基于深度学习的基坑开挖引起地表位移时序预测

唐浩然; 胡垚; 雷华阳; 路军富; 刘婷; 王凯

doi:10.11779/CJGE2024S20014

基于深度学习的基坑开挖引起地表位移时序预测 English Version

唐浩然^{1, 2,},
胡垚^{1, 2, ,},
雷华阳³,
路军富^{1, 2},
刘婷⁴,
王凯⁵

1.
地质灾害防治与地质环境保护国家重点实验室(成都理工大学), 四川成都 610059
2.
成都理工大学环境与土木工程学院, 四川成都 610059
3.
天津大学建筑工程学院, 天津 300350
4.
中国电建集团西北勘测设计研究院有限公司, 陕西西安 710065
5.
中铁建设集团有限公司华东分公司, 江苏昆山 215300

基金项目:

国家自然科学基金项目 42307260

四川省自然科学基金项目 2023NSFSC0882

地质灾害防治与地质环境保护国家重点实验室开发基金 SKLGP2023K024

详细信息

作者简介:
唐浩然（1999—），男，硕士研究生，主要从事岩土与地下工程方面的研究。E-mail：tanghaoran@stu.cdut.edu.cn

通讯作者:
胡垚, E-mail: hyjiaoliu@163.com

中图分类号: TU43
计量
- 文章访问数: 165
- HTML全文浏览量: 22
- PDF下载量: 46
出版历程
- 收稿日期: 2024-06-21
- 刊出日期: 2024-09-30

Time series prediction of surface displacement induced by excavation of foundation pits based on deep learning

TANG Haoran^{1, 2,},
HU Yao^{1, 2, ,},
LEI Huayang³,
LU Junfu^{1, 2},
LIU Ting⁴,
WANG Kai⁵

1.
State Key Laboratory of Geological Disaster Prevention and Geological Environment Protection, Chengdu University of Technology, Chengdu 610059, China
2.
College of Environment and Civil Engineering, Chengdu University of Technology, Chengdu 610059, China
3.
School of Civil Engineering, Tianjin University, Tianjin 300350, China
4.
PowerChina Northwest Engineering Corporation Limited, Xi'an 710065, China
5.
East China Branch of China Railway Construction Group Co, Ltd., Kunshan 215300, China

摘要

摘要: 为更精准预测基坑工程中数据的时间特性，结合卷积神经网络CNN模型与两种单一时间序列神经网络模型长短期记忆网络LSTM模型、门控循环单元GRU模型，建立混合时间序列神经网络CNN-LSTM模型、CNN-GRU模型。基于杭州某邻近既有车站基坑开挖工程，采用滚动预测方法建立基坑开挖引起邻近地铁车站地表沉降数据集。通过平均绝对误差MAE、平均相对误差MAPE和均方根误差RMSE3种评价指标对预测结果进行评价。结果表明：CNN-GRU模型预测效果最优，CNN-LSTM模型次之，其次是GRU模型，最后是LSTM模型。CNN-LSTM混合网络模型相较于LSTM模型对3种评价指标分别降低了24.4%，53.8%，4.1%，CNN-GRU混合网络模型相较于GRU模型分别降低了13.9%，49.1%，1%。
- 基坑开挖 /
- 深度学习 /
- 卷积神经网络 /
- 长短期记忆网络 /
- 门控循环单元
Abstract: To predict the time characteristics of data more accurately in foundation pit engineering, two single time series neural network models are combined, the convolutional neural network (CNN) and long short-term memory network (LSTM), as well as the gated recurrent unit (GRU), to establish a hybrid time series neural network model CNN-LSTM and CNN-GRU. An excavation project of a foundation pit adjacent to an existing station in Hangzhou is selected, and a rolling prediction method is used to create a dataset of surface settlement caused by excavation of the foundation pit in the adjacent subway stations. The predicted results are evaluated by three evaluation indexes: mean absolute error (MAE), mean relative error (MAPE) and root mean square error (RMSE). The results demonstrate that the CNN-GRU has the best prediction effects, followed by the CNN-LSTM, GRU and LSTM. Compared with the LSTM model, the CNN-LSTM hybrid network model reduces the three evaluation indexes by 24.4%, 53.8% and 4.1%, respectively, and the CNN-GRU hybrid network model decreases by 13.9%, 49.1% and 1%, respectively, compared with the GRU model.
- excavation of foundation pit /
- deep learning /
- convolutional neural network /
- long short-term memory network /
- gated recurrent unit

