Dear all,

I am currently trying to introduce geometric imperfections into my Abaqus model based on the eigenvalue buckling modes. The aim of this imperfections is to improve  agreement between experimental and FEA results.

I have been able to run the eigenvalue buckling analysis and  use the resulting buckling mode as an imperfection in a subsequent nonlinear analysis. However, the imperfection does not appear to have any noticeable effect on the response of my model.

This is how I have approached it presently,
1. set up the model for an eigenvalue buckling analysis using:

*Step, name=Buckling, nlgeom=NO, perturbation
*Buckle
5, , 10, 300

2. I  added the following to the input file so that the nodal displacement information for the buckling modes would be written to the results:

*NODE FILE, GLOBAL=NO,
MODE=1, LAST MODE=5
U

3. I introduced the buckling mode as an imperfection using this steps

*END ASSEMBLY:

*Imperfection,
file=S-BUCKLING, step=1
5,0.1

4.I then performed a nonlinear static analysis with geometric nonlinearity enabled:

The analysis is displacement controlled, with:
I also requested reaction force and displacement output so that I can compare the numerical response with my experimental results.

The analysis runs successfully after introducing the imperfection, but the imperfection does not appear to have a noticeable effect on the model response.

I will like to know if I have done this correctly,  and how I can observe a noticeable difference in results.

Any advice on whether I am approaching this correctly would be greatly appreciated.

Thanks,

Yemi