No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
An assessment of anomaly detection methods applied to microtunnelling
Abstract
The proliferation of data collected by modern tunnel boring machines presents a substantial opportunity for the application of data-driven anomaly detection (AD) techniques that can adapt dynamically to site specific conditions. Based on jacking forces measured during microtunnelling, this paper explores the potential for AD methods to provide more accurate and robust detection of incipient faults. A selection of the most popular AD methods proposed in the literature, comprising both clusteringand regression-based techniques, are considered for this purpose. The relative merits of each approach is assessed through comparisons to three microtunnelling case histories where anomalous jacking force behaviour was encountered. The results highlight an exciting potential for the use of anomaly detection techniques to reduce unplanned downtimes and operation costs.
... One-class support vector machine (OCSVM) denotes the case where the data comprises only one class and the task is to identify whether new measurements belong to that class. Schölkopf et al. (2001) framed the OCSVM approach by considering the origin as the only member of the second class (Sheil et al., 2020). A hyperplane is constructed in the feature space to separate the dataset from the origin, using a maximal margin (Fig. 1). ...
... Partition creation continues until all datapoints are isolated; in most cases a limit is placed on the maximum number of partitions. Multiple training datasets are produced by sampling with replacement randomly from the original dataset, and anomalies are ultimately identified by sorting datapoints according to their corresponding path lengths (Sheil et al., 2020). For a dataset of size n, the average, c(n), of each path length, h(x), is calculated as: ...
‘Clogging’ is a common issue encountered during tunnelling in clayey soils which can impede tunnel excavation, cause unplanned downtimes and lead to significant additional project costs. Clogging can result in a drastic reduction in performance due to reduced jacking speeds and the time needed for cleaning if it cannot be fully mitigated. The data acquired by modern tunnel boring machines (TBMs) have grown significantly in recent years presenting a substantial opportunity for the application of data-driven artificial intelligence (AI) techniques. In this study, a baseline assessment of clogging in slurry-supported pipejacking is performed using a combination of TBM parameters and the semi-empirical diagram proposed in the literature. The potential for one-class support vector machines (OCSVM), isolation forest (IForest) and robust covariance (Robcov) to assess the tendency to clayey clogging is then explored in this work. The proposed approach is applied to a pipejacking case history in Taipei, Taiwan, involving tunnelling in soft alluvial deposit. The results highlight an exciting potential for the use of OCSVM, IForest and Robcov to detect clogging during slurry-supported pipejacking.
... Data collected from modern TBMs presents a new opportunity for improved machine learning (ML) predictive methods (Sheil et al., 2020). For traditional large-diameter tunnelling, applications of ML to TBM performance prediction have soared, e.g. ...
Purpose
A common design driver for pipe-jacking projects is the jacking force required to advance the tunnel boring machine and pipe string. Empirical methods are popular in industry but are well known to lack accuracy, while there is a strong desire to supplement such approaches with robust data-driven techniques, typically small construction datasets present significant challenges.
Design/methodology/approach
To address this challenge, this paper develops a physics-constrained neural network predictive model for pipe-jacking forces. Information used as input into the model includes principal design information and soil type.
Findings
The physics constrained model was found to predict jacking force to a higher accuracy than current industry practice and better discern meaningful patterns in data than a purely data-driven artificial neural network. The results reveal promising performance for this initial dataset such that there is motivation, as a longer-term objective, to train the present approach on a more comprehensive drive database for more reliable and cost effective solutions for new projects.
Originality/value
Novel contributions include (a) a bespoke framework to constrain a neural network using a pipe-jacking mechanistic model which includes stoppage-induced friction increases, (b) built-in model uncertainty for greater confidence in model outputs, (c) new historical drive data for model training and (d) one-hot encoding of soil type as a model input. The model is calibrated and validated against 14 tunnel drives across four different sites with four distinctive ground types.
... Intelligent method gains rapid advancements in data collection and transmission technologies have enabled tunnel engineers to gather a vast amount of raw data generated by TBM operations, presenting significant opportunities for applying advanced machine learning techniques to predict TBM performance [27]. According to recent studies, random forest (RF) and support vector machine (SVM) are the two most extensively utilized models for predicting the on-site excavation performance of TBMs [28]. ...
