Skip to main content

post buckling behaviour of cylindrical shell under axial compression in abaqus

Submitted by sragrawal on

I am doing postbuckling analysis of cylindrical shell in abaqus.I have to plot a graph of reaction force vs lateral deflection of shell.The deflection pattern shown by experimental result is different than pattern shown by abaqus result.The experimental pattern shows punching type of deflection while abaqus shows other type of deflection..

Please tell me what should i do to get the result. 

Forums
Ask iMechanica
Akumar

Cylinderical shell post buckling behavior is highly sensitive to the initial imperfection. It is also it's quite challenging problem (that's why it is included in abaqus verification manual). Anyway, from your description, things are not quite clear. Could you show us the plot (experimental and simulation)? It will help a lot in order to figure out what is going on.

------

The world started with 0, is progressing with 0, but doesn't want 0.

Tue, 09/11/2012 - 21:57 Permalink
sragrawal

The experimental plot comes out to be  while the simulation graph looks like  . As you can see that experimental graph shows that it stablises after some time while simulation graph doesn't. Also the lateral deflection pattern is different as mention earlier.

 Other details:

Axial load is applied and static riks method was used for analysis.Boundary conditions are simply supported. Dimensions - diameter-200mm,length - 200mm and thickness -1mm.

Wed, 09/12/2012 - 06:06 Permalink
Akumar

I don't see any plot in relation with  your description. If you're having difficulty in uploading plots, you can view this thread:

http://imechanica.org/node/245

or, you can upload plots on some uploading site, and paste the link here.

------

The world started with 0, is progressing with 0, but doesn't want 0.

Wed, 09/12/2012 - 13:17 Permalink
sragrawal

Link for experimental graph: 
http://www.flickr.com/photos/87095362@N02/7978788843/

Link for simulation graph : http://www.flickr.com/photos/87095362@N02/7978789199/

 Link for simulated shell : http://www.flickr.com/photos/87095362@N02/7977392794/

Link for experimental shell : http://www.flickr.com/photos/87095362@N02/7977367142/ 

Also if i introduce imperfection in shell(based on linear combination of buckle mode) i won't get such type of deflection , instead i get the longitudinal wave.   

Wed, 09/12/2012 - 13:55 Permalink
Akumar

I have few questions?

  1. It seems force (y-axis) and strain/displacement (x-axis) are normalized. Could youtell me w.r.t. what?
  2. Why force reading start from value 1.0 @ eps11 0.0
  3. You're not using axisymmetric or cyclic symetry. Since experiment shows unsymmetric deformation.
  4. Did you compare your Pcr (from abaqus model) with experimental? Use full integration instead of reduced integration default.
  5. Does your model include plasticity? From experimental picture, it seems crimping is happening in the shell therefore that kind of stuff you've to account for.Therefore atleast plasticity should be included.

Your load-deflection (simulation) is typical for cylinderical shell buckling. So I think there is something from experiment you're missing.

------

The world started with 0, is progressing with 0, but doesn't want 0.

Wed, 09/12/2012 - 16:40 Permalink
naser354

Hi All

dose any body can help me in how to finde the eigenvalues of buckling of a comppresed beam

i want to know how this can be implemented using the co rotational beam element

i dont know if this is the right place to put this qustion 

thanks

Wed, 10/10/2012 - 08:13 Permalink
parvizs

I think your experimental shell hass more imperfection than what you think, whic make torsion in your buckling.

go to key word and change your imperfection type specially in higher eigenvalues. 

Thu, 10/11/2012 - 19:20 Permalink