Re: [deal.II] Re: Thermoelastic Problem

[Muhammad Mashhood]

Prof. Wolfgang,
 Thank you for the concern and response. I got your 
point there. I will try it in this way then. :)

On Wednesday, August 7, 2019 at 6:38:53 PM UTC+2, Wolfgang Bangerth wrote:
>
> On 8/5/19 3:40 AM, Muhammad Mashhood wrote: 
> > The only thing is that I was wondering if I might be using extra 
> processing 
> > and memory by defining the fe_temperature, dof_handler_temperature and 
> > fe_values_temperature by defining them again in the solid mechanics 
> program. 
> > And if I am, then would there be any alternative method where I can use 
> the 
> > *const SmartPointer > *and *const 
> SmartPointer > FE_Q > *or something similar**to return *fe_temperature* and 
> > *dof_handler_temperature* from *HeatEquation* class to *SolidMechanics* 
> class 
> > (where they are being used for evaluating q_point temperature as I 
> mentioned 
> > above) ? Thank you! 
>
> Muhammad, 
> you are in essence asking general C++ programming questions about the 
> difference between regular member variables and pointers, and how to 
> return 
> objects by value, by reference, or by pointer. The answer is that yes, you 
> can 
> do as you suggest above. But I would suggest you take a look at a good C++ 
> book or online resource to understand *why* the answer is yes :-) 
>
> Best 
>   W. 
>
> -- 
>  
> Wolfgang Bangerth  email: bang...@colostate.edu 
>  
> www: http://www.math.colostate.edu/~bangerth/ 
>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/fdccffe8-426e-48d3-a90a-cefcdfbe52b2%40googlegroups.com.

Re: [deal.II] Re: Thermoelastic Problem

[Wolfgang Bangerth]

On 8/5/19 3:40 AM, Muhammad Mashhood wrote:
> The only thing is that I was wondering if I might be using extra processing 
> and memory by defining the fe_temperature, dof_handler_temperature and 
> fe_values_temperature by defining them again in the solid mechanics program. 
> And if I am, then would there be any alternative method where I can use the 
> *const SmartPointer > *and *const SmartPointer FE_Q > *or something similar**to return *fe_temperature* and 
> *dof_handler_temperature* from *HeatEquation* class to *SolidMechanics* class 
> (where they are being used for evaluating q_point temperature as I mentioned 
> above) ? Thank you!

Muhammad,
you are in essence asking general C++ programming questions about the 
difference between regular member variables and pointers, and how to return 
objects by value, by reference, or by pointer. The answer is that yes, you can 
do as you suggest above. But I would suggest you take a look at a good C++ 
book or online resource to understand *why* the answer is yes :-)

Best
  W.

-- 

Wolfgang Bangerth  email: bange...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/b4bc06b5-3a20-c74c-69a3-23363d7d0f38%40colostate.edu.

Re: [deal.II] Re: Thermoelastic Problem

[Muhammad Mashhood]

Dear Prof. Wolfgang,
  Thank you for your concern and 
understanding. 
Yes you are right. I tried something similar instead of copying the 
triangulation. 
I used *const SmartPointer >* to transfer 
triangulation between both thermal and the solid mechanics classes and 
seemingly the objective of using same triangulation for both parts is full 
filled also with same refinement and coarsening. 
While for the sake of using different *dof_handler* for temperature and 
solid mechanics, I currently have to define again the 
temperature_dof_handler as well as the fe_temperature objects in the solid 
mechanics (class) part to do something like following:

  *dof_handler_temperature.distribute_dofs(fe_temperature);*

*  cell_solid_mech = dof_handler.begin_active();*
*  cell_temperature = dof_handler_temperature.begin_active();*
(for loop for *cell_solid_mech and **cell_temperature* )





*  fe_values_solid_mech.reinit(cell_solid_mech);  
fe_values_temperature.reinit(cell_temperature);  cell_matrix = 
0;  cell_rhs = 0;  
fe_values_temperature.get_function_values(temperature_solution, 
temperature_solution_qpoint);  *
*fe_values_solid_mech[displacement].get_function_symmetric_gradients(displacement_solution,
   
strain_tensor);*  

So in this way, I am successful in using the common triangulation for both 
parts but with different corresponding dof_handlers.
The only thing is that I was wondering if I might be using extra processing 
and memory by defining the fe_temperature, dof_handler_temperature and 
fe_values_temperature by defining them again in the solid mechanics 
program. And if I am, then would there be any alternative method where I 
can use the *const SmartPointer > *and *const 
SmartPointer > *or something similar to return * 
fe_temperature* and *dof_handler_temperature* from *HeatEquation* class to 
*SolidMechanics* class (where they are being used for evaluating q_point 
temperature as I mentioned above) ? Thank you!

