[b.simpson]

Hi all,

I am trying to set up an Age of Air function object in OpenFOAM 8. I am running buoyantSimpleFoam in Ubuntu 18.04.

I have tried implementing a function object from older threads (mainly Age of air with function object).

My controlDict code is:

AoA
	{
		type    scalarTransport;
		libs 	("libutilityFunctionObjects.so");
		enabled true;
		writeControl writeTime;
		log yes;
		nCorr 1;

		field AoA;
		active          true;
		autoSchemes     false;
		resetOnStartUp false;
		DT              1e-5;

		fvOptions
		{
			IncrementTime
			{
				type            semiImplicitSource;
				active          true;
				selectionMode all;
				volumeMode  specific;
				sources
				{
					injectionRateSuSp
					{
						AoA       (1 0);
					}
				}
			}
		}
	}

I am now getting the following error:

Selecting finite volume options model type semiImplicitSource
    Source: IncrementTime
    - selecting all cells
    - selected 1119784 cell(s) with volume 158.70844
[0] 
[0] 
[0] --> FOAM FATAL ERROR: 
[0] 
    request for regIOobject injectionRateSuSp from objectRegistry region0 failed
    available objects of type regIOobject are

Does anyone know what I need to put in place of "injectionRateSuSp"?

Or if there is a better method for calculating age of air?

Thanks in advance for any support.

Kind regards,
Ben

My 0/AoA file is:

FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
	location    "0";
    object      AoA;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 0 1 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    "xmin|xmax|ymin|ymax|zmax_north|zmax_south"
    {
        type            fixedValue;
        value           uniform 0;
    }
	"zmin|Ceiling|Floor|East_Wall|North_Wall|South_Wall|West_Wall|Computer|Person|zmax_ceiling"
    {
        type            zeroGradient;
    }
	Inlet
    {
        type            fixedValue;
        value           uniform 0;
    }
	
	#includeEtc "caseDicts/setConstraintTypes"
}

fvSolutions:

AoA
    {
        solver           smoothSolver;
        smoother         GaussSeidel;
        tolerance        1e-08;
        relTol           0.1;
        nsweep           1;
    }

fvSchemes:

div(phi,AoA) bounded Gauss upwind;

[Tobermory]

There were some changes in v8 to the semiImplicit source term syntax. Check out https://openfoam.org/release/8/ and the example in https://github.com/OpenFOAM/OpenFOAM...7e9747fe696aeb. I am guessing that you need to change:

				sources
				{
					injectionRateSuSp
					{
						AoA       (1 0);
					}
				}

to something like:

				sources
				{
					AoA
					{
						explicit 1;
						implicit 0;
					}
				}

Good luck!

[b.simpson]

Thank you for your reply Tobermory.

The changes you suggested worked and my AoA is now working.

Many thanks.

Ben

[HPE]

In case of anyone's interest,


Function object definitions for the `local mean age of air` and `normalised local mean age of air` metrics (where zero-value means fresh air):

scalarTransport1
{
    type            scalarTransport;
    libs            (solverFunctionObjects);
    field           LMA;
    schemesField    k;    // bounded Gauss limitedLinear 1;
    bounded01       false;
    alphaD          1;    // viscous contribution
    alphaDt         0;    // turbulent contribution
    // Adding fvOption source for residence time
    fvOptions
    {
        unitySource
        {
            type            scalarSemiImplicitSource;
            enabled         true;
            scalarSemiImplicitSourceCoeffs
            {
                selectionMode   all;
                volumeMode      specific;
                injectionRateSuSp
                {
                    LMA           (1 0);
                }
            }
        }
    }
    resetOnStartUp  false;
    timeStart       <timeStart>;
    executeControl  timeStep;
    executeInterval 1;
    writeControl    writeTime;
}

normalisedLocalMeanAge1
{
    type            reference;
    libs            (fieldFunctionObjects);
    field           LMA;
    result          nLMA;
    // normT = T / (V/Q)
    // normT: normalised local mean age of air
    // T: local mean age of air (Bartak et al., 2001, Eq. 1)
    // V: volume of air in the room [m3]
    // Q: volumetric flow rate through the inlet [m3/s]
    // V/Q: (estimated) mean age of fluid through a reference area (e.g. outlet)
    // scale = 1/(V/Q) = Q/V [1/s]
    scale           <Q/V>;
    offset          0;
    timeStart       <timeStart>;
    executeControl  writeTime;
    writeControl    writeTime;
}

The field `LMA` needs the initial/boundary condition file (e.g. in `0.orig/LMA`) and a setup in `fvSolution`:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  v2012                                 |
|   \\  /    A nd           | Website:  www.openfoam.com                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    object      LMA;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 0 0 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    inlet
    {
        type            fixedValue;
        value           uniform 0;
    }

    outlet
    {
        type            inletOutlet;
        inletValue      uniform 0;
        value           uniform 0;
    }

    ".*"
    {
        type            zeroGradient;
    }

    #includeEtc "caseDicts/setConstraintTypes"
}


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

solvers
{
    LMA
    {
        solver          PBiCGStab;
        preconditioner  DILU;
        tolerance       1e-8;
        relTol          0.01;
    }
}

relaxationFactors
{
    equations
    {
        LMA         0.7;
    }
}

[t2the0bi]

@HPE:

Thanks for providing the code for the function object, I successfully implemented it into my case. However, analysing the results uncovered some unrealistic values where the LMA value was bigger than the timestep (e.g. after 40 s the LMA showed a value of 60). Any idea where this mistake might originate from?

I am simulating a box-shaped room with natural ventilation (inlet and outlet window).

Thanks in advance!