<Fourier coefficients file path> :: Command not found

[mehrdad roustaei]

Hello,

I am trying to run a moving boundary simulation with an unsteady inlet flow using svFSI. I have defined the inlet flow in a file in the following format:

<NumberOfTimePoints> <NumberOfFourierCoefficients>
<Time_1> <Flow_1>
<Time_2> <Flow_2>
.
.
.
<Time_NumberOfTimePoints> <Flow_NumberOfTimePoints>

However, when I try running, I get the following error asking for the Fourier coefficients:

ERROR occurred, see below for more explanation
ERROR: Near LINE 102 at <Add BC: Inlet>, searched COMMAND <Fourier coefficients file path> :: Command not found
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
STOP All processors are forced to stop by a fatal error

Please let me know how to address this.

Thank you.

[David Parker]

Hello,

Please post the .inp file here so I can have a look.

Cheers,
Dave

[mehrdad roustaei]

Hi David,

Please find the inp file attached along with the inflow BC. Please note that in this model, the wall, inlet and outlet surfaces move in time and a separate dat file is defined for prescribing the motion of each of the surfaces. I could not attach the .inp and .flow format, so the formats are change to .txt.

Thank you.
Mehrdad

[David Parker]

Hi Mehrdad,

I had a look at your svFSI.inp file. The Temporal values file path command is not being recognized because it has a tab (\t) before it

Code: Select all

od -c svFSI.inp 
0000000               \n  \n               A   d   d       B   C   :    
0000020    I   n   l   e   t       {  \n                           T   y
0000040    p   e   :       D   i   r   i   c   h   l   e   t  \n        
0000060                    T   i   m   e       d   e   p   e   n   d   e
0000100    n   c   e   :       U   n   s   t   e   a   d   y  \n  \t    
0000120        #   V   a   l   u   e   :       0  \n \t          T   e
0000140    m   p   o   r   a   l       v   a   l   u   e   s       f   i
0000160    l   e       p   a   t   h   :       /   h   o   m   e   /   m
0000200    e   h   r   d   a   d   /   D   o   w   n   l   o   a   d   s
0000220    /   m   o   v   i   n   g       b   o   u   n   d   a   r   y
0000240    /   V   _   i   n   .   f   l   o   w  \n                    
0000260        P   r   o   f   i   l   e   :       P   a   r   a   b   o
0000300    l   i   c  \n               }  \n  \n                        

so the solver then looks for a Fourier coefficients file path command.

Careful with those tabs!

Btw, the svFSI.inp file has several problems (no solid mesh Domain 2 defined, etc.).

Cheers,
Dave

[mehrdad roustaei]

Hi David,

Thank you! It worked after I deleted the tab in the inp file.

Regarding the mesh domain 2, since this is a moving boundary simulation, we only have a fluid domain 1 here. I was wondering if a mesh domain 2 for solid should be defined in this case as well.

I am getting the following error now on the moving boundary .dat file for the wall, not sure which Entry the error is referring to. I am attaching the first 1000 lines of the moving boundary file as txt.

DEBUG-INFO: At LINE 147, searched COMMAND <Temporal and spatial values file path> under <Add BC: Wall> :: Opened file from path </home/mehrdad/Downloads/moving boundary/mouse/moving boundaries/svFSI_Wall.dat>
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
ERROR occurred, see below for more explanation
ERROR: Entry 1 is out of bound in General
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
STOP All processors are forced to stop by a fatal error

[David Parker]

Hi Mehrdad,

Your temporal and spatial values file has the wrong format, should look something like

Code: Select all

2   2   65
0.000000000
100.000000000
1
0.000000000000000000e+00 -0.000000000000000000e+00 
0.000000000000000000e+00 -0.000000000000000000e+00 
66
0.000000000000000000e+00 -9.801714032956060363e-02 
0.000000000000000000e+00 -9.801714032956060363e-02 
...

I think for a moving boundary simulation you need to do an ALE simulation.

Cheers,
Dave

[Mehrdad Roustaei]

Hi David,

Thank you.

I checked your template and the one described here: http://simvascular.github.io/docssvFSI-Fluid.html#pres
I couldn't find the difference in the format between my file and these two, other than the vertex index.
I am using the GlobalNodeID as the vertex index in my file so that the code can match it with the node ids in the surface mesh files. I was wondering if that is necessary or it should always start from 1.

Regards,
Mehrdad

[Vijay Vedula]

HI Mehrdad,

Yes, currently the code requires these node indices to be the GlobalNodeID and not the local face nodal index. If you specify a different index, it is likely that the code will throw an error or will apply the BC to an incorrect node in your mesh.

-- Vijay

[mehrdad roustaei]

Hi Vijay,

Thank you very much for your reply. I edited the dat file to incorporate the GlobalNodeID and that error was resolved.
However, I am receiving this error now about the inlet BC in the mesh equation although the Temporal and spatial values file path is defined for the inlet.

DEBUG-INFO: At LINE 170, searched COMMAND <Time dependence> under <Add BC: Inlet> :: Read char value <General>
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
ERROR occurred, see below for more explanation
ERROR: Input file for General BC (bct.vtp) is not provided

[David Parker]

Hi Mehrdad,

For Time dependence: General you need to specify a Temporal and spatial values file path value.

Btw, you can often look in the svFSI source code to figure out the cause of an error. For example, I searched https://github.com/SimVascular/svFSI/bl ... urce/svFSI with is not provided and saw

Code: Select all

           lPtr =>list%get(fTmp,
     2         "Temporal and spatial values file path")
            IF (.NOT.ASSOCIATED(lPtr)) err = "Input file for General"//
     2         " BC (bct.vtp) is not provided"

The https://github.com/SimVascular/svFSI/bl ... master.inp file is also very useful for checking commands.

Cheers,
Dave