FRP shear contribution prediction for U-wrapped RC T-beams using a soft computing tool

Abstract

The reinforced concrete (RC) T-beam is relatively stronger in shear than the rectangular RC beam, represents the most realistic condition and has more implications in the construction industry. However, no study encompassing a large number of results of tests conducted on the RC T-beams externally shear strengthened with fiber-reinforced polymer (FRP) has been found. Therefore, the current study targets to calculate the contribution of externally bonded FRP towards the shear capacity of RC T-beams by means of a soft computing tool i.e. adaptive neuro-fuzzy inference system (ANFIS). A total of 148 sets of data collected from the literature and used for the training and testing of the ANFIS model. A comparative study between the ANFIS predictions and the obtained experimental outcomes has revealed that the ANFIS forecasts are in fine agreement with the experimental outcomes. Additionally, evaluation of the efficacy of the current ANFIS model has been done by comparing the ANFIS predictions with the predictions of seven broadly utilized design guidelines. On the basis of the comparison of different performance measures, it has been deduced that the efficacy of the current ANFIS model is superior to the considered design guidelines. A parametric study has also been done to explore the impact of the combined effect of the two input parameters on the FRP shear contribution.