Enhancing Moment Capacity Prediction in FRP-Reinforced Concrete Beams through Soft Computing Models

Abstract

This article presents a comprehensive study employing artificial neural networks (ANNs) to forecast the moment capacity of fiber-reinforced polymer (FRP)-reinforced concrete beams. Using a data set of 116 data points from previous experiments by various researchers, six essential input parameters—beam width, beam overall depth, compressive strength of concrete, area of steel reinforcement at the bottom layer of the beam, and modulus of elasticity and ultimate strength of FRP—were integrated to predict moment capacity. The findings highlight the ANN model’s remarkable precision in forecasting moment capacity, emphasizing its high accuracy. The model analysis further reveals the relative importance of each input parameter, elucidating their significant roles in determining moment capacity. This research significantly advances the field of structural engineering by introducing a dependable and efficient method for predicting the moment capacity of concrete beams reinforced with FRP bars, thus greatly facilitating the design and assessment of such structures. The outcomes of this study offer valuable insights for engineers and researchers dedicated to improving the performance and safety of FRP-reinforced concrete beams in diverse structural applications.