Grey Wolf Optimizer-Based ANN to Predict Compressive Strength of AFRP-Confined Concrete Cylinders

One of the well-known and attractive solutions for strengthening of reinforced concrete columns are Fiber-Reinforced Polymer (FRP) confinements. However most of the models developed in literature are based on a general equation originally given for steel confinements. Because of the material differences between FRPs and steel, we aimed to model the compressive strength of aramid fiber-reinforced polymer confined concrete cylinders without any assumption on the form of the model. To be useful, Artificial Neural Networks (ANNs) must be trained using an optimization algorithm. In this study, we used a recently proposed optimization algorithm named Grey Wolf Optimizer for training. After developing ANN models, we compared them with five existing equations. The statistical parameters indicated the superiority of the proposed ANN models over existing ones