System reliability-based criteria for designing cold-formed steel structures by advanced analysis

Abstract

This doctoral thesis develops a system-reliability based approach for designing cold-formed steel structures by advanced analysis. Specifically, this doctoral thesis carries out the underpinning structural analyses and reliability studies that enable the implementation of the next generation of system-based design-by-analysis steel specifications, i.e., specifications that establish a design approach where analysis and capacity checks are carried out in a single step. The study begins by developing and validating finite element (FE) models and advanced analyses that are capable of accurately predicting the behaviour and system strength of cold-formed steel structures. The modelling scheme includes a rational methodology for incorporating frame, member and sectional geometric imperfections, the ability to assign semi-rigid behaviour to the joints, the flexibility to represent the geometry of complex cross-sections and takes into account the detrimental effect of local and distortional buckling while determining the ultimate capacity of the structure. It follows the survey and statistical characterisation of the variability of the geometric and material parameters affecting the strength of cold-formed structures. Based on the collected data and the developed FE models, Monte-Carlo type of simulations are performed to determine system strength distributions and statistics of several representative frame configurations and failure modes. Reliability studies using the first-order reliability method (FORM) are then carried out to derive system resistance factors s corresponding to certain system reliability indices  and representative frames are categorised on their structural reliability features. The study concludes by establishing guidelines for the design of cold-formed steel structures based on the nominal system strength as obtained by advanced analysis and category of structural reliability.