How to create a new simulation?

The model requires several mesh files (the file format can be .ply or .off):

• mandible:
  • This is the closed surface mesh of the mandible.
  • the boundary conditions will be applied in the global cartesian reference frame. Thus, this mesh should be aligned with its principal axes (see MeshLab: Filters / Normals, Curvatures and Orientation / Transform: Align To Principal Axis)
  • A gmsh file (.msh) of the volume mesh can also be given
  • If this mesh is given as a surface mesh file, it will be meshed with gmsh. Try to mesh it manually in gmsh first because it sometimes fail.
• teeth:
  • A set of vertices selected on the top of the teeth where contact with the food occurs.
  • Using MeshLab:
    • Select the vertices using vertex selector ("Select Vertices" button) and the mouse.
    • When the selection is finished, invert the selection (pressing "SHIFT-i").
    • Delete all the selected vertices using "Filters / Selection / "Delete selected vertices".
    • Check that your model only contains the desired number of vertices.
    • Save these vertices to a file (e.g. teeth.ply)
• LTMJ/RTMJ:
  • Use the same procedure as for the teeth to save 1 vertex representing the left temporomandibular joint (LTMJ) and the right one (RTMJ) to 2 separate files (e.g. LTMJ.ply and RTMJ.ply)
• A series of muscle attachement areas (you can name them as you like)
  • for each muscle, using MeshLab:
    • Select the faces on which the muscle is attached to the bone
    • When the selection is finished, invert the selection (pressing "SHIFT-i").
    • Delete all the selected faces and vertices using "Filters / Selection / "Delete selected faces and vertices".
    • Save this group of faces to a file (e.g. muscle.ply).
• Each muscle requires a focal node towards which the muscle force is directed.
  • If this focal point can be "picked" in MeshLab on another mesh, you can use the same procedure as for LTMJ/RTMJ.
  • If not you can write the coordinate of the point in a .off file (several examples ara available in this folder).

The mesh files are specified in the input file:

path = 'dolicorhynchops/10k'
p['bone'] = f'{path}/mandible.ply'
p['contact_pts'] = f'{path}/teeth.off'
p['axis_pt1'] = f'{path}/LTMJ.off'
p['axis_pt2'] = f'{path}/RTMJ.off'
p['muscles'] = [
    {
        'file': f'{path}/Lmuscle.ply',
        'force': 100.,
        'focalpt': f'{path}/LmuscleF.off',
        'method': 'T'
    },
    {
        'file': f'{path}/Rmuscle.off',
        'force': 100.,
        'focalpt': f'{path}/RmuscleF.off',
        'method': 'T'
    }
]

For each muscle, the "method" can be 'U', 'T' or 'T+N' referring to (see Grosse et al. 2007)

• U: Ad Hoc Uniform Traction Model: a traction exerted by the muscle on each face is directed towards the focal node and constant in amplitude whatever its orientientation is.
• T: Tangential-Traction Model: Same as 'U' but the traction on the faces which do not have a direct line of sight to the focal point are projected onto the face and become tangential.
• T+N: Tangential-Plus-Normal-Traction Model: Same as 'T' but a normal pressure component is added to the faces which do not have a direct line of sight to the focal point. This normal component depends on the local curvature and the distance to the muscle fiber origin.

List of folders

• CLIpro: (V.Gaudichon) Clidastes propython FHSM17576
• contact: files used to create a meshed cylindrical surface so that we could model a bite onto a meshed surface (UNFINISHED). T
• cylinder: files used as a demontrator of boneload (see also view_cylinder_T.py and also view_cylinder_TN.py)
• Dolico_bon: (V.Gaudichon) Dolicorhyncops_bonneri_KUVP40001
• Dolico_osb: (V.Gaudichon)
• dolicorhynchops: first meshes sent by V.Gaudichon at the beginning of his master thesis
• PLTtymFrankencarpus: first meshes sent by V.Gaudichon at the beginning of his master thesis
• TYLber: (V.Gaudichon) Tylosaurus bernardi
• TYLnep: (V.Gaudichon) TYLnep_FHSM2209

run simulations on warson

List the files you want to run in a file named simus.txt. For example:

ls *750k* > simus.txt

Edit run_sim.py and check Metafor executable path. Then, run the run_sim.py script through at (check .forward to get an e-mail notification at the end of the calculations):

echo ./run_sim.py | at now -m

Copy zip_all_folders.sh and send_tbz2_to_dox.sh to workspace.

cp *.sh workspace
cd workspace
./zip_all_folders.sh       # <= create a .tbz2 archive for each simulation
./send_tbz2_to_dox.sh      # <= send the tbz2 files to DoX

To be fixed: (too many time steps - wrong input data?)

• TYLber_250k_7_2_bad.py
• TYLber_250k_15_2_bad.py All the other simulations ran successfully.