[WIP] Thermal stress computation for Neo-Hookean material model#2398
[WIP] Thermal stress computation for Neo-Hookean material model#2398Vaish-W wants to merge 1 commit intosu2code:developfrom Vaish-W:feature-thermal-stress
Conversation
|
Thank you for the contribution, I was curious after you talked about this on slack and I have an implementation that is a little more polished. |
|
Hi @pcarruscag, I understand this might not be a good implementation as you expect. But could you please tell me the changes if possible? So that I could work on this further. Are there a lot of changes or is my formulation itself wrong? I want to learn doing it completely. 😄 |
pcarruscag
left a comment
There was a problem hiding this comment.
Sure, I left some comments and you can see more here: #2399
Let me know if you want to work on coupling the heat solver with the structural solver (read the nodal temperatures from the heat solver).
I can add you to the SU2 org.
| inline su2double ComputeAverageTemperature(void) const { | ||
| su2double sum_temperature = 0.0; | ||
| for (const auto &temp : nodal_temperatures) { | ||
| sum_temperature += temp; | ||
| } | ||
| return nodal_temperatures.empty() ? 0.0 : (sum_temperature / nodal_temperatures.size()); | ||
| } |
There was a problem hiding this comment.
Here it should be the temperature interpolated at the Gaussian integration points.
| /*--- Thermal stress parameters ---*/ | ||
| su2double alpha = config->GetThermal_Expansion_Coeff(); // Coefficient of thermal expansion | ||
| su2double delta_T = element->GetTemperature() - config->GetTemperature_Ref(); // Temperature difference |
There was a problem hiding this comment.
The FEA solver has a way to use different materials in different parts of the structure, so I introduced one alpha per material.
The reference temperature you use here is used by some solver for non dimensionalization, so it was better to use a dedicated temperature.
I also added this stress contribution to all material models.
Thank you for your comments and I got the changes from PR#2399. Also, I would like to work on coupling of heat and structural solver. |
|
Great, then I'll merge my PR, close this one, add you to the SU2 org (then you can create branches in the su2code repo directly and it's easier to collaborate), and I'll start a discussion with some notes on how you can start coupling the solvers. |
Proposed Changes
This pull request adds the capability to compute thermal stresses in Neo-Hookean materials in SU2. It includes:
CFEM_NeoHookean_Compclass to incorporate thermal stress contributions under CFEM_NeoHookean_Comp::Compute_Stress_Tensor.CElementclass for handling temperature at the element and nodal levels.Thermal Expansion Coefficient,Reference Temperature) to update stress tensorRelated Work
This feature addresses thermal stress computation for the nonlinear elasticity module. It sets the groundwork for future problems, such as thermoelasticity in bimetallic strips. No specific issues or PRs are related.
PR Checklist
pre-commit run --allto format old commits.