[dkingsley]

Niels,

I just sent you an updated patch that should handle OF2.2.0 and OF2.2.x. OF2.2.x did in fact change the name of twoPhaseMixture to incompressibleTwoPhaseMixture.

djk

[ngj]

Dear Norman, Dennis and others,

I have made a small correction to the compilation of waves2Foam, so now compilation of the same part of the source code should be achievable.

I defined an additional environmental variable WAVES_XVERSION, so please remember to update your waves2Foam/bin/bashrc file based on the changes to waves2Foam/bin/bashrc.org.

Also, the changes are done in the dark, as I do have neither 2.2.0 nor 2.2.x, so please report any difficulties.

All the best,

Niels

[gregjunqua]

Dear Neils,
As i had say earlier I am interested by adding the wind. But i am stuck with this problem is that i would like to apply a different wind condition than the one proposed in your module.
I would like to know (I had search but unfortunately not found) where you did modifies the module for adding the wind.

Greg

[shashi]

Dear Neils,

Originally Posted by ngj

Dear Norman, Dennis and others,

I have made a small correction to the compilation of waves2Foam, so now compilation of the same part of the source code should be achievable.

I defined an additional environmental variable WAVES_XVERSION, so please remember to update your waves2Foam/bin/bashrc file based on the changes to waves2Foam/bin/bashrc.org.

Also, the changes are done in the dark, as I do have neither 2.2.0 nor 2.2.x, so please report any difficulties.

All the best,

Niels

I am trying to compile the waves2Faom, and have encountered the 2 errors. There is some error with the bin/bashrc where the $WM_PROJECT_VERSION takes in a wrong value. Is this what you are referring to above. Can you please help with resolving the other error also which says wmake command not found used in the respective ./Allwmakes .

Thanks in advance,
shashank

[ngj]

Hi Shashank,

Either you have forgotten to source the OpenFoam bashrc script (see the OpenFoam installation REAME) or you have not installed OpenFoam at all.

Good luck,

Niels

[shashi]

Originally Posted by ngj

Hi Shashank,

Either you have forgotten to source the OpenFoam bashrc script (see the OpenFoam installation REAME) or you have not installed OpenFoam at all.

Good luck,

Niels

Dear Niels,

I have sourced the OpenFoam bashrc script, and I have run several tutorials of sloshing tank, floating object etc., without errors, although , I have not compiled the files on my machine but have copied the already compiled files.

And moreover, the error with version number seems to be in the code, as I am operating in 2.1.1, where the value of x is supposed to be 1, instead it takes the version number as 0 and searches for solvers0 folder in /applications/solvers .

Pl correct me if I am wrong I am new to OpenFaom.

Thanks in advance.
shashank

[ngj]

Hello,

The fact that you are told that wmake does not exist and that neither WAVES_APPBIN nor WAVES_LIBBIN are given any value suggests to me that something is wrong with your installation.

These things means that you have not compiled anything on your OpenFoam installation and that the native OpenFoam environmental variables FOAM_USER_APPBIN and FOAM_USER_LIBBIN are not set. With respect to the latter I have no clue, why this happens.

These are fundamental problems, which I do not have the answer for, but I would suggest that you search the forum for the "wmake command not found" string.

Kind regards

Niels

P.S. No, version 2.1.1. is not an x-version. X-versions are e.g. 1.7.x, 2.1.x and 2.2.x.

[javierjap]

Hello all

I'm new to OpenFOAM and to this Forum as well. I was wondering if the wave2foam utility or the wavefoam solver would solve the second-order non-linear wave equation?

Kind regards
Javier

[shashi]

Originally Posted by ngj

Hello,

The fact that you are told that wmake does not exist and that neither WAVES_APPBIN nor WAVES_LIBBIN are given any value suggests to me that something is wrong with your installation.

These things means that you have not compiled anything on your OpenFoam installation and that the native OpenFoam environmental variables FOAM_USER_APPBIN and FOAM_USER_LIBBIN are not set. With respect to the latter I have no clue, why this happens.

These are fundamental problems, which I do not have the answer for, but I would suggest that you search the forum for the "wmake command not found" string.

Kind regards

Niels

P.S. No, version 2.1.1. is not an x-version. X-versions are e.g. 1.7.x, 2.1.x and 2.2.x.

