CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   How to get a single value from a volScalarField including the dimension (http://www.cfd-online.com/Forums/openfoam/93820-how-get-single-value-volscalarfield-including-dimension.html)

safre October 27, 2011 07:17

How to get a single value from a volScalarField including the dimension
 
Hello

I have problems getting the dimension correct when I am doing a simple sum. My equation is:

aveTSlab = aveTSlab + T[n];

where

aveTSlab is defined as

dimensionedScalar aveTSlab
(
"aveTSlab",
dimensionSet( 0, 0, 0, 1, 0, 0, 0),
scalar( 0)
);

and

volScalarField T
(
IOobject
(
"T",
runTime.timeName(),
mesh,
IOobject::MUST_READ
),
mesh
);

Both dimensions are defined correctly but when I take a single value T[n] it does not come with any dimension but the error is as:

--> FOAM FATAL ERROR:
LHS and RHS of + have different dimensions
dimensions : [0 0 0 1 0 0 0] + [0 0 0 0 0 0 0]

Please give advise how to get a single value from a volScalarField including the dimension.

tomislav_maric October 27, 2011 08:17

Quote:

Originally Posted by safre (Post 329699)
Hello

...

and

volScalarField T
(
IOobject
(
"T",
runTime.timeName(),
mesh,
IOobject::MUST_READ
),
mesh
);

Nope, the dimensions of the volScalarField are not set correctly, try this:

Code:

volScalarField T
    (
            IOobject
            (
                    "T",
                    runTime.timeName(),
                    mesh,
                    IOobject::MUST_READ
            ),
            mesh,
            dimensionedScalar
            (
                  "0", // a dummy name for the scalar
                  dimTemperature, // 00051 extern const dimensionSet dimTemperature; - preset global constant dimensionSet for the temperature
                  pTraits<scalar>::zero  // trait for the zero value of scalar
            )
    );


safre October 27, 2011 09:34

Thank you for your suggestion but it did not help me out.
 
Thank you for your suggestion but it did not help me out.

In a way I still do not think it is a problem in the definition of T since that is only read from the ordinary Temperature file from my time directory and if I print out the complete T field the dimension is correct. The problem starts when I try to get one single value out of the field. I guess there is some command to do it another way than I am trying?

In fact what I really want to do is to take an average of a part of a volScalarField something like

TSlab = cmptAv(T(n0-nN));

or

TSlab = average(T(n0-nN));

including the dimension with n0 as the first value in a part of field T and nN being the last value in the part. Perhaps there is an even quicker solution to that problem?

Regards, Sam

tomislav_maric October 27, 2011 10:51

I think I've found the reason for this.

When you look in the following classes:

GeometricField<Type, PatchField, GeoMesh> -> DimensionedField< Type, GeoMesh> ->
Field< T > -> List < T > -> UList < T>, you will find that the access operator has not been redefined all the way since UList < T >:

Code:

T & operator[] (const label)
Wich will just return a scalar. So, each time you call T[label], you won't get a dimensionedScalar, but a scalar. To avoid this, you may define your own access operator [] for the GeometricField (this will probably mess up a lot of things), or you can deal with values...

meaning that this should help:

Code:

    scalar& aveTSlabVal = aveTSlab.value();
   
    aveTSlabVal = T[0];

You can even call
Code:

aveTSlab.val() = T[0]
if you like.

olesen October 28, 2011 02:41

Quote:

Originally Posted by safre (Post 329699)
Hello

I have problems getting the dimension correct when I am doing a simple sum. My equation is:

aveTSlab = aveTSlab + T[n];

where

aveTSlab is defined as

dimensionedScalar aveTSlab
(
"aveTSlab",
dimensionSet( 0, 0, 0, 1, 0, 0, 0),
scalar( 0)
);

and

volScalarField T
(
IOobject
(
"T",
runTime.timeName(),
mesh,
IOobject::MUST_READ
),
mesh
);

Both dimensions are defined correctly but when I take a single value T[n] it does not come with any dimension but the error is as:

--> FOAM FATAL ERROR:
LHS and RHS of + have different dimensions
dimensions : [0 0 0 1 0 0 0] + [0 0 0 0 0 0 0]

Please give advise how to get a single value from a volScalarField including the dimension.

Once you understand the volFields it will make more sense.
A volField is essentially a list of normal values that is augmented by a dimension. It is not a list of dimensioned values.
The means that any dimension checking is only done once for the entire field and not for each value.

