摘要
随着盾构隧道工程信息化水平的提升,隧道掘进设备作业过程监测技术日益完善,记录的工程数据蕴含了掘进设备内部信息及其与外部地层的相互作用关系。机器学习因其数据分析能力强,无需先验的理论公式和专家知识,相较于传统的建模统计分析方法具有更大的应用空间。通过机器学习方法对收集的信息与数据进行深度挖掘并分析其内在联系,有助于提升盾构隧道工程建设的效率和安全保障水平。简述机器学习方法的基本原理,总结和分析机器学习方法在盾构工程中的应用研究状况,综述基于机器学习的盾构设备状态分析、盾构设备性能预测、围岩参数反演、地表变形预测和隧道病害诊断等5个方面的进展,并分析当前研究的不足。最后,分析盾构隧道工程向智能化方向发展需重点攻克的难题。
随着经济水平的不断发展,交通运输需求日益增长,而隧道能有效利用地下空间、节约地面土地资源,在交通运输基础设施(包括铁路、公路和城市轨道交通)建设中的比重越来越大。盾构工法因具有对地面影响小、掘进速度快、作业面安全等优势,在隧道建设中被广泛采用。随着人工智能技术的涌现与发展,盾构隧道工程建设也从过去追求高速、机械化向高质量、智能化方向发展。
在“大数据”背景下,盾构隧道建设呈现出高容量数据存储能力、高效实时数据处理能力和高强多源异构适应性的“三高”需求,机器学习(Machine Learning,ML)方法开始成为分析隧道工程建设“大数据”的新工具。机器学习的吸引力源自它独特的信息处理能力,如具有非线性、高并行性及高容错性的学习和泛化能
回顾并总结机器学习方法在盾构隧道工程中的设备状态分析、盾构性能预测、地质参数研究、地表变形预测和隧道病害预测等5个方面的应用研究现状,分析相关研究的进展和不足,并对机器学习方法在盾构隧道工程中的应用研究前景进行分析与展望,旨在推动机器学习方法在盾构隧道工程中的应用研究。
基于人工智能的数据分析算法发展历程如

图1 基于人工智能的数据分析算法发展历
Fig. 1 Development history of data analysis algorithms based on artificial intelligenc
机器学习的核心原理是通过输入信息训练计算机模仿人与动物“从经验中学习成长”的天性,其基于直接从数据中“学习”信息的计算方法,而不依赖于预设的方程模型。当训练样本数量增加时,训练出的模型性能相应提升,从而能更好地解决实际问题。机器学习算法可分为4种基本类型:监督学习(Supervised Learning,SL)、非监督学习(Unsupervised Learning,UL)、半监督学习(Semi-Supervised Learning,SSL)和强化学习(Reinforcement Learning,RL
1)监督学习:通过人工预设的训练特征和输出结果来训练模型,使模型具有预测未来输出的能
2)非监督学习:从输入信息中解析出隐藏在数据中的内在结
3)半监督学习:将监督学习与非监督学习相结合的一种学习方法。一般半监督学习的目标是找到一个函数迎合(回归任务),然后用分类任务的信息去优化回归函数。
4)强化学习:训练模型通过与输入信息的反复交互来学习处理任务。这种学习方法使模型面对动态环境能够做出一系列决策,从而使任务奖励期望最大化。
机器学习模型的建立通常包含以下几个步骤:收集数据、预处理数据和提取特征、训练模型、调整模
1)收集数据。收集数据的常见方法是运用信号处理或聚类技术来汇聚数据,如通过模拟仿真或传感器测量获得目标设备或系统的相关数据。
2)预处理数据和提取特征。在提取特征之前,大多数数据集都需要进行预处理,包括删除异常值和异常趋势、处理丢失的数据以及对数据进行归一化。特征提取是机器学习最重要的步骤之一,它将原始数据转变成适合机器学习算法的信息。特征提取消除了各类测量数据中的冗余现象,有助于学习阶段的泛化,而泛化是避免对特定样本过拟合的关键。
3)训练模型。主要步骤包括选择训练和验证的数据、选择训练的算法、反复训练和评估分类模型。在盾构隧道工程中常用的模型有盾构设备状态预测模型、地表沉降预测模型及隧道病害预测模型等。
4)调整模型。通过技术手段提高模型性能,常用方法包括调节模型参数、添加或修改训练数据、变换或提取新特征。
盾构隧道工程中常使用的机器学习基本算法有人工神经网络(Artificial Neural Network,ANN)、支持向量机(Support Vector Machine,SVM)、回归树(Regression Tree,RT)、随机森林(Random Forests,RF)等,常使用的深度学习算法有卷积神经网络(Convolutional Neural Networks,CNN)、循环神经网络(Recurrent Neural Networks, RNN)、生成对抗网络(Generative Adversarial Network,GAN)等。同时,以上算法与其他人工智能方法相结合并改进得到的复合人工智能方法可使机器学习模型有更高的准确率。
在盾构掘进过程中,可以采集大量的数据资料,如围岩参数、盾构状态及施工数据等,部分数据能反映盾构掘进设备与周围环境的相互作用规律。通过这些数据训练得到的机器学习模型不仅有助于解决盾构工程中信息处理不足、集成化不足、分析水平差等问题,还能对工程相关信息进行汇总并解析其中的关联性,从而对盾构隧道的设计、施工与运维发挥重要的作用。
基于机器学习的盾构工程管理应用是通过对盾构隧道建设中的相关工程数据进行整理存储和分类关联,基于不同机器学习方式进行分析,将所得分析模型形成相关数据库,再使用编程软件构建机器学习应用管理信息平台(如

图2 机器学习应用管理示意图
Fig. 2 Schematic diagram of machine learning application management
盾构机的设备状态和掘进性能对隧道建设的施工效率、质量和安全有着决定性影响,而机器学习方法在盾构掘进机运行情况识别与相关性能预测两个方面具有较好的适应能力和较大的应用空间。
盾构机组成复杂,在施工过程中容易出现各种故障;且因其在地下空间挖掘前进,出现故障时排查异常困难。刀盘作为盾构机的主要组成部分,是盾构设备故障的主要来源。针对刀盘故障问题,研究人员重点研究了基于机器学习算法的刀盘故障诊断方法。Jin
由于隧道盾构机是实时运作,如果对监测数据分析只体现静态关联性,难以对实际工程进行管控指导。因此,需要对数据动态特征进行汇总解析,研究基于时序特征的状态分析方法。Rumelhart
盾构设备监测信息复杂、特征繁多,基于未调整的原始数据无法训练出高精度的预测模型,因此,机器学习模型预测的准确率在很大程度上取决于数据预处理的效果。循环神经网络对时序特征有着极强的学习能力,被广泛应用于盾构设备状态的分析中,但其存在训练优化慢、计算能力需求大等不足,在盾构设备状态分析与预测中仍有很大的拓展空间。
在盾构施工过程管理中,盾构机的推进速率、刀盘荷载及土仓压力等性能指标对工期管理和成本把控具有重要意义。传统研究主要通过理论模型、室内试验和模拟仿真等预测盾构机的性能,但通常仅能分析某一方面的性能。基于现场实测数据,运用回归分析、模糊数学或者神经网络等机器学习算法,可综合分析盾构施工过程中的设备状况、性能指标与围岩参数的内在联系等,从而达到较高的预测精度。
针对盾构的掘进效率问题,研究人员基于围岩信息等数据,利用机器学习方法训练掘进效率分析模型。Salimi
更多基于机器学习算法的盾构掘进性能预测研究和应用案例见
文献 | 算法 | 地质参数 | 运行参数 |
---|---|---|---|
Fattahi | ANFIS | DPW、ALPHA、UCS、BI | |
Armaghani | ANN | WZ、RMR、RQD、UCS、BTS | WT、CT |
Samaei | CART | DPW、ALPHA、UCS、BI、BTS | |
