Mechanical properties of steel mesh in anchor-mesh support for rocky tunnels

作     者：SUN Keguo JIA Jinglong XU Weiping ZHANG Yu WANG Jinjin WANG Yichao LIU Yongkui 

作者机构：Key Laboratory of Transportation Tunnel EngineeringMinistry of EducationSchool of Civil EngineeringSouthwest Jiaotong UniversityChengdu 610031China State Key Laboratory of Intelligent Geotechnics and TunnellingChengdu 610031China Chengdu Rail Transit Group Co.Ltd.Chengdu 610036China China Railway 14th Bureau Group 2nd Engineering Corporation LimitedChina Railway Construction Corporation LimitedTaian 271000China 

出 版 物：《Journal of Mountain Science》 (山地科学学报（英文）)

年 卷 期：2024年第21卷第10期

页      面：3487-3502页

核心收录：

学科分类：08[工学] 080502[工学-材料学] 0805[工学-材料科学与工程（可授工学、理学学位）] 

基　　金：funded by the National Natural Science Foundation of China(Grant No.52178396) 

主　　题：Tunnel Steel mesh BP neural network Anchor-mesh support Rock reinforcement technique 

摘      要：Underground geotechnical engineering encounters persistent challenges in ensuring the stability and safety of surrounding rock structures, particularly within rocky tunnels. Rock reinforcement techniques, including the use of steel mesh, are critical to achieving this goal. However, there exists a knowledge gap regarding the comprehensive understanding of the mechanical behavior and failure mechanisms exhibited by steel mesh under diverse loading conditions. This study thoroughly explored the steel mesh s performance throughout the entire loading-failure process, innovating with detailed analysis and modeling techniques. By integrating advanced numerical modeling with laboratory experiments, the study examines the influence of varying reinforcement levels and geometric parameters on the steel mesh strength and deformation characteristics. Sensitivity analysis, employing gray correlation theory, identifies the key factors affecting the mesh performance, while a BP (Backpropagation) neural network model predicts maximum vertical deformation with high accuracy. The findings underscore the critical role of steel diameter and mesh spacing in optimizing peak load capacity, displacement, and energy absorption, offering practical guidelines for design improvements. The use of a Bayesian Regularization (BR) algorithm further enhances the predictive accuracy compared to traditional methods. This research provides new insights into optimizing steel mesh design for underground applications, offering an innovative approach to enhancing structural safety in geotechnical projects.