Predicting the Capacity of RC Beams Strengthened in Shear with Side-Bonded FRP Reinforcements Using Artificial Neural Networks

Abstract

The application of artificial neural network (ANN) to predict the shear capacity of reinforced concrete (RC) beams retrofitted in shear by means of side-bonded fiber-reinforced polymer (FRP) is investigated in this paper. An extensive literature review has been carried out. In addition, ten shear deficient RC beams with different carbon fiber-reinforced polymer (CFRP) configurations were tested and added as data to the collected data. It was aimed to build an efficient and practical ANN model with parameters which can easily be obtained without any calculation and/or experimental investigation. The results are compared with the design guideline equations that emerge as predictions of the FRP contribution using the trained neural networks: these are in good agreement with the experimental results and better than those calculated from the theoretical guideline equations. Based on ANN results, a parametric study has been carried out to study the importance of different influencing parameters on the FRP contribution. Thereafter, a new simple expression is proposed for determining the contribution of externally bonded side-bonded FRP. Accordingly, the suggested design formula is capable of predicting the experimental FRP satisfactorily so that it can be admitted as an alternative to the existing guideline equations within the range of parameters covered in the study.

Shear strength prediction of reinforced concrete beams using machine learning

“…The newly developed models outperformed the prediction capacities of the other models. In a separate study, Tanarslan employed an ANN to predict the shear strength of concrete beams retrofitted with a side-bonded FRP [38]. Tanarslan reported good conformance with the experimental results, and the predictions were more accurate than those obtained using the theoretical guidelines.…”

“…Based on the results and observations of their investigation researchers proposed modification to the design equations suggested by ACI for the evaluation of the shear capacity of beams strengthened with FRP. Tanarslan [38] evaluated the potential of ANN in predicting the shear strength of RC beams strengthened in shear with side-bonded FRP reinforcements. In addition to the dataset collected from literature, experimental testing of ten shear deficient beams with different carbon fibre reinforced polymer (CFRP) arrangements were conducted and the results were added.…”

Shear Strength Prediction of RC Beams Using Adaptive Neuro-Fuzzy Inference System

“…They also presented design equations to calculate the shear capacity of FRP-strengthened RC beams in shear [49]. Tanarslan (2011) investigated the performance of ANN in predicting the shear capacity of the RC beams retro tted in shear by means of side-bonded FRP; obtained results demonstrated the higher accuracy of those values obtained by ANN [50]. Tanarslan et al (2012) used the back propagation network to determine the shear strength contribution of RC beams strengthened in shear by retro tting externally bonded wrapped and U-jacketed FRP reinforcement, and showed that ANN was a good tool for predicting [51].…”