Eftekhari | ANN | RMR、RQD、UCS、BTS、Qu | WT、CF、CT |
Oraee | ANFIS | RQD、DPW、UCS | |
Torabi | ANN | UCS、C、PHI、POI | |
Shao | GP、SVM、ANN | UCS、BTS、PSI、ALPHA | |
Martins | ANN、SVM | UCS、PSI、DPW、ALPHA | |
Pham | ANN | RQ、RF、RT | CD、CF、CT、CP、WT |
注: ANN为人工神经网络;ANFIS为自适应神经模糊推理系统;CART为分类回归树;GP为高斯过程;SVM为支持向量机;DPW为碎岩体平均间距;ALPHA为结构面方位角;UCS为单轴抗压强度; BI为岩石脆性系数;WZ为风化带指数;BTS为巴西抗拉强度;RMR为岩体等级;RQD为岩石质量指标;Qu为石英百分比;C为土层黏聚力;PHI为摩擦角;POI为泊松比;PSI为峰斜率指数;RQ为岩石质量指数;RF为岩石破裂指数;RT为岩石纹理指数;WT为刀盘转速;CT为刀盘转矩;CF为刀盘推力;CD为刀盘直径;CP为刀盘功率。
现场采集数据的速度往往大于现有机器学习方法的分析速度,为解决这一问题,近年来计算能力更加高效的算法,如深度神经网络(Deep Neural Network,DNN
研究人员也基于机器学习算法,对土仓/泥水压力、姿态控制等性能指标进行预测研究。对于泥水平衡式盾构
在盾构姿态预测与控制方面,Zhou
以上研究表明,目前机器学习方法对盾构机关键性能(如掘进效率、土仓/泥水压力、姿态调整等)预测的应用取得了一定的进展,多数预测模型输入信息以地层勘察数据为主,以盾构掘进过程中的设备参数为辅。通常先进行不同输入与预测结果的相关性分析,进而筛选出最具相关性的输入特征,再将该特征导入合适的机器学习回归算法训练预测模型。
实际工程验证表明,基于这一思路建立的预测模型比直接用非线性回归算法训练的预测模型的准确度更高。
盾构掘进过程中的监测数据包含了地质工况及周边环境的动态变化信息(如

图3 变形因素的相互关系
Fig. 3 Relationship among deformation factors
盾构隧道等地下工程存在于地下岩土体中,岩土材料具有非均质、非连续、非线性等特点,传统的勘察方法成本高,获取的岩土参数信息有限,而理论和数值计算方法不能很好地解决盾构掘进扰动影响下的地层岩土参数问
在盾构掘进过程中,准确获取掌子面地质信息有助于设置最佳盾构作业参数,使盾构机获得更好的掘进效率。然而,由于盾构机的封闭性设计及较窄的作业面使操作人员无法直接观察周围环境,利用机器学习方法间接识别地质条件成为研究热点之一。Yu
以上研究表明,基于机器学习方法的围岩参数反演和地质识别在盾构隧道工程中有广泛的应用前景。然而,目前反演的信息主要是地层的岩土类型与空间分布,能够实现岩土类型与力学参数同时反演和预测的研究还有待进一步开展。
盾构机在施工过程中会与地层发生较强的相互作用(如

图4 不同施工阶段的地面沉
Fig. 4 Ground settlement at different stages of constructio
基于机器学习的地表变形预测研究中,孙钧
文献 | 算法 | 主要输入特征参数 | 数据量 |
---|---|---|---|
Ocak | ANN、SVM、GP | Fp、Pr、Ex、Tvp、D、A、H、Dl、Dr、SPT、Uw、Gc、Gw | 230 |
Wang | ANN | H、Ef、St、Gw、Fp、 Pr、 Pa、Gp、Gf | 661 |
Hasanipanah | PSO-ANN | Hv、Co、Y | 143 |
Kohestani | RF、ANN | H、Ds、Gc、Gi、Iw、Fp、Pr、Pa、Tvp、Gp | 49 |
Goh | MARS | H、Pr、Ep、SPT、Mc、E、Gp | 148 |
Mahmoodzadeh |
LSTM、DNN、KNN、 SVM、GP、DT、LR | Tw、H、Cm、Co、Fa、E | 300 |
Zhang | ANN | Th、Ct、Pr、Gp、Gv、Fp、Td、Tv、KTS、Gw、BCT、SCT、RCT、SIT、RIT、KH、KD | 328 |
注: Fp为表面压力;Pr为渗透率;Ex为挖掘材料的数量;Tvp为尾部空隙灌浆填充百分比;D为隧道直径;A为隧道之间的距离;H为隧道深度;Dl为左侧隧道与地面监测点之间的距离;Dr为右侧隧道与地面监测点之间的距离;SPT为标准渗透测试;Uw为单位重量;Gc为地质条件;Gw为地下水的影响;Ef为开挖面到沉降标记;St为地层类型;Pa为俯仰角;Gp为灌浆压力;Gf为灌浆;Th为推力;Ct为刀盘扭矩;Td为隧道偏差;Tv为隧道空隙;KTS为喀斯特洞穴处理方案;BCT为隧道顶部回填土的厚度;SCT为隧道顶部砂土厚度;RCT为隧道顶部风化岩石的厚度;SIT为隧道底架下砂土厚度;RIT为隧道底架下的岩石厚度;KH为溶洞高度;KD为溶洞与隧道内底之间的距离;Ds为距轴的距离;Gi为倒置地质;Iw为转化为地下水位;Hv为水平与垂直应力比;Co为凝聚力;Y为弹性模量;Ep为土压力;Mc为平均水分含量;E为地层弹性模量;DT为决策树;LR为线性回归;Tw为隧道宽度;Cm为施工方法;Fa为摩擦
通过对基于机器学习的地质参数反演和地表变形预测研究的综述可知,盾构施工长期在复杂环境下进行,监测系统采集的数据大部分是相似的无特征信息。传感器在施工现场不仅布设困难,而且容易受到现场施工作业影响(如设备损坏、丢失和供电中断等问题),从而导致监控数据无效或缺
隧道健康状态是隧道建设过程及后期运营阶段的重要监测内容。目前,研究人员已经开展了基于机器学习算法的隧道病害监测研究,并通过分析隧道健康情况,建立了隧道病害预测模型,服务盾构隧道的管养。Cha
对于隧道运营期监测数据的处理,采用传统机器学习方法的缺点在于需要手动定义目标的特征,对于复杂场景中的数据来说,目标的特征并不具体,很难定量描述。深度学习的发展改变了此现状,它通过卷积神经网络等算法进行特征提取,有效实现监测和检测数据中异常信息的分类和位置信息的获取。由于隧道衬砌结构病害特征的相似性以及结构的复杂性,在隧道衬砌检测方面,目前用深度学习实现多种病害分类的相关研究较少。
学者们在基于机器学习方法(如

图5 参考文献涉及的算法统计
Fig. 5 Statistics of research methods in the literature involved in this paper

图6 参考文献涉及的研究目的统计
Fig. 6 Statistics of research purpose in the literature involved in this paper
目前,机器学习方法在盾构隧道工程中应用的主要难点与不足主要表现在以下几个方面。
1)机器学习的预测实用性因盾构工程实时采集信息能力不足而受限。实时监测数据能极大增强机器学习算法的即时预测能力,但盾构设备本身构造复杂,施工环境恶劣,隧道掘进过程中难以为大量监测仪器提供合适的安装空间;盾构设备狭长且位于地层中,监测设备采集到的数据难以实时传送到收集终端,这些因素都限制了机器学习预测方法的普及和推广。
2)盾构隧道工程实测信息的数据模态、样本类别、信息结构等特征差异大,现阶段主要是通过数据类型转化及人工修正等方式来进行数据归一化,但处理过程需要大量的人工标注,主观性大,可能会导致数据内部某些潜在特征被忽视。因此,需要深入挖掘数据背后的产生机制,识别异常样本的特征,探明关键性因素并进行人工标注,但目前面向机器学习的多源异构数据处理方法还有待进一步研究。
