Comparative Analysis of ARIMA and LSTM Model-Based Anomaly Detection for Unannotated Structural Health Monitoring Data in an Immersed Tunnel

Structural Health Monitoring(SHM)systems have become a crucial tool for the operational management of long tunnels.For immersed tunnels exposed to both traffic loads and the effects of the marine environment,efficiently identifying abnormal conditions from the extensive unannotated SHM data presents a significant challenge.This study proposed amodel-based approach for anomaly detection and conducted validation and comparative analysis of two distinct temporal predictive models using SHM data from a real immersed tunnel.Firstly,a dynamic predictive model-based anomaly detectionmethod is proposed,which utilizes a rolling time window for modeling to achieve dynamic prediction.Leveraging the assumption of temporal data similarity,an interval prediction value deviation was employed to determine the abnormality of the data.Subsequently,dynamic predictive models were constructed based on the Autoregressive Integrated Moving Average(ARIMA)and Long Short-Term Memory(LSTM)models.The hyperparameters of these models were optimized and selected using monitoring data from the immersed tunnel,yielding viable static and dynamic predictive models.Finally,the models were applied within the same segment of SHM data,to validate the effectiveness of the anomaly detection approach based on dynamic predictive modeling.A detailed comparative analysis discusses the discrepancies in temporal anomaly detection between the ARIMA-and LSTM-based models.The results demonstrated that the dynamic predictive modelbased anomaly detection approach was effective for dealing with unannotated SHM data.In a comparison between ARIMA and LSTM,it was found that ARIMA demonstrated higher modeling efficiency,rendering it suitable for short-term predictions.In contrast,the LSTM model exhibited greater capacity to capture long-term performance trends and enhanced early warning capabilities,thereby resulting in superior overall performance.

Application of transient Rayleigh wave in detection of tunnel lining void area

Transient Rayleigh wave detection is a high-precision nondestructive detection method.At present,it has been widely used in shallow exploration,but rarely used in tunnel lining quality detection.Through the tunnel lining physical model experiment,the layout defects of the double-layer reinforcement lining area were detected and the Rayleigh wave velocity profile and dispersion curve were analyzed after data process-ing,which finally verified the feasibility and accuracy of Rayleigh wave method in detecting the tunnel lining void area.The results show that the method is not affected by the reinforcement inside the lining,the shallow detection is less disturbed and the accuracy is higher,and the data will fluctuate slightly with the deepening of the detection depth.At the same time,this method responds quite accurately to the thickness of the concrete,allowing for the assessment of the tunnel lining’s lack of compactness.This method has high efficiency,good reliability,and simple data processing,and is suitable for nondestructive detection of internal defects of tun-nel lining structure.

Automatic Sensing and Detection for Subway Tunnel Pathologies

Subway tunnels often suffer from surface pathologies such as cracks,corrosion,fractures,peeling,water and sand infiltration,and sudden hazards caused by foreign object intrusions.Installing a mobile visual pathology sensing system at the front end of operating trains is a critical measure to ensure subway safety.Taking leakage as the typical pathology,a tunnel pathology automatic visual detection method based on Deeplabv3+(ASTPDS)was proposed to achieve automatic and high-precision detection and pixel-level morphology extraction of pathologies.Compared with similar methods,this approach showed significant advantages and achieved a detection accuracy of 93.12%,surpassing FCN and U-Net.Moreover,it also exceeded the recall rates for detecting leaks of FCN and U-Net by 8.33%and 8.19%,respectively.

Detection and treatment of water inflow in karst tunnel：A case study in Daba tunnel

In a karst tunnel, fissures or cracks that are filled with weathered materials are a type of potential water outlet as they are easily triggered and converted into groundwater outlets under the influence of high groundwater pressure. A terrible water inrush caused by potential water outlets can seriously hinder the project construction. Potential water outlets and water sources that surrounding the tunnel must be detected before water inflow can be treated. This paper provides a successful case of the detection and treatment of water inflow in a karst tunnel and proposes a potential water outlet detection(PWOD) method in which heavy rainfall(>50 mm/d) is considered a trigger for a potential water outlet. The Daba tunnel located in Hunan province, China, has been constructed in a karst stratum where the rock mass has been weathered intensely by the influence of two faults. Heavy rain triggered some potential water outlets, causing a serious water inrush. The PWOD method was applied in this project for the treatment of water inflow, and six potential water outlets in total were identified through three heavy rains. Meanwhile, a geophysical prospecting technique was also used to detect water sources. The connections between water outlets and water sources were identified with a 3-D graphic that included all of them. According to the distribution of water outlets and water sources, the detection area was divided into three sections and separately treated by curtain grouting.

