Performance and Design of Composite Modular System with Tenon Connections for Multi-Storey Buildings

Abstract

Modular building is an innovative construction method based on advanced manufacturing technologies, which is a more eco-friendly, effective, and cost-saving alternative than conventional methods. The primary objective of this thesis is to design a sufficient composite modular system for multi-storey applications and provide design recommendations based on system-level analyses under earthquakes, winds, and sudden column losses. In the course of this thesis, a numerical model is first created for an existing tenon-connected inter-module connection to investigate its effects on the building’s lateral resistance. A cohesive interface model is used to account for the weld fracture. Due to the semi-rigid connectivity, there are around 53% and 28% reductions in the yield and maximum capacity of the building, respectively. The displacement coefficient method per American guidelines FEMA-356 is then adopted to predict the maximum deformation of the modular buildings under different design seismic loads. To strengthen the modular buildings, a novel composite modular system is newly proposed, which consists of concrete-filled steel tubular columns, laminated double beams, and integrated composite slabs. The structural responses of the composite modular buildings are assessed under design wind actions per Australian Standards AS 1170.0-2. The results indicate that the proposed buildings have sufficient design capacity but insufficient deflection control. The progressive collapse analysis is performed on the buildings in sudden column loss scenarios per Unified Facilities Criteria UFC 4-023-03. The results show that alternate load paths are activated after the notional column removals, and the progressive collapse is unlikely for the scenarios under consideration. Finally, the suitable dynamic increase factors of 1.90 and 1.60 are recommended for the 4- and 12-storey modular buildings, respectively, allowing peak dynamic responses to be predicted using the static approach.