Dear Niels,

I understood what you have told above, and even installed openfoam2.2.0 again, which is working fine( i mean, I have been able to run tutorials successfully), but upon compiling waves2Faom, it still gives the same error I have mentioned to you before
( i.e WAVES_APPBIN WAVES_LIBBIN do not take any values), is there any other way to resolve this, I am unable to find any other direction, I even went through the wmake string. you asked me to.

[JanL]

Dear Niels,


I'm currently testing the spectral analysis in waveFoam.
Therefore I've set up a 2D-case and wanted to generate a JONSWAP spectrum. I've attached the waveProperties file, which I've used.
Strangely, to get setWaveField to work, I had to define amplitude, frequency, phaselag, omega and waveNumber, although all of those parameters are defined through the calculation of the wave spectra (as far as I understood from looking at JONSWAP.C). Furthermore, when setting writeSpectrum true, and having a look at the written waveProperties file, those parameters aren't written out anymore. Could you tell me what those parameters are needed for in case of wave spectra?


When running the case with the waveProperties file using initializationName inlet, I do get strange results looking at the waves. The amplitude is far too big (approximately by factor N, which is 10 in my case) and the resulting elevation doesn't quite suit the analytical result (I've attached both files for comparison, where I computed the analytical result by using the values for amplitude, frequency, phaselag and waveNumber written out into the waveProperties file). Using intitializationName outlet (potentialCurrent in my case) the amplitude is correct but still the surfaceElevation is wrong (might be a problem in the accuracy of the generated waves as it looks not too bad in the beginning but getting worse over time as can be seen in the attached figure!?). I suspect that my grid resolution isn't high enough to accurately considers the smallest generated waves. Or does something else come to your mind?


My last problem is, when initializing the wave field, there appears a strange vertical displacement of alpha at the inlet (see attached figure). Again, I have the feeling waveFoam uses a superposed amplitude or something similar. The amplitudes, which are generated in my spectrum are much smaller! I have the feeling, that this displacement afflicts the phase in the first time steps. Or is there something I'm doing wrong for the initialization?


Any hints would be highly appreciated!


Best regards
Jan

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  1.5                                   |
|   \\  /    A nd           | Web:      http://www.OpenFOAM.org               |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    object      environmentalProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

timeShift       0.0;

seaLevel    0.00;

wind        (0 0 0);

// A list of the relaxation zones in the simulation. The parameters are given
// in <name>Coeffs below.
relaxationNames (inlet outlet);

initializationName outlet;

pName p_rgh;

inletCoeffs
{
    waveType            irregular;
    spectrum            JONSWAP;
    N                   10;             
    Tsoft               0;
    writeSpectrum       true;           
    Hs                  0.2;         
    amplitude           uniform 0.1;        
    frequency           uniform 0.4266;         
    phaselag            uniform 0;              
    omega               2.6804;         
    waveNumber          uniform (0.7323 0 0);   
    Tp                  2.3441;         
    gamma               3.3;
    depth               6.25;
    direction           (1 0 0);
    
    // Specifications on the relaxation zone shape and relaxation scheme
    relaxationZone
    {
        relaxationScheme Spatial;
        relaxationShape  Rectangular;
        beachType        Empty;
    
        relaxType   INLET;
        startX              ( -13.6 0.0 -11.2 );
        endX                ( -4.6 0.0 0 );
        orientation         ( 1.0 0.0 0.0 );
    }
};

outletCoeffs
{
    waveType            potentialCurrent;
    U                   ( 0 0 0 );
    Tsoft               0;

    relaxationZone
    {
        relaxationScheme Spatial;
        relaxationShape  Rectangular;
        beachType        Empty;    

        relaxType   OUTLET;
        startX              ( 8 0.0 -11.2 );
        endX                ( 28.7 0.0 0 );
        orientation         ( 1.0 0.0 0.0 );
    }
};



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

[ngj]

@Jan:

I have just tested that I can run setWaveParameters and subsequently setWaveField (using the inlet parameters), and I get something which looks like an irregular wave train. Important to note that I did not have to do any modifications what so ever. I am using the most recent SVN revision, which is #2000.
Furthermore, I do actually get the spectrum written, so we are probably not synchronised in terms of version.

The only thing I can say about your time series analysis is that you should consider the effect of non-linear interactions inside the computational domain. This will make the wave train deviate from the analytical solution.

@Shashi:

You should first try and see, whether you can compile any of the solvers in OpenFoam. I suggest that you begin with icoFoam. You will probably get the same wmake-related error. I cannot help you on that, as it will be the fundamental OF-settings which are then wrong.

