Enhancing the lateral performance of modular buildings through innovative inter-modular connections

Abstract

Steel modular constructions involve the manufacture of fully equipped three-dimensional prefabricated modules in factory-controlled settings which are then transported to construction sites and assembled to form a complete structure. Adjacent modules are attached to each other only at their corners at inter-modular connections. Typical inter-modular connections are incapable of providing resistance against lateral dynamics loads. Current research shows that under lateral dynamic loads, steel modular buildings with rigid unyielding connectors are vulnerable to failure of the columns which result in either partial or complete collapse of the structure. Modular systems would therefore require additional in-situ lateral load resisting systems, such as shear cores, which would devalue the benefits of purely modular construction as they would need to be built in-situ. To address this shortcoming, this research proposes a novel steel inter-modular connection, with two variations, to achieve safe, reliable and ductile dynamic performance of a modular building under seismic actions. An extensive experimental program was undertaken to study the feasibility of the strength hierarchy and expected ductile failure patterns of the newly proposed inter-modular connections under monotonic and cyclic lateral loads. The experimental study revealed that the proposed inter-modular connections display superior dynamic behaviour with respect to response parameters such as moment-carrying capacity, energy dissipation and ductility. Ductile failure patterns within the connection region and away from the columns, which are critical members, were observed. This information will contribute to the design of safe and efficient inter-modular connections and enable enhanced lateral performance of steel modular buildings under dynamic loads. A comprehensive numerical model of the connection was also developed and validated for use in future parametric studies.