Ultra-high performance concrete (UHPC) and fiber-reinforced polymer (FRP) have been more and more widely used in large-scale constructions. Using FRP at the shear and bending section of concrete can increase… Click to show full abstract
Ultra-high performance concrete (UHPC) and fiber-reinforced polymer (FRP) have been more and more widely used in large-scale constructions. Using FRP at the shear and bending section of concrete can increase the strength and fatigue resistance of the concrete members. However, the performance of FRP-UHPC composite structure depends mainly on the interface connection between FRP and UHPC. Therefore, to prevent the bond-slip of FRP-UHPC composite structure from causing destructive structural damage, it is essential to detect the bond-slip of interfaces for providing early warning of composite structures. Glass fiber reinforced polymer (GFRP) and concrete can be combined through several interface bond methods to form innovative composite structures. This study experimentally investigated the bond-slip detection of the shear interface of GFRP-concrete composite members using piezoceramic smart aggregates (SAs). Two groups of eight GFRP-concrete composite members with different bond methods were fabricated and tested. Both UHPC and regular concrete materials were considered. Six kinds of bond types were employed, including bolted, epoxy bonded, bonded by GFRP stay-in-plane form and their combinations. The push-out experimental results were analyzed in detail, involving the load versus slip displacement curves and failure modes. Meanwhile, the bond-slip between GFRP and UHPC interfaces was detected by the SA-based active sensing approach. A pair of SAs attached at both sides of each composite member was employed as an actuator and a sensor, respectively. The wavelet packet-based analyses, including the energy indices and damage index, were applied. Using an SA-based active sensing approach, the initiation and development of bond-slip for GFRP and UHPC composite members with different bond methods were successfully captured and quantitatively evaluated.
               
Click one of the above tabs to view related content.