Due to the uncertainty of geological conditions during the tunneling process, advanced prediction of TBM tunneling parameters is significant for evaluating operational safety and efficiency, especially for real-time prediction of key tunneling parameters during the steady phase of TBM operation. At present, although there are studies on constructing predictive models based on machine learning algorithms, multiparameter prediction consistent with the actual tunneling process remains challenging due to the complexity of the TBM tunneling process and the numerous tunneling parameters. Therefore, this paper proposes a real-time multiple tunneling parameters prediction method of TBM steady phase based on dual recurrent neural networks. Firstly, the irregular multidimensional time series of tunneling parameters are analyzed and processed, which are divided into an idle-push phase, a rising phase, and a steady phase; secondly, the parameters of rising phase are analyzed using a recurrent neural network, and the parameters relevant for constructing a real-time prediction model are screened; then, based on the screened parameters, the Bayesian-optimized gated recurrent unit (GRU, a kind of recursive neural network) is proposed to construct a real-time prediction model for the four key tunneling parameters during the steady phase. Finally, the effectiveness and practicality of the proposed method are demonstrated by verification on real TBM tunnel datasets and comparing it with the models constructed by six commonly used machine learning algorithms. The results of this paper show that the designed prediction method is able to achieve a good combination of performance in terms of accuracy and computational time-consumption, with an average prediction accuracy of 91.1% for the four parameters for different rock grades of geology, the multiparameter prediction time for 100 samples is only 11 ms. In addition, three current similar studies using deep learning methods were compared to demonstrate the superiority of this proposal. As a method more closer to practical application, this work provides guidance for the forward-looking prediction of TBM tunneling parameters.
... misidentifying regular data as an outlier. In micro tunelling, Sheil et al. [20] proposed a comparison of various clustering-and regression-based methods to detect anomalies using jacking force. It is further described in the paper how the detected anomalies help to decrease unplanned downtimes and operation costs of the project. ...
A major concern in urban mechanised tunnelling projects is avoiding damage to the existing buildings and the tunnel boring machine (TBM), which may be adjusted by an advanced precise excavation simulation. Because a realistic simulation must account for multiple interactions between the boring machine and the subsurface, an exact representation of the ground’s geological profile must be created beforehand. Due to the limited monitoring and sampling, several geologic anomalies may have been overlooked when sketching the geologic profile. As a result, the geological profile should be updated alongside the construction phases when new information becomes available. To accomplish this, one can use the boring machine’s recorded data to detect any irregularities in the drilling process caused by changes in geological conditions. This research compares various cutting-edge anomaly detection approaches on time series. Due to a large amount of sensor data, the visualization of multiple sensors/features over time was first performed, and the critical features with the highest impact on the detection process for identifying anomalies were selected. Anomaly detection techniques include isolation forest, k-Means, k-Means Sequential Time Series Cluster, Auto-Regression Integrated Moving Average (ARIMA), and Convolutional Neural Network (CNN) Auto-encoders are among the main aspects. The methods presented here were applied to a given data set from an actual tunnelling operation in Germany to locate the location of some concrete slabs in a relatively homogeneous ground. The obtained results agree well with the exact location of anomalies. The performance of various methods is evaluated through error quantification measures.
... Today, regression and clustering-based methods were more commonly used in anomaly detection. Sheil et al. [23] used the methods based on clustering and regression to detect abnormal jacking forces in the process of micro-advancement and made their own evaluation of the two methods. Grima et al. [24] proposed the neuro-fuzzy method to model the tunnel boring machine, which pioneered the use of fuzzy theory combined with neural networks to solve geological engineering problems. ...
The shield machine (SM) is a complex mechanical device used for tunneling. However, the monitoring and deciding were mainly done by artificial experience during traditional construction, which brought some limitations, such as hidden mechanical failures, human operator error, and sensor anomalies. To deal with these challenges, many scholars have studied SM intelligent methods. Most of these methods only take SM into account but do not consider the SM operating environment. So, this paper discussed the relationship among SM, geological information, and control terminals. Then, according to the relationship, models were established for the control terminal, including SM rate prediction and SM anomaly detection. The experimental results show that compared with baseline models, the proposed models in this paper perform better. In the proposed model, the R2 and MSE of rate prediction can reach 92.2\%, and 0.0064 respectively. The abnormal detection rate of anomaly detection is up to 98.2\%.
