Large-size concrete-filled steel tubular (CFST) members are widely applied in long-span and heavily-loaded structures due to increasing engineering demands. However, due to the size effect, applying the existing design methods to large-size CFST members may overestimate the load capacity, leading to unsafe designs. The plastic stress distribution (PSD) method is widely used as the primary method in typical design codes for calculating the strength of CFST members that are not susceptible to local buckling, while the PSD method was mainly used previously to evaluate the strength of CFST members with relatively small size. In this work, the experimental database of short CFST members with compact sections was complied, and the applicability of the PSD method was evaluated. The results showed that the test-to-calculated ratios decreased with increasing size of CFST members, and unsafe designs emerged for large-size CFST members. Based on the observations, the mechanics-based modified plastic stress distribution (MPSD) method was proposed, which not only considered the size effect at the material level similar to the way treated in most of the previous calculation methods, but also incorporated the size effect of the interaction action between the steel tube and concrete at the member level. Finally, the MPSD method was evaluated against the experimental database, and good agreement was achieved. These results provide strong justification for changes to the PSD method in design standards.
Read full abstract