Analysis of Detection and Treatment Schemes of Highway Tunnel Lining Cracks

Highway tunnels play a very important role in people's daily life.Among them,lining is an essential part of tunnel engineering,and the quality of lining greatly affects the overall quality of the tunnel.On this basis,the causes of lining cracks and the detection methods of existing highway tunnel lining cracks are analyzed,and the treatment countermeasures for highway tunnel lining cracks are proposed.

Spin detection and manipulation with scanning tunneling microscopy

Over the past few decades, spin detection and manipulation at the atomic scale using scanning tunneling microcopy has matured, which has opened the possibility of realizing spin-based functional devices with single atoms and molecules.This article reviews the principle of spin polarized scanning tunneling microscopy and inelastic tunneling spectroscopy,which are used to measure the static spin structure and dynamic spin excitation, respectively. Recent progress will be presented, including complex spin structure, magnetization of single atoms and molecules, as well as spin excitation of single atoms, clusters, and molecules. Finally, progress in the use of spin polarized tunneling current to manipulate an atomic magnet is discussed.

Detection of Terahertz Photon by GaAs-Based Single Electron Transistor at Low Temperatures

A reproducible terahertz （THz） photocurrent was observed at low temperatures in a Schottky wrap gate single electron transistor with a normal-incident of a CH3OH gas laser with the frequency 2.54THz. The change of source-drain current induced by THz photons shows that a satellite peak is generated beside the resonance peak. THz photon energy can be characterized by the difference of gate voltage positions between the resonance peak and satellite peak. This indicates that the satellite peak exactly results from the THz photon-assisted tunneling. Both experimental results and theoretical analysis show that a narrow spacing of double barriers is more effective for the enhancement of THz response.

基于红外热成像技术的隧道隐性渗漏检测研究

因隧道安装防火板,导致渗漏水隐蔽在防火板背后而不易被检测。红外热成像技术是一种非接触无损检测技术,因其对表面热辐射敏感而常用来测量表面温度场。基于红外热成像技术,研究了隧道隐性渗漏车载非接触式快速检测的可行性。结合某长江隧道,将红外热成像设备安置在车顶,对隧道内表面实施移动扫描拍照,并通过红外热力图准确识别既有渗漏点。为了进一步验证检测效果,人为在防火板表面涂湿模拟渗漏,该渗漏过程可以被准确识别。通过试验研究,表明红外热成像技术用于隧道隐性渗漏车载非接触式快速检测是可行性的。

Sand deposit-detecting method and its application in model test of sand flow

Against the background of the sand-flow foundation treatment engineering of Guangzhou Zhoutouzui variable cross-section immersed tunnel, a kind of sand deposit-detecting method was devised on the basis of full-scale model test of sand-flow method. The real-time data of sand-deposit height and radius were obtained by the self-developed sand-deposit detectors. The test results show that the detecting method is simple and has high precision. In the use of sand-flow method, the sand-carrying capability of fluid is limited, and sand particles are all transported to the sand-deposit periphery through crater, gap and chutes after the sand deposit formed. The diffusion range of the particles outside the sand-deposit does not exceed 2.0 m. Severe sorting of sand particles is not observed because of the unique oblique-layered depositing process. The temporal and spatial distributions of gap and chutes directly affect the sand-deposit expansion, and the expansion trend of the average sand-deposit radius accords with quadratic time-history curve.

Tunnel boring machine vibration-based deep learning for the ground identification of working faces

Tunnel boring machine(TBM) vibration induced by cutting complex ground contains essential information that can help engineers evaluate the interaction between a cutterhead and the ground itself.In this study,deep recurrent neural networks(RNNs) and convolutional neural networks(CNNs) were used for vibration-based working face ground identification.First,field monitoring was conducted to obtain the TBM vibration data when tunneling in changing geological conditions,including mixed-face,homogeneous,and transmission ground.Next,RNNs and CNNs were utilized to develop vibration-based prediction models,which were then validated using the testing dataset.The accuracy of the long short-term memory(LSTM) and bidirectional LSTM(Bi-LSTM) models was approximately 70% with raw data;however,with instantaneous frequency transmission,the accuracy increased to approximately 80%.Two types of deep CNNs,GoogLeNet and ResNet,were trained and tested with time-frequency scalar diagrams from continuous wavelet transformation.The CNN models,with an accuracy greater than 96%,performed significantly better than the RNN models.The ResNet-18,with an accuracy of 98.28%,performed the best.When the sample length was set as the cutterhead rotation period,the deep CNN and RNN models achieved the highest accuracy while the proposed deep CNN model simultaneously achieved high prediction accuracy and feedback efficiency.The proposed model could promptly identify the ground conditions at the working face without stopping the normal tunneling process,and the TBM working parameters could be adjusted and optimized in a timely manner based on the predicted results.