3)相较于传统的数值解析法或经验公式法,基于盾构隧道工程实测数据的机器学习预测模型通常具有更高的拟合精度,但要达到高精度,需要耗费大量运算时间与计算能力进行模型训练。因此,限制机器学习算法在盾构工程中大面积推广应用的一大原因就是现场计算能力。在隧道掘进现场,由于数据采集或监测设备提供的平台计算能力不足,难以满足利用实测数据训练机器学习算法的需求,因此,需探索与云计算或硬件加速等相结合的技术。
机器学习方法是基于现有数据分析理论上的更高层次的分析方法,其在盾构隧道工程中的应用主要包括装备运行状态识别、关键参数关联分析、刀具故障预测、地层参数识别等,相关研究可提高施工管理水平、减少盾构施工对邻近环境的影响。随着5G传感、物联网、云计算、北斗通信等新技术的快速迭代,盾构机实测数据的存储数量和质量、实时性都将得到持续发展,这对机器学习方法来说是“如虎添翼”。然而,机器学习算法要真正达到在实际工程中广泛应用的水平,未来还需在以下方面进行探索和发展。
1)海量多源数据的汇聚。不同厂家生产的隧道掘进机监控设备存在差异,采集的信息不同源且不兼容。可通过远程服务器根据对应的端口协议汇总数据,以此集成不同工程、不同设备、不同隧道的监测信息,通过大数据训练来增强机器学习模型的泛化能力,而这需打破现有数据的管理壁垒。
2)基于云计算和5G技术的机器学习算法开发。盾构隧道工程相关的机器学习模型训练计算成本高,因此,现有大多数机器学习分析模型是采集数据后在实验室平台进行训练。与云计算相结合的远程训练模式是满足工地实时计算需求的可行途径,即工地监测端负责数据汇总,上传至云端进行机器学习训练、优化、预测,再将结果返回至工地端。在云计算模式下,与5G无线通信技术相结合的机器学习算法是盾构工程需要探索的方向。
3)盾构隧道工程智能管控平台构建。随着盾构隧道掘进数据的不断累积,以及智能算法能力的不断提升,可以构建以机器学习方法为核心的盾构隧道工程智能管理模式和平台(如

图7 基于机器学习的智能管理模式
Fig. 7 Intelligent management mode based on machine learning
作为人工智能方法的重要组成部分,机器学习是工程信息化的重要发展方向。机器学习算法能够深入挖掘盾构隧道监测大数据的隐含特征,为盾构装备状态识别与性能预测、地质识别与地表变形预测,隧道健康监测与预警等方面提供技术支持。笔者总结了当前机器学习技术在盾构隧道工程中应用研究的主要进展与不足,并结合当前实际技术水平展望后续应用研究的主要方向,以期为隧道工程智能化发展添砖加瓦。
参考文献
JAIN A K, MAO J C, MOHIUDDIN K M. Artificial neural networks: A tutorial [J]. Computer, 1996, 29(3): 31-44. [百度学术]
JAVAD G, NARGES T. Application of artificial neural networks to the prediction of tunnel boring machine penetration rate [J]. Mining Science and Technology (China), 2010, 20(5): 727-733. [百度学术]
NEAUPANE K M, ADHIKARI N R. Prediction of tunneling-induced ground movement with the multi-layer perceptron [J]. Tunnelling and Underground Space Technology, 2006, 21(2): 151-159. [百度学术]
ATTEWELL P, WOODMAN J P. Predicting the dynamics of ground settlement and its derivitives caused by tunnelling in soil [J]. Ground Engineering, 1982, 15(8): 13-22. [百度学术]
陈丹, 刘喆, 刘建友, 等. 铁路盾构隧道智能建造技术现状与展望[J]. 隧道建设(中英文), 2021, 41(6): 923-932. [百度学术]
CHEN D, LIU Z, LIU J Y, et al. State-of-art and prospects for intelligent construction technology for railway shield tunneling [J]. Tunnel Construction, 2021, 41(6): 923-932. (in Chinese) [百度学术]
ZHANG W G, LI H R, LI Y Q, et al. Application of deep learning algorithms in geotechnical engineering: A short critical review [J]. Artificial Intelligence Review, 2021, 54(8): 5633-5673. [百度学术]
翟嘉琪, 杨希祥, 程玉强, 等. 机器学习在故障检测与诊断领域应用综述[J]. 计算机测量与控制, 2021, 29(3): 1-9. [百度学术]
ZHAI J Q, YANG X X, CHENG Y Q, et al. Overview of application of fault detection and diagnosis based on machine learning [J]. Computer Measurement & Control, 2021, 29(3): 1-9. (in Chinese) [百度学术]
李杰其, 胡良兵. 基于机器学习的设备预测性维护方法综述[J]. 计算机工程与应用, 2020, 56(21): 11-19. [百度学术]
LI J Q, HU L B. Review of machine learning for predictive maintenance [J]. Computer Engineering and Applications, 2020, 56(21): 11-19. (in Chinese) [百度学术]
施清清, 周伯儒, 倪申. 基于机器学习的空调系统管路焊堵自动检测研究[J]. 计量与测试技术, 2019, 46(4): 31-33. [百度学术]
SHI Q Q, ZHOU B R, NI S. Research on automatic detection of pipeline welding blockage in air conditioning system based on machine learning [J]. Metrology & Measurement Technique, 2019, 46(4): 31-33. (in Chinese) [百度学术]
GOMES SOARES S, ARAÚJO R. An on-line weighted ensemble of regressor models to handle concept drifts [J]. Engineering Applications of Artificial Intelligence, 2015, 37: 392-406. [百度学术]
孙振川, 钱彤途, 任颖莹, 等. 隧道掘进机工程大数据管理平台关键技术及应用研究[J]. 隧道建设(中英文), 2020, 40(6): 783-792. [百度学术]
