the wall. The result is that the normal energy cascade to the smaller scales is suppressed, which leads to a reduction in turbulent skin friction drag because of the stabilized flow. Increasing the dimple depth from 1.5% to 5% of its diameter increases the streamwise vorticity introduced, which leads to a greater reduction in skin friction. However, increasing the dimple depth also results in flow separation which increases form drag. The net effect to the total drag depends on the relative dominance between the drag reducing streamwise vorticity and the drag increasing flow separation region. As the Reynolds number increases, the region of flow separation can shrink and result in increasing drag reduction. By understanding the flow physics of drag reduction in dimples, there is opportunity to minimize the form drag by passive contouring of the dimples using non-spherical shapes to optimize the dimple performance.

show abstract

“…Experimental data were obtained in a previous control experiment of wall bounded turbulence in a channel flow [35], which the reader is referred to for a full description of the experimental setup and the conditions. The half channel height h is equal to 10 mm and the span of the test section is such that the aspect ratio width/height is equal to 14.…”

Section: Experimental Setup and Definitionsmentioning

confidence: 99%