Dear all,
I am working on a buckling problem of a long cylindrical shell under bending moment. One end is fixed and the other one is free to move. A bending moment is acting on free end. I calculated the critical moment from a Linear buckling analysis in Abaqus as first step. Then I used Riks method to trace the buckling behaviour and I took as imperfection the 1% of thickness of critical mode. After bifurcation point is reached convergence is hard to achieved and the post buckling behaviour cannot be traced. I was expected the load factor to decrease but such a thing wasn't happened. I run the same without including any imperfections and things seem more normal since I got a degradation in load factor as I was expected. Any ideas abou this?
Note: Figures below are from the case where imperfections are included.