Prediction of shear capacity of RC beams strengthened with FRCM composite using hybrid ANN-PSO model

Abstract

The aim of this study is to develop a hybrid Artificial Neural Network- Particle Swarm Optimization (ANN-PSO) model for improving shear strength prediction of reinforced concrete (RC) beams strengthened with fiber reinforced cementitious matrix (FRCM). A set of 89 experimental test results of strengthening RC beams are collected and used for developing the ANN-PSO model. The performance results of ANN-PSO are compared with those of pure ANN model. Typical statistical properties including the coefficient of determination (R2), root mean squared error (RMSE), and the number of predicted data falling in a deviation of ± 20% compared with experimental data (a20−index) are calculated to evaluate the accuracy of those models. The comparisons reveal that ANN-PSO outperforms the ANN model with R2, RMSE, and a20−index values of 0.937, 6.02, and 0.842, respectively. Moreover, the effects of input parameters (i.e., beam geometry, concrete and reinforcement properties, and FRCM composite parameters) on the predicted shear strength are quantified. Additionally, an efficient graphical user interface (GUI) tool is developed for facilitating the practical design process of the strengthening RC beams.