Enhancing Progressive Collapse Resistance of Precast Building

Abstract

Progressive collapse is spread of an initial local failure from element to element, which eventually results in the collapse of an entire structure or a disproportionately large part of it. Progressive collapse of building structures is due to extreme loading imposed by earthquake, flood, explosion, vehicle impact etc. Progressive collapse of any structure causes catastrophic failure of structure and substantial loss of human lives as well as natural resources. To reduce progressive collapse, either local resistance against abnormal load is enhanced for critical members or alternate load path is developed by providing continuity and redundancy. Precast buildings lack structural continuity and redundancy in the load paths. They are more susceptible to progressive collapse than cast-in-situ monolithic buildings. Hence, effective and economical strengthening techniques to upgrade load carrying capacity of beam-column assembly are needed in existing precast concrete buildings for progressive collapse mitigation. Objective of present study is to evaluate progressive collapse resistance of beam-column assembly with monolithic and precast connections by performing experimental and numerical study. Further to enhance progressive collapse resistance of monolithic and precast beam-column assembly, use of stainless steel wire mesh (SSWM) is explored in this study. For experimental study two bay one storey beam column assembly has been extracted from 6 storey building. Six specimens of beam-column assembly with monolithic connection and two precast wet connections are cast for experimental study. For each of monolithic & precast wet connections one control specimen is without strengthening and other is strengthened with SSWM. For precast wet connection in-situ microconcrete is used at connection region for embedding overlapping reinforcements. To bond SSWM on concrete surface Sikadur 30LP epoxy is used. End columns of beam-column assembly are restrained from vertical and horizontal movement using hydraulic jack and triangular frame. Progressive collapse scenario is simulated by applying vertical downward load using hydraulic jack at middle column. Load-displacement behaviour of specimens are recorded. On comparison of load-displacement behaviour of control and SSWM strengthened specimens of beam-column assembly, it is found that stainless steel wire mesh (SSWM) is effective to enhance progressive collapse resistance of monolithic as well as precast beam column assembly. For numerical simulation of progressive collapse of beam-column assembly finite element based software ABAQUS is used. Progressive collapse condition is simulated by applying downward load at middle removed column. To simulate progressive collapse condition similar to experimental study in numerical modelling, at top of side columns hinged support is considered while at bottom of side columns small vertical displacement is applied in addition to displacement in vertical downward direction applied at middle column. Concrete damage plasticity (CDP) model is used for finite element modelling of concrete. Concrete, reinforcement and SSWM are modelled using solid, three-dimensional bar and shell elements respectively. From finite element analysis load-displacement results at removed middle column location is plotted for control specimen and SSWM strengthened specimens of beam-column assembly. Load-displacement behaviour of beam-column assembly with monolithic and precast wet connections as obtained from experimental study is compared with that obtained from numerical analysis. Further, simplified model based on plastic analysis presented by Jian and Zheng for understanding progressive collapse behaviour of reinforced concrete beam-column substructure is studied and implemented for RC beam-column assembly considered in present investigation. On comparison of behaviour of RC beam-column assembly with monolithic connection under progressive collapse scenario as observed during experimental study and as obtained using simplified model, it is found that simplified model accurately predict behaviour upto peak loading.