基于高斯Copula贝叶斯模型的盾构下穿既有隧道施工风险的分析

吴忠坦

doi:10.13204/j.gyjzG22103112

基于高斯Copula贝叶斯模型的盾构下穿既有隧道施工风险的分析

doi: 10.13204/j.gyjzG22103112

吴忠坦

武汉地铁集团有限公司, 武汉 430030

详细信息

作者简介:
吴忠坦,男,1983年出生,工程师。电子信箱:870852150@qq.com。

计量
- 文章访问数: 38
- HTML全文浏览量: 7
- PDF下载量: 2
- 被引次数: 0
出版历程
- 收稿日期: 2022-10-31

Risk Analysis of Construction with Tunnel Boring Machines Passing Under Existing Tunnels Based on Gaussian Copula Bayesian Network Model

WU Zhongtan

Wuhan Metro Group Co., Ltd., Wuhan 430030, China

摘要

摘要: 为对盾构下穿既有隧道施工工程安全风险进行分析和管控,提出一种基于高斯Copula贝叶斯(GCBN)模型的盾构下穿既有隧道施工风险分析方法。基于故障树建立了一套包括12个因素的施工安全风险指标体系,将贝叶斯网络的动态推理诊断与Copula理论的依赖性表达相结合,在不确定和不完全信息下构建盾构下穿既有隧道施工风险分析的GCBN模型。以武汉轨道交通12号线下穿既有7号线工程为例,利用高斯Copula识别各因素边际分布类型,计算各因素间相关系数并连接网络中结点。通过模型推理进行定性和定量分析,识别盾构下穿施工安全风险状态,分析各致险因素对风险结果的影响。最后,对敏感性高的因素采取措施进行防控。通过防控前、后模型计算结果对比,实现盾构下穿施工过程实时动态安全预警管控。实践表明:GCBN模型预测结果与专家评价结果吻合,验证了所构建GCBN风险分析模型的可靠性。
- 盾构施工 /
- 下穿工程 /
- 施工安全风险 /
- GCBN模型 /
- 敏感性 /
- 预警管控
Abstract: To analyze, manage and control safety risks of construction with tunnel boring machines passing under existing tunnels, a risk analysis method based on the Gaussian Copula Bayesian Network (GCBN) model was proposed. Based on the fault tree, a set of the construction safety risk index system including 12 factors was established, and the dynamic inferential diagnosis of Bayesian Network was combined with the dependence expression of Copula Theory. The GCBN model for risk analysis of construction with tunnel boring machines passing under existing tunnels was constructed in the condition of uncertain and incomplete information. Taking the project of Wuhan Rail Transit Line 12 passing under the existing Line 7 as an example, the Gaussian Copula was used to identify marginal distribution types of each factor, calculate correlation coefficients between each factor, and connect nodes in the network. Through the model reasoning, qualitative analysis and quantitative analysis were conducted, the safety state under construction with tunnel boring machines passing under existing tunnels could be identified, the impact the risk factors on results of risks was analyzed. Eventually, preventive and control measures for high sensitivity factors were implemented. By comparison with the calculating results of models before and after being prevented and controlled, the real-time dynamic security warning and control for construction with tummel boring machines passing under existing tunnels were achieved. The application results indicated that the prediction results by the GCBN model were consistent with the evaluation results by experts, which verified the reliability of the established risk analysis model of GCBN.
- tunneling with tunnel boring machines /
- tunneling passing under existing tunnel /
- construction safety risk /
- Gaussian Copula Bayesian Network model /
- sensitivity factor /
- warning control

HTML全文

参考文献(24)

