Nonlinear Buckling Simulation with Ansys Mechanical APDL

Steph_Curry1Steph_Curry1 Member Posts: 1

Hello everyone,

i study civil engineering and want to simulate the nonlinear buckling behavior of different I-Beam profiles. This is my first time working with ansys. To do this i wrote a universal simulation with Ansys mechanical. This simulation is basically structured as follows:

-Entering the cross sction parameters of the I-Beam, the length of the system, the target load

-automatic creation of the system and the mesh

-if turned on at the beginning, creating the pre-deformation (on elastic theory) by using upgeom

-if turned on at the beginning, inset initial stresses

-activate the material behavior selected at the beginning (elastic, bilinear by using TB,bkin, multilinear by using TB,plastic)

-if turned on at the beginning, apply the load linearly by time


I have several problmes where i get stuck. That's why i'm looking for help here.

First problem: After initialize the pre-deformation i get some warnings like "Brick element 1has an aspect ratio of 46.42, which exceeds the warning limit of 20.". I 'fixed' this problem by temporarily increasing the allowable number of warnings. Is that okay or too quick and dirty? In my opinion, the elements thats involved are not in the area of interest.

My second problem: With the right analysis parameters, my simulation collapses at the euler's buckling load with litte deviation (13 different system length, deviation between 1-13%). Pre-deformation, initial stresses and nonlinear material behavior is turned off. So far that's great and what i expected. After that, i turn on the pre-defromation. The maximum load reduced by 2-12% at long system length. However, with small length the maximum load increases by up to +25% which is illogical. initial stresses and nonlinear material behavior is still turned off. I dont know how to fix this. I think i did something wrong.

My third problem: If i activate all options and start a nonlinear buckling analysis, the simulation collapses too early. Compares to the EC3 buckling curves i have a deviation of -45%. This ist far too little maximum load. In direct comparison to "my second problem" i activated the nonlinear material behavior and the initial stresses.

Has anyone ever written a similar simulation and can help me with troubleshooting?

Please forgive me misspellings, i am not a nativ speaker. Thanks in adance!

Answers

  • Steph_Curry1Steph_Curry1 Posts: 2Member

    My pre-deformation part:

    Initial stresses:

    And the main calculation with material behavior:


  • jjdoylejjdoyle PittsburghPosts: 121Member

    You might have considered this already, but if not, Chapter 21 of the MAPDL Technology Showcase: Example Problems has a buckling and post buckling problem worked out using Stabilization Damping. Perhaps you could compare this with your setup for tips.

    Why do you say that it is illogical for max buckling load to increase by 25% when column length decreases?   Doesn’t the shorter column become more resistant to buckling?  Perhaps I am misunderstanding what you are reporting.

    Regarding the 45% deviation from expected results, it is hard to say for sure.  I cannot tell what the parameterized inputs for NSUBST,,,, command are prescribing,  but, maybe you need many more substeps to get closer to the point of instability before it fails to converge.

    Regards,

    John

Sign In or Register to comment.