Development of Efficient Prediction Model of FRP-to-Concrete Bond Strength Using Curve Fitting and ANFIS Methods

Abstract

Externally bonded fiber-reinforced polymer (FRP) plates or sheets have become a common retrofitting approach for sustaining old reinforced concrete structures in the modern era. The capacity of FRP-strengthened structures cannot be accurately estimated because the bond strength between FRP and concrete surface is accurately unpredictable. Various studies are available in the literature to predict the FRP-to-concrete bond strength (FRP-CBS), but they are based on limited experimental data sets and have lesser accuracy. To solve this problem, curve-fitting (CF) and adaptive neuro-fuzzy inference systems (ANFIS) models have been developed to predict the FRP-CBS using 935 datasets. The database was collected from published literature and the same was used to develop the ML model. Comparison with standard guidelines, including ACI, TR-55 fib, CNR, and JCI, and other analytical models, revealed that the ANFIS model outperformed the CF model and all other analytical models. The ANFIS model achieved a correlation coefficient of 0.9189 and a mean absolute error (MAE) of 2.43 kN, while the CF model achieved a correlation coefficient of 0.7303 and an MAE value of 4.30 kN. Moreover, a parametric study was conducted to identify the influence of each specific parameter on the bond strength. The developed ANFIS-based model can be readily utilized by structural engineers, FRP applicators, and researchers for estimating the FRP-to-concrete bond strength.