On Sunday, August 4, 2019 at 11:54:58 PM UTC+2, Wolfgang Bangerth wrote:
>
> On 8/2/19 4:01 AM, Muhammad Mashhood wrote: 
> > 
> > considering the above scenario as well as the concerns, I would be 
> grateful to 
> > receive any suggestion from your side. Hope I am clear in my 
> description. 
> > Waiting for your kind response. Thank you in advance! 
>
> Instead of copying triangulations, you always have the option of just 
> creating 
> the two objects the same, and then refining them in exactly the same way. 
> Would that solve your problem? 
>
> Best 
>   W. 
>
>
> -- 
>  
> Wolfgang Bangerth  email: bang...@colostate.edu 
>  
> www: http://www.math.colostate.edu/~bangerth/ 
>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/91710d89-ac27-4df9-bbb3-1298853d705f%40googlegroups.com.

Re: [deal.II] Re: Thermoelastic Problem

[Wolfgang Bangerth]

On 8/2/19 4:01 AM, Muhammad Mashhood wrote:
> 
> considering the above scenario as well as the concerns, I would be grateful 
> to 
> receive any suggestion from your side. Hope I am clear in my description.
> Waiting for your kind response. Thank you in advance!

Instead of copying triangulations, you always have the option of just creating 
the two objects the same, and then refining them in exactly the same way. 
Would that solve your problem?

Best
  W.


-- 

Wolfgang Bangerth  email: bange...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/e7e69c28-5bdb-ab5e-5e07-0c716cdf8f19%40colostate.edu.

[deal.II] Re: Thermoelastic Problem

[Muhammad Mashhood]

Hi Daniel,
   Thank you for the suggestion and I also think it can be one 
of the way to successfully run both solutions by sharing information on 
correct corresponding q_points. 
I tried the same approach (as I am also using partitioned approach for 
thermomechanical coupling) but so far my (mesh files being read for both 
thermal and solid mechanics parts are same but ) triangulation are 
different. 
So I also want to use the same triangulation with different dof_handler for 
both parts. 

*The scenario of workflow in my case is following:*
The thermal code uses the *Triangulation triangulation_thermal* while 
the solid mechanics code is using *parallel::distributed::Triangulation 
triangulation_solid*. Yes the thermal part is programmed in serial and the 
solid mechanics part in parallel. The solid mechanics code runs and makes 
the grid with it's triangulation, As I want to use the same triangulation 
so I try to copy this triangulation with function 
*triangulation_solid.copy_triangulation(triangulation_thermal) 
*and here the *triangulation_thermal *is the one which I will pass to the 
name space and class of thermal analysis part before running it. Moreover 
during my analysis of solid mechanics part I also refine the mesh (and 
would also want to use the same refined mesh for the thermal analysis too 
by copying it).  

After describing the scenario following are my concerns where I need your 
expert opinion : 
*1)* when I use the *copy_triangulation() *function, I come across the 
error (even here the mesh is not refined so far but still) mentioning : 




*  void dealii::Triangulation::copy_triangulation(const 
dealii::Triangulation&) [with int dim = 2; int spacedim = 
2]The violated condition was: (vertices.size() == 0) && (levels.size() 
== 0) && (faces == nullptr)Additional information: You are trying to 
perform an operation on a triangulation that is only allowed if the 
triangulation is currently empty. However, it currently stores 100 vertices 
and has cells on 1 levels.*

*2)* As far as the documentation of *tria.h* is concerned it also informs 
that the *copy_triangulation()* cannot copy the refined mesh (where in my 
case the mesh might also get refined during analysis based on certain 
criteria and I would like to use same refined triangulation by copying it 
for thermal analysis so that triangulation is same in both problems).

considering the above scenario as well as the concerns, I would be grateful 
to receive any suggestion from your side. Hope I am clear in my description.
Waiting for your kind response. Thank you in advance!

