Behaviour and design of cold-formed high strength steel tubular members

Abstract

Steel is categorised into mild steel and high strength steel according to the material strengths. The material properties of high strength steel usually cannot be simply represented by traditional mild steel stress-strain curves, thus the behaviour of high strength steel members are generally different from those for mild steel products. Tubular members have great potential in structural engineering due to their aesthetic appearance and strong resistance against torsional buckling. Cold-formed steel tubular sections are preferred in some applications as they are easier and less energy consuming to fabricate when compared to hot-rolled steel and built-up steel sections. The aims of this study are to investigate experimentally and numerically the behaviour of cold-formed high strength tubular structural members, and develop design guidance for these products. The experimental programme in this study included 66 tensile coupon tests, residual stress measurements on 3 sections, 25 stub column tests, 25 beam tests, 32 long beam-column tests and 83 short beam-column tests. The tubular sections consisted of 9 square hollow sections, 2 rectangular hollow sections and 6 circular hollow sections. The specimens were categorised into three series according to their nominal yield stresses: H-series (700 MPa), V-series (900 MPa) and S-series (1100 MPa). The stub columns were tested in fixed-ended boundary conditions. The behaviour of beams was investigated through four-point bending tests. The beam-columns were tested between a set of knife edges with equal load eccentricities at both ends, which provided single curvature and uniform end moment to the specimens. Finite element models were developed and validated against experiments using Abaqus software. The influences of material properties, residual stresses, local and global geometric imperfections and other input factors were analysed. The results showed that the models are capable of replicating the key test results, load-deformation histories and failure modes of the specimens. Upon validation of finite element models, a comprehensive parametric study programme was carried out to supplement the experimental data. The parametric study consisted of 321 stub columns, 423 beams and 540 beam-columns. The test and finite element data were further used to evaluate the codified design rules from the current American, Australian/New Zealand and European design codes for steel structures. The Direct Strength Method was also evaluated for the cold-formed high strength steel tubular members in this study. In the comparison, the compactness criteria for tubular sections were examined and the predicted design strengths were also compared to the test and finite element results. Improved design recommendations are given accordingly. Reliability analysis were performed for codified and proposed design rules.