Predicting Shear Strength of Steel Fiber-Reinforced Concrete Beams Using Artificial Neural Networks (ANN)
Saeed Hashim Saeed *
Department of Structural Engineering, Faculty of Civil Engineering, University of Latakia, Latakia, Syria.
*Author to whom correspondence should be addressed.
Abstract
The study presents a highly efficient computational model for estimating the shear strength of steel fiber-reinforced concrete (SFRC) beams using Artificial Neural Networks (ANN). An ANN model was developed using 105 experimental cases compiled from reference studies in this field. The target output was the shear resistance (Vu), while the 11 input parameters included: beam width, beam height, shear span, fiber shape factor, fiber tensile strength, shear span-to-effective depth ratio, fiber volume fraction, longitudinal reinforcement ratio, maximum aggregate size, concrete compressive strength (f'c), and effective depth.
This configuration achieved an exceptional correlation coefficient of R = 0.98306 and This methodological advancement represents a significant departure from prior reference studies, which relied exclusively on empirical and statistical models accounting only for fundamental variables such as Primary beam dimensions, Shear span characteristics, Specified concrete compressive strength (f'c), Properties of embedded steel fibers, Gradation characteristics of coarse aggregates within the concrete mixture matrix.
Keywords: Steel fibers, regression analysis, artificial neural network, shear strength, concrete beams