Journal Published Online: 15 October 2025
Volume , Issue

Experimental Study and Prediction of the Elastic Modulus of Spontaneous-Combustion Coal Gangue Aggregate Concrete

CODEN: JTEVAB

Abstract

Coal gangue, a byproduct of coal mining, constitutes a significant source of industrial solid waste and poses environmental challenges, particularly due to its tendency for spontaneous combustion. Although its use as spontaneous-combustion coal gangue aggregates (SCGAs) in concrete mitigates environmental risks and supports sustainable construction, the distinct physicochemical properties of SCGAs substantially influence the mechanical behavior of concrete, especially its elastic modulus. Despite prior research emphasizing the impact of SCGA replacement ratios on concrete properties, the underlying mechanisms and predictive capabilities remain inadequately explored. To address this gap, this study investigates how varying SCGA replacement ratios affect the elastic modulus of SCGA concrete (SCGAC), incorporating experimental analysis and machine learning (ML) techniques. Five sets of concrete samples with different replacement ratios were prepared to evaluate their elastic modulus at various curing ages. The microstructure of the concrete interfacial transition zone was observed to reveal the deterioration mechanism of the elastic modulus. Eight algorithms were used to develop models for predicting the SCGAC’s elastic modulus, whereas the SHapley Additive exPlanations methodology was used for feature importance analysis. The findings indicate that an increase in the aggregates replacement ratio leads to a gradual reduction in the SCGAC’s elastic modulus. Compared with natural aggregate concrete (NAC), the 28-day elastic modulus for 50 % and 100 % replacement ratios decreased by 28.6 % and 31.6 %, respectively, whereas at 90 days, the decreases were 20.0 % and 30.8 %. The proposed ML models effectively predict the specimens’ elastic modulus, with the XGBoost model exhibiting the best performance, achieving a comprehensive performance index of 0.20. The primary factor influencing the concrete’s elastic modulus is the replacement ratio of SCGA, which shows a negative correlation with the elastic modulus. The established XGBoost model is capable of predicting the SCGAC’s elastic modulus.

Author Information

Zhang, Tirui
School of Civil Engineering, Shenyang Jianzhu University, Shenyang, Liaoning, China
Wang, Qinghe
School of Civil Engineering, Shenyang Jianzhu University, Shenyang, Liaoning, China
Zhao, Zihao
School of Civil Engineering, Shenyang Jianzhu University, Shenyang, Liaoning, China
Fang, Yanfeng
School of Civil Engineering, Shenyang Jianzhu University, Shenyang, Liaoning, China
Zhang, Yuzhuo
School of Civil Engineering, Shenyang Jianzhu University, Shenyang, Liaoning, China
Pages: 14
Price: $25.00
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Stock #: JTE20240690
ISSN: 0090-3973
DOI: 10.1520/JTE20240690