土木与环境工程学报（中英文）2025，Vol.47Issue(5)：67-76,10.DOI:10.11835/j.issn.2096-6717.2024.063
盾构隧道壁后注浆智能检测的"云-边-端"架构及应用
Cloud-edge-end architecture and application for intelligent detection of backfill grouting in shield tunnels
摘要
Abstract
With the continuous development of urban rail transit,the demand and quantity of shield tunnel construction are constantly increasing.Backfill grouting behind the shield tunnel lining is a crucial technology for controlling engineering issues such as ground disturbance and segment misalignment,which is vital for ensuring the safety of shield construction and surface buildings.In order to achieve automation and intelligence in backfill grouting detection,and to address the low efficiency of traditional Ground Penetrating Radar(GPR)manual detection,a Loaded-to-Frame(LTF)equipment capable of rapid automated GPR detection of grouting quality is proposed.The hardware composition,operating mode,and main parameters of the LTF equipment are described in detail.A series of model tests conducted under various conditions yielded over 600 000 high-quality labeled A-scan data.Based on both A-scan and B-scan data types and training strategies,intelligent models with excellent performance for backfill grouting thickness recognition are developed.An intelligent detection method for backfill grouting based on a"cloud-edge-end"architecture is proposed,and the GPR-AI Master platform,based on partial"cloud-edge-end"architecture,is developed to achieve cloud deployment and rapid application of the artificial intelligence model.A dynamic feedback mechanism based on the LTF equipment and intelligent analysis results is proposed,realizing comprehensive monitoring throughout the tunnel excavation process.The application is derived from 16 distinct shield tunnel projects domestically and internationally validate the effectiveness of the intelligent detection method,providing valuable references for the secure and intelligent construction of shield tunnels.关键词
隧道工程/盾构隧道/壁后注浆/探地雷达/智能检测/云-边-端Key words
tunnel engineering/shield tunnel/backfill grouting/ground penetrating radar/intelligent detection/cloud-edge-end分类
交通工程引用本文复制引用
李康,谢雄耀,周彪,曾里,黄昌富..盾构隧道壁后注浆智能检测的"云-边-端"架构及应用[J].土木与环境工程学报（中英文）,2025,47(5):67-76,10.基金项目
国家重点研发计划(2023YFC3806705) （2023YFC3806705）
国家自然科学基金(52038008、52378408) （52038008、52378408）
上海市科委项目(22dz1203004) （22dz1203004）
国网上海市电力公司项目(52090W23000B)National Key R&D Program of China(No.2023YFC3806705) （52090W23000B）
National Natural Science Foundation of China(Nos.52038008,52378408) （Nos.52038008,52378408）
Science and Technology Innovation Plan of Shanghai Science and Technology Commission(No.22dz1203004) （No.22dz1203004）
State Grid Shanghai Municipal Electric Power Company(No.52090W23000B) （No.52090W23000B）
