A nonlinear prediction model of the debonding process of an FRP-concrete interface under fatigue loading. (10th March 2023)

• Record Type: Journal Article
• Authors: Min, Xinzhe; Zhang, Jiwen; Li, Xing; Wang, Chao; Tu, Yongming; Sas, Gabriel; Elfgren, Lennart
• Is Part Of: Construction & building materials. Volume 369(2023)
• Publication Date: 2023-03-10
• Subjects: Fiber Reinforced Polymer -- Concrete -- Interface -- Fatigue -- Debond; Building materials -- Periodicals; 624.18
• DOI: 10.1016/j.conbuildmat.2023.130583
• Languages: English
• ISSNs: 0950-0618

Abstract

Highlights: The fatigue debonding rate was determined by the amplitude and level of the load. The proposed model explicitly took into account the noted factors. The d a /d N and the S ¯ · Δ S show a linear relationship in double logarithmic coordinates. The debonded FRP has a significant effect on the subsequent debonding development. Abstract: Externally bonded Fiber Reinforced Polymer (FRP) strengthening has been proven to be an efficient and reliable method for structural strengthening of reinforced concrete (RC) members. However, the beneficial effects of this method can be diminished due to the debonding of the FRP laminates. The mechanism of FRP debonding still requires further research, especially for strengthened members under fatigue loading. To understand and predict the FRP fatigue debonding process better, eleven FRP-concrete joint specimens were tested under static or fatigue loading. Both the theoretical derivation and the experimental study indicated that the debonding growth rate of the FRP laminate depended not only on the mean level ( S ¯ ), but also the amplitude ( Δ S ) of the applied fatigue load. In addition, the debonded portion of the FRP laminate had a significant impact on the following debonding process due to the friction and mechanical interaction between the debonded FRP and the concrete surface. Therefore, a new nonlinear prediction model is proposed in this paper. The proposed model explicitly took into account the amplitude and the mean level of the fatigue loading, which enabled the effect of both to be modelled. Meanwhile, a correction term was also introduced into the model to account for the influence of the previously debonded FRP laminate. The predicted results of the debonding growth rate and the debonding length agreed well with the experimental results.