基于机器学习的大盾构管片上浮缺失值填补方法

陈少林; 靳军伟; 李新潮; 李明宇; 杨钊

doi:10.20174/j.JUSE.2026.01.24

地下空间与工程学报 >

2026 , Vol. 22 >Issue 1: 230 - 238

DOI: https://doi.org/10.20174/j.JUSE.2026.01.24

设计、施工、监测

基于机器学习的大盾构管片上浮缺失值填补方法

陈少林 ,
靳军伟 ,
李新潮 ,
李明宇 ,
杨钊

展开

1.中交二航局(成都)建设工程有限公司,成都 610218;
2.郑州大学土木工程学院,郑州 450001;
3.龙湖集团控股有限公司, 北京 100012

陈少林(1987—),男,湖南临湘人,硕士,高级工程师,主要从事隧道工程方面的研究。E-mail: 87354032@qq.com

靳军伟(1986—),男,河南林州人,博士,讲师,主要从事隧道工程智能建造方面的研究。 E-mail:jinjunwe@zzu.edu.cn

收稿日期: 2025-05-15

网络出版日期: 2026-03-03

基金资助

河南省科技攻关项目(232102241011);河南省重点研发专项(231111322100)

收起

Machine Learning-Based Methods for Filling the Missing Values of Large Shield Segment Uplift

Chen Shaolin ,
Jin Junwei ,
Li Xinchao ,
Li Mingyu ,
Yang Zhao

Expand

1. China Communications Second Aviation Bureau (Chengdu) Construction Engineering Co., Ltd., Chengdu 610218, P.R. China;
2. School of Civil Engineering, Zhengzhou University, Zhengzhou 450001, P.R. China;
3. Longfor Group Holdings Co., Ltd., Beijing 100012, P.R. China

Received date: 2025-05-15

Online published: 2026-03-03

Fold

摘要

隧道施工期内管片上浮监测环境复杂,容易导致大量的监测数据缺失,影响管片结构受力及其他相关研究工作的开展。依托济南黄河隧道工程,基于机器学习方法建立管片上浮量缺失值填补方法。采用R-reliefF算法和主成分分析法对实测管片上浮量、盾构机掘进参数、地层参数等组成的数据集进行特征分析,分别使用随机森林算法和XGBoost算法对处理过的数据集展开机器学习训练和预测,在此基础上建立基于上述两种机器学习方法的管片上浮量缺失值填补方法。研究表明:本文算法对管片上浮的每一个阶段都能进行较好的缺失值填补工作;与随机森林算法相比基于XGBoost算法的缺失值填补方法能够更准确;同时,在实测缺失数据达到40%的情况下,本文方法所填补数据仍与实际接近,填补效果良好。研究成果对实际工程中的管片上浮测量及管片相关研究具有重要的借鉴价值。

关键词： 大直径隧道; 盾构隧道; 管片上浮; 机器学习; 缺失值填补; XGBoost算法

本文引用格式

陈少林 , 靳军伟 , 李新潮 , 李明宇 , 杨钊 . 基于机器学习的大盾构管片上浮缺失值填补方法[J]. 地下空间与工程学报, 2026 , 22(1) : 230 -238 . DOI: 10.20174/j.JUSE.2026.01.24

Abstract

The monitoring environment for segment uplift during tunnel construction is highly complex, often resulting in significant data loss. This issue adversely impacts research on the stress development of segment structures and related studies. Based on the Jinan Yellow River tunnel project, a method to fill the missing value of segment uplift was established based on machine learning. R-reliefF algorithm and principal component analysis method were used to carry out feature engineering on the dataset composed of measured segment uplift, shield machine driving parameters, formation parameters, etc. Random Forest algorithm and XGBoost algorithm were used to carry out machine learning training and prediction on the processed dataset, respectively. The missing value filling method is established based on the two machine learning methods above. The results show that: The proposed algorithm could fill the missing value in each stage of the segment uplift. Compared with the Random Forest algorithm, the missing value filling method based on the XGBoost algorithm is more accurate. At the same time, even if the measured missing data reaches 40%, the data filled by this method is still close to reality, and the filling effect is good. The results of this project have important reference values for the measurement of segment uplift and related research of segments in practical engineering.

Key words： large diameter tunnel; shield tunnel; segment uplift; machine learning; missing value filling; XGBoost algorithm

