DNS and LES of turbulent flow in a closed channel featuring a pattern of hemispherical roughness elements

Significance Statement

A detailed simulation campaign based on high fidelity LES and DNS was performed to investigate the effect of hemispherical roughness elements on fully developed turbulent flow between parallel plates. Variations in the shear Reynolds number (Reτ=180-400), element height (k+=10-20), element spacing (s+/k+=2-6) and distribution pattern (regular square lattice vs. random pattern) were explored to assess their effect on the friction factor, mean velocity and turbulent stresses profiles. The present LES/DNS study differs from the abundant published works, centering on large sharp-edged roughness obstructions (k+=40-100), in that it deals with the transitional roughness regime, where the Reynolds number is relatively high (for a DNS), and the roughness elements are small and of round shape, and could thus be randomly distributed. Such a situation is relevant to various energy systems such as fossil boilers and nuclear reactors, in which vapor bubbles are attached to the wall in subcooled flow boiling and effectively behave like small (<100 μm), near-hemispherical, roughness elements. In such cases, using the laws of smooth wall channel flow would give an under-prediction of the friction factor.

Overall the DNS results show a clear separation between the inner wall-layer, which is affected by the presence of the roughness elements, and the outer layer, which remains relatively unaffected. Roughness element height has a strong effect on the friction factor and on the mean velocity profile. The friction factor increases proportionally to the roughness element height, while the mean velocity profile shifts downward proportionally to the roughness element height. The type of roughness dealt with here also affects the turbulent stresses. In particular, the study reveals that the presence of roughness elements of this shape promotes locally the instantaneous flow motion in the lateral direction in the wall layer, which was found to cause a transfer of energy from the streamwise Reynolds stress to the lateral component; the wall-normal stress component, however, remains unaffected regardless of the roughness height or arrangement. Consequently, the shape of the turbulent kinetic energy profile changes, featuring a lower peak value and forward shift away from the wall as compared to the smooth channel case.

Element spacing changes the point of re-attachment of the boundary layer downstream of an element; at low spacing, recirculation cells spanning the gap between adjacent elements appear. However, for given element height, spacing has a relatively weak effect on friction factor and mean velocity profile, which is somewhat surprising, given the previous results for channels with two-dimensional ribs reported in the literature. Finally, a random distribution pattern of the elements does not affect either the friction factor or the mean velocity appreciably


About The Author

Professor  and  Associate  Department  Head,  Nuclear  Science  and  Engineering
Director,  Center  for  Advanced  Nuclear  Energy  Systems  (CANES)
Massachusetts  Institute  of  Technology
Email:  [email protected];
Office:  (617)  253–‐7316

Jacopo  Buongiorno  is  Professor  and  Associate  Department  Head  of  Nuclear  Science  and  Engineering  at  the  Massachusetts  Institute  of  Technology  (MIT).  He  earned  a  B.S.  and  a  Ph.D.  in  nuclear  engineering  from  the  Polytechnic  of  Milan  and  MIT,  respectively.  He  teaches  a  variety  of  undergraduate  and  graduate  courses  in  thermo‐  fluids  engineering  and  nuclear  reactor  engineering.  His  areas  of  technical  expertise  and  research  interest  are  reactor  safety  and  design,  nanofluid  technology,  and  two‐phase  flow  and  heat  transfer.  For  his  work  in  these  areas  and  his  teaching  at  MIT  Prof.  Buongiorno  won  several  awards,  including,  recently,  the  MacVicar  Faculty  Award  (MIT,  2014),  and  Landis  Young  Member  Engineering  Achievement  Award  (American  Nuclear  Society,  2011).  Professor  Buongiorno  is  the  Director  of  the  Center  for  Advanced  Energy  Systems  (CANES);  Co‐Director  of  the  Reactor  Technology  Course  for  Nuclear  Utility  Executives,  which  is  offered  jointly  by  MIT  and  the  Institute  of  Nuclear  Power  Operations  (INPO);  and  a  consultant  for  the  nuclear  industry  (e.g.  AREVADCNS,  B&W,  Westinghouse,  South  Texas  Project)  in  the  area  of  reactor  thermal  hydraulics.  He  served  on  the  ANS  Special  Committee  on  Fukushima,  and  is  on  the  Accrediting  Board  of  INPO’s  National  Academy  of  Nuclear  Training  (NANT).  He  is  a  member  of  the  American  Nuclear  Society  (ANS),  the  American  Society  of  Mechanical  Engineers  (ASME),  and  the  Defense  Science  Study  Group  (DSSG).


About The Author

Despoina  Chatzikyriakou,
PhD  Senior  Engineer  Energy  Research  Group
Research  &  Development  Toyota  Motor  Europe  Brussels,  Belgium
Email:  [email protected]

Dr.  Chatzikyriakou  earned  a  Diploma  in  Mechanical  Engineering  and  a  Master’s  degree  in  Computational  Mechanics  from  the  National  Technical  University  of  Athens  in  Greece.  In  2010,  she  graduated  from  Imperial  College,  London  with  a  PhD  in  Mechanical  Engineering.  She  then  worked  as  a  researcher  in  the  Nuclear  Science  and  Engineering  Department  at  MIT  applying  advanced  computational  methods  to  investigate  two‐phase  flow  and  heat  transfer  phenomena  for  nuclear  reactor  technology  applications.

Since  2012,  Dr.  Chatzikyriakou  is  a  Senior  Engineer  and  Researcher  in  the  Energy  Research  Group  of  Toyota  Motor  Europe.  Her  current  work  focuses  on  the  development  of  innovative,  multidisciplinary  projects  and  business  models  to  address  the  specificity  of  the  automotive  EU  market.  She  is  responsible  for  energy  modelling  activities  such  as  energy  demand/supply  and  vehicle  emissions  projections  up  to  2030  as  well  as  for  new  energy  management  solutions  for  hybrid  powertrains.   