*Best regards,*
Muhammad

On Thursday, March 3, 2016 at 10:49:37 AM UTC+1, Daniel Arndt wrote:
>
> Hello Sudharsan,
>
> what you want to do is to step through all the cells and reinitialize both 
> DoFHandlers simultaneously, i.e. something like
>
> cell_elastic=dh_elastic.begin_active();
> cell_temperature=dh_temperature.begin_active();
> for (;cell_elastic != dh_elastic.end(); ++cell_elastic, ++cell_temperature)
> {
>   fe_values_elastic.reinit(cell_elastic);
>   fe_values_temperature.reinit(cell_temperature);
>   fe_values_temperature.get_function_values(temperature, 
> local_temperature);
> ...
> }
>
> This of course can only work if both DoFHandlers are based on the same 
> Triangulation.
>
> Best regards,
> Daniel
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/1ca1a6d7-81a2-4353-af83-2d8e4d002c30%40googlegroups.com.

Re: [deal.II] Re: Thermoelastic Problem

[Wolfgang Bangerth]

On 5/24/19 2:46 AM, Muhammad Mashhood wrote:
> Thank you for informative reply and posting this concern on the forum. I am 
> also interested in thermoelastic problem and new use of deal.ii.
> My question is that other than the tutorial steps 26 & 18 or 20,21 & 22, is 
> there pre-developed application at "https://www.dealii.org/; for this field 
> of 
> study?

None that immediately implements the thermoelastic problem. But there are of 
course many building blocks you can find in a variety of tutorial programs and 
code gallery programs.

Best
  W.

-- 

Wolfgang Bangerth  email: bange...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/1c66e935-f866-a40e-b311-7e9239c7633a%40colostate.edu.
For more options, visit https://groups.google.com/d/optout.

[deal.II] Re: Thermoelastic Problem

[Muhammad Mashhood]

Hi Joerg and Lisa!
Thank you for informative reply and posting this concern on the forum. I am 
also interested in thermoelastic problem and new use of deal.ii. 
My question is that other than the tutorial steps 26 & 18 or 20,21 & 22, is 
there pre-developed application at "https://www.dealii.org/; for this field 
of study? Thank you!   

On Wednesday, August 27, 2014 at 12:06:40 PM UTC+2, deal.II-MailingList 
wrote:
>
> Hello Lisa,
>
> I suppose that it depends on how you want to solve your thermoelastic 
> problem.
> Do you want to assemble one big system which contains one block for the 
> elastic equation,
> another block for the heat equation and coupling blocks? Then you could 
> solve that system
> in a monolithic way. Therefore you should have a look at step 20, 21 and 
> 22.
> If you want to solve the elastic equation and the heat equation 
> alternately a look to step 18
> and 26 might be enough.
> step 18 shows an example for how to solve an elastic equation as a time 
> depend problem.
> And step 26 gives an example for how to deal with the heat equation.
>
> Since I am currently working on a thermoelastic problem let me know if you 
> have further
> questions.
>
> Best,
> Joerg
>
> Am Dienstag, 26. August 2014 15:21:25 UTC+2 schrieb Lisa Collins:
>>
>> Hi,
>>
>> I want to start deal.ii and model a thermoelastic problem in 2D & 3D.
>>
>> Can you please guide me which steps I should study.
>> Which step should be considered as the base of my work and by modifying 
>> it I can model a thermoelastic problem?
>>
>> Best  regards
>> Lisa
>>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/2a596416-2d5e-4aed-a80b-efc5c0bb66b7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[deal.II] Re: Thermoelastic Problem

[ha . badnava]

 


I have solve a thermoelastic problem using block matrices and block vectors 
and a same dofhandler for the displacement and temperature fields.