... In [36], an artificial neural network was trained to predict tunneling faults. However, abnormalities in time series data are also significant when analyzing the status of a TBM [37]; these abnormalities are normally called outliers. The existence of outliers in a time series can have a dramatic influence on the time series segmentation because only a few outliers are sometimes sufficient to distort the global performance of a mathematical model and the trend of the time series [38]. ...
The segmentation of tunnel boring machine (TBM) time series plays a crucial role in analyzing TBM operating statuses and mining potential information from the collected signals. In this paper, a novel algorithm named SDP-RLR is proposed to segment multivariate TBM time series with outliers caused by harsh operating environments and changeable tunneling statuses. In this algorithm, the 3σ rule is extended to externally studentized residuals of a linear regression and used to identify the outliers in each segment of the input time series. An outlier removal penalty term is added to the segment errors to avoid regarding borderline data points as outliers. As a segmentation optimization algorithm, dynamic programming (DP) is improved to scalable dynamic programming (SDP) in this paper by combining time series in two stages to reduce computational costs. In the first stage, consecutive time points are integrated to reduce the calculations required for DP, while the second stage refines the segmentation results obtained from the first stage. Experiments are conducted on synthetic datasets to evaluate the performance of the SDP-RLR algorithm, and a complete TBM time series in terms of three key variables validates its effectiveness for segmentation and outlier detection tasks. The SDP-RLR algorithm can segment a TBM time series into four different statuses, which assists in analyzing the operating statuses of TBMs. In addition, comparative experiments indicate that the proposed algorithm can handle the segmentation of complex TBM time series when they cannot be successfully processed using the segmentation approach after outlier detection.
... The problem of pile penetration and its disturbance to the surrounding soil have been studied and discussed by many scholars (Randolph and Wroth, 1979;Hwang et al., 2001;Yetginer et al., 2006;Salgado and Prezzi, 2007;Sheil et al., 2018;Cheng et al., 2018aCheng et al., , 2018bCheng et al., , 2020c. The jacked-pile penetration is a complex process which involves local large deformation, discontinuities and the change of stress state. ...
Non-intrusive observation of the spatial deformation field of the soil has been a difficult problem for model test measurements. Based on the advantage that the visualization test method of transparent soil can observe the two-dimensional (2D) deformation inside the soil, this study proposes a new automatic tomographic scanning measuring device to observe the three-dimensional (3D) spatial soil deformation inside the transparent soil model. A series of 2D laser speckle images of different vertical cross sections before and after deformation were obtained, and an improved 3D reconstruction algorithm was used to reconstruct the 3D displacement field of soil after deformation. Different types of jacked-pile penetration model tests were carried out to investigate the spatial disturbance of the soil around the pile caused by the squeezing effect of the jacked-pile. The test results showed that the developed novel automatic tomographic scanning measuring device with the modified 3D reconstruction procedure could be an innovative tool in geotechnical physical model experiments. The model test results visually revealed the mechanism of the soil squeezing effect of jacked-piles with different pile head forms. Moreover, the spatial disturbance effect caused by different penetration stages was also discussed herein.
... Many researchers have studied the leakage of infrastructure. Some researchers have studied the seepage impact on the interface behavior of tunnel [6][7][8]. Peng and Wang [9] and Zhao et al. [10] studied the inductive factors and nondestructive detection methods of earth dam leakage. The seepage and scour mechanism of the contact surface between culvert pipe and dam body has also been studied by some researchers [11,12]. ...