About The Author

Djamel  Lakehal
CEO  of  ASCOMP  AG  Switzerland  &  ASCOMP  Inc.  USA.
Research  scientist,  Department  of  Nuclear  Science  &  Engineering,
D‐NSE,  Massachusetts  Institute  of  Technology
Email:  [email protected][email protected]

Djamel  Lakehal  is  the  founder  and  CEO  of  ASCOMP.  He  has  been  hosted  at  the  MIT  (D‐NSE)  as  invited/external  Research  Scientist,  and  acts  as  an  adjunct  lecturer  at  ETH  Zurich  and  ENS  Paris  France.  Djamel  obtained  a  Master  degree  and  a  Ph.D  in  Fluid  Mechanics  from  Ecole  Centrale  of  Nantes,  France.  In  the  period  1995‐1997,  he  collaborated  with  Prof.  W.  Rodi  at  the  University  of  Karlsruhe  as  a  post‐doctoral  researcher  working  on  advanced  turbulence  modelling.  As  a  Research  Associate  at  the  Institute  of  Fluid  Mechanics  at  TU‐Berlin  (1997‐1998),  he  collaborated  with  Prof.  F.  Thiele.  In  1998  he  joined  the  Institute  of  Energy  Technology  of  the  ETH  Zurich  as  a  Group  Leader  and  Senior  Lecturer.  He  initiated  the  creation  of  the  Computational  Multi‐fluid  Dynamics  Group,  hosting  doctoral  and  post‐doctoral  scientists,  performing  cutting‐edge  research  in  computational  multi‐fluid  dynamics  (CMFD).  In  June  2004  he  was  awarded  the  French  Habilitation  Degree  jointly  from  the  Ecole  Centrale  of  Lyon  and  ETH  Zurich.  In  January  2004,  he  founded  ASCOMP,  a  Company  specialized  in  industrial  fluid  dynamics  and  heat  transfer  for  energy‐related  technologies.  Dr  Lakehal  was  hosted  as  invited  Professor  in  various  universities,  including  KTH  Stockholm,  UC  Sta  Barbara,  ENS  Cachan  Paris,  and  Imperial  College  London.  Dr  Lakehal  acts  as  a  reviewer  of  research  projects  for  the  EU  (DG  Energy).  He  authored  about  70  journal  papers  and  more  than  120  conference  papers  in  various  areas  related  to  fluid  mechanics  and  heat  transfer.  He  was  invited  ten  times  as  a  keynote  speaker  in  international  conferences.       

About The Author

Daniel  Caviezel
Head  of  Product  Development,  ASCOMP  AG  Switzerland
Email:  [email protected]

Daniel  Caviezel  is  the  Head  of  Product  Development  at  ASCOMP,  more  precisely;  he  leads  the  development  group  of  the  TransAT  CFD/CMFD  platform.  Daniel  earned  his  Masters  in  Computational  Science  from  ETH  Zurich  in  2008.  He  joined  ASCOMP  immediately  after  his  MS  degree,  and  was  initially  dealing  with  the  HPC  porting  of  the  platform.  Daniel  is  an  expert  in  computational  physics  and  simulation  methods  with  a  broad  understanding  of  software  engineering  principles.

(Left) Instantaneous streamwise velocity contours for slices at the hemispheres crest and in between hemispheres.
(Right) Instantaneous velocity contours at a slice in the middle of the hemispheres. The recirculation regions in between the hemispheres can be clearly seen.

DNS and LES of turbulent flow in a closed channel featuring a pattern of hemispherical roughness elements advances in engineering






Journal Reference

D. Chatzikyriakou1, J. Buongiorno1, D. Caviezel2, D. Lakehal1, 2

1 Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
2 ASCOMP GmbH, Zurich, Switzerland


Direct Numerical Simulations (DNS) and Large Eddy Simulations (LES) were performed for fully-developed turbulent flow in channels with smooth walls and walls featuring hemispherical roughness elements at shear Reynolds numbers Reτ = 180 and 400, with the goal of studying the effect of these roughness elements on the wall-layer structure and on the friction factor. The LES and DNS approaches were verified first by comparison with existing DNS databases for smooth walls. Then, a parametric study for the hemispherical roughness elements was conducted, including the effects of shear Reynolds number, normalized roughness height (k+ = 10–20) and relative roughness spacing (s+/k+ = 2–6). The sensitivity study also included the effect of distribution pattern (regular square lattice vs. random pattern) of the roughness elements on the walls. The hemispherical roughness elements generate turbulence, thus increasing the friction factor with respect to the smooth-wall case, and causing a downward shift in the mean velocity profiles. The simulations revealed that the friction factor decreases with increasing Reynolds number and roughness spacing, and increases strongly with increasing roughness height. The effect of random element distribution on friction factor and mean velocities is however weak. In all cases, there is a clear cut between the inner layer near the wall, which is affected by the presence of the roughness elements, and the outer layer, which remains relatively unaffected. The study reveals that the presence of roughness elements of this shape promotes locally the instantaneous flow motion in the lateral direction in the wall layer, causing a transfer of energy from the streamwise Reynolds stress to the lateral component. The study indicates also that the coherent structures developing in the wall layer are rather similar to the smooth case but are lifted up by almost a constant wall-unit shift y+ (∼10–15), which, interestingly, corresponds to the relative roughness k+ = 10.

Go To International Journal of Heat and Fluid Flow