SUN Z C, QIAN T T, REN Y Y, et al. Study on key technologies and application of engineering big data management platform of tunnel boring machine [J]. Tunnel Construction, 2020, 40(6): 783-792. (in Chinese) [百度学术]
JIN Y R, QIN C J, TAO J F, et al. An accurate and adaptative cutterhead torque prediction method for shield tunneling machines via adaptative residual long-short term memory network [J]. Mechanical Systems and Signal Processing, 2022, 165: 108312. [百度学术]
GUO J W, LAO Z P, HOU M, et al. Mechanical fault time series prediction by using EFMSAE-LSTM neural network [J]. Measurement, 2021, 173: 108566. [百度学术]
MAHMOODZADEH A, MOHAMMADI M, HASHIM IBRAHIM H, et al. Machine learning forecasting models of disc cutters life of tunnel boring machine [J]. Automation in Construction, 2021, 128: 103779. [百度学术]
HAN H, GAO X. Fault prediction of shield machine based on rough set and BP neural network [C]//2017 4th International Conference on Information Science and Control Engineering (ICISCE). July 21-23, 2017, Changsha, China. IEEE, 2017: 994-998. [百度学术]
ELBAZ K, SHEN S L, ZHOU A N, et al. Prediction of disc cutter life during shield tunneling with AI via the incorporation of a genetic algorithm into a GMDH-type neural network [J]. Engineering, 2021, 7(2): 238-251. [百度学术]
ZOU L, LIANG L. Fault diagnosis of shield machine based on SOM-BP neural network fusion [C]//2018 International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC). August 15-17, 2018, Xi'an, China. IEEE, 2018: 232-237. [百度学术]
RUMELHART D E , HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors [J]. Nature, 1986, 323(6088): 533-536. [百度学术]
ZHANG L, YI Z. Selectable and unselectable sets of neurons in recurrent neural networks with saturated piecewise linear transfer function [J]. IEEE Transactions on Neural Networks, 2011, 22(7): 1021-1031. [百度学术]
GAO X J, SHI M L, SONG X G, et al. Recurrent neural networks for real-time prediction of TBM operating parameters [J]. Automation in Construction, 2019, 98: 225-235. [百度学术]
SUN Z H, ZHENG H Y, SHI B H. Faults prediction and diagnoses of shield machine based on LSTM [C]//2019 Chinese Control and Decision Conference (CCDC). June 3-5, 2019, Nanchang, China. IEEE, 2019: 5254-5259. [百度学术]
HUANG X, ZHANG Q T, LIU Q S, et al. A real-time prediction method for tunnel boring machine cutter-head torque using bidirectional long short-term memory networks optimized by multi-algorithm [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2022, 14(3): 798-812. [百度学术]
QIN C J, SHI G, TAO J F, et al. Precise cutterhead torque prediction for shield tunneling machines using a novel hybrid deep neural network [J]. Mechanical Systems and Signal Processing, 2021, 151: 107386. [百度学术]
SALIMI A, ROSTAMI J, MOORMANN C, et al. Application of non-linear regression analysis and artificial intelligence algorithms for performance prediction of hard rock TBMs [J]. Tunnelling and Underground Space Technology, 2016, 58: 236-246. [百度学术]
FATTAHI H, BABANOURI N. Applying optimized support vector regression models for prediction of tunnel boring machine performance [J]. Geotechnical and Geological Engineering, 2017, 35(5): 2205-2217. [百度学术]
STYPULKOWSKI J B, BERNARDEAU F G, JAKUBOWSKI J. Descriptive statistical analysis of TBM performance at Abu Hamour Tunnel Phase I [J]. Arabian Journal of Geosciences, 2018, 11(9): 1-11. [百度学术]
