The Application of Machine Learning Algorithms to Bond Strength between Steel Rebars and Concrete Using Bayesian Optimization DOI Open Access
Huajun Yan,

Nan Xie,

Dandan Shen

et al.

Materials, Journal Year: 2024, Volume and Issue: 17(18), P. 4641 - 4641

Published: Sept. 21, 2024

The purpose of this study is to estimate the bond strength between steel rebars and concrete using machine learning (ML) algorithms with Bayesian optimization (BO). It important conduct beam tests determine since it affected by stress fields. A approach for based on 401 six impact factors presented in paper. model composed three standard algorithms, including random forest (RF), support vector regression (SVR), extreme gradient boosting (XGBoost), combined BO technique. Compared empirical models, BO-XGB`oost was found be most accurate method, values R2, MAE, RMSE 0.87, 0.897 MPa, 1.516 MPa test set. development a simplified that contains input variables (diameter rebar, yield reinforcement, compressive strength) has been proposed make more convenient apply. According prediction, Shapley additive explanation (SHAP) can help explain why ML-based predicts particular outcome does. By utilizing predict complex interfacial mechanical behavior, possible improve accuracy model.

Language: Английский

Research on the Influence of Recycled Fine Powder on Chloride Ion Erosion of Concrete in Different Chloride Salt Environments DOI Open Access
Lijun Chen, Gang Zhao, Ying Li

et al.

Materials, Journal Year: 2025, Volume and Issue: 18(9), P. 2018 - 2018

Published: April 29, 2025

The Qinghai-Tibet Plateau features a high-altitude, cold, and arid climate, with harsh environmental conditions. It is also one of the regions in China where chloride-rich salt lakes are abundant. These circumstances pose significant challenges to durability concrete. This study explored impact recycled fine powders (RFP) on resistance concrete chloride ion erosion. To evaluate this, 3.5% sodium solution Qarhan Salt Lake brine were employed as erosion media. depth concentration penetration, free diffusion coefficient (Df), microstructure measured. results demonstrated that when replacement rate RFP was 20%, displayed excellent both brine. XRD analysis SEM images revealed addition enabled bind more Cl- form Friedel's salt, which filled pores reduced within Moreover, soaking time extended continuously, damage effects severe than those solution.

Language: Английский

Citations

0
Machine learning in concrete durability: challenges and pathways identified by RILEM TC 315-DCS towards enhanced predictive models DOI Creative Commons
Woubishet Zewdu Taffese, Benoît Hilloulin, Yury Villagrán Zaccardi

et al.

Materials and Structures, Journal Year: 2025, Volume and Issue: 58(4)

Published: May 1, 2025

Language: Английский

Citations

0
Artificial intelligence in the design, optimization, and performance prediction of concrete materials: a comprehensive review DOI Creative Commons
Dayou Luo,

Kejin Wang,

Dongming Wang

et al.

npj Materials Sustainability, Journal Year: 2025, Volume and Issue: 3(1)

Published: May 17, 2025

Language: Английский

Citations

0
The Application of Machine Learning Algorithms to Bond Strength between Steel Rebars and Concrete Using Bayesian Optimization DOI Open Access
Huajun Yan,

Nan Xie,

Dandan Shen

et al.

Materials, Journal Year: 2024, Volume and Issue: 17(18), P. 4641 - 4641

Published: Sept. 21, 2024

The purpose of this study is to estimate the bond strength between steel rebars and concrete using machine learning (ML) algorithms with Bayesian optimization (BO). It important conduct beam tests determine since it affected by stress fields. A approach for based on 401 six impact factors presented in paper. model composed three standard algorithms, including random forest (RF), support vector regression (SVR), extreme gradient boosting (XGBoost), combined BO technique. Compared empirical models, BO-XGB`oost was found be most accurate method, values R2, MAE, RMSE 0.87, 0.897 MPa, 1.516 MPa test set. development a simplified that contains input variables (diameter rebar, yield reinforcement, compressive strength) has been proposed make more convenient apply. According prediction, Shapley additive explanation (SHAP) can help explain why ML-based predicts particular outcome does. By utilizing predict complex interfacial mechanical behavior, possible improve accuracy model.

Language: Английский

Citations

1