Presolver scriptfile error

[Tejas Canchi]

Hi guys

I am running a simple aorta model without the iliac bifurcations. So, it is essentially a cylinder. In the svpre file, the setwallBC contains the walls_combined.vtp address. But I just have a single wall and when I wrote the mesh files, it did not write the walls_combined.vtp file at all.

So, I pointed it to wall.vtp in mesh-surfaces in the svpre file. When i ran the presolver, I got the error "Line does not intersect vtkPolyData" . Can anybody say why that error is showing up? I get it after the write_restart line. I have attached the log file as well.

Thanks

Tejas

[Adam Updegrove]

Hi Tejas,

The walls_combined file is only written if there are multiple faces that are named with wall in it.

It looks to me like it is probably with the prescribed velocities on the inlet face. The file you have listed for the prescribed_velocities_vtp in the log file attached isn't complete.

If that isn't the issue, I would open the inlet file in paraview and make sure everything looks okay.

SimVascular Development Team

[Tejas Canchi]

Hi Adam,

I did check the inlet.vtp file in paraview. There doesnt seem to be anything wrong per se.

The inlet line you mentioned is showing up in the window when i load the presolver file but in the log, as I see it, it is not writing the extension.

Do you have any idea why it is doing that.

Thanks

Tejas

[Adam Updegrove]

Hi Tejas,

Were you able to solve this issue? If you haven't, something you can try is creating the bct files separately in the GUI. Follow the steps in the quick guide. Create the bct files from the GUI, and the restart and the geombc from the presolver. Let me know if you have the same issue.

SimVascular Development Team

[Eileen Hwuang]

Hi,

I am having trouble using svPre in linux command line but the exact same input file and svPre works successfully in Windows command prompt. Can you explain how to fix this error? Thank you.

Error:

Code: Select all

ubuntu@ip###:~/SV/Test6/pulsatile-sim-files$ ls
12-procs_case  bct.dat  geombc.dat.1   numstart.dat     rcr_f1.txt  restart.0.1
Aorta.flow     bct.vtp  mesh-complete  pulsatile.svpre  rcrt.dat    solver.inp
ubuntu@ip###:~/SV/Test6/pulsatile-sim-files$ cd mesh-complete/
ubuntu@ip###:~/SV/Test6/pulsatile-sim-files/mesh-complete$ ls
mesh-complete.exterior.vtp  mesh-surfaces
mesh-complete.mesh.vtu      walls_combined.vtp
ubuntu@ip###:~/SV/Test6/pulsatile-sim-files/mesh-complete$ cd ../
ubuntu@ip###:~/SV/Test6/pulsatile-sim-files$ svpre pulsatile.svpre
LINE 0000: mesh_and_adjncy_vtu mesh-complete/mesh-complete.mesh.vtu

).ERROR:  command could not be processed./mesh-complete.mesh.vtu
ubuntu@ip###:~/SV/Test6/pulsatile-sim-files$

This is the svpre file I am using:

Code: Select all

mesh_and_adjncy_vtu mesh-complete/mesh-complete.mesh.vtu
set_surface_id_vtp mesh-complete/mesh-complete.exterior.vtp 1
set_surface_id_vtp mesh-complete/mesh-surfaces/Aorta.vtp 2
set_surface_id_vtp mesh-complete/mesh-surfaces/LExtIl.vtp 3
set_surface_id_vtp mesh-complete/mesh-surfaces/LUtA.vtp 4
set_surface_id_vtp mesh-complete/mesh-surfaces/RExtIl.vtp 5
set_surface_id_vtp mesh-complete/mesh-surfaces/RUtA.vtp 6
fluid_density 1.05
fluid_viscosity 0.035
initial_pressure 133322
initial_velocity 0.0001 0.0001 0.0001
prescribed_velocities_vtp mesh-complete/mesh-surfaces/Aorta.vtp
bct_analytical_shape parabolic
bct_period 0.7228
bct_point_number 201
bct_fourier_mode_number 10
bct_create mesh-complete/mesh-surfaces/Aorta.vtp Aorta.flow
bct_write_dat bct.dat
bct_write_vtp bct.vtp
pressure_vtp mesh-complete/mesh-surfaces/LExtIl.vtp 6666
pressure_vtp mesh-complete/mesh-surfaces/LUtA.vtp 6666
pressure_vtp mesh-complete/mesh-surfaces/RExtIl.vtp 6666
pressure_vtp mesh-complete/mesh-surfaces/RUtA.vtp 6666
noslip_vtp mesh-complete/walls_combined.vtp
write_geombc geombc.dat.1
write_restart restart.0.1
write_numstart 0

