围岩智能化分级及其工程应用  

作　　者：钟庆华 孙鑫 张铁军 乔洪磊 杨新安[2] 

机构地区：[1]浙江杭温铁路公司,杭州311200 [2]同济大学交通运输工程学院,上海201804

出　　处：《科技创新与应用》2024年第12期77-80,89,共5页Technology Innovation and Application

基　　金：浙江省交通运输厅科研计划项目(20211050)。

摘　　要：由于地质条件的复杂多变、勘察方法和技术缺陷等原因,勘察阶段的山岭隧道围岩分级与施工阶段存在差异。为动态获得施工中杭温铁路隧道前方围岩等级,该文基于岩体基本质量指标(BQ)的综合评判法,利用现场回弹试验、超前地质预报及掌子面图像识别等手段,通过人工智能算法获得前方揭露围岩岩石强度及岩体完整性系数;再结合隧道现场水文情况、软弱结构面、初始地应力状态等修正因素,对围岩进行精细分级。现已在杭温铁路隧道古塘源二号隧道进口DK90+292、金竹坪隧道出口DK86+184、木匪岭隧道出口DK84+565及松坞尖隧道进口DK86+752等断面围岩等级变更工作中实现成功应用,得到与专业团队相吻合的揭露围岩级别判别结果,具有进一步推广与研究价值。Due to the complex and changeable geological conditions and the defects of survey methods and techniques,there is a difference between the surrounding rock classification of mountain tunnel in the investigation stage and the construction stage.In order to dynamically obtain the grade of surrounding rock in front of the tunnel of Hangzhou-Wenzhou Railway under construction,based on the comprehensive evaluation method of rock mass basic quality(BQ),by means of on-site rebound test,advanced geological prediction and palm face image recognition,the rock strength and rock mass integrity coefficient of front exposed surrounding rock are obtained by artificial intelligence algorithm.Based on the correction factors such as hydrological condition,weak structural plane and initial ground stress state of the tunnel,the surrounding rock is classified finely.It has been successfully applied in the surrounding rock grade change of Gutangyuan No.2 tunnel entrance DK90+292,Jinzhuping tunnel exit DK86+184,Mugangling tunnel exit DK84+565 and Songwujian tunnel entrance DK86+752 of Hangzhou-Wenzhou Railway tunnel.The results are consistent with the professional team to reveal the surrounding rock grade,which is worth further popularizing and studying.

关 键 词：山岭隧道 人工智能 图像识别 围岩精细分级 围岩等级变更 

分 类 号：U25[交通运输工程—道路与铁道工程]