Shear capacity estimation of FRP-reinforced concrete beams using computational intelligence

Abstract

Fibre Reinforced Polymer (FRP) bars have great potential for the strengthening of the existing buildings. They also can be used as a suitable replacement for steel bars. Therefore, the structural behavior and predicting the strength of the constructed elements by this type of material is an important issue. However, the existing equations to determine the shear capacity of concrete elements reinforced with FRP are conservative and usually derived from developing the available relationships for steel-reinforced members, especially for beams. In this study, an Artificial Neural Network (ANN) model was trained to extract a new equation to predict the shear strength of concrete beams reinforced with FRP bars. To this end, a large number of experimental data was applied to the proposed ANN to predict the shear strength. Also, to investigate the effective percentage of each independent parameter on the considered output, sensitivity analyses were performed and discussed in detail. Finally, the accuracy of the proposed equation was investigated in comparison with the existing models. It could be approved that the provided equation with high accuracy presented better results than the other relationships.