AFRADI A, EBRAHIMABADI A, HALLAJIAN T. Prediction of the penetration rate and number of consumed disc cutters of tunnel boring machines (TBMs) using artificial neural network (ANN) and support vector machine (SVM)—Case study: Beheshtabad water conveyance tunnel in Iran [J]. Asian Journal of Water, Environment and Pollution, 2019, 16(1): 49-57. [百度学术]
FATTAHI H, BAZDAR H. Applying improved artificial neural network models to evaluate drilling rate index [J]. Tunnelling and Underground Space Technology, 2017, 70: 114-124. [百度学术]
GAO B Y, WANG R R, LIN C J, et al. TBM penetration rate prediction based on the long short-term memory neural network [J]. Underground Space, 2021, 6(6): 718-731. [百度学术]
FATTAHI H. Adaptive neuro fuzzy inference system based on fuzzy c-means clustering algorithm, a technique for estimation of TBM penetration rate [J]. Iran University of Science & Technology, 2016, 6: 159-171. [百度学术]
ARMAGHANI D J, MOHAMAD E T, NARAYANASAMY M S, et al. Development of hybrid intelligent models for predicting TBM penetration rate in hard rock condition [J]. Tunnelling and Underground Space Technology, 2017, 63: 29-43. [百度学术]
SAMAEI M, RANJBARNIA M, NOURANI V, et al. Performance prediction of tunnel boring machine through developing high accuracy equations: A case study in adverse geological condition [J]. Measurement, 2020, 152: 107244. [百度学术]
EFTEKHARI M, BAGHBANAN A, BAYATI M. Predicting penetration rate of a tunnel boring machine using artificial neural network [C]//Isrm International Symposium-6th Asian Rock Mechanics Symposium: Onepetro, 2010. [百度学术]
ORAEE K, KHORAMI M, HOSSEINI N. Prediction of the penetration rate of TBM using adaptive neuro fuzzy inference system (ANFIS) [C]//Proceeding of Sme Annual Meeting and Exhibit, 2012: 297-302. [百度学术]
TORABI S R, SHIRAZI H, HAJALI H, et al. Study of the influence of geotechnical parameters on the TBM performance in Tehran-Shomal highway project using ANN and SPSS [J]. Arabian Journal of Geosciences, 2013, 6(4): 1215-1227. [百度学术]
SHAO C J, LI X L, SU H Y. Performance prediction of hard rock TBM based on extreme learning machine [M]//Intelligent Robotics and Applications. Berlin, Heidelberg: Springer, 2013: 409-416. [百度学术]
MARTINS F F, MIRANDA T. Prediction of hard rock TBM penetration rate based on data mining techniques [C]//Proceedings of the 18th International Conference on Soil Mechanics and Geotechnical Engineering, Paris, 2013, 2: 1751-1754. [百度学术]
PHAM H V, YUJI F, KAMEI K. Hybrid artificial neural networks for TBM performance prediction in complex underground conditions [C]//2011 IEEE/SICE International Symposium on System Integration (SII). December 20-22, 2011, Kyoto, Japan. IEEE, 2011: 1149-1154. [百度学术]
KOOPIALIPOOR M, TOOTOONCHI H, JAHED ARMAGHANI D, et al. Application of deep neural networks in predicting the penetration rate of tunnel boring machines [J]. Bulletin of Engineering Geology and the Environment, 2019, 78(8): 6347-6360. [百度学术]
SALIMI A, FARADONBEH R S, MONJEZI M, et al. TBM performance estimation using a classification and regression tree (CART) technique [J]. Bulletin of Engineering Geology and the Environment, 2018, 77(1): 429-440. [百度学术]
YAGIZ S, GOKCEOGLU C, SEZER E, et al. Application of two non-linear prediction tools to the estimation of tunnel boring machine performance [J]. Engineering Applications of Artificial Intelligence, 2009, 22(4/5): 808-814. [百度学术]
