Optimization of anchorage support parameters for soft rock tunnel based on displacement control theory

作     者：LI Gan MA Wei-bin YU Chang-yi TAO Zhi-gang WANG Feng-nian LI Gan;MA Wei-bin;YU Chang-yi;TAO Zhi-gang;WANG Feng-nian

作者机构：School of Civil&Environmental Engineering and Geography ScienceRock mechanics Research InstituteNingbo UniversityNingbo 315000China Railway Engineering Research InstituteChina Academy of Railway Sciences Corporation LimitedBeijing 100081China CCCC First Harbor Engineering CompanyCo.Ltd.Tianjin 300461 CCCC-Tianjin Port Engineering InstituteCo.Ltd.Tianjin 300222 State Key Laboratory for Geomechanics&Deep Underground EngineeringBeijing 100083China Shanxi Transportation Technology Research and Development Co.LTDTaiyuan 030032China 

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

年 卷 期：2023年第20卷第7期

页      面：2076-2092页

核心收录：

学科分类：0830[工学-环境科学与工程（可授工学、理学、农学学位）] 081406[工学-桥梁与隧道工程] 08[工学] 0814[工学-土木工程] 082301[工学-道路与铁道工程] 0823[工学-交通运输工程] 

基　　金：supported by the Open Fund of State Key Laboratory of High speed Railway Track Technology(2022YJ127-1) National Natural Science Foundation of China(52104125,41941018) the Natural Science Basic Research Plan in Shaanxi Province of China(2022JQ-304) the Young Elite Scientists Sponsorship Program by CAST(No.2021QNRC001) 

主　　题：Displacement control theory Anchorage support parameters Numerical simulation PSOLSSVM Tunnel construction 

摘      要：In the construction of a soft rock tunnel,it is critical to accurately estimate the pre-stressed anchor support parameters for surrounding rock reinforcement;otherwise,engineering disasters may *** paper presents a support parameter selection method that aims to allow deformation as a control objective,which was applied to the tunnel located in Muzailing Highway,Min County,Dingxi City,Gansu Province,*** theoretical analysis,we have identified five factors that influence pre-stressing *** selection of mechanical parameters for the rock mass was carried out using an inverse analysis *** with the measured data,the maximum displacement error of the numerical simulation results was only 0.07 *** length of anchor cable,circumferential spacing of anchor cable,longitudinal spacing,and pre-stress index are adopted as the input parameters for the support vector machine neural network model based on particle swarm optimization(PSO-LSSVM).Besides,the vault subsidence and the maximum deformation of surrounding rock are considered as output values(performance indices).The goodness of fit between the predicted values and the simulated values exceeds ***,all support parameters within the acceptable deformation range are *** optimal support variables are derived by considering the construction cost and *** field application results show that it is feasible to construct the sample database utilizing the numerical simulation approach by taking the displacement as the control target and using the neural network to specify the appropriate support parameters.