HTML全文

图 1 CNN神经网络

Figure 1. CNN neural network

下载: 全尺寸图片幻灯片

图 2 LSTM单元结构图

Figure 2. Structure of LSTM cell

下载: 全尺寸图片幻灯片

图 3 GRU模型结构

Figure 3. Structure of GRU model

下载: 全尺寸图片幻灯片

图 4 CNN-LSTM和CNN-GRU结构

Figure 4. Structure of CNN-LSTM and CNN-GRU

下载: 全尺寸图片幻灯片

图 5 杭州某邻近既有隧道基坑开挖工程概况

Figure 5. Overview of project adjacent to excavation of existing tunnel in Hangzhou

下载: 全尺寸图片幻灯片

图 6 滚动预测法

Figure 6. Rolling forecasting method

下载: 全尺寸图片幻灯片

图 7 地表位移预测值对比

Figure 7. Comparison of predicted surface displacements

下载: 全尺寸图片幻灯片

图 8 地表位移预测精度对比

Figure 8. Comparison of prediction accuracies of surface displacement

下载: 全尺寸图片幻灯片

表 1 预测模型评价指标

Table 1 Evaluation indexes of prediction models

评判指标	LSTM	CNN-LSTM	优化效果/%	GRU	CNN-GRU	优化效果/%
MAE	0.1019	0.0819	24.4	0.0902	0.0782	13.9
MAPE	1.2779	0.5909	53.8	1.0970	0.5581	49.1
RMSE	0.1251	0.1199	4.1	0.1185	0.1174	1.0

下载: 导出CSV

参考文献(13)

[1]	LIU Bo, WU Wenwen, LIU Haipei, et al. Effect and control of foundation pit excavation on existing tunnels: a state-of-the-art review[J]. Tunnelling and Underground Space Technology, 2024, 147: 105704. doi: 10.1016/j.tust.2024.105704
[2]	王卫东. 软土深基坑变形及环境影响分析方法与控制技术[J]. 岩土工程学报, 2024, 46(1): 1-25. doi: 10.11779/CJGE20231146 WANG Weidong. Analytical methods and controlling techniques for deformation and environmental influence of deep excavations in soft soils[J]. Chinese Journal of Geotechnical Engineering, 2024, 46(1): 1-25. (in Chinese) doi: 10.11779/CJGE20231146
[3]	HU Y, LEI H Y, ZHENG G, et al. Assessing the deformation response of double-track overlapped tunnels using numerical simulation and field monitoring[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2022, 14(2): 436-447. doi: 10.1016/j.jrmge.2021.07.003
[4]	XU Q W, XIE J L, LU L H, et al. Numerical and theoretical analysis on soil arching effect of prefabricated piles as deep foundation pit supports[J]. Underground Space, 2024, 16: 314-330. doi: 10.1016/j.undsp.2023.09.011
[5]	胡之锋, 陈健, 邱岳峰, 等. 挡墙水平变位诱发地表沉降的显式解析解[J]. 岩土力学, 2018, 39(11): 4165-4175. HU Zhifeng, CHEN Jian, QIU Yuefeng, et al. Analytical formula for ground settlement induced by horizontal movement of retaining wall[J]. Rock and Soil Mechanics, 2018, 39(11): 4165-4175. (in Chinese)
[6]	薛艳杰. 基于机器学习算法的土岩复合地层深基坑变形时序预测[J]. 现代隧道技术, 2022, 59(增刊2): 77-85. XUE Yanjie. Deformation time series prediction of deep foundation pit in soil-rock composite stratum based on machine learning algorithm[J]. Modern Tunnelling Technology, 2022, 59(S2): 77-85. (in Chinese)
[7]	赵华菁, 张名扬, 刘维, 等. 基于神经网络算法的深基坑地连墙变形动态预测[J]. 地下空间与工程学报, 2021, 17(增刊1): 321-327. ZHAO Huajing, ZHANG Mingyang, LIU Wei, et al. Dynamic prediction of deformation of diaphragm wall in deep foundation pit based on neural network algorithm[J]. Chinese Journal of Underground Space and Engineering, 2021, 17(S1): 321-327. (in Chinese)
[8]	张生杰, 谭勇. 基于LSTM算法的基坑变形预测[J]. 隧道建设(中英文), 2022, 42(1): 113-120. ZHANG Shengjie, TAN Yong. Deformation prediction of foundation pit based on long short-term memory algorithm[J]. Tunnel Construction, 2022, 42(1): 113-120. (in Chinese)
[9]	徐长节, 李欣雨. 基于人工神经网络的深基坑支护结构侧移预测[J/OL]. 上海交通大学学报, 1-20. [2024-09-11]. https://doi.org/10.16183/j.cnki.jsjtu.2023.109. XU Changjie, LI Xinyu. Lateral Deformation Prediction of Deep Foundation Retaining Structures Based on Artificial Neural Network[J/OL]. Journal of Shanghai Jiao Tong University, 1-20[2024-09-11]. https://doi.org/10.16183/j.cnki.jsjtu.2023.109. (in Chinese)
[10]	方庆, 陈胜, 刘雪珠, 等. 基于变分模态分解的CNN-LSTM模型在基坑变形预测中的应用[J/OL]. 力学与实践, 1-8[2024-09-11]. http://kns.cnki.net/kcms/detail/11.2064.o3.20240314.2043.002.html. FANG Qin, CHEN Sheng, LIU Xuezhu, et al. Application of the variational mode decomposition-based CNN-LSTM model in predicting excavation deformation[J]. Mechanics in Engineering, 1-8[2024-09-11]. http://kns.cnki.net/kcms/detail/11.2064.o3.20240314.2043.002.html. (in Chinese)
[11]	满轲, 武立文, 刘晓丽, 等. 基于CNN-LSTM模型的TBM隧道掘进参数及岩爆等级预测[J/OL]. 煤炭科学技术, 1-19[2024-09-11]. http://kns.cnki.net/kcms/detail/11.2402.TD.20231026.1344.003.html. MAN Ke, WU Liwen, LIU Xiaoli, et al. The prediction of TBM tunnel boring parameters and rockburst grade based on CNN-LSTM model[J/OL]. Coal Science and Technology, 1-19[2024-09-11]. http://kns.cnki.net/kcms/detail/11.2402.TD.20231026.1344.003.html. (in Chinese)
[12]	王锋. 基于SSA-LSTM模型的软岩隧道变形特征智能预测及应用研究[J]. 现代隧道技术, 2024, 61(1): 56-66. WANG Feng. Study on intelligent prediction of the deformation characteristics of soft rock tunnel based on SSA-LSTM model and its application[J]. Modern Tunnelling Technology, 2024, 61(1): 56-66. (in Chinese)
[13]	洪宇超, 钱建固, 叶源新, 等. 基于时空关联特征的CNN-LSTM模型在基坑工程变形预测中的应用[J]. 岩土工程学报, 2021, 43(增刊2): 108-111. doi: 10.11779/CJGE2021S2026 HONG Yuchao, QIAN Jiangu, YE Yuanxin, et al. Application of CNN-LSTM model based on spatio-temporal correlation characteristics in deformation prediction of foundation pit engineering[J]. Chinese Journal of Geotechnical Engineering, 2021, 43(S2): 108-111. (in Chinese) doi: 10.11779/CJGE2021S2026