XU H, ZHOU J, ASTERIS P G, et al. Supervised machine learning techniques to the prediction of tunnel boring machine penetration rate [J]. Applied Sciences, 2019, 9(18): 3715. [百度学术]
ARMAGHANI D J, KOOPIALIPOOR M, MARTO A, et al. Application of several optimization techniques for estimating TBM advance rate in granitic rocks [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2019, 11(4): 779-789. [百度学术]
FENG S X, CHEN Z Y, LUO H, et al. Tunnel boring machines (TBM) performance prediction: A case study using big data and deep learning [J]. Tunnelling and Underground Space Technology, 2021, 110: 103636. [百度学术]
MAHDEVARI S, SHAHRIAR K, YAGIZ S, et al. A support vector regression model for predicting tunnel boring machine penetration rates [J]. International Journal of Rock Mechanics and Mining Sciences, 2014, 72: 214-229. [百度学术]
SUN W, SHI M L, ZHANG C, et al. Dynamic load prediction of tunnel boring machine (TBM) based on heterogeneous in-situ data [J]. Automation in Construction, 2018, 92: 23-34. [百度学术]
LI J H, LI P X, GUO D, et al. Advanced prediction of tunnel boring machine performance based on big data [J]. Geoscience Frontiers, 2021, 12(1): 331-338. [百度学术]
王潇. 盾构施工喷涌及地表变形的预警预测方法研究[D]. 北京: 北京交通大学, 2020. [百度学术]
WANG X. Research on method of pre-warning and prediction for spewing and ground surface movements in shield construction [D]. Beijing: Beijing Jiaotong University, 2020. (in Chinese) [百度学术]
WANG Q, XIE X Y, SHAHROUR I, et al. Use of deep learning, denoising technic and cross-correlation analysis for the prediction of the shield machine slurry pressure in mixed ground conditions [J]. Automation in Construction, 2021, 128: 103741. [百度学术]
LI X F, GONG G F. Predictive control of slurry pressure balance in shield tunneling using diagonal recurrent neural network and evolved particle swarm optimization [J]. Automation in Construction, 2019, 107: 102928. [百度学术]
ZHOU C, DING L Y, HE R. PSO-based Elman neural network model for predictive control of air chamber pressure in slurry shield tunneling under Yangtze River [J]. Automation in Construction, 2013, 36: 208-217. [百度学术]
ZHOU C, XU H C, DING L Y, et al. Dynamic prediction for attitude and position in shield tunneling: A deep learning method [J]. Automation in Construction, 2019, 105: 102840. [百度学术]
WANG R H, LI D Q, CHEN E J, et al. Dynamic prediction of mechanized shield tunneling performance [J]. Automation in Construction, 2021, 132: 103958. [百度学术]
WANG P, KONG X G, GUO Z K, et al. Prediction of axis attitude deviation and deviation correction method based on data driven during shield tunneling [J]. IEEE Access, 7: 163487-163501. [百度学术]
ZHANG P, CHEN R P, WU H N. Real-time analysis and regulation of EPB shield steering using random forest [J]. Automation in Construction, 2019, 106: 102860. [百度学术]
刘靖. 小仟隧道围岩参数智能反分析及施工应用研究[D]. 北京: 北京交通大学, 2018. [百度学术]
LIU J. Research on intelligent back analysis of rock parameters and construction application of Xiao Qian tunnel [D]. Beijing: Beijing Jiaotong University, 2018. (in Chinese) [百度学术]
韩小川. 夏村隧道施工监控量测与围岩参数位移反分析[D]. 南昌: 南昌工程学院, 2015. [百度学术]
HAN X C. Xiacun tunnel construction monitoring and back analysis of rock parameters from measured displacements [D]. Nanchang: Nanchang Institute of Technology, 2015. (in Chinese) [百度学术]