Plain reservoir plays an important role in alleviating water shortage in plain areas which are generally crowded with large populations. As an effective and cheap anti-seepage measure, geomembrane is widely applied in plain reservoirs. Therefore, it is necessary to investigate the seepage discharge caused by composite geomembrane leakage. The laboratory test and numerical calculation are carried out in this paper to analyze the influence of three factors (i.e., water head, leakage size, and leakage location) on seepage discharge. It is found from the results of the orthogonal and single-factor analysis that the impact order of the three factors on the seepage discharge of plain reservoir is: distance from dam toe > water head > leakage size. Moreover, the seepage discharge increases as the water head, leakage size, and leakage quantity increase, in a linear relation. The opposite trend can be sawed in the seepage discharge when the distance from dam toe rises. Furthermore, a threshold distance is innovatively presented based on the results of numerical analysis. The ranking of three factors has enlightening significance for future scholars to track and study key issues of the leakage of composite geomembrane. The threshold distance presented in this paper is beneficial for engineers to manage and maintain the reservoir. Generally, the findings of this study can be beneficial to deepen the understanding of the influence of composite geomembrane leakage on the plain reservoirs.
... To examine the causes, Terzaghi [1] developed the so-called trapdoor apparatus to conduct model testing. The trapdoor apparatus is considered as an effective tool and, since then, has been either replicated to study the arching effect [2][3][4][5]or modified in many studies for different purposes or test conditions, such as a depleting petroleum reservoir [6], buried pipelines [7], surface subsidence due to underground tunneling [8], sinkhole development [9], and microtunnelling and pipe-jacking [10][11][12][13][14]. In the latter situation, the arching effect is considered in the calculation of friction resistance along the pipe string because as the microtunnel boring machine advances into the ground, the soil being driven is yielded and the active case of arching can be used to estimate the new vertical stresses [11]. ...
Arching is regarded as a ubiquitous phenomenon in geomaterials where a new stress field is created following partial, induced displacements in the geomaterial. In the past, less attention has been paid to the deformations associated with the arching effect. The Digital Image Correlation (DIC) technique was implemented to map the shear and volumetric strain fields where arching occurred in a dense sand layer placed in a trapdoor apparatus. Model tests were conducted under active and passive arching along with surcharge loading. It was observed that shear bands comprising of two sections with different orientation angles were formed. The orientation angles of the shear bands were associated with the dilation angle of the sand (the deformation related angle of the soil) rather than its internal angle of friction (the stress related angle of the soil). The magnitude of shear strains increased with trapdoor displacement. It was also found that strain localization was associated with dilation. Contraction zones were also detected in the sand layer specifically at the surface and on the trapdoor element. In addition, it was realized the shear bands developed in the passive arching condition possessed greater width and inclination angles compared to bands of the active arching situation. The local surcharge used in the experiments markedly affected the strain localization and volumetric strain patterns and magnitudes.
... Further, the feature space of the RBF kernel is capable of extending to an infinite number of dimensions. The RBF kernel is, therefore, selected here (Sheil et al. 2020). ...
There frequently exists inadequacy regarding the number of boreholes installed along tunnel alignment. While geophysical imaging techniques are available for pre-tunnelling geological characterization, they aim to detect specific object (e.g. water body and karst cave). There remains great motivation for the industry to develop a real-time identification technology relating complex geological conditions with the existing tunnelling parameters. This study explores the potential for the use of machine learning-based data driven approaches to identify the change in geology during tunnel excavation. Further, the feasibility for machine learning-based anomaly detection approaches to detect the development of clayey clogging is also assessed. The results of an application of the machine learning-based approaches to Xi’an Metro line 4 are presented in this paper where two tunnels buried in the water-rich sandy soils at depths of 12-14 m are excavated using a 6.288 m diameter EPB shield machine. A reasonable agreement with the measurements verifies their applicability towards widening the application horizon of machine learning-based approaches.
... Zhou et al. (2019) proposed an approach for predicting the load on tunneling machine, which can correctly indicating the geological variation during excavation. Sheil et al. (2020) has carried out incipient fault detection of tunnel driving based on clustering and regression, and found that the abnormal detection techniques has potential in reducing the unplanned downtimes and operation costs. Zhang et al. (2020) used hybrid meta-heuristic and machine learning algorithms, reinforcement learning to study the tunneling-induced settlement. ...
