[xuhuang]

Hi foamers,

I hope you are all doing well.

I would like to solve radaition heat transfer in combustion. When I use P1 model, it works fine.
But when I set fvDOM in radiationProperties dictionary, there was an error when the code goes into greyDiffusiveRadiationMixedFvPatchScalarField.C.

    const scalarField& Tp =
        patch().lookupPatchField<volScalarField, scalar>(TName_);

    const radiationModel& radiation =
        db().lookupObject<radiationModel>("radiationProperties");

    const fvDOM& dom(refCast<const fvDOM>(radiation));

The error is from the red part. It reports

--> FOAM FATAL ERROR:

    lookup of radiationProperties from objectRegistry region0 successful
    but it is not a radiationModel, it is a dictionary

    From function objectRegistry::lookupObject<Type>(const word&) const
    in file /opt/OpenFOAM/OpenFOAM-2.3.1/src/OpenFOAM/lnInclude/objectRegistryTemplates.C at line 181.

I have no idea why. I didn't change radaition codes and I have my own combustion model.

Any ideas?

Best regards,
Xu

[TomasDenk]

Hi,

what application are you using? Is it chtMultiRegionFoam?
Could you post content of radiationProperties dictionary for the region0 and snippet of the BC dictionary that uses greyDiffusiveRadiation?

I have used fvDOM for several radiation problems and I've never encountered such error so I will need information to locate the source of the issue.

[xuhuang]

Originally Posted by TomasDenk

Hi,

what application are you using? Is it chtMultiRegionFoam?
Could you post content of radiationProperties dictionary for the region0 and snippet of the BC dictionary that uses greyDiffusiveRadiation?

I have used fvDOM for several radiation problems and I've never encountered such error so I will need information to locate the source of the issue.

Hi Tomas,

Thanks very much for your reply.
I am not using chtMultiRegionFoam. I have my own solver and combustion model for comustion modelling. I have also implemented radiation model in reactingFoam, and fvDOM model is working fine. But for my own solver when using fvDOM it reports the error, but P1 is working properly.

Actually, in my BC I don't have region0. The below is my BC for radiation and radiationProperties settings.
IDefault BC:

boundaryField
{
    FUEL
    {
        type            greyDiffusiveRadiation;
        T               T;
        emissivityMode  lookup;
        emissivity      uniform 1.0;
        value           uniform 0.0;
    }
    AIR
    {
        type            greyDiffusiveRadiation;
        T               T;
        emissivityMode  lookup;
        emissivity      uniform 1.0;
        value           uniform 0.0;
    }
    WALL
    {
        type            greyDiffusiveRadiation;
        T               T;
        emissivityMode  lookup;
        emissivity      uniform 1.0;
        value           uniform 0.0;
    }    
    OUTLET
    {
        type            greyDiffusiveRadiation;
        T               T;
        emissivityMode  lookup;
        emissivity      uniform 1.0;
        value           uniform 0.0;
    }    
}

radiationProperties:

radiation on;

radiationModel  fvDOM;

fvDOMCoeffs
{
    nPhi        2;          // azimuthal angles in PI/2 on X-Y.(from Y to X)
    nTheta      4;          // polar angles in PI (from Z to X-Y plane)
    convergence 1e-3;   // convergence criteria for radiation iteration
    maxIter     10;         // maximum number of iterations
    cacheDiv    false;
}

// Number of flow iterations per radiation iteration
solverFreq 5;

absorptionEmissionModel constantAbsorptionEmission;

constantAbsorptionEmissionCoeffs
{
   absorptivity    absorptivity    [ m^-1 ]         0.5;
   emissivity      emissivity      [ m^-1 ]         0.5;
   E               E               [ kg m^-1 s^-3 ] 0;
}

scatterModel    none;

sootModel       none;

Thanks in advance.

Best regards,

Xu

[TomasDenk]

Hi Xu,

I'm afraid I won't of much help here. It all seems pretty normal to me and I don't see any reason why it shouldn't work. You might have to wait for someone smarter to figure it out.

Good luck with your modelling,
Tomas

[xuhuang]

Hi Tomas,

Thanks anyway.

Cheers,

Xu

[xuhuang]

Problem solved. It is a problem about objectRegistry in openfoam. In my solver, the radiationProperties dictionary is only registered as a dictionary type, but not a radiationModel type.

For more information about objectRegistry in openfoam, you may refer to https://openfoamwiki.net/index.php/O...objectRegistry.

Hope this help.

Cheers,

Xu

[TomasDenk]

Thanks for sharing the cause of the problem - I will read more about objectRegistry.

Bye,
Tom

[Raza Javed]

Originally Posted by TomasDenk

Hi,

what application are you using? Is it chtMultiRegionFoam?
Could you post content of radiationProperties dictionary for the region0 and snippet of the BC dictionary that uses greyDiffusiveRadiation?

I have used fvDOM for several radiation problems and I've never encountered such error so I will need information to locate the source of the issue.

Hi Tomas,


I have a question related to this thread.



I am using chtMultiRegionSimpleFoam and my Openfoam version is 4.1.


In my geometry, I have two regions (solid and air).


On solid, I am putting, 500K of temperature, and Air is at 300K temperature.


Now, I want to see the heat transfer through radiation from solid to air.


Just like we have heaters in our home and they heat up the room.


I have put velocity of air equals to zero, so that there would be no convection.


