Experimental, numerical simulation and design methodology of axial compression performance of combined steel columns for modular buildings

Abstract

Load-supporting columns of modular buildings are their key load-bearing components, but fewer research studies have been conducted so far. To investigate the effects of initial defects of components, bolt-hole sizes, end stiffness, and component slenderness ratios on the axial compressive performance of load-supporting columns, this paper carries out a full-scale modeling experimental study and parametric finite element analysis. Subsequently, the test and finite element results were compared with the predicted results of the specifications in China (GB 50017-2017), the United States (ANSI/AISC 360-22), and Australia (AS 4100-2020) to assess the appropriateness of the specifications. The results show: It is recommended that small-section high-strength bolts be used to connect the load-supporting columns. When the overall and local defect magnitude meets the accuracy requirements, the magnitude size has little effect on the component bearing capacity. The direction of the overall bending defect has little effect on the initial stiffness and peak load-carrying capacity of the steel tube spandrel column. When the regularized slenderness ratio of the components is less than 0.4, the reduction of the hinged restraint capacity is less than 10% compared with that of the stiffened restraint capacity. The Chinese specification can be used for component design and its predictions are on the conservative side. As the section width-to-thickness ratio becomes larger, the predictions of the U.S. specification change from an accurate prediction to one with too large an error. The Australian specification predictions were all on the unsafe side, but most components had errors of 10% or less. The results of this paper help to promote the development of modular building technology.