Axial Compressive Performance and Design Method for Reinforced Hollow High-Strength/Ultra-High-Strength Concrete-Filled Square Steel Tubular Columns

Reinforced hollow high-strength/ultra-high strength concrete-filled square steel tubular (RHHS-CFST) columns are a kind of prefabricated members with advantages such as mature production techniques, low cos, light weight, and high bearing capacity. As prefabricated components, they hold broad application potential in engineering structures, particularly for long-span foundation excavation support systems and multi-level parking facilities. However, there is no design method capable of accurately predicting the axial compressive bearing capacity of this member. The available design methods cannot account for the influence of high concrete strength (exceeding C80 grade) or the early buckling of the steel tube caused by the high hollow ratio and insufficient wall thickness. To address this issue, this paper conducted a systematic parametric investigation with a reliable finite element modeling method (FEM). Subsequently, a new expression was proposed to predict the axial compressive bearing capacity based on available test data and finite element analytical results. The results showed that the hollow ratio and width-to-thickness ratio of the steel tube significantly affected the axial compressive bearing capacity. For the members with a hollow ratio greater than 0.25 and a width-to-thickness ratio unsatisfying the requirements of current design codes for concrete-filled steel tubes, local buckling of the steel tube might occur before reaching the peak load, leading to a reduction in strength. The predicted results from the proposed expression matched well with the FEM results and test data. The average ratio of predicted to measured/FEM results was 0.935, with a coefficient of variation of 0.108. This indicates that the proposed design method can safely predict the axial compressive bearing capacity of RHHS-CFST members. The highest overestimation by the proposed design method was only 8.6%.