Frequency-Domain Full-Waveform Inversion Based on Tunnel Space Seismic Data

Tunnel seismic detection methods are effective for obtaining the geological structure around the tunnel face,which is critical for safe construction and disaster mitigation in tunnel engineering.However,there is often a lack of accuracy in the acquired geological information and physical properties ahead of the tunnel face in the current tunnel seismic detection methods.Thus,we apply a frequency-domain acoustic full-waveform inversion(FWI)method to obtain high-resolution results for the tunnel structure.We discuss the influence of the frequency group selection strategy and the tunnel observation system settings regarding the inversion results and determine the structural imaging and physical property parameter inversion of abnormal geological bodies ahead of the tunnel face.Based on the conventional strategies of frequency-domain acoustic FWI,we propose a frequency group selection strategy that combines a low-frequency selection covering the vertical wavenumber and a high-frequency selection of antialiasing.This strategy can effectively obtain the spatial structure and physical parameters of the geology ahead of the tunnel face and improve the inversion resolution.In addition,by linearly increasing the side length of the tunnel observation system,we share the influence of the length of the two sides of the observation systems of different tunnels on the inversion results.We found out that the inversion results are the best when the side length is approximately five times the width of the tunnel face,and the influence of increasing the side observation length beyond this range on the inversion results can be ignored.Finally,based on this approach,we invert for the complex multi-stratum model,and an accurate structure and physical property parameters of the complex stratum ahead of the tunnel face are obtained,which verifies the feasibility of the proposed method.

Application of high-density resistivity method to evaluate safety around Minyan Tunnel,Helong City

By determining the distribution and extent of geological structures surrounding the Mingyan Tunnel,Xicheng Town,Helong City,Jilin Province,we can evaluate the stability of the rock mass and assess potential hazards during tunnel construction.We use the high-density resistivity method to analyze the subsurface structure of the study area.Conductive anomalies are likely to represent joint and fissure systems within strongly weathered host rocks,and the bedrock surrounding the tunnel is relatively stable and does not contain well-developed faults.High-density resistivity analysis can provide valuable information in the context of tunnel engineering and safety.

Measurement of T wave in magnetocardiography using tunnel magnetoresistance sensor

Several critical clinical applications of magnetocardiography(MCG)involve its T wave.The T wave’s accuracy directly affects the diagnostic accuracy of MCG for ischemic heart disease and arrhythmogenic.Tunnel magnetoresistance(TMR)attracts attention as a new MCG measurement technique.However,the T waves measured by TMR are often drowned in noise.The accuracy of T waves needs to be discussed to determine the clinical value of MCG measured by TMR.This study uses an improved empirical mode decomposition(EMD)algorithm and averaging to eliminate the noise in the MCG measured by TMR.The MCG signals measured by TMR are compared with MCG measured by the optically pumped magnetometer(OPM)to judge its accuracy.Using the MCG measured by OPM as a reference,the relative errors in time and amplitude of the T wave measured by TMR are 3.4%and 1.8%,respectively.This is the first demonstration that TMR can accurately measure the time and amplitude of MCG T waves.The ability to provide reliable T wave data illustrates the significant clinical application value of TMR in MCG measurement.

Full-area Exploration Test of Deep Concealed Ore Bodies based on Tunnel Gravity

Metal mineral resources are the product of the deep material and energy exchange with its deep power process(Wan,2017).It is the foundation of contemporary national economic development,a priority area for the country’s implementation of strategic development.

Application of high-density resistivity method for assessing construction safety of Shimodong tunnel in Helong City of Jilin Province

Some unfavorable geological conditions can affect the construction of tunnels.In order to evaluate the damage degree of tunnel construction and determine the surrounding rock grade and stability of the tunnel,the authors used high-density resistivity method to detect the surrounding rocks of Shimodong tunnel in Xicheng Town of Helong City.The underground resistivity structures of the entrance,exit and middle parts of the tunnel are obtained.Through analysis,it is found that there are no bedrock faults near the tunnel,although some joints and fissures are developed in some locations,which are characterized by low-resistivity anomalies.The tunnel structures are stable overall,favorable for safe and efficient construction.The study also proves the good application effect of the high-density resistivity method in tunnel safety detection.

大直径盾构隧道成型质量巡检方法研究

针对因工业应用成本限制,中、小盾构隧道成型质量无损检测技术迁移至大直径盾构隧道时精度、速度折损严重的问题,以巡检车为载体,集成二维激光扫描仪、编码器和计算机等设备,研制了大盾构隧道成型质量巡检车,并提出一种基于数字图像的盾构质量非对称巡检方法.分析大直径盾构的施工环境,滤除地面、车体点云,并采用邻域向量法提取中轴线,建立隧道中心坐标系.偏心布置巡检路线,按照点云密度将采样点云分为稠密侧和稀疏侧点云,通过不同方法实现对管片接缝特征的拾取:将稠密侧点云绕中轴线展开为二维灰度图像,并通过缩放、归一化、梯度阈值分割等方法实现接缝图像分割;基于直线方程对接缝进行分类,结合管片结构、布置点位,推导出稀疏侧接缝与稠密侧接缝的线性分布公式,间接拾取稀疏侧接缝.根据接缝特征点计算两侧管片边缘点云簇,计算管片错台量;剔除接缝点云簇,使用最小二乘法拟合隧道点云,计算隧道椭圆度.最后在某大直径盾构隧道进行巡检试验,试验结果表明:成型质量巡检车在十四米盾构隧道中巡检速度为3 km·h-1,与传统方法的错台量检测偏差小于2 mm,椭圆度检测偏差小于0.1%,可以满足大直径盾构隧道成型质量巡检的高速度、高精度、低成本需求.

