Machine Learning-Based Prediction of Compressive Performance in Circular Concrete Columns Confined with FRP

This article presents a comprehensive investigation, focusing on the prediction and formulation of the design equation of compressive strength of circular concrete columns confined with Fiber Reinforced Polymer (FRP) using advanced machine learning models. Through an extensive analysis of 170 experimental data specimens, the study examines the effects of six key parameters, including concrete cylinder diameter, concrete cylinder-FRP thickness, compressive strength of concrete without FRP, initial compressive strain of concrete without FRP, elastic modulus and tensile strength of FRP, on the compressive strength of the circular concrete columns confined with FRP. The predictive model and design equation of compressive strength is developed using a machine learning technique, specifically the artificial neural networks (ANN) model. The results demonstrates strong correlations between the compressive strength of the circular concrete columns confined with FRP and certain factors, such as the compressive strength of the concrete and compressive strain of the concrete column without FRP, elastic modulus of FRP, and tensile strength of FRP. The ANN model specifically developed using Neural Designer, exhibits superior predictive accuracy compared to other constitutive models, showcasing its potential for practical implementation. The study's findings contribute valuable insights into accurately predicting the compressive performance of circular concrete columns confined with FRP, which can aid in optimizing and designing civil engineering structures for enhanced performance and efficiency.