Non-linear material

Discussion about building and installing Elmer
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Non-linear material

Post by kevinarden »

I meant to say UMAT is not supported for transient, I don't know of any plans to add it.
Sayan
Posts: 112
Joined: 06 Dec 2020, 12:44
Antispam: Yes

Re: Non-linear material

Post by Sayan »

ohhhhhhhh....
I was just checking the loading unloading of bilinear curve which you posted and also my own result. I found one thing that the bilinear curve is not behaving accurately in the unloading part..During unloading it should follow linear path with permanent deformation but here it's going down with different displacement from loading path but not completely parallel to elastic part during the initiation of unloading..Is there any way by which I can get accurate linear unloading with permanent deformation which is parallel to elastic zone?
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Non-linear material

Post by kevinarden »

What you describe is true in a stress-strain curve plot, not sure if that is accurate for a load displacement plot.
Sayan
Posts: 112
Joined: 06 Dec 2020, 12:44
Antispam: Yes

Re: Non-linear material

Post by Sayan »

I have checked it using axial force also to plot stress strain curve, but I have found same observation...For restoring model some equation need to be followed I believe.. Is it programmed like that here? Otherwise I have to code it the way I discussed in previous message ..For that also I need your support...
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Non-linear material

Post by kevinarden »

It is coded to decide the stiffness based on the current location of the stress-strain curve. To unload as you describe the code would have to know that the structure is being unloaded and use the original stiffness for unloading.
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Non-linear material

Post by kevinarden »

I found a way to do the loading/unloading. I do not think it is an elegant solution but it is workable. I could not find a way in the subroutine to automatically tell if the model was unloading. I tried several ways to determine if mises was increasing or decreasing from the previous converged increment.

I decided to tell the subroutine if the model was loading or unloading using material constants. In this case the loading or unloading is known from the force function of time. So I passed load and unload flags to the subroutine.
ycase.sif
(3.01 KiB) Downloaded 175 times
UMATbi_linear.F90
(14.34 KiB) Downloaded 177 times
Attachments
results.png
results.png (26.53 KiB) Viewed 2032 times
Post Reply