The soil discharged by an earth pressure balanced (EPB) shield machine reflects ground loss and controls surface settlement. In engineering, the discharge was mainly estimated by weighing the tunnelling spoil, however, human disturbance and information lag bring serious hidden security risks to tunnel construction. Through the analysis of conveying flow and force of screw conveyor, an energy model of interaction between screw conveyor and soil discharged was proposed and verified experimentally. Then, the energy model was combined with the discharge equation and the force balance equation, and a semi-analytical method to fast estimate the soil discharged of the EPB shield machine was proposed. Taking an EPB shield construction project as an example, according to the data of earth pressure, the torque and the rotation of the screw conveyor recorded by EPB shield machine, the real-time soil discharged of EPB shield machine can be accurately estimated. The mean absolute error (MAE) of this method is 3.177 × 103 kg compared with the measured weight of tunnelling spoil, which verifies the rationality of the method and provides a method for further effective evaluation and control of ground loss.
... It generated relatively high excess pore water pressure to lead to liquefaction in the sliding surface. A large body of research has studied the interface behaviour (Cheng et al. 2018(Cheng et al. , 2019a(Cheng et al. , b, 2020bSheil et al. 2020;Wang et al. 2020;Duan et al. 2021). The result given by Sassa and Wang (2005) explained the landslides could occur in very dry season, with remarkably high mobility. ...
The Loess Plateau can be considered as a landslide-prone area in northwest China. The genera consensus about the interaction between landslide deposit and terrace sediments is not well studied; this paper summarised 40 loess landslides in the South Jingyang Platform, Shaanxi Province, China to help understand of this issue. Four of the loess landslides with high mobility have been analysed in detail. Three trenches T1, T2, and T3 dug after the loess landslides LD37, LD11 and LD38 highlighted the landslide-induced changes in geomorphology and internal geometry of geology, respectively. Furthermore, observation of upwards seepage flow on the profile of trench T3 is believed to be the trigger of the high speed, and long runout flowslides in the study area. A newly developed sandbox apparatus is used to reproduce the landslide kinematics due to a mass travelling over an inclined plane. The sandbox experiments show that the sediments are sheared and pushed upwards after the collision with the deposits. The deposits are then wrapped in a space between sediments, which tends to form the 'sandwich' structure. The distal sediments are thrust when the loess deposits' kinetic energy consistently dissipates, developing the accumulated folded strata. These results reveal the deposits' interactions with the sediments in the study area and provide key guideposts regarding prevention and mitigation of loess landslide hazards.
... Before starting these tunnel construction applications, experimental studies are carried out in the laboratory environment to determine the strength and other engineering properties of soil samples brought from the field. According to these results, necessary design parameters are created, and land applications are started (Cheng et al., 2018a(Cheng et al., , 2018bSheil et al., 2020). ...
Recently, utilization of wastes has become increasingly important in aspects of sustainability and waste disposal. The aim of this study is to investigate the feasibility of using glass manufacturing waste (GMW) for the first time in the relevant literature, geogrids and GMW-geogrid reinforcement to increase the load bearing capacity of a clayey soil (CS) by large-scale model experiments. Initially, optimum water ratios were determined for different ratios of GMW and CS. Secondly, the optimum mixing ratio of GMW was determined. Next, the optimum soil depth improved with GMW and the optimum number of geogrids were investigated. Consequently, a bearing capacity ratio (BCR) increase of 2.23 times was acquired and the optimum mixture ratio was attained for 25% GMW addition. Improvement depth (Rd) and optimum number of geogrids (N) were obtained as 1.75D (D: diameter of the model foundation) and 3, respectively. The highest degree of improvement was reached for geogrids-GMW combination and a minimum waiting period of 7 days was advised in the sequel of such soil improvement work. Out of the observation of micro-structural analyses results, both physical and chemical interactions were encountered throughout the occurrence period of bearing capacity improvements.
... The discusser also disagrees that geological prediction is the only viable role for forecasting techniques in tunnelling. Accurate forecasting of TBM performance in the short-term is desirable for TBM anomaly detection [12] as well as boulder detection [9]. In this case, the ML model learns 'healthy' behaviour thereby developing a model of 'normality' from which anomalous behaviour may then be inferred. ...
