Machine Learning-Based Prediction of Load Capacity in Corrugated Concrete-Filled Double Skin Tubes under Axial Load

Concrete-filled double-skin steel tubular (CFDST) columns have become widely utilized in building construction and bridges, thanks to their exceptional structural capabilities. This study investigates the axial compression behavior of corrugated CFDST columns, analyzing the impact of flat or corrugated external/internal plates and varying internal steel tube widths. Furthermore, basic regression analysis fails to precisely forecast the complicated relation between the column properties and their compressive strength. To overcome these challenges, this study suggests two machine learning (ML) models, including the Gaussian process (GPR) and the extreme gradient boosting model (XGBoost). To estimate their strength, these models employ a range of input variables, such as the geometric and material properties of corrugated CFDST columns. The models are trained and evaluated based on two datasets consisting of 243 axially loaded corrugated CFDST columns have done by authors and past researches. This study proposed two machine-learning models to estimate the ultimate compressive strength of corrugated CFDST columns. The findings indicated that the GPR model outperformed the XGBoost model in predicting the bearing capacity of corrugated CFDST columns. Additionally, the Shapley Additive Explanation technique was employed for feature analysis. The outcomes of this analysis revealed that parameters such as section width and concrete strength positively influence the compressive strength index. This suggests that optimizing these parameters could significantly improve the design and performance of corrugated CFDST columns. Future research should focus on further refining these models and exploring additional factors that may affect compressive strength.