Two artificial neural networks (ANN), back‐propagation neural network (BPNN) and the radial basis function neural network (RBFNN), are proposed to predict the carbonation depth of prestressed concrete. In order to generate the training and testing data for the ANNs, an accelerated carbonation experiment was carried out, and the influence of stress level of concrete on carbonation process was taken into account especially. Then, based on the experimental results, the BPNN and RBFNN models which all take the stress level of concrete, water‐cement ratio, cement‐fine aggregate, cement‐coarse aggregate ratio and testing age as input parameters were built and all the training and testing work was performed in MATLAB. It can be found that the two ANN models seem to have a high prediction and generalization capability in evaluation of carbonation depth, and the largest absolute percentage errors of BPNN and RBFNN are 10.88% and 8.46%, respectively. The RBFNN model shows a better prediction precision in comparison to BPNN model.

The substitution conventional steel reinforcement with glass fiber-reinforced polymer (GFRP) bars is a widely adopted strategy used to improve the durability of concrete structures in chloride environments, offering benefits such as enhanced corrosion resistance, reduced maintenance needs, and increased service life. This study investigates the bond behavior between glass fiber-reinforced polymer (GFRP) bars and concrete under long-term chloride dry–wet cycling exposure. Pull-out tests were conducted on various specimens subjected to exposure durations of 0, 3, 6, 9, and 12 months. The experimental results indicate that, after 12 months of chloride dry–wet cycling, the bond strength retention rates of threaded ribbed GFRP with a bond length of 5d, sand-coated GFRP with a bond length of 5d, and threaded ribbed GFRP with a bond length of 7d were 57.9%, 62.2%, and 63.8%, respectively. To predict the GFRP–concrete bond performance after chloride exposure, a novel bond strength model for GFRP bars embedded in concrete, considering the mechanical interlocking effect of ribs, was proposed and validated by the test results. The overall prediction errors for RG-5d, SG-5d, and RG-7d specimens were 0.98, 0.81, and 0.93, respectively. Additionally, a sensitivity analysis was conducted on the main parameters in the model. Finally, the long-term GFRP–concrete bond performance deterioration was estimated using the proposed model. These findings are expected to provide valuable insights into the long-term bond performance and service life prediction of GFRP–concrete members in chloride environments.

The substitution of conventional steel reinforcement with fiber-reinforced polymer (FRP) bars is a widely adopted strategy for enhancing the durability of concrete structures in chloride environments. However, due to the distinct material properties of FRP and steel bars, conventional degradation theory and formulas of bonding strength cannot be directly applied. This study investigates the bond behavior between Glass fiber-reinforced polymer (GFRP) bars and concrete under long-term chloride dry-wet cycling exposure. Pull-out tests were conducted on various specimens subjected to exposure durations of 0, 3, 6, 9, and 12 months. The experimental results demonstrated that factors such as surface textures and effective bond length significantly influenced bond behaviors and the sustained deterioration was observed with increased exposure duration. To predict the GFRP-concrete bond performance after chloride exposure, a novel bond strength model for GFRP bars embedded in concrete, considering the mechanical interlocking by ribs, was proposed and validated by the test results. Additionally, a sensitivity analysis was conducted on the main parameters in the model. Finally, the long-term GFRP-concrete bond performance deterioration was estimated using the proposed model. These findings are expected to provide valuable insights into the long-term bond performance and service life prediction of GFRP-concrete members in chloride environments.