• 中国科学引文数据库(CSCD)来源期刊
  • 中文核心期刊中文科技核心期刊
  • Scopus RCCSE中国核心学术期刊
  • 美国EBSCO数据库 俄罗斯《文摘杂志》
  • 《日本科学技术振兴机构数据库(中国)》
二维码

隧道建设(中英文) ›› 2022, Vol. 42 ›› Issue (S1): 331-341.DOI: 10.3973/j.issn.2096-4498.2022.S1.038

• 研究与探索 • 上一篇    下一篇

基于数据增强和机器学习算法的盾构隧道引发地面沉降预测及应用

管浩1, 刘维1, *, 王峰2, 赵华菁1, 张功3, 张高海4   

  1. (1. 苏州大学轨道交通学院, 江苏 苏州 215000 2. 中铁十八局集团有限公司, 广东 珠海 5190003. 北京住总集团有限公司, 北京 100000 4. 中国铁建投资集团有限公司, 广东 珠海 519000)
  • 出版日期:2022-07-22 发布日期:2022-08-23
  • 作者简介:管浩(1998—),男,江苏扬州人,苏州大学道路与铁道工程专业在读硕士,研究方向为隧道与地下空间。Email: 20205246014@stu.suda.edu.cn。*通信作者: 刘维, Email: ggoulmmeng@suda.edu.cn。

Prediction of Surface Settlement Caused by Shield Tunneling Based on Data Enhancement and Machine Learning Algorithm

GUAN Hao1, LIU Wei1, *, WANG Feng2, ZHAO Huajing1, ZHANG Gong3, ZHANG Gaohai4   

  1. (1. School of Rail Transportation, Soochow University, Suzhou 215000, Jiangsu, China; 2. China Railway 18th Bureau Group Co., Ltd., Zhuhai 519000, Guangdong, China; 3. Beijing Uni-Construction Group Co., Ltd., Beijing 100000, China; 4. China Railway Construction Investment Group Co., Ltd., Zhuhai 519000, Guangdong, China)
  • Online:2022-07-22 Published:2022-08-23

摘要: 为解决地铁盾构隧道施工引起的地表沉降预测过程中数据样本不足、对数据预处理较为粗糙的问题,选取北京地区32个地铁盾构隧道施工引起的地面沉降数据作为数据库,采用合成少数类过采样技术(synthetic minority oversampling technique, SMOTE)算法对数据库进行扩增,并在此基础上选取BP神经网络(back propagation, BP)、随机森林(random forest, RF)、支持向量机(support vector machine, SVM)K近邻(K-nearest neighbor, KNN)4种机器学习模型对沉降进行预测分析。分析结果表明: 1)经过预处理后的数据集预测能力显著增强,其中,KNN模型表现最佳,测试集平均绝对误差仅为1.60 mm 2)采用KNN模型对北京轨道交通12号线西坝河—三元桥区间地层沉降进行预测,基于该模型预测值的Peck公式与实测值拟合度较高; 3)基于数据增强下的KNN模型对于盾构施工引起的地面沉降变形有良好的预测效果。

关键词: 盾构隧道, 地面沉降, 机器学习算法, 数据扩增, 地面沉降预测

Abstract: The prediction of surface settlement induced by shield tunneling has many problems such as insufficient data samples and relatively rough data preprocessing. As a result, a case study is conducted on Beijing metro shield tunneling projects, and the surface settlement induced by shield tunneling of 32 cases are selected. Then, the synthetic minority oversampling technology is used to enlarge the database, and four machine learning models, namely back propagation, random forest, support vector machine, and knearest neighbor (KNN), are employed to predict the surface settlement. The comparative analysis shows the following: (1) The prediction ability of the preprocessed database is significantly enhanced, and KNN shows a smallest average absolute error of the test set, which is 1.60 mm. (2) The bestperforming model is applied to XibaheSanyuanqiao shield tunneling section of Beijing metro line 12, and the predicted results agree well with the monitored results. (3) The KNN model based on the data enhancement has a good surface settlement prediction effect.

Key words: shield tunnel, surface settlement, machine learning algorithm, data enhancement, surface settlement prediction