Application of end moments to an I beam (modelled with shell elements) in Abaqus

[ConwaysGlider]

Hello,

I am using Abaqus to analyze I beams (with shell elements), for various load cases. I have some problems with applying end moments. Until now, I have applied shell edge loads at the end of the beam with a linear stress distribution (along the height of the profile) in order to simulate the application of a bending moment. This worked quite well but is not exact enough, since after yielding, the stress distribution is no longer linear. I have also tried to apply a constant shell edge load (compression on one half, tension on the other) in order to simulate the fully plastic distribution of stresses. I am using a Riks step because I am interested in buckling limit loads, taking into account plasticity, etc.

Technically, I think it may be possible to add a stiff end plate of some sorts and apply the end moments through an eccentricity of applied loads, however I would prefer not to do this, since I am using Python scripts due to the large number of simulations and this would complicate things quite a bit.

Would anyone have a suggestion on how to apply an end moment, in this type of case? I have tried a bit to apply BC rotations at the end and I have also tried using a reference point, but with no success.

Any help would be much appreciated. Thanks in advance,

Cristi

 

[IceBreakerSours]

Have you considered using *Coupling (Kinematic)? See documentation for more.

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[ConwaysGlider]

Thanks for the reply IceBreakerSours!

Yes, what I meant when I said I tried using a reference point was using it with a kinematic coupling. I have read the documentation over and over, but I cannot get it to work properly. So I have a normal I profile modelled with shell elements, with point supports (U2) at the middle of the web, in both ends. This becomes a problem in and of itself, because as I understand, I cannot have a node with a dof restrained and coupled to a RP. I tried coupling all nodes, except these two (on which I have the BC's) at the end of the member to a RP, through both types of coupling (kinematic and distributed). It still didn't work properly. Applying forced rotations to all end nodes, without using couplings didn' work either. I mean, I get weird values for the maximum LPF in the Riks step, since I know in what range the maximum rotations at the end have to be. Furthermore, in some cases, the analysis does not converge, no matter how low I set the minimum arc length increment...