__/\\\________/\\\__/\\\\\\\\\\\__/\\\\\\\\\\\\\\\_____/\\\\\\\\\\_____________/\\\\\_
 _\/\\\_______\/\\\_\/////\\\///__\///////\\\/////____/\\\///////\\\________/\\\\////__
  _\/\\\_______\/\\\_____\/\\\___________\/\\\________\///______/\\\______/\\\///_______
   _\/\\\\\\\\\\\\\\\_____\/\\\___________\/\\\_______________/\\\//_____/\\\\\\\\\\\____
    _\/\\\/////////\\\_____\/\\\___________\/\\\______________\////\\\___/\\\\///////\\\__
     _\/\\\_______\/\\\_____\/\\\___________\/\\\_________________\//\\\_\/\\\______\//\\\_
      _\/\\\_______\/\\\_____\/\\\___________\/\\\________/\\\______/\\\__\//\\\______/\\\__
       _\/\\\_______\/\\\__/\\\\\\\\\\\_______\/\\\_______\///\\\\\\\\\/____\///\\\\\\\\\/___
        _\///________\///__\///////////________\///__________\/////////________\/////////_____

Pseudo-spectral code for DNS of Homogenous isotropic turbulence.

The kernel of the code is based on hit3d. The source code has been then rearranged and modified.

Main Additions:

• Paraview files (fields and particles).
• Automatic compilation and setup (compile.sh).
• Paralelizzation strategy modified.
• Optimization and code organization.
• Terminal output.
• Organization of the results folder.
• Compile.sh (automatic setup of input.f90 file).

Main deletions:

• LES part removed.

Future developments:

• Allen-Cahn equation for phase-field modeling.
• GPU-ready version.

Systems supported:

• MAC OS X (Intel)
• UNIX (ubuntu 18.04)

Things to be done/updated (ASAP)

• Particles tracking (input, output files, remove cint? only trilinear?)

To run a simulation:

• A) Setup the input.f90 file.
• B) execute ./compile.sh

Parallelization strategy

Four (three) possible strategies depending on the type of flow considered: single-phase flow or particles-laden flow.

• Single-phase flow:
  • Split=0 (no MPI communicator splitting): all task are solving the Eulerian fields and doing the stats (suggested strategy)
  • Split=1 (MPI communicator splitting): 2/3 of tasks solve the Eulerian fields and 1/3 do the stats.
• Particles-laden flow:
  • Split=0 (no MPI communicator splitting): not possible at the moment.
  • Split=1 (MPI communicator splitting): 1/2 of tasks solve the Eulerian fields and 1/4 do the stats and 1/4 track the particles.

Output and restart files.

Files containing the Eulerian fields (u_***,v_****,w_****,sc_***, etc.) and the particle positions (p_***) are stored in set_run/results

Visualization with Paraview

Two possible cases: single-phase and particles-laden flow.

• Single-phase flow:
  • Go to set_run/results/paraview_fields and run go.sh, paraview files are generated in the output folder.
• Particles-laden flow:
  • Go to set_run/results/paraview_fields and run go.sh, paraview files are generated in the output folder.
  • Go to set_run/results/paraview_particles and run go.sh, paraview files are generated in the output folder.

Validation database:

https://torroja.dmt.upm.es/turbdata/agard/chapter3/HOM03/

References:

• [1] S.G. Chumakov, "A priori study of subgrid-scale flux of a passive scalar in turbulence", Phys. Rev. E, 78 15563.
• [2] S.G. Chumakov, "Scaling properties of subgrid-scale energy dissipation", Phys. Fluids, 19 058104.
• [3] L. Machiels, "Predictability of Small-Scale Motion in Isotropic Fluid Turbulence", Phys. Rev. Lett. 79, 3411.
• [4] J. Jimenez, A.A. Wray, P.G. Saffman and R.S. Rogallo, "The structure of intense vorticity in isotropic turbulence", J. Fluid Mech., 255, 65-90.