[Kun_zheng]

Hi all,

I want to add a new wave theory to waves2Foam. Did anyone know how to do it?

The wave theory is not analytic, I have a fortran code to calculated it every time step. What I want is to add the subroutines of fortran code to the waves2Foam.

I think it is possible to compile the fortran code into an object file. Then link the Cpp file and object file. But I failed.

If anyone know how to do it, could you please give me some guidances.
Thank you for your time!

Kun

[ngj]

Hi Kun,

You most likely need to create an interface between fortran and C++; at least this was the approach taken for OceanWave3D. Consequently, I can only recommend that you look into both the source code of OceanWave3D and the external wave theory oceanWave3D to get a feel for the coupling.

In the present implementation, OceanWave3D exists as a library, which is executed through OpenFoam.

May I ask what the external method does?

Kind regards,

Niels

[Kun_zheng]

Hi Niels,

I find it. I think that is what I need. Thank you !
The external model I used is named Green Naghdi equations. It is a strong nonlinear wave model.

Best regards,
Kun

[ngj]

Hi Kun,

That is very interesting. If you get it to work, it would be great to include in waves2Foam. Please consider.

Kind regards

Niels

[Kun_zheng]

Originally Posted by ngj

Hi Kun,

That is very interesting. If you get it to work, it would be great to include in waves2Foam. Please consider.

Kind regards

Niels

Hi Niels,

After a minor modification, I get trouble in running the case " couplingOceanWave3D ". Could you see what I did and help me fix the trouble ?

Under your suggestion, I find the source code of OceanWave3D in " **/waves2Foam/ThirdParty/OceanWave3D-Fortran90/src/main ".
So I add a fortran code of " hello world " to test what would happen if a new subroutine is added in " oceanWave3D.C ". And then the test-subroutine is called in "oceanWave3D.C" like other previous subroutines.
After that I re-compile the " **/waves2Foam/ThirdParty " and the " **/waves2Foam/src ".
No errors came out till now.

Then I run the case "couplingOceanWave3D", the log of setWaveField says:
" symbol lookup error: **/platforms/linux64GccDPInt32Opt/lib/libwaves2Foam.so: undefined symbol: test "
But there was no errors when I compile the lib. I do not understand what was wrong. Could you help me ?
Thank you very much for your time !


Best regards,
Kun

[ngj]

Hi Kun,

The compilation gets a new pull from the git for OceanWave3D every time, so that might overwrite your modifications. I did not make the fortran side of the coupling, so I can only hypothesise on how things work. Right now, I do not have a good suggestion besides making a simpler test outside of OceanWave3D source.

Kind regards

Niels

[Kun_zheng]

Hi Niels,

Yes, I have checked the modification. It is still there. Thank you so much, you really help a lot ! I think I have to do more tests to find out where is the problem. Hope to make a progress soon.


Best regards,

Kun