... Urbanization, assisted by the innovative approaches (Cheng et al., 2019a(Cheng et al., , 2019bQiao et al., 2019;Song et al., 2019;Tian et al., 2019;Cheng et al., 2020a;Sheil et al., 2020), appears to be the first priority for many megacities worldwide. However, the urbanization process is typically environmentally damaging and can lead to geo-hazards despite the application of many countermeasures (Modoni et al., 2006;Modoni and Bzòwka, 2012;Modoni et al., 2016;Cheng et al., 2018aCheng et al., , 2018bCheng et al., , 2018cDuan et al., 2018;Chen et al., 2018;Cheng et al., 2020b;Gu et al., 2019;Shao, 2019a, 2019b). ...
The Loess Plateau has been deemed as a landslide-prone area in northwest China because of the unique platform geomorphology and the wetting-induced loess collapse. The interactions of landslide deposit and terrace sediment have been under-explored in the literature. This lack of research has inhibited the prevention and mitigation of loess landslide. This study summarises a total of 40 loess landslides in the South Jingyang Platform, Shaanxi Province; 4 out of the 40 loess landslides are investigated in detail, with an emphasis on the geomorphology feature and the internal geometry of geology. While the sandbox experiments and the discrete element modelling primarily aimed to reproduce the kinetic process of landslide deposit falling from the platform edge and colliding with terrace sediments. The field observation distinguished three domains that represent varying degrees of interaction between the landslide deposit and the terrace sediments, namely push forward domain, shear up/out domain and original terrace sediment domain. The push forward domain is defined as an area completely contained the loess deposit, with most distinct surface upheaval, while the shear up/out domain is defined as an area that significantly interacts with the push forward domain and possesses remarkable evidence of interactions. The original terrace sediment is defined as an area that is thoroughly not disturbed by the interactions. The internal geometry change and the geomorphology features, induced by the interactions of the deposit with the sediments, are reproduced using the sandbox experiments and the discrete element modelling. The sediments shearing upwards and the occurrence of shear liquefaction are interpreted from perspectives of the velocity of deposit movement and the apparent friction angle. The results are deemed to be useful in enhancing our understanding about the interactions of the landslide deposit with the terrace sediments and countermeasures against loess flowslides in the study area in future.
... A wide range of ML techniques have been developed for tunnelling applications. Research areas have included TBM automation(Mokhtari and Mooney, 2019), tunnel condition assessmentLi et al., 2017;Zhu et al., 2020), anomaly detection (e.g.Sheil et al., 2020;Yu et al., 2018), tunnel profile measurement (e.g.Xue and Zhang, 2019), resilience assessment (e.g.Khetwal et al., 2019), structural defect identification (e.g.Ding et al., 2019), tunnel face stability (e.g. Hayashi et al., 2019), rockburst prediction (e.g. ...
The proliferation of data collected by modern tunnel boring machines (TBMs) presents a substantial opportunity for the application of machine learning (ML) to support the decision-making process on site with timely and meaningful information. The observational method is now well-established in geotechnical engineering and has a proven potential to save time and money relative to conventional design. ML advances the traditional observational method by employing data analysis and pattern recognition techniques, predicated on the assumption of the presence of enough data to describe the modelled system’s physics. This paper presents a comprehensive review of recent advances and applications of ML to inform tunnelling construction operations with a view to increasing their potential for uptake by industry practitioners. This review has identified four main applications of machine learning to inform tunnelling, namely TBM performance prediction, tunnelling-induced settlement prediction, geological forecasting and cutterhead design optimisation. The paper concludes by summarising research trends and suggesting directions for future research for ML in the tunnelling space.
The “regional advantage” hypothesizes that inference uncertainty/prediction error of geotechnical or geological properties at a target site (a site contains a group of records measured from different locations/depths using a variety of tests) can be smaller if we use a quasi-regional cluster (includes two or more database sites with geotechnical or geological properties similar to the target site) instead of the entire database. A tailored clustering enabled regionalization (TCER) framework has been proposed to verify this “regional advantage” hypothesis. TCER requires the target site should not be an outlier site relative to the database. However, it remains a challenge on how to detect an outlier site (or data group) from a database. In this paper, we modify the original TCER by introducing a novel outlier site detection step called maximum site similarity (MSS) into the original TCER. The capability of MSS is verified using synthetic and real examples. Additionally, three inference methods [e.g., probabilistic multiple regression (PMR), classical Bayesian model (CBM), and hierarchical Bayesian model (HBM)] are studied for the purpose of determining the optimal inference method for the modified TCER in terms of achieving the minimum inference uncertainty/prediction error with reasonable computational time. It is shown that the modified TCER with CBM outperforms other inference methods for the examples shown in this paper.
