The Application of Particle Swarm Optimization and Artificial Neural Networks to Estimating the Strength of Reinforced Concrete Flexural Members

Abstract

The aim of this paper is a determination of the shear strength of fiber reinforced polymer reinforced concrete flexural members without stirrups. For this purpose, a neural network approach was used. The weights and biases of the considered network determined based on best values which were optimized from the particle swarm optimization algorithm (PSO). For training the model, a collection of 108 datasets which was published in literature was applied. Six inputs including the compressive strength of concrete, flexural FRP reinforcement ratio, modulus of elasticity for FRP, shear span-to-depth ratio, member web width and adequate member depth used for creating the model while the shear strength considered as the output. The best structure for the network was obtained by a network with one hidden layer and ten nodes. The results indicated that artificial neural networks based on particle swarm optimization algorithm could be able to predict the strength of the considered RC elements.