[1]	ZHAN G D M, HUANG Z K, LI Z L, et al. Analytical solution for the response of an existing tunnel to a new tunnel excavation underneath[J]. Computers and Geotechnics, 2019, 108:197-211.
[2]	陶宇,梁伟桥,方五军.盾构下穿近邻既有隧道稳定性研究[J].工业建筑, 2020, 50(4):66-70.
[3]	YIN M, JIANG H, JIANG Y, et al. Effect of the excavation clearance of an under-crossing shield tunnel on existing shield tunnels[J]. Tunnelling and Underground Space Technology, 2018, 78:245-258.
[4]	黄向阳,章邦超,梁基冠,等.上软下硬地层盾构下穿施工中既有隧道应力和变形研究[J].建筑安全, 2022, 37(7):4-8.
[5]	MA S K, SHAO Y, LIU Y, et al. Responses of pipeline to side-by-side twin tunnelling at different depths:3D centrifuge tests and numerical modelling[J]. Tunnelling and Underground Space Technology, 2017, 66:157-173.
[6]	阿卜杜拉,雷春明,田雨,等.盾构下穿对既有隧道影响的模型试验研究[J].地下空间与工程学报, 2020, 16(增刊2):540-544.
[7]	ZHOU Z, GOH Y M, LI Q. Overview and analysis of safety management studies in the construction industry[J]. Safety Science, 2015, 72:337-386.
[8]	曾铁梅,刘茜,冯宗宝,等.基于PCBN模型盾构下穿既有隧道施工安全风险评价[J].隧道建设(中英文), 2021, 41(10):1692-1699.
[9]	PAN Y, ZHANG L, KOH J, et al. An adaptive decision making method with copula Bayesian network for location selection[J]. Information Sciences, 2021(544):56-77.
[10]	王双成,高瑞,杜瑞杰.基于高斯Copula的约束贝叶斯网络分类器研究[J].计算机学报, 2016, 39(8):1612-1636.
[11]	JUN H B, KIM D. A Bayesian network-based approach for fault analysis[J]. Expert Systems with Applications, 2017, 81:332-348.
[12]	周圆媛,何一韬,赵璟璐,等.基于贝叶斯网络的岩溶区盾构隧道施工安全性分析[J].现代隧道技术, 2018, 55(增刊2):764-771.
[13]	任超,吴伟,黄征凯,等.基于AIC准则的RBF神经网络在GPS高程拟合中的应用[J].测绘科学, 2013, 38(2):77-79.
[14]	刘文,刘文黎,翟世鸿.基于Copula相依模型的地铁结构安全可靠性分析[J].中国安全科学学报, 2019, 29(8):164-171.
[15]	勾红叶,冷丹,王涵玉,等.基于混合Copula函数的风雨联合概率分布模型[J].中国公路学报, 2021, 34(2):309-316.
[16]	GUO X, JI J, KHAN F, et al. Fuzzy Bayesian network based on an improved similarity aggregation method for risk assessment of storage tank accident[J]. Process Safety and Environmental Protection, 2021(149):817-846.
[17]	花玲玲,郑伟.基于复杂网络理论的铁路事故致因分析[J].中国安全科学学报, 2019, 29(增刊1):114-122.
[18]	刘文黎,吴贤国,张文静,等.地铁健康监测PCBN模型的参数相关性分析[J].土木与环境工程学报(中英文), 2019, 41(2):45-52.
[19]	BALRAM D, LIAN K Y, SEBASTIAN N. Air quality warning system based on a localized PM2.5 soft sensor using a novel approach of Bayesian regularized neural network via forward feature selection[J/OL]. Ecotoxicology and Environmental Safety, 2019, 182[2022-10-31]. https://doi.org/10.1016/j.ecoenv.2019.109386.
[20]	樊学平,杨光红,肖青凯,等.大跨桥梁主梁失效概率分析的最优R-Vine Copula[J].吉林大学学报(工学版), 2021, 51(4):1296-1305.
[21]	LIU W, WU X, ZHANG L, et al. Sensitivity analysis of structural health risk in operational tunnels[J]. Automation in Construction, 2018, 94:135-153.
[22]	LIANG R, XIA T, HUANG M, et al. Simplified analytical method for evaluating the effects of adjacent excavation on shield tunnel considering the shearing effect[J]. Computers and Geotechnics, 2017(81):167-187.
[23]	LIN X T, CHEN R P, WU H N, et al. Deformation behaviors of existing tunnels caused by shield tunneling undercrossing with oblique angle[J]. Tunnelling and Underground Space Technology, 2019(89):78-90.
[24]	LIU X, FANG Q, ZHANG D, et al. Behaviour of existing tunnel due to new tunnel construction below[J]. Computers and Geotechnics, 2019, 110:71-81.