buckling analysis of copressed column

[Aldus]

I am trying to model compressed steel columns with partial continuity connections at mid-height.
I guess that a proper modeling scheme may be two elastic beam elements linked by a "hinge" connector with non-linear behavior. Displacement of the hinged node can then model imperfection.

Anyone already involved in this problem?

Do the displacement of central node have to be applied in a linear perturbation step?

Thanks in advance

Aldo