HSP超前探测技术在煤矿TBM掘进巷道中的应用研究

随着全断面掘进机TBM(Tunnel Boring Machine)逐渐应用于煤矿岩巷掘进,对不良地质构造进行超前准确快速预测的需求日益迫切。通过对主动源地震波超前探测方法的特点和TBM破岩震源超前探测技术的适用性进行分析,结合煤矿巷道地质和生产条件,提出了适用于煤矿巷道TBM掘进的HSP超前探测方法。以河南平顶山首山一矿TBM掘进底板瓦斯治理巷道为工程背景,选用防爆硬件一体化设计的探测仪器在煤矿巷道中进行应用。构建了空间型观测方式对煤矿巷道近水平煤线进行探测,优化了双护盾TBM掘进巷道狭小空间检波器阵列式布置参数;基于时频分析、互相关干涉处理、反射与散射联合反演等方法处理原始信号并进行探测结果成像。研究表明:采用空间型观测方式可实现与巷道小角度斜交煤线的识别,设计震源与首检波器间距离为15 m时最优。通过时频分析提取有效信号,利用互相关干涉法获取虚拟震源道和反射特征曲线,并结合反射与散射联合反演成像得到探测区域地层反射能量分布图,能够较准确地推测得到围岩存在的不良地质构造。通过比较现场开挖揭露情况与探测结果发现两者吻合度较高,表明HSP超前探测方法可实现掘进工作面前方100 m范围内超前无损地质预测,有助于提高煤矿岩巷TBM掘进速度。

综合物探方法在贵南高铁隧道基底隐伏岩溶探测中的应用

高铁隧道路线经过岩溶地区,勘察阶段由于地形和物探方法局限性的影响,不能完全勘察清楚岩溶发育情况。本文结合高密度地震映像、地质雷达、二维瞬态面波和钻探方法,详细研究了隧道施工阶段岩溶探查中各种方法的外业数据采集、处理方法、数据分析的注意事项。经过研究,将地震映像的主频提高到800 Hz,可分辨深度5~25 m、宽度10 m左右的空溶洞和充填溶洞;地质雷达有效能量提高到600 ns,可分辨深度30 m内岩溶通道。结合贵南高铁隧道基底岩溶探查中的各种典型图像特征,以及地质施工人员的要求,探讨了隧道基底岩溶探测资料物探地质解释的四类划分解释原则,满足了地质人员及设计施工对物探资料可读性的要求,为岩溶探测资料解释提供了参考。

改进Mask RCNN的盾构隧道渗漏水检测方法

渗漏水是盾构隧道结构存在潜在损伤或缺陷的重要表征,快速、准确检测出渗漏水位置,对隧道安全运营和维护具有重要意义。现有的方法大多采用光学影像对隧道渗漏水进行检测,受隧道内空间和光线条件限制,难以获得高质量病害图片。因此,本文提出了一种基于激光点云数据与改进Mask RCNN相结合的渗漏水检测方法。首先对激光点云反射强度进行修正;然后生成灰度图像并建立渗漏水病害数据集;最后在Mask RCNN算法中引入空洞卷积和变形卷积,实现了隧道渗漏水病害的快速检测。利用某地铁采集的数据进行验证,结果表明,本文提出的改进Mask RCNN算法相较于原始算法和FCN算法检测精度均有明显提升,在盾构隧道渗漏水识别方面性能表现较好。

基于三维激光扫描技术的竣工盾构隧道渗漏水检测

渗漏水是隧道常见的病害,长期渗漏会导致隧道结构开裂、裂缝、钢筋腐蚀,威胁到隧道运营安全,因此对其检测一直受到重视。本文利用三维激光扫描技术实现了竣工盾构隧道渗漏水的位置及面积的自动检测。首先采用架站式三维激光扫描仪采集竣工隧道点云,基于修正后的反射强度值生成隧道衬砌表面灰度图,再采用膨胀法与腐蚀法对灰度图像进行预处理;然后利用连通域算法计算渗漏水位置及其面积;最后结合实际工程验证本文方法的实用性及准确性。结果表明,应用本文方法竣工隧道的渗漏水检测准确率达92%。