LIANG Y C, FENG D P, LIU G R, et al. Neural identification of rock parameters using fuzzy adaptive learning parameters [J]. Computers & Structures, 2003, 81(24/25): 2373-2382. [百度学术]
SUWANSAWAT S, EINSTEIN H H. Artificial neural networks for predicting the maximum surface settlement caused by EPB shield tunneling [J]. Tunnelling and Underground Space Technology, 2006, 21(2): 133-150. [百度学术]
郝哲, 刘斌. 基于差分法及神经网络的硐室围岩力学参数反分析[J]. 岩土力学, 2003, 24(Sup2): 77-80. [百度学术]
HAO Z, LIU B. Back analysis of mechanics parameters of rocks surrounding openings on the basis of calculus of difference and neural network [J]. Rock and Soil Mechanics, 2003, 24(Sup2): 77-80. (in Chinese) [百度学术]
高攀科, 谢永利. 隧道软弱围岩的改进BP神经网络位移反分析[J]. 郑州大学学报(工学版), 2013, 34(1): 23-26. [百度学术]
GAO P K, XIE Y L. Displacement back analysis of tunnels in soft and weak rocks based on improved BP neural network method [J]. Journal of Zhengzhou University (Engineering Science), 2013, 34(1): 23-26. (in Chinese) [百度学术]
YU H G, TAO J F, QIN C J, et al. A novel constrained dense convolutional autoencoder and DNN-based semi-supervised method for shield machine tunnel geological formation recognition [J]. Mechanical Systems and Signal Processing, 2022, 165: 108353. [百度学术]
ZHANG Q L, LIU Z Y, TAN J R. Prediction of geological conditions for a tunnel boring machine using big operational data [J]. Automation in Construction, 2019, 100: 73-83. [百度学术]
YANG H Q, SONG K L, ZHOU J Y. Automated recognition model of geomechanical information based on operational data of tunneling boring machines [J]. Rock Mechanics and Rock Engineering, 2022, 55(3): 1499-1516. [百度学术]
ERHARTER G H, MARCHER T, REINHOLD C. Rock mechanics for natural resources and infrastructure development [M]: CRC Press, 2019: 2426-2432. [百度学术]
LIU Q S, WANG X Y, HUANG X, et al. Prediction model of rock mass class using classification and regression tree integrated AdaBoost algorithm based on TBM driving data [J]. Tunnelling and Underground Space Technology, 2020, 106: 103595. [百度学术]
KIM T H, KWAK N S, KIM T K, et al. A TBM data-based ground prediction using deep neural network [J]. Journal of Korean Tunnelling and Underground Space Association, 2021, 23(1): 13-24. [百度学术]
SHI M L, SUN W, ZHANG T C, et al. Geology prediction based on operation data of TBM: Comparison between deep neural network and soft computing methods [C]//2019 1st International Conference on Industrial Artificial Intelligence (IAI). July 23-27, 2019, Shenyang, China. IEEE, 2019: 1-5. [百度学术]
YU H G, TAO J F, QIN C J, et al. Rock mass type prediction for tunnel boring machine using a novel semi-supervised method [J]. Measurement, 2021, 179: 109545. [百度学术]
ZHAO J H, SHI M L, HU G, et al. A data-driven framework for tunnel geological-type prediction based on TBM operating data [J]. IEEE Access, 2019, 7: 66703-66713. [百度学术]
孙钧, 袁金荣. 盾构施工扰动与地层移动及其智能神经网络预测[J]. 岩土工程学报, 2001, 23(3): 261-267. [百度学术]
SUN J, YUAN J R. Soil disturbance and ground movement under shield tunnelling and its intelligent prediction by using ANN technology [J]. Chinese Journal of Geotechnical Engineering, 2001, 23(3): 261-267. (in Chinese) [百度学术]
KOHESTANI V R, BAZARGAN-LARI M R, MARNANI J A. Prediction of maximum surface settlement caused by earth pressure balance shield tunneling using random forest [J]. Journal of AI and Data Mining, 2017, 5: 127-135. [百度学术]
