[aranha92]

Hi all,

EDIT 1: Ok, I found out this is a parallelization issue. If I run the solver without parallelization it works perfectly. Now for making it work in parallel.

Things I tried for the parallelization:

1) Using reduce() with 2 approaches:


1.1) This one compiles, but the output stays the same.

scalar cellPhase1 = phase1[cellID];


 reduce(cellPhase1, sumOp<scalar>());


Info << cellPhase1 << std::endl;

1.2) This one does not compile.

 reduce(phase1, sumOp<volScalarField>());


Info << phase1[cellID]; << std::endl;

Error:

required from here
/opt/openfoam6/src/OpenFOAM/lnInclude/gatherScatter.C:63:15:  error: no matching function for call to  ‘Foam::GeometricField<double, Foam::fvPatchField,  Foam::volMesh>::GeometricField()’
             T value;

2) Using Pstream:


2.1) Does not compile

List<Field<volScalarField>> gatheredPhase1(Pstream::nProcs());


gatheredPhase1[Pstream::myProcNo()] = phase1;


Pstream::gatherList(gatheredPhase1);


Pstream::scatterList(gatheredPhase1);


phase1 = ListListOps::combine<Field<volScalarField>>
               (
                gatheredPhase1,
                accessOp<volScalarField>()
               );


Info << phase1[cellID]; << std::endl;

Error:

error:  no match for ‘operator=’ (operand types are ‘Foam::volScalarField {aka  Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh>}’  and ‘Foam::Field<Foam::GeometricField<double, Foam::fvPatchField,  Foam::volMesh> >’)
         );

I've been trying to modify the multiphaseInterFoam solver to read during runtime the fluid phase(the value between 0 and 1) of a specific mesh cell, but with no success. The problem I am dealing with consists of 3 phases, and this is what I tried doing at the moment:


First I store a reference of the volScalarField of each phase by iterating through the phases dictionary in the multiphaseMixture Library.

volScalarField &phase1 = mixture.getAlpha('phase1')
volScalarField &phase2 = mixture.getAlpha('phase2')
volScalarField &phase3 = mixture.getAlpha('phase3')

After that, a header file is ran every timestep that should give me the phase content of a specific cell, however I am finding it extremely difficult to understand what I am doing wrong.

phase1[specificCellID]
phase2[specificCellID]
phase3[specificCellID]

The Output comes out the same like this, for all the simulation timesteps:

phase1 phase2 phase3
7.46E-07 1.54E-05 1.48E-05

I don't know if I should read each phase independently or find some way to read all alphas at the same time. If someone more competent could provide me with a direction I would appreciate a lot.


Thank you!