Prediction of Shear Strength of CFRP-Reinforced Concrete Beams using Soft Computing Technique

Abstract

In this study, an adaptive neuro-fuzzy inference system (ANFIS) and particle swarm optimization (PSO) models were developed to predict the shear carrying capacity of reinforced concrete (RC) beams externally strengthened by carbon fiber reinforced polymer (CFRP) sheets. The models were developed using a database collected from 150 experiments available in the existing literature. Width of beam, effective depth of beam, shear span to depth ratio, strength of concrete, shear reinforcement ratio, and CFRP properties including depth, thickness, number of layers, elastic modulus, width /spacing, configuration, and orientation of principal fiber were considered as input parameters while the output parameter was the shear carrying capacity. A parametric study has been carried out using the ANFIS model to study the effect of different input parameters on the shear capacity of the strengthened RC beams. The study also presents a PSO model to predict the maximum shear carrying capacity of strengthened RC beams. PSO model was conducted by establishing an appropriate objective function estimated by neural network approach. The results show that the ANFIS model is a powerful approach for predicting the shear capacity compared to the results obtained from the design guideline equations (ACI 440.2R, fib-TG9.3, and CSA-S806). Additionally, the results show that the PSO model is a good tool to predict the optimal shear capacity of RC beams strengthened with CFRP sheet. Finally, the proposed models can serve as an accurate and simple predictive tools for the prediction of shear carrying capacity of the RC beam with externally bonded CFRP.