Progressive collapse response of steel and composite buildings

Abstract

Subsequent to the World Trade Centre collapses in 2001, general interest into structural robustness and progressive collapse has been significantly heightened. It is recognised that the current prescriptive rules employed in the design of building structures to resist progressive collapse need to be replaced by performance-based approaches. A continuing research program at Imperial College London aims at the development of a complete design method that will address the basic features of progressive collapse whilst being tractable in terms of complexity for routine use in practice. An important step towards that objective is the proper treatment and understanding of the fundamental mechanics of the problem. The current study is motivated by that requirement and seeks to build on previous developments at Imperial to explore the progressive collapse response of steel and composite buildings on a quantitative basis. At first, the study is dedicated to the development of a simplified model for representation of the connection behaviour. The model is incorporated into a slope-deflection approach and an analytical method for prediction of the nonlinear static response of steel and composite beams following column removal – i.e. a common design scenario for progressive collapse – is derived. The method is carefully validated and applied in a detailed study of the response of axially restrained beams in progressive collapse, where the most important structural parameters and their effects on performance are identified. Based on those outcomes, the behaviour of bare steel and composite grillage systems following sudden column loss is subsequently appraised. It is concluded that progressive collapse resistance depends on the interplay between the connection moment capacities and ductility. Performance may also be enhanced by compressive arching action in the presence of axial restraint; however, for average values of connection ductility, failure typically occurs prior to the development of significant tensile catenary action. Therefore, it is suggested that design methods for progressive collapse should be primarily oriented towards the prediction of appropriate values for the connection moment capacities.