underworldcode / underworld2

underworld2: A parallel, particle-in-cell, finite element code for Geodynamics.
http://www.underworldcode.org/
Other
168 stars 58 forks source link

Need some help on a cartesian convection model #382

Closed Prasanna2989 closed 4 years ago

Prasanna2989 commented 5 years ago

Hi,

I want to set up a mantle convection model in cartesian domain with a rigid non-deformable non-movable block above the fluid, is there a way to do it without applying a sticky layer method? This block should be able to transfer heat by conduction. Please share whether this is possible to implement using simple convection example in UW. The attachments contain required set up. I am not sure how to apply the rigid block without emerge in the fluid.

cheers

image

image

julesghub commented 5 years ago

You can define a group of nodes (a region of the domain) to have a zero velocity Dirichlet condition. To do this you must: 1) create an FeMesh_IndexSet containing the nodes of the region. 2) Then use that set with the DirichletConditions object.

Note be aware of the pressure gradients that could result.

lmoresi commented 5 years ago

Julian,

Just be careful about the whole issue of ill-defined pressure constraints if the velocity is going to be zero in a large region where T is being solved.

That part is not 100% clear from the diagram.

Prof Louis Moresi

louis.moresi@unimelb.edu.aumailto:louis.moresi@unimelb.edu.au

(w) +61 3 8344 1217

(m) +61 4 0333 1413

(us) +1 505 349 4425

www.moresi.infohttp://www.moresi.info/

www.facebook.com/underworldcodehttp://www.facebook.com/underworldcode

@LouisMoresihttps://twitter.com/LouisMoresi

On 17 May 2019, 2:48 PM +1000, Julian Giordani notifications@github.com, wrote:

You can define a group of nodes (a region of the domain) to have a zero velocity Dirichlet condition. To do this you must:

  1. create an FeMesh_IndexSet containing the nodes of the region.
  2. Then use that set with the DirichletConditions object.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/underworldcode/underworld2/issues/382?email_source=notifications&email_token=ADABPI6HQXXJ74UFRC7EYFLPVY2HVA5CNFSM4HNR5T32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVTXAVQ#issuecomment-493318230, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ADABPIZN5P54AWU3K4ZU7YTPVY2HVANCNFSM4HNR5T3Q.

Prasanna2989 commented 5 years ago

So I need to define FeMesh_IndexSet like following function. Could you please give a simple example how should I give the coordinates.

def MinJ_VertexSet( mesh ):
       return _uw.mesh.FeMesh_IndexSet( object           = mesh,
                                                                    topologicalIndex = 0,
                                                                     size             = 
                                                                     fromObject       =      )

IWalls = mesh.specialSets["MinI_VertexSet"] + mesh.specialSets["MaxI_VertexSet"]
JWalls = mesh.specialSets["MinJ_VertexSet"] + mesh.specialSets["MaxJ_VertexSet"]

uw.conditions.DirichletCondition(     variable=velocityField, 
                                              indexSetsPerDof=(IWalls,JWalls) )
jmansour commented 5 years ago

This hasn't seen any activity for a while. @Prasanna2989 did you make any progress?

Prasanna2989 commented 5 years ago

Sorry. I was away for a while. I am going to do this test soon.

On Thu, Jun 27, 2019 at 5:07 PM jmansour notifications@github.com wrote:

This hasn't seen any activity for a while. @Prasanna2989 https://github.com/Prasanna2989 did you make any progress?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/underworldcode/underworld2/issues/382?email_source=notifications&email_token=AGGDHQWV3O3WNEA2KBGVVEDP4RRJLA5CNFSM4HNR5T32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYWFFHA#issuecomment-506221212, or mute the thread https://github.com/notifications/unsubscribe-auth/AGGDHQUKLPUIMYAUE3B3KIDP4RRJLANCNFSM4HNR5T3Q .

-- Prasanna M. Gunawardana

PhD candidate Earth, Atmosphere and Environment Science Monash University, Melbourne Australia

jmansour commented 5 years ago

Ok, no probs. Let us know how you go.