Data-Driven Prediction and Probabilistic Modeling of the Axial Capacity of Concrete-Filled Steel Tubular (CFST) Columns Using Artificial Neural Networks

Document Type : Research Paper

Authors

Department of Civil Engineering, FSh. C., Islamic Azad University, Fouman, Iran

10.22124/jcr.2026.33264.1738

Abstract

This study aims to develop an Artificial Neural Network (ANN) model to predict the axial capacity of Concrete-Filled Steel Tubular (CFST) columns under axial loading. A comprehensive database comprising 95 experimental specimens from reliable sources was compiled to capture a wide range of geometric and material characteristics. A multilayer feedforward ANN with optimized architecture was designed and trained to model the nonlinear relationships between input parameters and the axial capacity of CFST columns. The model demonstrated high accuracy, with correlation coefficients of 0.9968, 0.9663, and 0.9612 for the training, validation, and test datasets, respectively. Evaluation using an independent dataset further confirmed the model’s reliability, yielding a correlation coefficient of 0.9994. Sensitivity analysis based on Garson’s method indicated that the diameter-to-thickness ratio had the greatest influence (31.39%), while the column slenderness ratio had the least (13.99%). Additionally, a practical computational framework was developed based on the trained ANN weights and biases, allowing direct calculation of axial capacity without rerunning the network. Finally, goodness-of-fit tests suggested that the Generalized Extreme Value distribution provides the best match to the experimental data, serving as an optimal probability distribution function for the axial capacity of CFST columns. The proposed ANN model and computational framework provide an efficient, accurate, and generalizable tool for predicting the axial performance of CFST columns in structural design and assessment.

Keywords

Main Subjects

References

[1] Yang, C., Gao, P., Wu, X., Chen, Y. F., Li, Q., Li, Z. Practical Formula for Predicting Axial Strength of Circular-CFST Columns Considering Size Effect. Journal of Constructional Steel Research, 168 (2020), 105979.
[2] Sobhani, J., Ejtemaei, M., Sadrmomtazi, A., Mirgozar, M. A. Modeling Flexural Strength of EPS Lightweight Concrete Using Regression, Neural Network and ANFIS. International Journal of Optimization in Civil Engineering, 9(2) (2019), 313–329.
[3] Li, J., Pang, Y., Mu, Q., Zhang, X., Shi, Y., Wang, H. Post-Blast Capacity Evaluation of Concrete-Filled Steel Tubular (CFST) Column Based on Machine Learning Technique. Advances in Structural Engineering, 26(11) (2023), 1953–1972.
[4] Ci, J., Ahmed, M., Tran, V. L., Jia, H., Chen, S. Axial Compressive Behavior of Circular Concrete-Filled Double Steel Tubular Short Columns. Advances in Structural Engineering, 25(2) (2022), 259–276.
[5] Wang, C., Chan, T. M. Machine Learning (ML) Based Models for Predicting the Ultimate Strength of Rectangular Concrete-Filled Steel Tube (CFST) Columns under Eccentric Loading. Engineering Structures, 276 (2023), 115392.
[6] Ma, L., Zhou, C., Lee, D., Zhang, J. Prediction of Axial Compressive Capacity of CFRP-Confined Concrete-Filled Steel Tubular Short Columns Based on XGBoost Algorithm. Engineering Structures, 260 (2022), 114239.
[7] Megahed, K. Symbolic Regression for Strength Prediction of Eccentrically Loaded Concrete-Filled Steel Tubular Columns. Scientific Reports, 15(1) (2025), 3085.
[8] Asteris, P. G., Sivenas, T., Gkantou, M., Formisano, A., Le, T. T. Estimation of Axial Load-Carrying Capacity of Elliptical Concrete Filled Steel Tubular Columns Using Computational Intelligence. Journal of Building Engineering, 113738 (2025).
[9] Roy, D., Das, D., Islam, K., Billah, A. M. Machine Learning Assisted Axial Strength Prediction Models for Concrete Filled Stainless Steel Tubular Columns. Structures, 73 (2025), 108329.
[10] Lai, B. L., Zheng, X. F., Fan, S. G., Chang, Z. Q. Behavior and Design of Concrete Filled Stainless Steel Tubular Columns under Concentric and Eccentric Compressive Loading. Journal of Constructional Steel Research, 213 (2024), 108319.
[11] Qi, H. H., Li, G. Q., Lou, G. B. Behavior of Concrete-Filled Steel Tubular Column at Large Axial Deformation. Structures, 70 (2024), 107603.
[12] Abramski, M. Load-Carrying Capacity of Axially Loaded Concrete-Filled Steel Tubular Columns Made of Thin Tubes. Archives of Civil and Mechanical Engineering, 18(3) (2018), 902–913.
[13] Xu, L. H., Xu, F. Z., Zhou, P. H., Gu, Y. S., Wu, M. Experimental Research on Axial Compression Performance of Medium-and-Long Steel Tubular Columns Filled with High Strength Self-Stressing Self-Compacting Concrete. China Civil Engineering Journal, 49(11) (2016), 26–34, 44.
[14] Ekmekyapar, T., Al-Eliwi, B. J. Experimental Behaviour of Circular Concrete Filled Steel Tube Columns and Design Specifications. Thin-Walled Structures, 105 (2016), 220–230.
[15] Dundu, M. Compressive Strength of Circular Concrete Filled Steel Tube Columns. Thin-Walled Structures, 56 (2012), 62–70.
[16] De Oliveira, W. L. A., De Nardin, S., de Cresce El, A. L. H., El Debs, M. K. Influence of Concrete Strength and Length/Diameter on the Axial Capacity of CFT Columns. Journal of Constructional Steel Research, 65(12) (2009), 2103–2110.
[17] Han, L. H., Yao, G. H. Experimental Behaviour of Thin-Walled Hollow Structural Steel (HSS) Columns Filled with Self-Consolidating Concrete (SCC). Thin-Walled Structures, 42(9) (2004), 1357–1377.
[18] Tan, K. F., Pu, X. C. Study on Behavior and Load Bearing Capacities of Slender Steel Tubular Columns and Eccentrically Loaded Steel Tubular Columns Filled with Extra Strength Concrete. Journal of Building Structures, 21(2) (2000), 12–19.
[19] Cai, S. H., Gu, W. L. Behaviour and Ultimate Strength of Long Concrete-Filled Steel Tubular Columns. Journal of Building Structures, 3 (1985), 32–40.

Files

Share

How to cite

Statistics