Advanced predictive machine and deep learning models for round-ended CFST column

Abstract Confined columns, such as round-ended concrete-filled steel tubular (CFST) columns, are integral to modern infrastructure due to their high load-bearing capacity and structural efficiency. The primary objective of this study is to develop accurate, data-driven approaches for predicting the...

Full description

Saved in:
Bibliographic Details
Main Authors: Feng Shen, Ishan Jha, Haytham F. Isleem, Walaa J.K. Almoghayer, Mohammad Khishe, Mohamed Kamel Elshaarawy
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-90648-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849737782223699968
author Feng Shen
Ishan Jha
Haytham F. Isleem
Walaa J.K. Almoghayer
Mohammad Khishe
Mohamed Kamel Elshaarawy
author_facet Feng Shen
Ishan Jha
Haytham F. Isleem
Walaa J.K. Almoghayer
Mohammad Khishe
Mohamed Kamel Elshaarawy
author_sort Feng Shen
collection DOAJ
description Abstract Confined columns, such as round-ended concrete-filled steel tubular (CFST) columns, are integral to modern infrastructure due to their high load-bearing capacity and structural efficiency. The primary objective of this study is to develop accurate, data-driven approaches for predicting the axial load-carrying capacity (P cc​) of these columns and to benchmark their performance against existing analytical solutions. Using an extensive dataset of 200 CFST stub column tests, this research evaluates three machine learning (ML) models – LightGBM, XGBoost, and CatBoost – and three deep learning (DL) models – Deep Neural Network (DNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM). Key input features include concrete strength, column length, cross-sectional dimensions, steel tube thickness, and yield strength, which were analysed to uncover underlying relationships. The results indicate that CatBoost delivers the highest predictive accuracy, achieving an RMSE of 396.50 kN and an R 2 of 0.932, surpassing XGBoost (RMSE: 449.57 kN, R 2: 0.906) and LightGBM (RMSE: 449.57 kN, R2: 0.916). Deep learning models were less effective, with the DNN attaining an RMSE of 496.19 kN and R 2 of 0.958, while the LSTM underperformed substantially (RMSE: 2010.46 kN, R 2: 0.891). SHapley Additive exPlanations (SHAP) identified cross-sectional width as the most critical feature, contributing positively to capacity, and column length as a significant negative influencer. A user-friendly, Python-based interface was also developed, enabling real-time predictions for practical engineering applications. Comparison with 10 analytical models demonstrates that these traditional methods, though deterministic, struggle to capture the nonlinear interactions inherent in CFST columns, thus yielding lower accuracy and higher variability. In contrast, the data-driven models presented here offer robust, adaptable, and interpretable solutions, underscoring their potential to transform design and analysis practices for CFST columns, ultimately fostering safer and more efficient structural systems.
format Article
id doaj-art-e5428f75e28740c089e1a73349baca73
institution DOAJ
issn 2045-2322
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e5428f75e28740c089e1a73349baca732025-08-20T03:06:49ZengNature PortfolioScientific Reports2045-23222025-02-0115112910.1038/s41598-025-90648-2Advanced predictive machine and deep learning models for round-ended CFST columnFeng Shen0Ishan Jha1Haytham F. Isleem2Walaa J.K. Almoghayer3Mohammad Khishe4Mohamed Kamel Elshaarawy5College of Civil Engineering, Huaqiao UniversityDepartment of Civil Engineering, Indian Institute of Technology-ISMJadara University Research Center, Jadara UniversitySchool of business, Nanjing University of Information Science & TechnologyDepartment of Electrical Engineering, Imam Khomeini Naval Science University of NowshahrCivil Engineering Department, Faculty of Engineering, Horus University-EgyptAbstract Confined columns, such as round-ended concrete-filled steel tubular (CFST) columns, are integral to modern infrastructure due to their high load-bearing capacity and structural efficiency. The primary objective of this study is to develop accurate, data-driven approaches for predicting the axial load-carrying capacity (P cc​) of these columns and to benchmark their performance against existing analytical solutions. Using an extensive dataset of 200 CFST stub column tests, this research evaluates three machine learning (ML) models – LightGBM, XGBoost, and CatBoost – and three deep learning (DL) models – Deep Neural Network (DNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM). Key input features include concrete strength, column length, cross-sectional dimensions, steel tube thickness, and yield strength, which were analysed to uncover underlying relationships. The results indicate that CatBoost delivers the highest predictive accuracy, achieving an RMSE of 396.50 kN and an R 2 of 0.932, surpassing XGBoost (RMSE: 449.57 kN, R 2: 0.906) and LightGBM (RMSE: 449.57 kN, R2: 0.916). Deep learning models were less effective, with the DNN attaining an RMSE of 496.19 kN and R 2 of 0.958, while the LSTM underperformed substantially (RMSE: 2010.46 kN, R 2: 0.891). SHapley Additive exPlanations (SHAP) identified cross-sectional width as the most critical feature, contributing positively to capacity, and column length as a significant negative influencer. A user-friendly, Python-based interface was also developed, enabling real-time predictions for practical engineering applications. Comparison with 10 analytical models demonstrates that these traditional methods, though deterministic, struggle to capture the nonlinear interactions inherent in CFST columns, thus yielding lower accuracy and higher variability. In contrast, the data-driven models presented here offer robust, adaptable, and interpretable solutions, underscoring their potential to transform design and analysis practices for CFST columns, ultimately fostering safer and more efficient structural systems.https://doi.org/10.1038/s41598-025-90648-2Concrete-filled steel tubular columnsAxial load predictionMachine learning modelsDeep learning architecturesStructural engineering applicationsSHAP Analysis
spellingShingle Feng Shen
Ishan Jha
Haytham F. Isleem
Walaa J.K. Almoghayer
Mohammad Khishe
Mohamed Kamel Elshaarawy
Advanced predictive machine and deep learning models for round-ended CFST column
Scientific Reports
Concrete-filled steel tubular columns
Axial load prediction
Machine learning models
Deep learning architectures
Structural engineering applications
SHAP Analysis
title Advanced predictive machine and deep learning models for round-ended CFST column
title_full Advanced predictive machine and deep learning models for round-ended CFST column
title_fullStr Advanced predictive machine and deep learning models for round-ended CFST column
title_full_unstemmed Advanced predictive machine and deep learning models for round-ended CFST column
title_short Advanced predictive machine and deep learning models for round-ended CFST column
title_sort advanced predictive machine and deep learning models for round ended cfst column
topic Concrete-filled steel tubular columns
Axial load prediction
Machine learning models
Deep learning architectures
Structural engineering applications
SHAP Analysis
url https://doi.org/10.1038/s41598-025-90648-2
work_keys_str_mv AT fengshen advancedpredictivemachineanddeeplearningmodelsforroundendedcfstcolumn
AT ishanjha advancedpredictivemachineanddeeplearningmodelsforroundendedcfstcolumn
AT haythamfisleem advancedpredictivemachineanddeeplearningmodelsforroundendedcfstcolumn
AT walaajkalmoghayer advancedpredictivemachineanddeeplearningmodelsforroundendedcfstcolumn
AT mohammadkhishe advancedpredictivemachineanddeeplearningmodelsforroundendedcfstcolumn
AT mohamedkamelelshaarawy advancedpredictivemachineanddeeplearningmodelsforroundendedcfstcolumn