Kind regards

Niels

[Abdel]

Hi everyone,

I'm studying a wave Flume case and I have a little problem. I just want to generate my waves during 10s but let running my simulation during 60s.
I don't know how I should proceed ... Can somebody help me please ?
Thank You in advance

I wrote this scipt in waveProperties file :

timeShift 0.0;

seaLevel 0.00;

// A list of the relaxation zones in the simulation. The parameters are given
// in <name>Coeffs below.
relaxationNames ();

initializationName init;

pName p_rgh;

inletCoeffs
{
// Wave type to be used at boundary "inlet" and in relaxation zone "inlet"
waveType stokesFirst;

// Ramp time of 2 s
Tsoft 2;

// Water depth at the boundary and in the relaxation zone
depth 0.700000; //0.4

// Cyclic wave frequency
omega 3,141593; //0.546882617

// Phase shift in the wave
phi 0.000000;

// Wave number vector, k.
waveNumber (1.70048 0.0 0.0);

// Wave height
height 0.05; //0.10

// Tstart 0; // Not required (Default 0.0);

Tend 5; // Not required (Default GREAT);

// Specifications on the relaxation zone shape and relaxation scheme
/* relaxationZone
{
relaxationScheme Spatial;
relaxationShape Rectangular;
beachType Empty;

relaxType INLET;
startX (0 0.0 -1);
endX (0.5 0.0 1);
orientation (1.0 0.0 0.0);
} */
};

initCoeffs
{
waveType potentialCurrent;
U (0 0 0);
Tsoft 2;
};

//

[ngj]

Hi Abdel,

Such functionality is not available, but you should be able to modify the code in order to achieve what you want.

Kind regards

Niels

[Abdel]

Originally Posted by ngj

Hi Abdel,

Such functionality is not available, but you should be able to modify the code in order to achieve what you want.

Kind regard

Niels

Hi Niels,

First and foremost I'd like to thank You for your response. I already had this idea and start to write in the src files : wavestheories/regular/stokesFirst .C and .H scripts. and in setwavesProperties/regular/stokesFirstProperties .C and .H scipts.
But I haven't the results I want ...

I wrote this : stokesFirst.C

// * * * * * * * * * * * * * * * * Constructors * * * * * * * * * * * * * * //

stokesFirst::stokesFirst
(
const word & subDictName,
const fvMesh & mesh_
)
:
waveTheory(subDictName, mesh_),
H_(readScalar(coeffDict_.lookup("height"))),
h_(readScalar(coeffDict_.lookup("depth"))),
omega_(readScalar(coeffDict_.lookup("omega"))),
period_(2 * PI_ / omega_),
phi_(readScalar(coeffDict_.lookup("phi"))),
k_(vector(coeffDict_.lookup("waveNumber"))),
K_(mag(k_)),
Twend_(readScalar(coeffDict_.lookup("Twend"))), //ajout

Tsoft_(coeffDict_.lookupOrDefault<scalar>("Tsoft", period_))
{}


void stokesFirst:rintCoeffs()
{
Info << "Loading wave theory: " << typeName << endl;
}

// * * * * * * * * * * * * * * * Member Functions * * * * * * * * * * * * * //


scalar stokesFirst::factor(const scalar & time) const
{
scalar factor(1.0);
/* if (Tsoft_ > 0.0)
if ( time = 0.0) //Tend_(readScalar(coeffDict_.lookup("Twend"))), //ajout
else if ( Twend_ < 10.1)
factor = Foam::sin(2 * PI_ / (4.0 * Tsoft_) * Foam::min(Tsoft_, time));


return factor; */

while (Twend_ < 10.1)
factor = Foam::sin(2 * PI_ / (4.0 * Tsoft_) * Foam::min(Tsoft_, time));

return factor;