In accordance with this, the
Code:

operator[]
will return the plain, undimensioned value. Thus the easiest and most efficient is to use the plain values directly
Code:

aveTSlab.value() += T[n];
If you really prefer to work with dimensioned values (and have dimensioning checking for each addition operation within the loop!!), try something like this:

Code:

aveTSlab += dimensionedScalar("T", T.dimensions(), T[n]);
For ease and generality, you will probably want to declare aveTSlab using the same dimensions as "T":
Code:


dimensionedScalar aveTSlab
(
    "aveTSlab",
    T.dimensions(),
    pTraits<Scalar>::zero
);


safre October 28, 2011 03:44

Hello

Yes that works or even simpler using an ordinary double definition. It was just that I wanted to keep the dimensions in order to keep the dimension check in my coding.

The way I solved it now is to use a double inside the calculation and then do a nn.write(); to a correctly dimensioned volScalarFieldaveTz in the end as

aveTz[k]= some value without dimension
aveTz.write();

It works but it could have been more elegant.

Thanks for all your help, Sam

tomislav_maric October 28, 2011 04:41

Quote:

Originally Posted by safre (Post 329841)
Hello

Yes that works or even simpler using an ordinary double definition. It was just that I wanted to keep the dimensions in order to keep the dimension check in my coding.

The way I solved it now is to use a double inside the calculation and then do a nn.write(); to a correctly dimensioned volScalarFieldaveTz in the end as

aveTz[k]= some value without dimension
aveTz.write();

It works but it could have been more elegant.

Thanks for all your help, Sam

Dimension checks, based on this information, are kept only for the fields, since as soon as one wants to access a DimensionedField<Type, GeoMesh> element, a Type is returned, instead of Dimensioned<Type>. This is good when one assembles field equations and performs computations on fields, not on specific elements. This means that +,/,* operators on GeometricField have built in dimension checks.

What you are describing here is an addition of value, not a complex equation involving fields: OpenFOAM is all about working with fields. :) I'm doing a lot of cell local explicit computation myself and the farther I get from manipulating the fields using their operators and numerics that is already there, the more problems I get into.

Anyway, good luck. :)

calim_cfd March 1, 2012 10:41

hihi!

within this scope.. can any1 please tell me what am i missing?

i wanna change the dimensions of wallHeatFlux so i can operate with T.

I figured this should work:
Code:

volScalarField wallHeatFlux(wallHeatFluxheader, mesh);

        wallHF += dimensionedScalar("wallHF", T.dimensions(), wallHeatFlux);

i get the error:
Code:

conVh.C: In function ‘int main(int, char**)’:
conVh.C:166:3: error: ‘wallHF’ was not declared in this scope

thanks in advance!

laurensvd March 1, 2012 10:50

You never defined what type wallHF is, you need to add volScalarField in front of wallHF +=...

what you can also do is reset the dimensions of wallHeatFlux to something different like so:
wallHeatFlux.dimensions().reset(T.dimensions());

calim_cfd March 1, 2012 14:45

Quote:

Originally Posted by laurensvd (Post 347097)
You never defined what type wallHF is, you need to add volScalarField in front of wallHF +=...

what you can also do is reset the dimensions of wallHeatFlux to something different like so:
wallHeatFlux.dimensions().reset(T.dimensions());


hi lauren!

Thanks a lot for replying..

but i guess the units error im getting comes from the sum (T - 300)

I guess i gotta set dimension to that 300... i've tried to find sth on the forums but so far nothing.. any ideas?

i tried a few syntax too but few would work... this one works:

scalar Tam = 300; and i set the sum to (T- Tam) .. the app compiles but the error is there.. ill keep digging :(

thanks once again!



EDIT:

i guess i got it..:

dimensionedScalar Tam ("Tam", dimensionSet(0, 0, 0, 1, 0, 0, 0),300.0);

now im getting other erros,

but again TY!!!!!

sfigato March 25, 2013 05:33

Hi Foamers,

I would like to implement a new cavitation solver in openFoam! I added a transport equation for mass fraction where I need to compare my local pressure (p) with a scalar experimental value pEq!

During the compiling of the code I got a dimension error!
Can anyone help me to fix it?

Here is the transport equation code:


Quote:
// Solve Mass Fraction Equation for the Gas Air (undissolved)
{
#include "sourceTerms.H"

//Scalar Transport Equation for Mass Fraction
if (pEq > p) //p)
{
fvScalarMatrix f4Eqn
(
fvm::ddt(rho, f4)
+ fvm::div(rhoPhi /*phi*/, f4)
- fvm::laplacian(Gamma4,f4)
//+ fvm::Sp(AaL,f4)
+ fvm::Sp(AdL,f4)

