34

2D Flow Around a Cylinder

To use high fidelity Computational Fluid Dynamic Software to generate data to train Machine Learning algorithms for model Reduction.

1.Theory and Background:

We want to tackle a classical Fluid Dynamic Problem - Laminar flow of a 2D Cylinder

Critical Reynolds Number would be used to determine accuaracy of our implimentation as there is consensus on this value. Problem Description image Source : Fezia 2018

Choice of Software: Ansys Fluent vs OpenFoam:
Parameters to Consider -

  • Automation
  • Meshing
  • Community Support and Code Ba

2.Simulation:

Inlet Velocity is maintained at 5 m/s while we change Kinematics Viscosity

Boundary Conditions - At Cylinder wall u = v = 0 m/s At inlet u = 5 m/s V = 0 m/s

Problem Description image

Max Skewness 0.629 < 0.8

Cell count ~ 18,000

Mesh

2. Simulation:

Using OpenFoam for ease of Automation / Solver - icoFoam / The solver uses the PISO algorithim to solver the continuity equation

u=0\nabla \cdot u = 0

and momnentum equation

ut+(uu)(νu)=p \frac{\partial u}{\partial t} + \nabla \cdot (u \otimes u) - \nabla \cdot (\nu \nabla u) = - \nabla p

where uu is Velocity and pp is Kinematic pressure

Integration Time: 0.001 Simulation Time: 5 sec

3. Implementation Challenges

  • Boundary Conditions on Top and Bottom Walls and Outlet

    • Symmetry Plane vs No Slip
  • Step Time and Simulation Time

    • Sampling Frequency to detect Strouhal Frequency
  • Mesh Clarity for Boundary Layer

    • Balancing Mesh Thickness near Boundary layer and Skewness
  • Post Processing

4. Results and Discussion

Reynolds Graph The middle circle is peak frequency where as the 2 cricles abovce and below it are to judge the interval size

Total Simulation Run Time: ~2 hours

Mean Error 8.977 %

Critical Reynold Number: 46 - 49

Click to watch Video for Reynolds number of 100 video

6. Project Takeaways

  • Strouhal Number is highly sensitive parameter

  • Computational Fluid Dynamics Require

    • Appropriate Meshing
  • Appropriate Time Step

  • Understanding Dynamics of the Problem

7. References

  1. Fezai, Salwa & Oueslati, Fakher & Ben-Cheikh, Nader & Beya, Brahim. (2018). Sensitivity of wake parameters to diameter changes for a circular cylinder. International Journal of Modern Physics C. 29. 25. 10.1142/S0129183118500870.

  2. https://www.openfoam.com/documentation/guides/latest/doc/guide-applications-solvers-incompressible-icoFoam.html

  3. Williamson, C.H., & Brown, G.L. (1998). A SERIES IN 1/√Re TO REPRESENT THE STROUHAL–REYNOLDS NUMBER RELATIONSHIP OF THE CYLINDER WAKE. Journal of Fluids and Structures, 12, 1073-1085.

  4. Griffith, Martin & Leontini, J. & Thompson, Mark & Hourigan, Kerry. (2011). Vortex shedding and three-dimensional behaviour of flow past a cylinder confined in a channel. Journal of Fluids and Structures - J FLUID STRUCTURE. 27. 855-860. 10.1016/j.jfluidstructs.2011.02.007.