参考文献

[1] 王凯, 曾垂刚, 陈馈, 等. 水下隧道复合地层超大直径泥水盾构刀盘刀具设计及应用实例[J]. 铁道标准设计, 2024, 68 (11): 141-148. (Wang Kai, Zeng Chuigang, Chen Kui, et al. Design and application of cutterhead and cutting tool for super-large diameter slurry shields in composite strata of underwater tunnel [J]. Railway Standard Design, 2024, 68 (11): 141-148. (in Chinese))
[2] 陈林,秦银平,张文新,等. 超大直径泥水盾构水中接收关键技术研究—以汕头海湾隧道为例[J]. 铁道标准设计, 2024, 68 (8): 130-136. (Chen Lin, Qin Yinping, Zhang Wenxin, et al. Key technologies for underwater receiving of super large diameter slurry shield - taking shantou bay tunnel as an example [J]. Railway Standard Design, 2024, 68 (8): 130-136. (in Chinese))
[3] 张建勇, 李明宇, 陈健,等.基于双面弹性地基梁的大直径盾构隧道管片上浮预测方法[J].现代隧道技术,2023,60(2):159-167. (Zhang Jianyong, Li Mingyu, Chen Jian et al. Prediction methods for segment uplift in large-diameter shield tunnels based on double elastic foundation beams [J]. Modern Tunnelling Technology, 2023, 60 (2): 159-167. (in Chinese))
[4] 田应飞, 李明宇, 陈健, 等.大直径盾构隧道斜螺栓-凹凸榫环缝剪切特性足尺试验研究[J].铁道标准设计,2022,66(11):135-141. (Tian Yingfei, Li Mingyu, Chen Jian, et al. Full-scale experimental study on shear characteristics of oblique bolt and tenon circumferential joint of large diameter shield tunnel [J]. Railway Standard Design, 2022, 66 (11): 135-141. (in Chinese))
[5] 李明宇, 余刘成, 陈健, 等.粉质黏土中大直径泥水盾构隧道管片上浮及错台现场测试分析[J].铁道科学与工程学报,2022,19(6):1705-1715. (Li Mingyu, Yu Liucheng, Chen Jian, et al. In situ test analysis of segment uplift and dislocation of large-diameter slurry shield tunnel in silty clay [J]. Journal of Railway Science and Engineering, 2022, 19(6): 1705-1715. (in Chinese))
[6] 叶俊能,班勇婷,叶肖伟,等.盾构隧道施工期管片上浮计算方法研究[J].地下空间与工程学报,2025,21(3):816-823.(Ye Junneng,Ban Yongting,Ye Xiaowei,et al.Research on calculation method of segment floating during construction of shield tunnel[J].Chinese Journal of Underground Space and Engineering,2025,21(3):816-823.(in Chinese))
[7] 张小颖, 李继波, 黄业胜, 等.基于机器学习的电力隧道施工工法预测[J].建筑技术,2023,54(6):657-660. (Zhang Xiaoying, Li Jibo, Huang Yesheng, et al. Prediction of power tunnel excavation method based on machine learning [J]. Architecture Technology, 2023, 54 (6): 657-660. (in Chinese))
[8] 乔国华. 盾构掘进中地层特征的实时识别方法 [J]. 地下空间与工程学报, 2023, 19 (6): 2039-2044,2071. (Qiao Guohua. Real-time identification method of stratum characteristics during shield tunnelling [J]. Chinese Journal of Underground Space and Engineering, 2023, 19 (6): 2039-2044,2071. (in Chinese))
[9] 王湘怡, 周小雄, 卢建炜,等.基于机器学习的TBM隧道掘进岩爆预测[J].施工技术(中英文),2022,51(20):1-7. (Wang Xiangyi, Zhou Xiaoxiong, Lu Jianwei, et al. Rockburst prediction of TBM tunneling based on machine learning [J]. Construction Technology (Chinese sand English), 2022, 51 (20): 1-7. (in Chinese))
[10] 魏力峰, 贾思桢, 朱牧原, 等.泥水盾构施工参数相关性及预测策略优化研究 [J]. 地下空间与工程学报, 2022, 18 (6): 1996-2004. (Wei Lifeng, Jia Sizhen, Zhu Muyuan, et al. Study on the correlation and prediction strategy optimization of construction parameters of slurry balanced shield [J] Chinese Journal of Underground Space and Engineering, 2022, 18 (6): 1996-2004. (in Chinese))
[11] 陶怡汐, 牛彦敏, 刘馨媛.认知诊断中缺失数据的插补方法比较研究[J].安阳工学院学报,2023,22(4):61-67, 84. (Tao Yixi, Niu Yanmin, Liu Xinyuan. Comparison of imputation methods for missing data in cognitive diagnosis [J]. Journal of Anyang Institute of Technology, 2023, 22 (4): 61-67, 84. (in Chinese))
[12] 邓明星,欧阳含笑,钱枫,等. 基于改进LSTM的重型柴油车远程监测NOx浓度缺失数据填补[J]. 环境科学学报, 2023, 43 (11): 245-257. (Deng Mingxing, OuYang Hanxiao, Qian Feng, et al. Filling in missing nox concentration data for remote monitoring of heavy-duty diesel vehicles based on improved LSTM [J]. Acta Scientiae Circumstantiae, 2023, 43 (11): 245-257. (in Chinese))
[13] 马思远, 焦佳辉, 任晟岐,等.基于注意力机制的城市多元空气质量数据缺失值填充[J].计算机工程与科学,2023,45(8):1354-1364. (Ma Siyuan, Jiao Jiahui, Ren Shengqi, et al. Missing value filling for multi-variable urban air quality data based on attention mechanism [J]. Computer Engineering and Science, 2023, 45 (8): 1354-1364. (in Chinese))
[14] 房旭.改进注意力机制方法对能源系统缺失值插补的研究[J].计算机时代,2023(7):11-14. (Fang Xu. Research on the missing value imputation of energy system by improved attention mechanism [J]. Computer Era, 2023 (7): 11-14. (in Chinese))
[15] 陈健, 靳军伟, 李新潮,等.基于XGBoost算法的大直径穿黄隧道施工期管片上浮研究[J]. 隧道建设, 2023, 43(增1): 72-80. (Chen Jian, Jin Junwei, Li Xinchao, et al. Segment uplift of large-diameter tunnel crossing yellow river during construction based on xgboost algorithm[J]. Tunnel Construction, 2023, 43(Supp.1): 72-80. (in Chinese))
[16] 游凤, 李代伟, 张海清, 等.基于归一化KNNI的随机森林填补算法[J].成都信息工程大学学报,2021,36(1):32-40. (You Feng, Li Daiwei, Zhang Haiqing, et al. A random forest approach for missing data imputation based on normalized KNNI [J]. Journal of Chengdu University of Information Technology, 2021, 36 (1): 32-40. (in Chinese))
[17] Chen T, Guestrin C. XGBoost: A scalable tree boosting system[A] // The 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining[C]. Texas: Association for Computing Machinery, 2016: 785-794.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献