Menu

Discussion Forum

New Discussion
Filter

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Is it possible to apply global von-Mises stress constraint in Topology Optimization?

Posted Jul 28, 2021, 10:12 a.m. EDT Optimization, Structural Mechanics Version 5.6 3 Replies

Vishal Dolas
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

I have an objective of increasing stiffness/ reducing strain energy.

But my main constraint is not about mass (how much mass I want to retain), but I want to put constraint over the maximum global von-Mises stress. Is it possible to set a problem like this in COMSOL?

3 Replies Last Post Jan 28, 2022, 3:58 a.m. EST
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 years ago Jul 28, 2021, 4:16 p.m. EDT

Hi,

this may work with a pointwise constraint. Another option may be to set up a maximum operator maxop1 in the nonlocal couplings and use maxop1(vanMises) in a global contraint.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hi, this may work with a pointwise constraint. Another option may be to set up a maximum operator maxop1 in the nonlocal couplings and use maxop1(vanMises) in a global contraint. Cheers Edgar
Vishal Dolas
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 years ago Jul 30, 2021, 5:16 a.m. EDT

Hello Edgar,

Thanks for the reply. I tried both the options you suggested.

I found "point-sum inequality constraint" to be working very nicely in general for all kinds of constraint but solver is not following "Global inequality constraint". I mean the results are not as expected when I used global constraint. I applied it the way you suggested, from non-linear coupling by choosing maximum operator maxop.

Do you have any other sugestions for applying a global constraint?

Hello Edgar, Thanks for the reply. I tried both the options you suggested. I found "point-sum inequality constraint" to be working very nicely in general for all kinds of constraint but solver is not following "Global inequality constraint". I mean the results are not as expected when I used global constraint. I applied it the way you suggested, from non-linear coupling by choosing maximum operator maxop. Do you have any other sugestions for applying a global constraint?
Joe Alexandersen
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 years ago Jan 28, 2022, 3:58 a.m. EST

maxop is not differentiable, you need to make an integral over the domain of the VM stress to the power p. This makes a continuous p-norm, which is an approximation of the max function for p --> infinity.

maxop is not differentiable, you need to make an integral over the domain of the VM stress to the power p. This makes a continuous p-norm, which is an approximation of the max function for p --> infinity.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.

Suggested Content