POURTAGHI A, LOTFOLLAHI-YAGHIN M A. Wavenet ability assessment in comparison to ANN for predicting the maximum surface settlement caused by tunneling [J]. Tunnelling and Underground Space Technology, 2012, 28: 257-271. [百度学术]
CHEN R P, ZHANG P, WU H N, et al. Prediction of shield tunneling-induced ground settlement using machine learning techniques [J]. Frontiers of Structural and Civil Engineering, 2019, 13(6): 1363-1378. [百度学术]
CAO Y, ZHOU X K, YAN K. Deep learning neural network model for tunnel ground surface settlement prediction based on sensor data [J]. Mathematical Problems in Engineering, 2021, 2021: 9488892. [百度学术]
ZHANG W G, ZHANG R H, WU C Z, et al. State-of-the-art review of soft computing applications in underground excavations [J]. Geoscience Frontiers, 2020, 11(4): 1095-1106. [百度学术]
赵勇, 王敏, 高文旗, 等. 分布式光纤传感新技术在盾构隧道结构变形监测中的应用[J]. 公路, 2017, 62(7): 326-329. [百度学术]
ZHAO Y, WANG M, GAO W Q, et al. Application of distributed optical fiber sensing technology in deformation monitoring of shield tunnel [J]. Highway, 2017, 62(7): 326-329. (in Chinese) [百度学术]
OCAK I, SEKER S E. Calculation of surface settlements caused by EPBM tunneling using artificial neural network, SVM, and Gaussian processes [J]. Environmental Earth Sciences, 2013, 70(3): 1263-1276. [百度学术]
WANG F, GOU B C, QIN Y W. Modeling tunneling-induced ground surface settlement development using a wavelet smooth relevance vector machine [J]. Computers and Geotechnics, 2013, 54: 125-132. [百度学术]
HASANIPANAH M, NOORIAN-BIDGOLI M, JAHED ARMAGHANI D, et al. Feasibility of PSO-ANN model for predicting surface settlement caused by tunneling [J]. Engineering with Computers, 2016, 32(4): 705-715. [百度学术]
GOH A T C, ZHANG W G, ZHANG Y M, et al. Determination of earth pressure balance tunnel-related maximum surface settlement: A multivariate adaptive regression splines approach [J]. Bulletin of Engineering Geology and the Environment, 2018, 77(2): 489-500. [百度学术]
MAHMOODZADEH A, MOHAMMADI M, DARAEI A, et al. Forecasting maximum surface settlement caused by urban tunneling [J]. Automation in Construction, 2020, 120: 103375. [百度学术]
ZHANG K, LYU H M, SHEN S L, et al. Evolutionary hybrid neural network approach to predict shield tunneling-induced ground settlements [J]. Tunnelling and Underground Space Technology, 2020, 106: 103594. [百度学术]
CHA Y J, CHOI W, BÜYÜKÖZTÜRK O. Deep learning-based crack damage detection using convolutional neural networks [J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(5): 361-378. [百度学术]
BESAW L E, STIMAC P J. Deep learning algorithms for detecting explosive hazards in ground penetrating radar data [C]//SPIE Defense + Security. Proc SPIE 9072, Detection and Sensing of Mines, Explosive Objects, and Obscured Targets XIX, Baltimore, Maryland, USA. 2014, 9072: 224-234. [百度学术]
MAKANTASIS K, PROTOPAPADAKIS E, DOULAMIS A, et al. Deep convolutional neural networks for efficient vision based tunnel inspection [C]//2015 IEEE International Conference on Intelligent Computer Communication and Processing. September 3-5, 2015, Cluj-Napoca, Romania. IEEE, 2015: 335-342. [百度学术]
CHEN Z Y, ZHANG Y P, LI J B, et al. Diagnosing tunnel collapse sections based on TBM tunneling big data and deep learning: a case study on the Yinsong Project, China [J]. Tunnelling and Underground Space Technology, 2021, 108: 103700. [百度学术]