[be_inspired]

Hi all,

I have just implemented a generic canopy model over OF2.1.1 based on kEpsilon turbulence model ( 4 parameters controls the type of canopy model) based on the work of Mochida and others. The solver simpleFoam has been also modified to take into account the modification over U equation.

Until now the solver runs correctly and the results looks fine when comparing with the benchmarking of Mochida. It can runs on serial or parallel but when I run "blockMesh" or "reconstructPar" or whatever utility different to the solver I always receive the following warning:

--> FOAM Warning :
From function dlOpen(const fileName&, const bool)
in file POSIX.C at line 1175
dlopen error : /home/mmontero/OpenFOAM/mmontero-2.1.1/platforms/linux64Gcc45DPOpt/lib/libmyIncompressibleRASModel.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv
--> FOAM Warning :
From function dlLibraryTable:pen(const fileName&, const bool)
in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96
could not load "libmyIncompressibleRASModel.so"

I think that the problem is located and it is because there are 4 new parameters and 2 new volScalarFields (canopyDensity and canopyLeaf) but RASModel.C/H does not recognize them.

.H

            const volScalarField canopyDensity_;
            const volScalarField canopyLeaf_;

.C

    canopyDensity_
    (
        IOobject
        (
            "canopyDensity",
            runTime_.timeName(),
            mesh_,
            IOobject::READ_IF_PRESENT,
            IOobject::AUTO_WRITE
        ),
        mesh_
    ),

    canopyLeaf_
    (
        IOobject
        (
            "canopyLeaf",
            runTime_.timeName(),
            mesh_,
            IOobject::READ_IF_PRESENT,
            IOobject::AUTO_WRITE
        ),
        mesh_
      )

I hope someone could help me to clean this warning. The solver and the new turbulent model runs, converges and the results looks valid but...

Thank you very much
Marcelino

[Lieven]

Hi Marcelino,

Do you load the library in the controlDict-file of the case?

Cheers,

L

[be_inspired]

Yes, it is included.
The point is that the new turbulent model and the new solver are working correctly. I have compared the results with experimental data and looks "valid" ( it is quite difficult to know how is the density leaf of pines).
The warning only happens when using reconstructPart, paraFoam, blockMesh.... but no error or warning when performing the simulation.

I have looked for the same error in the forum and there is an old post about it but it seems that never was solved.