ActuatorDiskExplicitForce in OF2.1. Help
 

Hi all,

Resume:

Are there a guide to now how to change a solver code prepared for OF1.5dev to a newer OF versions(OF2.1)?

Long version:

I have been checking the solver windSimpleFoam of the version 2.0.1 and due to the fact than it also takes into account the thrust to modify the momentum equation.

I investigate and I found the following report:

http://www.tfd.chalmers.se/~hani/kur...ningReport.pdf

It seems very interesting because it can be modelled the trust and the swirl with a radial distribution based on a analytical approach. It can be fixed the thrust and the torque of your wind turbine.

Although it was done based on OF 1.5dev, I have been trying to implement the same approach for OF 2.1 but I am not able to do it. Are there a guide to now how to change a solver between 1.5dev a newer OF versions?

When I try to compile, I receive the following error:

This is my Ueqn.H code:

I have commented some lines because I receive error with eqnResidual and maxResidual due to the fact, I think, that in OF2.1 there are not included:
"readSIMPLEControls.H","initConvergenceCheck.H " and "convergenceCheck.H"

This is my first/second approach to the code of OF, so I will try to continue with the problem and If a find the mistake, I will post the solution.

Best Regards

With this new Ueqn.H code, I have received the next error:


actuatorDiskExplicitForceSimpleFoam.C

I have compiled the new code, but....
 

But it seems that it does not work properly.

It computes the forces (axial and tangential), but the volumeForce is always zero. There is not a source.

Please, I think that I am near to solve the problem but I do not find where it is.

Thank you

hello
 

hi all

i am very interested in this case of actuator dis of Erik Svenning
but i found the same problem of you :( ???
are you solved??
please contact me

hello drsattar and be_inspired,

about the Erik Svenning code there's a couple of things you ahve to take into account,

first that code was made for OpenFoam 1.5 dev, and the coding the solvers has changed from that version to 2.1.x. so that's why the simplefoam solver that comes with the code won't compile in more recent versions of openfoam. if you insist on using it you would have to cut and paste the parts of the code into the newer version of simplefoam (it can be a bit tricky - it took me a while)

second, that code was made on a openfoam workshop, and not for a phd or masters thesis, i've personally updated the code and made it run in the new version of openfoam and i've encounter a couple of mistakes in the code:

- first in the code he sets the axial and tangential force to be in the same direction, which is not true, even though openfoam works in cartesian coordiantes you have to decompose the tagential force in Z and Y.

-also the way the code selects the cells to insert the bodyforce form points and faces leads to a erratic distribution of the force.

i finally gave up on that code :s

i would refer you to this thread:

http://www.cfd-online.com/Forums/ope...rce-model.html

here they show a very simple and practical way to generate and insert a bodyforce into a solver :),

and depending on you C++ it turns out to be very flexible to insert mathematical expresions to calculate the force.

i hope this helps and good luck!

Oscar

bodyForce
 

dear oscar_urri

i am very thanful for your reply, if you like send me your email to send you the final case which i created it of actuator disk model. i wish if you able to check it...
also the case of Fan (body force mode) is not work with me??? :o
my email (hdrsattar@yahoo.com)
thank again :)

dear drsattar,

it's not necessary for you to send me an e-mail :),

my recomendation is not to use the actuador disk model form Erik Svenning, as i told you it has mistakes that i was not able to correct,

you should go with the bodyforce model, i'm going to attach a sample solver and a sample case, (it's only set to add an axial force, you're gonna have to modify it to include the tangential force),

to make a good case,

the first thing you should do is to validate your code against the analytical solution, here's and article that talks about that, and gives the formulas for the anlytical solution,

http://orbit.dtu.dk/fedora/objects/o...190785/content

after you can set up an actual real case and compare it against experimental data, or against other already validated model.

i wish i could help you more, but i also have do my own simulations, i have 2 semesters before i finish my masters :) and i have to concentrate on that.

if you have any other questions ask, but it might take me a while to answer,

the best of luck on your research, i know it's not easy

Oscar

actuator disk (bodyForce)
 

Dear oscar_urri

I am very grateful to you about your answer on my problem, now i will try to use and modify the attachment files ......:)
i wish to be contact with me in the future.....
thanks again

Dear oscar_urri,

With your solver, pressure wiggles are not fixed:

http://www.cfd-online.com/Forums/ope...l-wiggles.html

This is quite quite important for me.
The report that you have linked is my reference point but I do not know how to modify simpleFoam and rotorDiskSource to manage body forces without pressure wiggles.

The problem happens when body force is directly applied over the UEqn and velocity and pressure are not coupled.
It seems like you have not implemented the correction proposed by your link

Do you have a modified version of your simpleFoamForce to solve that?

Hello everybody,
i tried to use the case and the solver from oscar_urr.

When I try to compile the solver simple Foam Force, there comes an error:



simpleFoamForce.C:34:22: fatal error: RASModel.H: No such file or directory
#include "RASModel.H"
^
compilation terminated.
make: *** [/mnt/apps_linux/home/abz860/OpenFOAM-in-Box/OpenFOAM-in-Box-18.06v1/OpenFOAM-dev/platforms/linux64Gcc62DPInt32Opt/applications/solvers/incompressible/simpleFoamForce/simpleFoamForce.o] Error 1



does anybody has an idea how to handle this?


kind Regards
Dirk

do you have

-I$(LIB_SRC)/turbulenceModels/incompressible/RAS/lnInclude \

in Make/options ?

This error pops up if the linker does not find the header files.