}



stokesFirst.H

// Member Functions
Switch conflictTSoftInitialise() const
{
/* if ( Tsoft_ > 0 )
return true;
else
return false; */
while (Twend_ < 10.1)
return true;
else
return false;



stokesFirstProperties.C

// Info << "\nConstructing: " << this->type() << "(Dummy)"<< endl;

period_ = 0.0;
depth_ = 0.0;
omega_ = 0.0;
Twend_ = 0.0;//ajout
}

stokesFirstProperties::stokesFirstProperties
(
const Time & rT,
dictionary & dict,
bool write
)
:
setWaveProperties(rT, dict, write)
{
Info << "\nConstructing: " << this->type() << endl;

period_ = readScalar( dict.lookup("period") );
depth_ = readScalar( dict.lookup("depth") );
omega_ = 2.0 * PI_ / period_ ;
Twend_ = readScalar( dict.lookup("Twend"); //ajout
}

stokesFirstProperties::stokesFirstProperties
(
const Time & rT,
dictionary & dict,
bool write,
word string
)
:
setWaveProperties(rT, dict, write)
{
Info << "\nConstructing: " << this->type() << " (Used by another wave theory)";

period_ = readScalar( dict.lookup("period"+string) );
depth_ = readScalar( dict.lookup("depth") );
omega_ = 2.0 * PI_ / period_;
Twend_ = readScalar( dict.lookup("Twend") );
}

// * * * * * * * * * * * * * * * Member Functions * * * * * * * * * * * * * //

void stokesFirstProperties::set( Ostream & os )
{
scalar k = linearWaveNumber();

// Write the beginning of the sub-dictionary
writeBeginning( os );

// Write the already given parameters
writeGiven( os, "waveType" );

if ( dict_.found( "Tsoft" ) )
writeGiven( os, "Tsoft");

writeGiven( os, "depth");
writeGiven( os, "period" );
writeGiven( os, "direction" );
writeGiven( os, "phi");
writeGiven( os, "height");
writeGiven(os, "Twend"); //ajout
if ( write_ )
{
vector direction( vector(dict_.lookup("direction")));
direction /= Foam::mag(direction);
direction *= k;

writeDerived( os, "waveNumber", direction );
writeDerived( os, "omega", omega_);




and stokesFirstProperties.H

namespace Foam
{

class stokesFirstProperties
:
public setWaveProperties
{
private:

scalar period_;
scalar omega_;
scalar depth_;
scalar Twend_; //Ajout


.................................................. .......................

I'm sorry for this very long post.

Kind regards,

Abdel

[ngj]

Hi Abdel,

It is hard to help you, when you do not state, what type of result you get, now that you do not get the one you want.

Secondly, I am very scared over your use of

while

-operator, as you will easily end up in an infinite loop.

Thirdly, please do use the

[ CODE ]
[ /CODE ]

brackets (without the spaces) around all code you put in the forum, as this will keep the formatting and not it easier to read.

Kind regards

Niels

[Abdel]

Hi Niels,
thank You for your post !

modification I added are useless : in fact, I obtain the same results as always.
I'm not a professional in programmation language but I made some modifications again ..
I know "while" isn't the best idea so I replaced them by "if"

I wrote this but I don't see the results I want when I use paraFoam :

/ * * * * * * * * * * * * * * * * Constructors * * * * * * * * * * * * * * //

stokesFirst::stokesFirst
(
const word & subDictName,
const fvMesh & mesh_
)
:
waveTheory(subDictName, mesh_),
H_(readScalar(coeffDict_.lookup("height"))),
h_(readScalar(coeffDict_.lookup("depth"))),
omega_(readScalar(coeffDict_.lookup("omega"))),
period_(2 * PI_ / omega_),
phi_(readScalar(coeffDict_.lookup("phi"))),
k_(vector(coeffDict_.lookup("waveNumber"))),
K_(mag(k_)),
Twend_(readScalar(coeffDict_.lookup("Twend"))), //ajout

Tsoft_(coeffDict_.lookupOrDefault<scalar>("Tsoft", period_))
{}


void stokesFirst:rintCoeffs()
{
Info << "Loading wave theory: " << typeName << endl;
}

// * * * * * * * * * * * * * * * Member Functions * * * * * * * * * * * * * //


scalar stokesFirst::factor(const scalar & time) const
{
scalar factor(1.0);
/* if (Tsoft_ > 0.0)
if ( time = 0.0) //Tend_(readScalar(coeffDict_.lookup("Twend"))), //ajout
else if ( Twend_ < 10.1)
factor = Foam::sin(2 * PI_ / (4.0 * Tsoft_) * Foam::min(Tsoft_, time));


return factor; */

if (Twend_ < 10.1)
factor = Foam::sin(2 * PI_ / (4.0 * Tsoft_) * Foam::min(Tsoft_, time));
//ajout

return factor;

