assessment of turbulence modeling for compressible flow
play

Assessment of Turbulence Modeling for Compressible Flow Around - PowerPoint PPT Presentation

Assessment of Turbulence Modeling for Compressible Flow Around Stationary and Oscillating Cylinders by Alejandra Uranga A p p l i e d V e h i c l e Technologies August 21, 2006 Supervisors: Drs Nedjib Djilali and Afzal Suleman Dept. of


  1. Assessment of Turbulence Modeling for Compressible Flow Around Stationary and Oscillating Cylinders by Alejandra Uranga A p p l i e d V e h i c l e Technologies August 21, 2006 Supervisors: Drs Nedjib Djilali and Afzal Suleman Dept. of Mechanical Engineering - University of Victoria

  2. Outline  Introduction  Simulation Methodology  Stationary Cylinder  Oscillating Cylinder  Conclusions

  3. Introduction Ká rmán Vortex Street Periodic pattern of counter-rotating vortices caused by unsteady separation from a bluff body Introduction Methodology Stationary Oscillating Conclusions

  4. Introduction Ká rmán Vortex Street Periodic pattern of counter-rotating vortices caused by unsteady separation from a bluff body NASA satellite image 1999 Introduction Methodology Stationary Oscillating Conclusions

  5. Introduction Flow Around a Circular Cylinder  Interaction between 3 shear layers - Boundary layer - Free shear layer - Wake Introduction Methodology Stationary Oscillating Conclusions

  6. Introduction Flow Around a Circular Cylinder  Interaction between 3 shear layers - Boundary layer - Free shear layer - Wake  Transition to turbulence in - Wake Re D 200 → 400 - Free Shear layer Re D 400 → 150x10 3 - Boundary layer Re D 150x10 3 → 8x10 6 Introduction Methodology Stationary Oscillating Conclusions

  7. Introduction Scope  Simulation of turbulent flow around circular cylinders - Stationary Re D = 3900 - Oscillating Re D = 3600  Compare accuracy of turbulence models using same numerical procedure with respect to experiments and other simulations Introduction Methodology Stationary Oscillating Conclusions

  8. Methodology Numerical Simulation of Turbulent Flows

  9. Methodology The Need for Turbulence Models Example: Incompressible Momentum Equation Applying an average or filter operator (overbar) to the momentum equation yields  The terms , , are solved for  The cross terms are unknown closure problem Introduction Methodology Stationary Oscillating Conclusions

  10. Methodology Simulation of Turbulence LES DNS URANS Unsteady Large Eddy Simulation Direct Numerical Simulation Reynolds Averaged Navier-Stokes (One-point closure) large scale Solve all mean Subgrid-Scale Scales u i fluctuating Solve large Solve mean scale eddies u i very thin quantities u i grid required Model subgrid-scale Model Reynolds stress stresses Introduction Methodology Stationary Oscillating Conclusions

  11. Methodology Turbulence Models Considered  URANS - One equation Spalart-Allmaras - K-tau Speziale et al.  Large Eddy Simulation (LES) - Smagorinsky-Lilly  Very Large Eddy Simulation (VLES) - Adaptive k-tau Magagnato & Gabi (uses a URANS type subgrid-scale model) Introduction Methodology Stationary Oscillating Conclusions

  12. Methodology Computational Code  SPARC Structured PArallel Research Code  Finite Volume, Cell Centered, Block- Structured, Multigrid  Simulations are 3D Unsteady Compressible Viscous Introduction Methodology Stationary Oscillating Conclusions

  13. Stationary Cylinder Stationary Circular Cylinder in a Uniform Flow

  14. Stationary Cylinder Problem Setup  Cylinder diameter D = 1m  Flow velocity U 0 = 68.63m/s  Mach number Mach 0.2  Reynolds number Re D = 3900 Introduction Methodology Stationary Oscillating Conclusions

  15. Stationary Cylinder Computational Domain 2D figures: x-y plane at span center Introduction Methodology Stationary Oscillating Conclusions

  16. Stationary Cylinder URANS : Average Fields u / U 0 ω z D / U 0 SA Sp Introduction Methodology Stationary Oscillating Conclusions

  17. Stationary Cylinder URANS : Average Streamlines SA Sp Introduction Methodology Stationary Oscillating Conclusions

  18. Stationary Cylinder URANS : Average Profiles c p c f u / U 0 around around at x/D = 1.54 cylinder cylinder Introduction Methodology Stationary Oscillating Conclusions

  19. Stationary Cylinder LES-VLES : Streamlines LES VLES Introduction Methodology Stationary Oscillating Conclusions

  20. Stationary Cylinder LES-VLES : Average Fields u / U 0 ω z D / U 0 LES VLES Introduction Methodology Stationary Oscillating Conclusions

  21. Stationary Cylinder LES-VLES : Average Profiles c p c f u/U 0 around around at x/D=1.54 cylinder cylinder Introduction Methodology Stationary Oscillating Conclusions

  22. Stationary Cylinder 3-Dimensionality Streamwise velocity iso-surfaces LES URANS Sp Introduction Methodology Stationary Oscillating Conclusions

  23. Stationary Cylinder Comparison Introduction Methodology Stationary Oscillating Conclusions

  24. Oscillating Cylinder Circular Cylinder in Cross-Flow Oscillations

  25. Oscillating Cylinder Motion and Cases  Vertical sinusoidal motion  2D URANS k-tau Speziale  Reynolds number 3600  Lock-in: vortex shedding frequency matches cylinder motion frequency Introduction Methodology Stationary Oscillating Conclusions

  26. Oscillating Cylinder URANS Sp Fields Case IV f c / f 0 = 0.800 u/U 0 ω z D/U 0 Introduction Methodology Stationary Oscillating Conclusions

  27. Oscillating Cylinder Lock-in 63 % shedding frequency f S /f 0 32 % 1 % 2 % 40 % 46 % motion frequency f c /f 0 Introduction Methodology Stationary Oscillating Conclusions

  28. Conclusions Summary and Further Work

  29. Conclusions  comparison of results from different turbulence models with same numerical procedure  Spalart-Allmaras model - error in separation point  flow remains attached too long small recirculation zone  low back pressure  large drag  - Accurate Strouhal number Introduction Methodology Stationary Oscillating Conclusions

  30. Conclusions  K-tau Speziale model - Good mean global quantities  Strouhal number, drag, back pressure, separation point velocity profiles along the wake   LES and VLES - reveal secondary eddies - LES resolves dynamics in boundary layer  Oscillating Cylinder - No other numerical results in same regime - Lock-in over large range of motion frequencies - Further investigation required Introduction Methodology Stationary Oscillating Conclusions

  31. Conclusions Further Work  Better averages on LES and VLES  LES with Dynamic and Dynamic Mixed subgrid- scale models  LES of oscillating cylinder Introduction Methodology Stationary Oscillating Conclusions

  32. Questions

Recommend


More recommend