Prediction of the ultimate strength of reinforced concrete beams FRP-strengthened in shear using neural networks

Abstract

In the last years, a great number of experimental tests have been performed to determine the ultimate strength of reinforced concrete beams retrofitted in shear by means of externally bonded fibre-reinforced polymers (FRP). Most of design proposals for shear strengthening are based on a regression analysis from experimental data corresponding to specific configurations which makes very difficult to capture the real interrelation among the involved parameters. To avoid this, an artificial neural network has been developed to predict the shear strength of concrete beams reinforced with this method from previous tests. Furthermore, a parametric study has been carried out to determine the influence of some beam and external reinforcement parameters on the shear strength with the purpose of reaching more reliable designs. Finally, some modifications of the design expressions are proposed and checked with experimental results.