摘要
对既有隧道的保护是城市地区深基坑施工的关注重点,土体、支护结构和隧道结构等多种因素之间的复杂相互作用,使得预测基坑开挖引起的隧道变形极为复杂和困难。首先利用神经网络学习土体、支护结构以及隧道之间复杂相互作用,形成基坑围护墙最大水平变形的预测模型。将围护墙最大水平位移输入基坑开挖诱发临近隧道变形的位移控制两阶段分析方法,得到基坑开挖诱发的隧道变形。在此基础上,结合开挖过程产生的数据,通过反演分析综合考虑施工过程不可控因素逐步减少土体参数不确定性,优化变形预测结果,实时预测后续隧道变形。通过数值模拟和现场测量,对所提出的预测模型的有效性进行了评估。结果表明,基于开挖过程产生的信息,将人工智能技术与基本预测模型相结合的策略可以有效的对基坑开挖诱发的隧道变形进行实时的高精度预测。
Potential damages to existing tunnels represent a major concern for constructing deep excavations in urban areas.The nonlinear interactions between soils,support structures,and tunnel structures make the prediction of the response of tunnel induced by adjacent excavations a rather difficult and complex task.Proposing an initiative to predict tunnel displacement by using process-based model,which including consisting artificial neural network(ANN)module,inverse modelling module and tunnel displacement module.First the ANN module is trained to learn and recognize the patterns of the complex interactions between soil,support constructions and tunnels.Based on back analysis method and measured data during construction processes to reduce the uncertainty associated with soil characterizations and optimize the prediction.With optimized data and displacement-controlled two-stage model,tunnel displacement could be predicted properly.The effectiveness of the proposed process-based model is evaluated against highfidelity numerical simulations and field measurements.These evaluations suggest that the strategy of combining artificial intelligence techniques with information generated during interaction processes can represent a promising approach to solve complex engineering problems in conventional industries.
作者
木林隆
林剑鸿
康兴宇
谷志旺
李帅
Mu Lin-long;Lin Jian-hong;Kang Xing-yu;Gu Zhi-wang;Li Shu-ai(Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education,Tongji University,Shanghai 200092,China;Department of Geotechnical Engineering,Tongji University,Shanghai 200092,China;Shanghai Construction No.4(Group)Co.,Ltd.,Shanghai 200080;China Machinery Industry Information Institute,Beijing 100037)
出处
《建筑科学》
CSCD
北大核心
2020年第S01期227-232,共6页
Building Science
关键词
基坑
隧道变形
神经网络
反演
智能预测
excavation
tunnel deformation
neural network
inverse analysis
intelligent prediction
作者简介
木林隆,[联系方式]E-mail:mulinlong@tongji.edu.cn
