Reliability assessment of built-up steel beams subjected to bending

Resumo

Although built-up cold-formed steel members are commonly used under bending loads, the structural design of such sections requires further investigation. This study proposes an assessment of the reliability of built-up cold-formed steel members subjected to bending and designed using the Direct Strength Method (DSM). A comprehensive database was assembled, comprising 87 specimens with open cross-sections and 61 with closed cross-sections, formed from singly and doubly stiffened C-sections, all exhibiting double symmetry. Based on this database, a statistical analysis of the experimental results was carried out to identify the probability distribution that best fits the model error variable, along with its corresponding parameters. In addition to model error, the statistical parameters of material properties, cross-sectional geometry, loading type, and load ratio were considered in the reliability study, and a sensitivity analysis was performed to evaluate the relative importance of these variables. Subsequently, the reliability indices of members designed according to load combinations from the North American specification (AISI S100, 2016) in the LRFD (Load and Resistance Factor Design) and LSD (Limit States Design) formats, as well as from the Brazilian standard (NBR 14762, 2010), were compared. The results showed that the LRFD load combination yielded reliability indices closer to the target value, while the LSD format presented lower indices, and the Brazilian load combination yielded intermediate performance. Furthermore, it was observed that closed sections demonstrated higher reliability levels compared to open sections.