Application of Artificial Neural Networks for Predicting Axial Strain of FRP-Confined Concrete

Abstract

Multiple research studies have developed frameworks to forecast the ability of concrete structural elements to withstand compression along their length. However, further exploration is required to refine predictions for the axial compressive strain, as existing strain models lack precision. The earlier models were created with restricted and noisy data sets and basic modelling methods, underscoring the necessity for a more meticulous approach to introduce a more accurate strain model and to evaluate its forecasts against those of current models.This study wants to fill in the gap by creating models for how much concrete reinforced with fiber-reinforced polymer (FRP) can stretch using computer simulations called artificial neural networks (ANN). This approach is based on a substantial database comprising 570 sample points. The comprehensive investigation of these estimates robustly validates the accuracy and practicality of the suggested ANN models for predicting the axial strain of FRP -confined concrete compression members.

Keywords

• fiber reinforced polymer (frp)
• confined concrete
• artificial neural networks
• strain model