1. Computational Investigation of Buoyancy Influenced Flows and Benchmarking exercise Presenter: Yacine Addad (PDRA) (KNOO Team: D. Laurence, M. Cotton, A. Keshmiri, S. Rolfo) (Collaborators: S. Benhamadouche, R. Howard, F. di Mare) School of Mechanical, Aerospace & Civil Engineering (MACE) The University of Manchester 7th Int. ERCOFTAC Symposium-ETMM7- 4-6 June 2008,Limassol, Cyprus

2. AGR working scheme The Heated pipe: Relevance to AGR and VHTR

3. RANS models predictions buoyancy aiding

4. Results for Fully-Developed Forced Convection. (no buoyancy) Expt. of Polyakov & Shindin

5. Test Case Description Re=180 based on R and u . Boussinesq Approximation. Periodic Flow. Lz=30R Grid1: 1.61 million Grid2: 4.83 million Grid1 Resolution:  r + min =1.0,  + =6.28,  z + =18. Grid2 Resolution:  r + min =1.0,  +=6.28,  z+=7.03. Buoyancy effects on turbulence in VHTR channels: WR-LES

6. Nusselt number as a function of the buoyancy parameter from You et al. For RANS models predictions, see the paper by Amir Keshmiri et al. Ratio of SGS to molecular viscosity

7. Case1: Forced convection Gr/Re 2 =0.0 Buoyancy effects on turbulence in VHTR channels: WR-LES Data normalized by the bulk velocity

8. Cases from Gr/Re 2 =0.063 to Gr/Re 2 =0.241 Buoyancy effects on turbulence in VHTR channels: WR-LES Data normalized by the bulk velocity

9. NEARLY-HORIZONTAL SHALLOW CAVITY TEST CASE Ra= 4.16  10 8 NCELL= 3 million Boussinesq approximation Pr=0.71 (Air)  =5° LES Grid (Case1) 0.8h Plan Y-Z

10. NEARLY-HORIZONTAL SHALLOW CAVITY TEST CASE LES RESULTS Q=0.05

11. NEARLY-HORIZONTAL SHALLOW CAVITY TEST CASE LES RESULTS

12. NEARLY-HORIZONTAL SHALLOW CAVITY TEST CASE RANS RESULTS Low-Re. models tested: k-  Lien et al. (1996) k-  Wilcox (1998) k-  SST Menter (1993) v 2 -f Lien & Durbin (1996) Buoyancy term included in k-equation thermal fluxes

13. Energy conservation: Taylor-Green vortices test case.

14. Taylor-Green vortices test case: mesh generation. Resolution: 60 x 60 Time step: 0.01 => CFL max < 0.2

15. Effects of the velocity/pressure coupling

16. Comparison between the codes for RR=0.75 (6/8)‏ RR=0.75  6/8

17. N xNyNz LES68 to 2004642 to 100 DNS256193192 Re=395 Domain 2   2    LES Ncells= 443,272 DNS Ncells = 9,486,336 (Ref: Moser et al. 1999) Channel flow test case

18. Channel flow test case: Mean Velocity predictions

19. Channel flow test case: Reynolds Stresses

20. Budgets of the Reynolds stress components Ref: T. Omori et al. Int. J. Heat and fluid flow 2008

21. Viscus diffusion term Dissipation term Channel flow test case: Budget for k

22. Budget for k STAR: Grid (141x184x177) = 4.59 million DNS : Grid (256x193x192)= 9.48 million

23. STAR: Grid (141x184x177) = 4.59 million DNS : Grid (256x193x192)= 9.48 million Budget for uu

24. STAR: Grid (141x184x177) = 4.59 million DNS : Grid (256x193x192)= 9.48 million Budget for vv

25. STAR: Grid (141x184x177) = 4.59 million DNS : Grid (256x193x192)= 9.48 million Budget for ww

26. STAR: Grid (141x184x177) = 4.59 million DNS : Grid (256x193x192)= 9.48 million Budget for uv

27. Conclusions Acknowledgements: TSEC programme 'KNOO' EPSRC funding under grant EP/C549465/1 for PDRA and Computer resources