Abstract
This investigation introduces advanced predictive models for estimating axial strains in Carbon Fiber-Reinforced Polymer (CFRP) confined concrete cylinders, addressing critical aspects of structural integrity in seismic environments. By synthesizing insights from a substantial dataset comprising 708 experimental observations, we harness the power of Artificial Neural Networks (ANNs) and General Regression Analysis (GRA) to refine predictive accuracy and reliability. The enhanced models developed through this research demonstrate superior performance, evidenced by an impressive R-squared value of 0.85 and a Root Mean Square Error (RMSE) of 1.42, and significantly advance our understanding of the behavior of CFRP-confined structures under load. Detailed comparisons with existing predictive models reveal our approaches' superior capacity to mimic and forecast axial strain behaviors accurately, offering essential benefits for designing and reinforcing concrete structures in earthquake-prone areas. This investigation sets a new benchmark in the field through meticulous analysis and innovative modeling, providing a robust framework for future engineering applications and research.