I am applying Generalized Displacement Control method for predicting the nonlinear inelastic local buckling behavior of steel frames.
I proposed some formulations to considering local buckling for both compact and slender sections as shown in Fig. 1.
However my proposed program can not capture the post-buckling path of a clamped column under axial compression. The solution procedure is interrupted due to the whole section almost yielded. May be tangent stiffness matrix is singular.
In this case, how to capture the post-buckling path?
Thanks much for your comments
Did you consider using a modified Riks method (arc length method)? It allows to study post-buckling bahviour of instable structures. This algorithm is implemented in many FEM tools, like Abaqus, and it's used in many appplications.
As an example, it has been used for the study of stability of grid shells in this paper (Figure 10). https://www.researchgate.net/publication/282151602_Buckling_of_Elastic_gridshells
Best regards,
Romain Mesnil
Article Buckling of Elastic gridshells
hi Romain Mesnil,
Thanks for your ideas. Both GDC and Arc length methods can capture problems with many critical points.
I think that my problem is how materials behave at critical points. Load will reduce or change from loading to unloading at critical points. However, displacements often continuously increase. Please see the attached figure (an example).
Best regards,
Phu-Cuong
i would use an implicit solver using Riks (Arc method). with fracture data for collapse. you may have convergence problemn due to negative Jacobian. due to my small experince in post buckling Riks method considered to be the best!.
- Badges
- Science topic
- Similar topics
- Engineering
- Mechanical Engineering