Predicting the axial load capacity of circular concrete-filled steel tube columns confined with fiber-reinforced polymer using machine learning models

In this study, a new model was developed to predict the axial load-carrying capacity of circular concretefilled steel tube (CFST) columns externally confined with fiber-reinforced polymer (FRP). For this purpose, 227 experimental data points collected from the literature were split, with 75% for training and 25% for testing. A new equation was then derived using Gene Expression Programming (GEP). Additionally, prediction models were developed using several machine learning (ML) algorithms, including MLP (Multilayer Perceptron), KNN (KNearest Neighbors), BAG (Bootstrap Aggregating), RF (Random Forest), GBM (Gradient Boosting Machine), LightGBM (Light Gradient Boosting Machine), XGBoost (Extreme Gradient Boosting), and CatBoost (Categorical Boosting). A 10-fold cross-validation approach was employed during the grid search to identify the optimal hyperparameter combination for the ML models. The predictive performances of the proposed models were statistically evaluated and compared with existing equations in the literature. CatBoost demonstrated the best predictive performance on the test data, with a MAPE of 4.075, an RMSE of 180.509, an R2 of 0.988, and a COV of 0.059. SHAP analysis was used to evaluate the contribution of each input parameter to the prediction results.