Thanks for your help,
Eileen

[David Parker]

Hi Eileen,

If you copied the .svpre file from your Windows computer then it will have the DOS '\r' end of line characters in it. This will cause svpre to fail when reading the file. You can remove the '\r' by editing the file using vim, type :set ff=unix return and save the file.

Cheers,
Dave

[Eileen Hwuang]

It works! Thanks Dave!

[Charlie Bright]

Dear SimVascular Development Team,

Sorry to drag up an old forum, but I am experiencing a similar issue as Tejas did.

I am working on a coronary model, my model has several bypasses within it but otherwise it was created using the same workflow as the Coronary Normal tutorial.
I have been able to generate a model and a mesh file for the simulation.

In the simulation set up section, when I press the 'Create Data Files for Simulation' button the process runs but is extremely slow compared to how long it took on the Coronary Normal tutorial (though I think the physical size of this model is larger due to me miss-scaling the data).

However, once the process is completed, when I press the 'Show Details...' button the dialogue box ends in a string of errors similar to Tejas'. I cannot attach a separate file as he has done, but I'll type it in code at the bottom of the post. The end error message repeats around 260 times.

Comparing this message to that which you get upon a successful creation I'd expect the process to be complete after LINE 0039. Looking at the .svpre file seems to be correctly set up but running the simulation does not work.

I input the same .flow file and plv file from the tutorial to eliminate those variables causing the error.
Do you have any suggestions for how to fix this?

I have tried this on SimVascular versions: 2020.04.06 and 2019.08.09.
Both of these are using MITK 2018.4.2, VTK8.1.1, Qt 5.11.3.
The svSolver version I am using is 2019-05-28.
All running on Windows 10 using the GUI and MPI.

Thanks in advance,
Charlie

Code: Select all

LINE 0000: mesh_and_adjncy_vtu mesh-complete/mesh-complete.mesh.vtu

LINE 0001: set_surface_id_vtp mesh-complete/mesh-complete.exterior.vtp
LINE 0002: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_Aorta.vtp
LINE 0003: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_Aorta_2.v
LINE 0004: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_LCA_Main.
LINE 0005: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_LCA_Sec.v
LINE 0006: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_LCA_Sub_1
LINE 0007: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_RCA_Main.
LINE 0008: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_RCA_Sec.v
LINE 0009: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_RCA_Sub_1
LINE 0010: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_Top_Arter
LINE 0011: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_Top_Arter
LINE 0012: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_Top_Arter
LINE 0013: set_surface_id_vtp mesh-complete/mesh-surfaces/cap_Top_Arter
LINE 0014: fluid_density 1.06

LINE 0015: fluid_viscosity 0.04

LINE 0016: initial_pressure 0

LINE 0017: initial_velocity 0.0001 0.0001 0.0001

LINE 0018: prescribed_velocities_vtp mesh-complete/mesh-surfaces/cap_Ao
LINE 0019: bct_analytical_shape parabolic

LINE 0020: bct_period 1

LINE 0021: bct_point_number 201

LINE 0022: bct_fourier_mode_number 10

LINE 0023: bct_create mesh-complete/mesh-surfaces/cap_Aorta.vtp cap_Aor
LINE 0024: bct_write_dat bct.dat

