[jeanbvb]

Hi guys.

I'm running a case that consist in a rectangular box with dimensions height , for the streamwise direction and in the spanwise direction, with about 700 K cells in total.

The OpenFoam version used is the 2.4.0 running over Debian 8. The solver used is the buoyantPimpleFoam.

The simulation runs well, and I've concluded 1 hour of simulation. Now I want to change some parametres like temperature and continue the simulation from 1 hour to another few hours.

As I already have the full fields of the initial 1 hour of simulation, the subsequent full fields are no longer necessary. I wonder if I could write only the probe locations instead of the whole fields after i restart the simulation from 1 hour, and write the full fields after a certain number of write intervals (like at every 500 write intervals), in case I need to restart the simulation.

I'm asking that because i'm running out of disk quota on the cluster, which is about 7.5 TB.

I'm hopefull that someone can help me, since I'm not very familiar with object-oriente programming.

Thanks in advance.

[akidess]

http://www.cfdsupport.com/OpenFOAM-T...t/node138.html

[jeanbvb]

Originally Posted by akidess

http://www.cfdsupport.com/OpenFOAM-T...t/node138.html

Hi akidess, thanks for you interest in help me.

I friend of mine made some changes in the buoyantPimpleFoam.C file as you can see bellow:

\*---------------------------------------------------------------------------*/

#include "fvCFD.H"
#include "rhoThermo.H"
#include "turbulenceModel.H"
#include "radiationModel.H"
#include "fvIOoptionList.H"
#include "pimpleControl.H"
#include "fixedFluxPressureFvPatchScalarField.H"

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

int main(int argc, char *argv[])
{
    #include "setRootCase.H"
    #include "createTime.H"
    #include "createMesh.H"
    #include "readGravitationalAcceleration.H"
    #include "createFields.H"
    #include "createFvOptions.H"
    #include "createRadiationModel.H"
    #include "initContinuityErrs.H"
    #include "readTimeControls.H"
    #include "compressibleCourantNo.H"
    #include "setInitialDeltaT.H"

    pimpleControl pimple(mesh);

    // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

/*    Info<< "\n#Starting time loop\n" << endl;*/

    IOdictionary controlDict(IOobject("controlDict", runTime.system(),mesh,IOobject::MUST_READ,IOobject ::AUTO_WRITE));
    scalar wi = readScalar(controlDict.lookup("writeInterval"));
    scalar deltat = readScalar(controlDict.lookup("deltaT"));
    scalar st = readScalar(controlDict.lookup("startTime"));
    Info<< "#wi = " << wi << endl;
    Info<< "#deltat = " << deltat << endl;
    Info<< "#st = " << st << endl;


    while (runTime.run())
    {
        #include "readTimeControls.H"
        #include "compressibleCourantNo.H"
        #include "setDeltaT.H"

/******************************************/ 
IOdictionary pLocs 
( 
IOobject 
( 
"probeLocations", 
runTime.constant(), 
mesh, 
IOobject::MUST_READ, 
IOobject::NO_WRITE 
) 
); 
const pointField& probeLocations(pLocs.lookup("probeLocations")); 

labelList probeCells(probeLocations.size(), -1); 

forAll(probeLocations, pI) 
{ 
probeCells[pI] = mesh.findCell(probeLocations[pI]); 
} 
/******************************************/ 

        runTime++;

        Info<< "#Time = " << runTime.timeName() << nl << endl;

        #include "rhoEqn.H"

        // --- Pressure-velocity PIMPLE corrector loop
        while (pimple.loop())
        {
            #include "UEqn.H"
            #include "EEqn.H"

            // --- Pressure corrector loop
            while (pimple.correct())
            {
                #include "pEqn.H"
            }

            if (pimple.turbCorr())
            {
                turbulence->correct();
            }
        }

        rho = thermo.rho();

/*        runTime.write();*/

/*        Info<< "#ExecutionTime = " << runTime.elapsedCpuTime() << " s"
            << "#  ClockTime = " << runTime.elapsedClockTime() << " s"
            << nl << endl;*/

/*******************************************/ 
    Info<< "#passo = " << (runTime.value()-st)/deltat << endl;

if ( (int)((runTime.value()-st)/deltat) % (int)wi == 0)
forAll(probeCells, pI) 
{ 
if(probeCells[pI] != -1) 
{ 
label cellNo = probeCells[pI]; 
Info << runTime.timeName()
<< " "
<< pI << ". Location: " 
<< probeLocations[pI]
<< ", alphaSgs: " << alphaSgs[cellNo]
<< ", k: " << k[cellNo]
<< ", muSgs: " << muSgs[cellNo]
<< ", p: " << p[cellNo]
<< ", phi: " << phi[cellNo]
<< ", pMean: " << pMean[cellNo]
<< ", pPrime2Mean: " << pPrime2Mean[cellNo]
<< ", p_rgh: " << p_rgh[cellNo]
<< ", T: " << T[cellNo]
<< ", U: " << U[cellNo]
<< ", UMean: " << UMean[cellNo]
<< ", UPrime2Mean: " << UPrime2Mean[cellNo] 
//add any other fields here 
<< endl; 
} 
} 

/*******************************************/ 


    }

/*    Info<< "End\n" << endl;*/

    return 0;
}


// ************************************************************************* //

The whole solver can be found in the following link:

https://drive.google.com/open?id=0B-...0JxZjlpWV9pSk0

The new solver was named buoyantPimpleFoamD and compiles ok.

The intention was to make the solver read a file called probeLocations inside the constant folder, this file contains the fields and locations that I want to probe, then write only the probes in a text file through the command "buoyantPimpleFoamD | grep Location > data".

When I try to run the case with the new solver it gives me the following error:

#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigFpe::sigHandler(int) at ??:?
#2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
#3  ? at ??:?
#4  __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#5  ? at ??:?

The case runs fine when I use the original buoyantPimpleFoam solver.

Could some one help me debug this error?

[akidess]

Did you even have a look at the link I provided? The changes made to the solver are unnecessary.

[jeanbvb]

Originally Posted by akidess

Did you even have a look at the link I provided? The changes made to the solver are unnecessary.

Yes, i did read your post.

At first it was still writting the whole fields in each time steps together whit the probes. Then a set a purgeTime in the controlDict and now is working the way I wanted.

Your solution was a way easier than what I had in my mind, at first i couldnt belive it.

Thank you so much.

[akidess]

Try setting writeInterval to whatever large value you want to have, and adding "outputInterval 1;" to the secion where you define the probe instead of purging the field data.

Edit: You also need to set "outputControl" accordingly (probably "timeStep").

[jeanbvb]

I added "outputControl timeStep;" and "outputInterval 1;" in the probes function and set "writeInterval 600;", so my probes will be written according my deltaT and the time folder will be written at every 600s of simulation, once I've set "writeControl runTime;".

Correct me if i'm wrong.