Fully coupled thermal-stress analysis is needed when the stress analysis is 
dependent on the temperature distribution and the temperature distribution 
depends on the stress solution. For example, metalworking problems may 
include significant heating due to inelastic deformation of the material 
which, in turn, changes the material properties. Some problems require a 
fully coupled analysis in the sense that the mechanical and thermal 
solutions evolve simultaneously, but with a weak coupling between the two 
solutions. In other words, the components in the off-diagonal submatrices 
are small compared to the components in the diagonal submatrices. For the 
thermoelastic problem the off-diagonal terms are zero and solution is 
straight and stable. You can solve the block systems simultaneously or 
using staggered solution technique.  


Setup system as follow:


dof_handler.distribute_dofs (fe);

DoFRenumbering::Cuthill_McKee(dof_handler);

BlockDynamicSparsityPattern dsp(2, 2);

std::vector<*unsigned* *int*> block_component(3, 0); 

block_component[*dim*] = 1; // temperature

DoFRenumbering::component_wise(dof_handler, block_component);

DoFTools::count_dofs_per_block(dof_handler, dofs_per_block,block_component);

*const* types::global_dof_index n_dofs_u = dofs_per_block[0];

// determine number of temperature DOFs if requested.

types::global_dof_index n_dofs_temp;

n_dofs_temp = dofs_per_block[1];

dsp.block(0, 0).*reinit*(n_dofs_u, n_dofs_u);

dsp.block(0, 1).*reinit*(n_dofs_u, n_dofs_temp);

dsp.block(1, 0).*reinit*(n_dofs_temp, n_dofs_u);

dsp.block(1, 1).*reinit*(n_dofs_temp, n_dofs_temp);

dsp.collect_sizes();

...


use the following approach for system matrix and RHS:

*for* (*unsigned* *int* q=0; qget_dof_indices (local_dof_indices);

*for* (*unsigned* *int* i=0; i *local_quadrature_points_data

= *reinterpret_cast**>(cell->user_pointer());


// extract nodal total and incremental displacement

// will be used for strain and delta_strain calculation

fe_values[u_fe].get_function_gradients (solution_delta, 
displacement_increment_grads);

fe_values[u_fe].get_function_gradients (solution, displacement_grads);

// extract temperature values and gradients

*if*(parameters.is_temperature){

fe_values[temp_fe].get_function_values (solution, temp_value);

fe_values[temp_fe].get_function_values (solution_n, temp_n_value);

fe_values[temp_fe].get_function_gradients (solution, temp_grads);

}

// Then we can loop over all degrees of freedom on this cell and

// compute local contributions to the right hand side:

*for* (*unsigned* *int* q=0; qget_dof_indices (local_dof_indices);

*for* (*unsigned* *int* i=0; i
> Hi,
>
> I want to start deal.ii and model a thermoelastic problem in 2D & 3D.
>
> Can you please guide me which steps I should study.
> Which step should be considered as the base of my work and by modifying it 
> I can model a thermoelastic problem?
>
> Best  regards
> Lisa
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[deal.II] Re: Thermoelastic Problem

[Daniel Arndt]

Hamed,

assuming that both DoFHandlers are based on the same triangulation, you can 
do something like the following:

FEValues fe_stress_values(...);
stress_cell = dof_handler_stress.begin_active();
temperature_cell = dof_handler_temperature.begin_active();
for (; stress_cell != dof_handler_stress.end(); ++stress_cell, 
++temperature_cell)
{
  fe_stress_values.reinit(stress_cell);
  fe_stress_values.get_function_values(solution_stress, local_stress);
  //do all the temperature related stuff and use local_stress
}

Instead of having two iterators separated, you can also use 
SynchronousIterators [1]. They arec partly used in step-35 for that purpose.