);

f4Eqn.relax();
solve( f4Eqn /*== AaL*fGlim*/);
}
else
{
fvScalarMatrix f4Eqn
(
fvm::ddt(rho, f4)
+ fvm::div(rhoPhi /*phi*/, f4)
- fvm::laplacian(Gamma4,f4)
+ fvm::Sp(AaL,f4)
//+ fvm::Sp(AdL,f4)

);

f4Eqn.relax();
solve( f4Eqn == AaL*fGlim);
}

//Update of All Phases Coeffcients
//rho == scalar(1)/((f1+f4)/rho1 + f2/rho2 + f3/rho3);
//rhoPhi == rho*phi;
}
Here the declarations for pEq and p

Quote:
Info<< "Reading thermodynamicProperties\n" << endl;

IOdictionary thermodynamicProperties
(
IOobject
(
"thermodynamicProperties",
runTime.constant(),
mesh,
IOobject::MUST_READ,
IOobject::NO_WRITE
)
);

dimensionedScalar pEq(thermodynamicProperties.lookup("pEq"));
Quote:
Info<< "Reading field p\n" << endl;
volScalarField p
(
IOobject
(
"p",
runTime.timeName(),
mesh,
IOobject::MUST_READ,
IOobject::AUTO_WRITE
),
mesh
The compiling error is:

Quote:
Making dependency list for source file ifasFCMfoam.C
Making dependency list for source file incompressibleFourPhaseMixture/fourPhaseMixture.C
SOURCE=ifasFCMfoam.C ; g++ -m64 -Dlinux64 -DWM_DP -Wall -Wextra -Wno-unused-parameter -Wold-style-cast -Wnon-virtual-dtor -O3 -DNoRepository -ftemplate-depth-100 -I/opt/OpenFOAM/OpenFOAM-2.1.x/src/turbulenceModels/incompressible/turbulenceModel -I/opt/OpenFOAM/OpenFOAM-2.1.x/src/transportModels -I/opt/OpenFOAM/OpenFOAM-2.1.x/src/transportModels/incompressible/singlePhaseTransportModel -I/opt/OpenFOAM/OpenFOAM-2.1.x/src/finiteVolume/lnInclude -I/opt/OpenFOAM/OpenFOAM-2.1.x/src/meshTools/lnInclude -IlnInclude -I. -I/opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude -I/opt/OpenFOAM/OpenFOAM-2.1.x/src/OSspecific/POSIX/lnInclude -fPIC -c $SOURCE -o Make/linux64GccDPOpt/ifasFCMfoam.o
In file included from ifasFCMfoam.C:76:0:
f4Eqn.H: In function ‘int main(int, char**)’:
f4Eqn.H:6:11: error: no match for ‘operator>’ in ‘pEq > p’
f4Eqn.H:6:11: note: candidates are:
In file included from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/VectorSpace.H:164:0,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/dimensionedType.H:41,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/dimensionedScalar.H:38,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/TimeState.H:38,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/Time.H:47,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/finiteVolume/lnInclude/fvCFD.H:6,
from ifasFCMfoam.C:36:
/opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/VectorSpaceI.H:650:13: note: template<class Form, class Cmpt, int nCmpt> bool Foam::operator>(const Foam::VectorSpace<Form, Cmpt, nCmpt>&, const Foam::VectorSpace<Form, Cmpt, nCmpt>&)
/opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/VectorSpaceI.H:650:13: note: template argument deduction/substitution failed:
In file included from ifasFCMfoam.C:76:0:
f4Eqn.H:6:11: note: ‘Foam::dimensionedScalar {aka Foam::dimensioned<double>}’ is not derived from ‘const Foam::VectorSpace<Form, Cmpt, nCmpt>’
In file included from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/dimensionedType.H:298:0,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/dimensionedScalar.H:38,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/TimeState.H:38,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/Time.H:47,
from /opt/OpenFOAM/OpenFOAM-2.1.x/src/finiteVolume/lnInclude/fvCFD.H:6,
from ifasFCMfoam.C:36:
/opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/dimensionedType.C:450:6: note: template<class Type> bool Foam::operator>(const Foam::dimensioned<Type>&, const Foam::dimensioned<Type>&)
/opt/OpenFOAM/OpenFOAM-2.1.x/src/OpenFOAM/lnInclude/dimensionedType.C:450:6: note: template argument deduction/substitution failed:
In file included from ifasFCMfoam.C:76:0:
f4Eqn.H:6:11: note: ‘Foam::volScalarField {aka Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh>}’ is not derived from ‘const Foam::dimensioned<Type>’
In file included from ifasFCMfoam.C:49:0:
createFields.H:106:30: warning: unused variable ‘rho4’ [-Wunused-variable]
make: *** [Make/linux64GccDPOpt/ifasFCMfoam.o] Fehler 1


All times are GMT -4. The time now is 10:26.