Numerical Modeling of Cold-Formed Steel Members Based on Experimental Tests at Normal and Elevated Temperatures

Finite element modeling is widely used to investigate the structural behavior of cold-formed steel (CFS) members, interpret experimental observations, and perform parametric studies. However, the numerical response may be significantly affected by the adopted constitutive model, initial geometric imperfections, residual stresses, corner strength enhancement, and other modeling assumptions. The lack of consistent modeling procedures can therefore limit the reproducibility and direct comparison of results reported in different studies. This paper presents a finite element modeling methodology for cold-formed steel members subjected to normal and elevated temperatures. The methodology is implemented in SAFIR and validated against experimental programs involving columns, beams, and beam-columns with lipped-channel, sigma, and Z cross-sections. The validation covers members subjected to compression, bending, and combined compression and bending. The effects of the constitutive material model, initial geometric imperfections, residual stresses, and corner strength enhancement are also examined. The results show that the initial geometric imperfections have the greatest influence among the modeling assumptions considered. When combined local, distortional, and global imperfections are introduced, the additional influence of residual stresses and corner strength enhancement is comparatively small. The numerical models provide satisfactory agreement with the experimental ultimate loads, load-displacement responses, deformation patterns, failure times, critical temperatures, and governing instability modes at normal and elevated temperatures. The resulting modeling procedure provides a consistent basis for nonlinear analyses and parametric studies of cold-formed steel members under normal- and elevated-temperature conditions.