Direct numerical simulation of spatially developing turbulent boundary layer for skin friction drag reduction by wall surface-heating or cooling

Yukinori Kametani, Koji Fukagata

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Direct numerical simulation (DNS) of zero-pressure-gradient spatially developing turbulent boundary layer with uniform heating (UH) or cooling (UC) is performed aiming at skin friction drag reduction. The Reynolds number based on the free-stream velocity, U∞, the 99% boundary layer thickness at the inlet, δ0, and the kinematic viscosity, ν, is set to be 3000 and the Prandtl number is 0.71. The computational domain is set to be 9πδ0 × 3δ0 × πδ0 in the streamwise, wall-normal, and spanwise directions, respectively. A constant temperature is imposed on the wall. The Richardson number Ri for the buoyancy is varied in the range of -0.1 ≤ Ri ≤ 0.1. The DNS results show that UC reduces skin friction drag with a maximum drag reduction rate of 65%, while UH enhances it. The trend is similar to that in channel flow studied by Iida and Kasagi in 1997 and Iida et al. in 2002 and that in spatially developing boundary layer flow by Hattori et al. in 2007. Dynamical decomposition of skin friction drag using the identity equation (FIK identity, discussed by Fukagata et al. in 2002) quantitatively shows that drag reduction by UC is due to reduced Reynolds shear stress (RSS), while drag increase by UH is augmentation of RSS. The control efficiency of UC, however, is found to be largely negative; namely, net power saving is not achieved.

Original languageEnglish
Pages (from-to)1-20
Number of pages20
JournalJournal of Turbulence
Volume13
Publication statusPublished - 2012

Fingerprint

friction drag
drag reduction
Drag reduction
skin friction
Skin friction
turbulent boundary layer
Direct numerical simulation
direct numerical simulation
Boundary layers
Cooling
Drag
cooling
Heating
heating
Reynolds stress
shear stress
Shear stress
Richardson number
boundary layer thickness
boundary layer flow

Keywords

  • Control
  • Direct numerical simulation
  • Drag reduction
  • Turbulent boundary layer

ASJC Scopus subject areas

  • Mechanics of Materials
  • Computational Mechanics
  • Physics and Astronomy(all)
  • Condensed Matter Physics

Cite this

@article{4c6bbf8f02d541f29687784acb3bdc52,
title = "Direct numerical simulation of spatially developing turbulent boundary layer for skin friction drag reduction by wall surface-heating or cooling",
abstract = "Direct numerical simulation (DNS) of zero-pressure-gradient spatially developing turbulent boundary layer with uniform heating (UH) or cooling (UC) is performed aiming at skin friction drag reduction. The Reynolds number based on the free-stream velocity, U∞, the 99{\%} boundary layer thickness at the inlet, δ0, and the kinematic viscosity, ν, is set to be 3000 and the Prandtl number is 0.71. The computational domain is set to be 9πδ0 × 3δ0 × πδ0 in the streamwise, wall-normal, and spanwise directions, respectively. A constant temperature is imposed on the wall. The Richardson number Ri for the buoyancy is varied in the range of -0.1 ≤ Ri ≤ 0.1. The DNS results show that UC reduces skin friction drag with a maximum drag reduction rate of 65{\%}, while UH enhances it. The trend is similar to that in channel flow studied by Iida and Kasagi in 1997 and Iida et al. in 2002 and that in spatially developing boundary layer flow by Hattori et al. in 2007. Dynamical decomposition of skin friction drag using the identity equation (FIK identity, discussed by Fukagata et al. in 2002) quantitatively shows that drag reduction by UC is due to reduced Reynolds shear stress (RSS), while drag increase by UH is augmentation of RSS. The control efficiency of UC, however, is found to be largely negative; namely, net power saving is not achieved.",
keywords = "Control, Direct numerical simulation, Drag reduction, Turbulent boundary layer",
author = "Yukinori Kametani and Koji Fukagata",
year = "2012",
language = "English",
volume = "13",
pages = "1--20",
journal = "Journal of Turbulence",
issn = "1468-5248",
publisher = "IOP Publishing Ltd.",

}

TY - JOUR

T1 - Direct numerical simulation of spatially developing turbulent boundary layer for skin friction drag reduction by wall surface-heating or cooling

AU - Kametani, Yukinori

AU - Fukagata, Koji

PY - 2012

Y1 - 2012

N2 - Direct numerical simulation (DNS) of zero-pressure-gradient spatially developing turbulent boundary layer with uniform heating (UH) or cooling (UC) is performed aiming at skin friction drag reduction. The Reynolds number based on the free-stream velocity, U∞, the 99% boundary layer thickness at the inlet, δ0, and the kinematic viscosity, ν, is set to be 3000 and the Prandtl number is 0.71. The computational domain is set to be 9πδ0 × 3δ0 × πδ0 in the streamwise, wall-normal, and spanwise directions, respectively. A constant temperature is imposed on the wall. The Richardson number Ri for the buoyancy is varied in the range of -0.1 ≤ Ri ≤ 0.1. The DNS results show that UC reduces skin friction drag with a maximum drag reduction rate of 65%, while UH enhances it. The trend is similar to that in channel flow studied by Iida and Kasagi in 1997 and Iida et al. in 2002 and that in spatially developing boundary layer flow by Hattori et al. in 2007. Dynamical decomposition of skin friction drag using the identity equation (FIK identity, discussed by Fukagata et al. in 2002) quantitatively shows that drag reduction by UC is due to reduced Reynolds shear stress (RSS), while drag increase by UH is augmentation of RSS. The control efficiency of UC, however, is found to be largely negative; namely, net power saving is not achieved.

AB - Direct numerical simulation (DNS) of zero-pressure-gradient spatially developing turbulent boundary layer with uniform heating (UH) or cooling (UC) is performed aiming at skin friction drag reduction. The Reynolds number based on the free-stream velocity, U∞, the 99% boundary layer thickness at the inlet, δ0, and the kinematic viscosity, ν, is set to be 3000 and the Prandtl number is 0.71. The computational domain is set to be 9πδ0 × 3δ0 × πδ0 in the streamwise, wall-normal, and spanwise directions, respectively. A constant temperature is imposed on the wall. The Richardson number Ri for the buoyancy is varied in the range of -0.1 ≤ Ri ≤ 0.1. The DNS results show that UC reduces skin friction drag with a maximum drag reduction rate of 65%, while UH enhances it. The trend is similar to that in channel flow studied by Iida and Kasagi in 1997 and Iida et al. in 2002 and that in spatially developing boundary layer flow by Hattori et al. in 2007. Dynamical decomposition of skin friction drag using the identity equation (FIK identity, discussed by Fukagata et al. in 2002) quantitatively shows that drag reduction by UC is due to reduced Reynolds shear stress (RSS), while drag increase by UH is augmentation of RSS. The control efficiency of UC, however, is found to be largely negative; namely, net power saving is not achieved.

KW - Control

KW - Direct numerical simulation

KW - Drag reduction

KW - Turbulent boundary layer

UR - http://www.scopus.com/inward/record.url?scp=84871145919&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871145919&partnerID=8YFLogxK

M3 - Article

VL - 13

SP - 1

EP - 20

JO - Journal of Turbulence

JF - Journal of Turbulence

SN - 1468-5248

ER -