Best,
Daniel

[1] https://dealii.org/8.4.1/doxygen/deal.II/structSynchronousIterators.html

Am Dienstag, 7. Juni 2016 20:30:27 UTC+2 schrieb Hamed Babaei:
>
> Hello Daniel,
>
> Since although we can solve two elastic and heat equations separately, we 
> still need to enter stress computed from elastic equation into the heat 
> equation in every time step and keep updating stress field.
> I was wondering when we have two different DoFHandlers for each of the 
> equations, how to use, for instance, stress field stored on a quadrature 
> point related to elastic problem DofHandler, in order to solve 
> stress-induced heat equation problem. In other words, I don't know how to 
> exchange information between two equation.
>
> Thanks,
> Hamed
>
> On Friday, June 3, 2016 at 3:40:28 AM UTC-5, Daniel Arndt wrote:
>>
>> You're right, Hamed. There is FEValuesViews::Vector< dim, spacedim 
>> >::symmetric_gradient [1] 
>> and FEValuesViews::Vector< dim, spacedim >::gradient [2] instead. The 
>> first should be exactly what you are looking for.
>>
>> Code that I have written normally uses multiple DoFHandlers if I solve 
>> for multiple equations separately.
>>
>> Best,
>> Daniel
>>
>> [1] 
>> https://dealii.org/8.4.1/doxygen/deal.II/classFEValuesViews_1_1Vector.html#a4e5dfbb49d284a368acac6ef5dd4b2ef
>>  
>> 
>> [2] 
>> https://dealii.org/8.4.1/doxygen/deal.II/classFEValuesViews_1_1Vector.html#ad7b4df64147989229f566eff541ddebc
>>  
>> 
>>
>>
>>
>>
>>
>>
>> Am Freitag, 3. Juni 2016 01:33:34 UTC+2 schrieb Hamed Babaei:
>> Daniel,
>>
>> Since in the FEValuesViews::Vector< dim, spacedim > Class there isn't any 
>> shape_grad_component function, I am not sure that I can update the 
>> get_strain as I mentioned.
>> I was wondering if you have already written a code for a thermoelastic 
>> problem in dealii, solving heat equation and elastic equation separately.
>>
>> Best,
>> Hamed
>>
>> On Thursday, June 2, 2016 at 3:31:48 PM UTC-5, Daniel Arndt wrote:
>> Hamed,
>>
>>
>> There shouldn't be a problem to do what you following step-22. In 
>> particular, you can modify step-18::get_strain as you proposed.
>> Did you face any difficulties so far?
>>
>>
>> Best,
>> Daniel
>>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[deal.II] Re: Thermoelastic Problem

[Hamed Babaei]

Hello Daniel,

Since although we can solve two elastic and heat equations separately, we 
still need to enter stress computed from elastic equation into the heat 
equation in every time step and keep updating stress field.
I was wondering when we have two different DoFHandlers for each of the 
equations, how to use, for instance, stress field stored on a quadrature 
point related to elastic problem DofHandler, in order to solve 
stress-induced heat equation problem. In other words, I don't know how to 
exchange information between two equation.

Thanks,
Hamed

On Friday, June 3, 2016 at 3:40:28 AM UTC-5, Daniel Arndt wrote:
>
> You're right, Hamed. There is FEValuesViews::Vector< dim, spacedim 
> >::symmetric_gradient [1] 
> and FEValuesViews::Vector< dim, spacedim >::gradient [2] instead. The 
> first should be exactly what you are looking for.
>
> Code that I have written normally uses multiple DoFHandlers if I solve for 
> multiple equations separately.
>
> Best,
> Daniel
>
> [1] 
> https://dealii.org/8.4.1/doxygen/deal.II/classFEValuesViews_1_1Vector.html#a4e5dfbb49d284a368acac6ef5dd4b2ef
>  
> 
> [2] 
> https://dealii.org/8.4.1/doxygen/deal.II/classFEValuesViews_1_1Vector.html#ad7b4df64147989229f566eff541ddebc
>  
> 
>
>
>
>
>
>
> Am Freitag, 3. Juni 2016 01:33:34 UTC+2 schrieb Hamed Babaei:
> Daniel,
>
> Since in the FEValuesViews::Vector< dim, spacedim > Class there isn't any 
> shape_grad_component function, I am not sure that I can update the 
> get_strain as I mentioned.
> I was wondering if you have already written a code for a thermoelastic 
> problem in dealii, solving heat equation and elastic equation separately.
>
> Best,
> Hamed
>
> On Thursday, June 2, 2016 at 3:31:48 PM UTC-5, Daniel Arndt wrote:
> Hamed,
>
>
> There shouldn't be a problem to do what you following step-22. In 
> particular, you can modify step-18::get_strain as you proposed.
> Did you face any difficulties so far?
>
>
> Best,
> Daniel
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.