Can you please tell me, which radiation model will work for my case?

what I did uptill now is that I put radiation model OpaqueSolid on the solid region, and viewFactor on the air region. but I didn't see any difference.


It was exactly the same, as it is when I don't use radiation.


I shall be very thankful, if you can help me out in this.


Thank you

[TomasDenk]

Dear Raza,


I have to admit that I don't have much experience with viewFactor radiation model. My understanding is that it is more effective than DOM on simple geometries. Most of my models are quite complex and viewFactor can lead to high memory demands.


However, based on what you wrote, I don't see any reason why viewFactor shouldn't work. Any radiation model (DOM, P1,...) should show at least some impact on the heat transfer so I suppose there's a glitch in BCs or numerics. Could you share setup of your case (e.g. link to a zip)? For now, I won't need mesh.


Have a nice day,
Tom

[Aarav]

Originally Posted by TomasDenk

Hi,

what application are you using? Is it chtMultiRegionFoam?
Could you post content of radiationProperties dictionary for the region0 and snippet of the BC dictionary that uses greyDiffusiveRadiation?

I have used fvDOM for several radiation problems and I've never encountered such error so I will need information to locate the source of the issue.

I am using chtMultiregionFoam and added fvDOM. I just copied the boundary condition given in multiRegionHeater tutorials for viewFactor model. Can you please suuggest the correct temperature coupling boundary condition for this case?

[TomasDenk]

Hi Aarav,


Please have a look at compressible::turbulentTemperatureRadCoupledMixed.


If you are not happy with the results, I would need more info about what you're trying to do. Ideal is simplified case with few boxes, which you can share with me (e.g. post download link to .zip file) including all setup files and mesh.

[Aarav]

Originally Posted by TomasDenk

Hi Aarav,


Please have a look at compressible::turbulentTemperatureRadCoupledMixed.


If you are not happy with the results, I would need more info about what you're trying to do. Ideal is simplified case with few boxes, which you can share with me (e.g. post download link to .zip file) including all setup files and mesh.

Thank you sir for responding. I also used this one. The cases are running successfully. But the problem is I don't know what it is meant by qrNbr and qr in interface boundary condition. I am using FvDOM. Also Iam confused with G and Idefault BCs

[TomasDenk]

If you want to understand the variables, the best way is to look into the source code and analyze deeply definitions and operations performed on them.


It has been a while since I've done some programming in that direction, but I'll try to provide couple hints. The treatment of the boundary should be consistent from both sides. Therefore, you often need to access fields from the other domain/side, which usually has "Nbr" suffix.

valueFraction() = KDeltaNbr/(KDeltaNbr + KDelta);
refValue() = TcNbr;
refGrad() = (qr + qrNbr)/kappa(Tp);

In the code above, you need conductivity (KDelta) and incident radiation (qr) from both, your and neighbor sides to ensure consistency oftotal heat flux through the boundary.


As for G and IDefault BCs, you would need to start with basic thermodynamic laws and use lots of complex math (and some dirty tricks too) to develop the formulas. However, you should be fine understanding that IDefault defines BC for arbitrary number of ordinates (directions) and G is sum of radiative intensities from all ordinates (directions).

IRay_[rayI].addIntensity();
G_ += IRay_[rayI].I()*IRay_[rayI].omega();

The first line of the code above adds up intensities from all spectral bands, and the second line adds these "per ray" intensities into G. Note that IRay only denotes intensity in one ordinate and has nothing to do with radiation model casting rays based on Monte Carlo method.

[Aarav]

Originally Posted by TomasDenk

If you want to understand the variables, the best way is to look into the source code and analyze deeply definitions and operations performed on them.


It has been a while since I've done some programming in that direction, but I'll try to provide couple hints. The treatment of the boundary should be consistent from both sides. Therefore, you often need to access fields from the other domain/side, which usually has "Nbr" suffix.

valueFraction() = KDeltaNbr/(KDeltaNbr + KDelta);
refValue() = TcNbr;
refGrad() = (qr + qrNbr)/kappa(Tp);

In the code above, you need conductivity (KDelta) and incident radiation (qr) from both, your and neighbor sides to ensure consistency oftotal heat flux through the boundary.


As for G and IDefault BCs, you would need to start with basic thermodynamic laws and use lots of complex math (and some dirty tricks too) to develop the formulas. However, you should be fine understanding that IDefault defines BC for arbitrary number of ordinates (directions) and G is sum of radiative intensities from all ordinates (directions).

IRay_[rayI].addIntensity();
G_ += IRay_[rayI].I()*IRay_[rayI].omega();

The first line of the code above adds up intensities from all spectral bands, and the second line adds these "per ray" intensities into G. Note that IRay only denotes intensity in one ordinate and has nothing to do with radiation model casting rays based on Monte Carlo method.

Thank you sir. In viewFactor model we are defining Qr. But in fvDOM model we have to define IDefault. So my problem is that is the neighbouring field is Qr ?

[TomasDenk]

I'm not sure if I understand your question correctly, but I'll try with this...


Qr = sum_all_ordinates(incident_radiation * cos(incident_angle, surface_normal) * surface_area

[Aarav]

Originally Posted by TomasDenk

I'm not sure if I understand your question correctly, but I'll try with this...


Qr = sum_all_ordinates(incident_radiation * cos(incident_angle, surface_normal) * surface_area

Thank you sir