}

scalar stokesFirst::eta
(
const point & x,
const scalar & time
) const
{

scalar eta = H_ / 2.0 * Foam::cos(omega_ * time - (k_ & x) + phi_) * factor(time) + seaLevel_;
return eta;
}

scalar stokesFirst::ddxPd
(
const point & x,
const scalar & time,
const vector & unitVector
) const
{

scalar Z(returnZ(x));
scalar arg(omega_ * time - (k_ & x) + phi_);

scalar ddxPd(0);

ddxPd = (
rhoWater_ * mag(g_) * K_ * H_ / 2.0 * Foam::cosh(K_ * (Z + h_)) / Foam::cosh(K_ * h_) * Foam::sin(arg)
) * factor(time);

// ddxPd += rhoWater * Foam::mag(G) * k_ * height_ / 2 * Foam::cosh(k_ * (c[cI].component(1) - seaLevel_ + depth_))
// / Foam::cosh(k_ * depth_) * Foam::sin(omega_ * db().time().value() + mathematicalConstant:i / 2) * factor;
// Info << "ddxPd still isn't implemented. Need to think about the gradient on arbitrary directed mesh with arbitrary wave number vector! and arbitrary g-direction!!!" << endl;
return ddxPd;
}

vector stokesFirst::U
(
const point & x,
const scalar & time
) const
{
scalar Z(returnZ(x));

scalar Uhorz = PI_ * H_ / period_ *
Foam::cosh(K_ * (Z + h_)) / Foam::sinh(K_ * h_) *
Foam::cos(omega_ * time - (k_ & x) + phi_);
Uhorz *= factor(time);

scalar Uvert = - PI_ * H_ / period_ *
Foam::sinh(K_ * (Z + h_)) / Foam::sinh(K_ * h_) *
Foam::sin(omega_ * time - (k_ & x) + phi_);
Uvert *= factor(time);

return Uhorz * k_ / K_ - Uvert * direction_; // Note "-" because of "g" working in the opposite direction
}

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

} // End namespace waveTheories
} // End namespace FoamThank You very much for your advices
Kind regards,

Abdel

[Ya_Squall2010]

Hi Niels,

I had a test run using stokeFifth of waves2Foam. The domain is around 8 wave lengths long, with a relaxation zone of 1.8 wave length long at the outlet side. Just notice the wave trains are beautifully produced at the initial tens of seconds, but start to diminish afterward and completely smeared out at the late stage of the simulation. Please have a look of the pictures attached. I wonder whether this is due to the domain is not quite long enough?

1.0002.jpg

1.0015.jpg

1.0030.jpg

1.0170.jpg

[Ya_Squall2010]

And here's the waveProperties:

timeShift 0;

seaLevel 0;

wind ( 0 0 0 );

relaxationNames ( inlet outlet );

initializationName inlet;

pName p_rgh;

inletCoeffs
{
waveType stokesFifth;
height 15;
period 10;
depth 100;
stokesDrift 0;
direction ( 1 0 0 );
Tsoft 0;
phi 0;
waveNumber (0.0378937298104 0 0);
waveLength 165.810685267;
omega 0.628318530718;

relaxationZone
{
relaxationScheme Spatial;
relaxationShape Rectangular;
beachType Empty;
relaxType INLET;
startX ( -500 -300 -105 );
endX ( -400 300 105 );
orientation ( 1 0 0 );
}
}

outletCoeffs
{
waveType potentialCurrent;
U ( 0 0 0 );
Tsoft 0;

relaxationZone
{
relaxationScheme Spatial;
relaxationShape Rectangular;
beachType Empty;
relaxType OUTLET;
startX ( 400 -300 -105 );
endX ( 700 300 105 );
orientation ( 1 0 0 );
}
}

[kilroy]

Hi Niels,

I had a test run using stokeFifth of waves2Foam. The domain is around 8 wave lengths long, with a relaxation zone of 1.8 wave length long at the outlet side. Just notice the wave trains are beautifully produced at the initial tens of seconds, but start to diminish afterward and completely smeared out at the late stage of the simulation. Please have a look of the pictures attached. I wonder whether this is due to the domain is not quite long enough?

Ya_Squall201;

I am having exactly the same problem.

Thanks,

kilroy

[dahicke]

I am currently using Star CCM+ for my wave work, but I noticed something that may help. I had a region of larger mesh cell size next to a region with smaller cell size perpendicular to the wave. When running the wave, the wave in the larger cells moved slower and appeared to dissipate compared to the smaller cell region. Could your problem be with cell size?