施引文献(28)

期刊类型引用(12)

1.	李贝贝. 强动压巷道底板变形破坏特征及控制技术研究. 山东煤炭科技. 2025(01): 1-4+11 . 百度学术
2.	毕智强，于振亚，张涛. 深井动压软岩巷道底鼓变形破坏机理与防治技术. 陕西煤炭. 2025(04): 84-88+113 . 百度学术
3.	伊丙鼎. 煤层水仓围岩变形破坏特征及控制技术研究. 煤炭工程. 2024(03): 117-123 . 百度学术
4.	肖同强，王泽源，刘发义，代晓亮，赵帅，余子豪. 深部强动压巷道底鼓控制机理及技术研究. 采矿与安全工程学报. 2024(04): 666-676 . 百度学术
5.	丁维波，王丹影，闫医慧. 回采巷道底鼓机理及防治技术研究. 煤炭技术. 2024(08): 23-27 . 百度学术
6.	朱国保，董垠枫. 非均质围岩隧道断面的施工关键技术研究. 四川建筑. 2024(05): 200-201 . 百度学术
7.	耿铭，孙静. 厚硬顶板悬顶致灾机理及切顶控制技术研究. 工矿自动化. 2024(11): 132-141 . 百度学术
8.	柴敬，韩志成，雷武林，张丁丁，马晨阳，孙凯，翁明月，张有志，丁国利，郑忠友，张寅，韩刚. 回采巷道底鼓演化过程的分布式光纤实测研究. 煤炭科学技术. 2023(01): 146-156 . 百度学术
9.	吕情绪，曹军，高亮. 重复采动回采巷道变形机理及稳定控制. 中国矿业. 2023(05): 96-103 . 百度学术
10.	丁自伟，王少轩，王庆阳，王耀声，王春斌，李军岐，邸广强，李亮. 软岩巷道底鼓机理及其稳定性控制研究. 煤炭工程. 2023(07): 102-109 . 百度学术
11.	吕志强，黄亚军，景明，徐啸川. 深部软岩巷道变形破坏机理及支护技术. 能源与环保. 2023(11): 280-286 . 百度学术
12.	杨晓炜. 深部软岩大断面巷道大变形控制技术研究. 能源与环保. 2023(11): 312-318 . 百度学术