LINE 0025: bct_write_vtp bct.vtp

LINE 0026: pressure_vtp mesh-complete/mesh-surfaces/cap_Aorta_2.vtp 0

LINE 0027: pressure_vtp mesh-complete/mesh-surfaces/cap_LCA_Main.vtp 0

LINE 0028: pressure_vtp mesh-complete/mesh-surfaces/cap_LCA_Sec.vtp 0

LINE 0029: pressure_vtp mesh-complete/mesh-surfaces/cap_LCA_Sub_1.vtp 0
LINE 0030: pressure_vtp mesh-complete/mesh-surfaces/cap_RCA_Main.vtp 0

LINE 0031: pressure_vtp mesh-complete/mesh-surfaces/cap_RCA_Sec.vtp 0

LINE 0032: pressure_vtp mesh-complete/mesh-surfaces/cap_RCA_Sub_1.vtp 0
LINE 0033: pressure_vtp mesh-complete/mesh-surfaces/cap_Top_Artery_1.vt
LINE 0034: pressure_vtp mesh-complete/mesh-surfaces/cap_Top_Artery_2.vt
LINE 0035: pressure_vtp mesh-complete/mesh-surfaces/cap_Top_Artery_3.vt
LINE 0036: pressure_vtp mesh-complete/mesh-surfaces/cap_Top_Artery_4.vt
LINE 0037: noslip_vtp mesh-complete/walls_combined.vtp

LINE 0038: write_geombc geombc.dat.1

LINE 0039: write_restart restart.0.1


ERROR: Line does not intersect vtkPolyData!
ERROR: Line does not intersect vtkPolyData!

[David Parker]

Hi Charlie,

The

Code: Select all

ERROR: Line does not intersect vtkPolyData!

message is coming from svPre code that creates the inflow velocity BCT file. It seems to mean there is a problem with the mesh inlet geometry, maybe it is not planar? If you can put the mesh files someplace I can download them I'll have a look.

I've opened a GitHub Issue for the lack of error handling in svpre (https://github.com/SimVascular/svSolver/issues/76).

Cheers,
Dave

[Charlie Bright]

Hi Dave,

Thanks for the quick response, and sorry for the delay getting back to you.

You were absolutely right, the inlet was not completely flat.
I managed to achieve a flat inlet by meshing without a boundary layer, in this case the files for simulation are created correctly.
When I come to run this simulation though, even with a pretty coarse mesh, I get a pop up saying the simulation was finished very quickly and under 'Show Details..'' is the error code:
"Fatal: Total Element Block Allocation Exceeded.
Use more processors for this problem or modify MAXBLK in the code."
Here is a link to a Google Drive folder that contains the coarse mesh with no boundary layer, and the extremely fine mesh that was broken in the first post.

https://drive.google.com/drive/folders/ ... sp=sharing

Do you have any suggestions for fixing this?
Could this be that the refinement of the model it is reading from is too high and I need to make that coarser along with the mesh?
I have uploaded the model into that folder as well.

Ideally I would like the mesh to have a boundary layer.
I have been experiencing some difficulties with meshing with a boundary layer within SimVascular recently, either the meshing procedure fails or flat faces become bumpy.
I have read some of the other threads where users also encounter the error message:
"PLC Error: A segment and facet intersect at a point (X, Y, Z), ERROR TetGen quit and returned error code 3".

I did some investigation into what was causing the issue and have attached a Google Doc with some of my findings on which you might find interesting, especially in regard to another forum in which you suggested meshing the model in parts to identify the issue causing piece.

https://docs.google.com/document/d/1wbW ... sp=sharing

In this test, the mesher failed at meshing a model containing 3 lofted segments, but the mesher succeeded in meshing the same model with 1 extra segment under the same meshing inputs.

I shall be installing Ubuntu today as I read in a forum post this may be a Windows SV version issue. I'll get back to you if I'm able to find anything out in regards to that handing a boundary layer mesh.
Thanks for your efforts in the forum, I've read some really thorough and in depth responses that've been very useful.

Thanks again,
Charlie