Fatigue Life Prediction Model of FRP-Concrete Interface Based on Gene Expression Programming

Area of Science:

• Civil Engineering
• Materials Science
• Structural Engineering

Background:

• Accurate prediction of interfacial fatigue life in fiber-reinforced polymer (FRP)-concrete composites is crucial for assessing the performance of strengthened concrete structures.
• Existing fatigue life prediction models for FRP-concrete interfaces often suffer from inconsistencies and limited accuracy.
• Understanding the fatigue behavior of externally bonded (EB) and near-surface-mounted (NSM) reinforcement is vital for structural retrofitting.

Purpose of the Study:

• To develop a more accurate and reliable model for predicting the interfacial fatigue life of FRP-concrete composites.
• To address the limitations of existing fatigue life prediction models by employing advanced computational techniques.
• To establish distinct fatigue life calculation formulas for both EB and NSM reinforcement methods.

Main Methods:

• Collected and analyzed 219 sets of interfacial fatigue test data for both EB and NSM reinforcement configurations.
• Utilized Pearson correlation analysis to identify key factors influencing interfacial fatigue life.
• Employed gene expression programming (GEP) to develop and optimize fatigue life prediction models, exploring various input parameter forms.
• Conducted parameter sensitivity and variable importance analyses to validate the GEP model's robustness.

Main Results:

• Developed novel fatigue life calculation formulas specifically for EB and NSM FRP-concrete interfaces.
• The GEP model demonstrated superior prediction accuracy compared to existing models, achieving a coefficient of determination (R²) of 0.819.
• The GEP model exhibited lower average absolute error and other statistical indicators, signifying enhanced predictive performance.
• The model effectively captured the intrinsic relationships between fatigue life and influential factors.

Conclusions:

• The developed GEP model provides a significant advancement in predicting the interfacial fatigue life of FRP-concrete.
• The model's high accuracy and robustness make it a valuable tool for the design and assessment of FRP-strengthened concrete structures.
• The study highlights the potential of GEP in addressing complex material behavior and improving engineering predictions in structural retrofitting.