Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning Framework

Bond behavior between steel bars and concrete is fundamental to the structural integrity and durability of reinforced concrete. However, corrosion-induced deterioration severely impairs bond performance, highlighting the need for advanced and reliable assessment methods. This paper pioneers an algor...

Full description

Saved in:
Bibliographic Details
Main Authors: Jin-Yang Gui, Zhao-Hui Lu, Chun-Qing Li
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Corrosion and Materials Degradation
Subjects:
Online Access:https://www.mdpi.com/2624-5558/6/2/24
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849432806738886656
author Jin-Yang Gui
Zhao-Hui Lu
Chun-Qing Li
author_facet Jin-Yang Gui
Zhao-Hui Lu
Chun-Qing Li
author_sort Jin-Yang Gui
collection DOAJ
description Bond behavior between steel bars and concrete is fundamental to the structural integrity and durability of reinforced concrete. However, corrosion-induced deterioration severely impairs bond performance, highlighting the need for advanced and reliable assessment methods. This paper pioneers an algorithm for an advanced ensemble learning framework to predict bond strength between corroded steel bars and concrete. In this framework, a novel Stacked Boosted Bond Model (SBBM) is developed, in which a Fusion-Based Feature Selection (FBFS) strategy is integrated to optimize input variables, and SHapley Additive exPlanations (SHAP) are employed to enhance interpretability. A merit of the framework is that it can effectively identify critical factors such as crack width, transverse confinement, and corrosion level, which have often been neglected by traditional models. The proposed SBBM achieves superior predictive performance, with a coefficient of determination (<i>R</i><sup>2</sup>) of 0.94 and a mean absolute error (MAE) of 1.33 MPa. Compared to traditional machine learning and analytical models, it demonstrates enhanced accuracy, generalization, and interpretability. This paper provides a reliable and transparent tool for structural performance evaluation, service life prediction, and the design of strengthening measures for corroded reinforced concrete structures, contributing to safer and more durable concrete structures.
format Article
id doaj-art-a2af77261b9b4c21b64bbaaec891f1b0
institution Kabale University
issn 2624-5558
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Corrosion and Materials Degradation
spelling doaj-art-a2af77261b9b4c21b64bbaaec891f1b02025-08-20T03:27:15ZengMDPI AGCorrosion and Materials Degradation2624-55582025-06-01622410.3390/cmd6020024Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning FrameworkJin-Yang Gui0Zhao-Hui Lu1Chun-Qing Li2Key Laboratory of Urban Security and Disaster Engineering of Ministry of Education, Beijing University of Technology, Beijing 100124, ChinaKey Laboratory of Urban Security and Disaster Engineering of Ministry of Education, Beijing University of Technology, Beijing 100124, ChinaKey Laboratory of Urban Security and Disaster Engineering of Ministry of Education, Beijing University of Technology, Beijing 100124, ChinaBond behavior between steel bars and concrete is fundamental to the structural integrity and durability of reinforced concrete. However, corrosion-induced deterioration severely impairs bond performance, highlighting the need for advanced and reliable assessment methods. This paper pioneers an algorithm for an advanced ensemble learning framework to predict bond strength between corroded steel bars and concrete. In this framework, a novel Stacked Boosted Bond Model (SBBM) is developed, in which a Fusion-Based Feature Selection (FBFS) strategy is integrated to optimize input variables, and SHapley Additive exPlanations (SHAP) are employed to enhance interpretability. A merit of the framework is that it can effectively identify critical factors such as crack width, transverse confinement, and corrosion level, which have often been neglected by traditional models. The proposed SBBM achieves superior predictive performance, with a coefficient of determination (<i>R</i><sup>2</sup>) of 0.94 and a mean absolute error (MAE) of 1.33 MPa. Compared to traditional machine learning and analytical models, it demonstrates enhanced accuracy, generalization, and interpretability. This paper provides a reliable and transparent tool for structural performance evaluation, service life prediction, and the design of strengthening measures for corroded reinforced concrete structures, contributing to safer and more durable concrete structures.https://www.mdpi.com/2624-5558/6/2/24bond strengthcorrosionmachine learningfeature selectionensemble learning framework
spellingShingle Jin-Yang Gui
Zhao-Hui Lu
Chun-Qing Li
Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning Framework
Corrosion and Materials Degradation
bond strength
corrosion
machine learning
feature selection
ensemble learning framework
title Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning Framework
title_full Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning Framework
title_fullStr Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning Framework
title_full_unstemmed Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning Framework
title_short Enhanced Prediction of Bond Strength in Corroded RC Structures Using Advanced Feature Selection and Ensemble Learning Framework
title_sort enhanced prediction of bond strength in corroded rc structures using advanced feature selection and ensemble learning framework
topic bond strength
corrosion
machine learning
feature selection
ensemble learning framework
url https://www.mdpi.com/2624-5558/6/2/24
work_keys_str_mv AT jinyanggui enhancedpredictionofbondstrengthincorrodedrcstructuresusingadvancedfeatureselectionandensemblelearningframework
AT zhaohuilu enhancedpredictionofbondstrengthincorrodedrcstructuresusingadvancedfeatureselectionandensemblelearningframework
AT chunqingli enhancedpredictionofbondstrengthincorrodedrcstructuresusingadvancedfeatureselectionandensemblelearningframework