Artificial Neural Network Algorithms to Predict the Bond Strength of Reinforced Concrete: Coupled Effect of Corrosion, Concrete Cover, and Compressive Strength

Abstract

Degradation of the bond between reinforcement steel bars and concrete poses a huge challenge to the design of sustainable infrastructure; thus, it is important to assess the key factors that influence the bond strength of reinforced concrete (RC) structures. Studies relating to the effects of concrete compressive strength and cover, steel embedment length, and diameter on RC bond strength have been achieved primarily through time-consuming (and expensive) experiments. In this study, an effort is made toward developing generic Artificial Neural Network (ANN) models to predict the bond strength between steel reinforcement and concrete. To assess the efficiency of the ANN prediction capability under a case of limited experimental data, the ANN models were activated through Softplus, Rectified Linear unit (ReLU), or Sigmoid functions and their results were compared. The experimental/test data used in the modeling study covered corrosion levels from 0 to 20% of the reinforcement bars' weight, concrete compressive strengths of 23 and 51 MPa, and concrete covers ranging between 15 and 45 mm. A comparison was made between the bond strength values predicted by the ANN models, linear/non-linear statistical regression equations, and other analytical equations available in the literature. The model results indicated that amongst the three parameters considered in the study, the level of corrosion predominantly controls the bond strength. Moreover, the ANN(Softplus) model with a mean squared error (J) of 2.89 and a coefficient of determination (R2) of 96% demonstrated an accurate prediction of the bond strength in comparison with the ANN(Sigmoid), ANN(ReLu), and statistical regression models.

Keywords: Bond strength modeling, Reinforced concrete, Artificial neural networks, Corrosion.