This study proposes a novel method that addresses a nontraditional class of outlier detection problems. The purpose of most outlier detection methods in the literature is to detect outliers within a dataset. A record can be considered an outlier if it is distinct from the regular records in the dataset. However, the purpose of the novel outlier detection method proposed in this study is to detect outlier data groups (a data group may denote a site or a project) with respect to a soil/rock property "MUSIC" database. A data group is an outlier group if its characteristics (mean, variance, correlation, or higher order dependency) are distinct from the regular data groups in the database. This study frames the outlier detection problem into a formal hypothesis testing problems with the null hypothesis that “the target data group is identically distributed as the regular groups in the database.” With the hierarchical Bayesian model previously developed by the first two authors, the p-value for this hypothesis testing problem can be estimated rigorously. Numerical and real examples show that the p-value can effectively detect outlier data groups as well as outlier records with respect to a database.
Robust optimization is an ideal solution for enhancing safety in tunnel construction in the presence of unpredictable soil conditions, especially in large-diameter tunnel construction, since it requires the least amount of information about uncertainties. However, the application of robust optimization to real-world projects is greatly hampered by its dependence on mathematical models. To address this issue, this study builds a pipeline machine learning model to forecast tunnel-induced damage that can be addressed using the robust optimization (RO) algorithm with high accuracy. The optimization process is integrated into a building information modeling (BIM) platform and analyzed using the Shapley Additive ExPlanations (SHAP) technique, allowing the designer to understand and interact with the algorithm. The average improvement of testing samples using an ellipsoidal uncertainty set with a size of 0.05 is 23.8 and 4.9% on the two selected criteria, which is more conservative than using deterministic optimization (DO) and stochastic optimization (SO). This study establishes an interactive and explainable optimization platform that enables designers to make judgments under the most unfavorable soil conditions with the least amount of accessible information about the uncertainties during tunneling.
The failure criteria of practical soil mass are very complex, and have significant influence on the safety factor of slope stability. The Coulomb strength criterion and the power-law failure criterion are classically simplified. Each one has limited applicability owing to the noticeable difference between calculated predictions and actual results in some cases. In the work reported here, an analysis method based on the least square support vector machine (LSSVM), a machine learning model, is purposefully provided to establish a complex nonlinear failure criterion via iteration computation based on strength test data of the soil, which is of more extensive applicability to many problems of slope stability. In particular, three evaluation indexes including coefficient of determination, mean absolute percentage error, and mean square error indicate that fitting precision of the machine learning-based failure criterion is better than those of the linear Coulomb criterion and nonlinear power-law criterion. Based on the proposed LSSVM approach to determine the failure criterion, the limit equilibrium method can be used to calculate the safety factor of three-dimensional slope stability. Analysis of results of the safety factor of two three-dimensional homogeneous slopes shows that the maximum relative errors between the proposed approach and the linear failure criterion-based method and the power-law failure criterion-based method are about 12% and 7%, respectively.
Living in the twenty-first century means designing, constructing, operating, and decommissioning infrastructure in ways that are good for planet and people. Over the last decade, research and discussion of this imperative has evolved from calls to ‘green the built environment’, through to calls for ‘sustainable design and construction’, and more recently ‘low-carbon design and construction’ for sustainable built environment.
The mechanistic behavior of a tunnel boring machine (TBM) starts drawing great attention from scientists and engineers in recent years because it appears to affect the efficiency of tunnel excavations and project costs. Factors affecting the mechanistic behavior of tunnel boring machine primarily include geology and jamming phenomena. The following section details how the two factors influence the mechanistic behavior of TBM.
Machine learning models have been widely used for landslide susceptibility assessment (LSA) in recent years. The large number of inputs or conditioning factors for these models, however, can reduce the computation efficiency and increase the difficulty in collecting data. Feature selection is a good tool to address this problem by selecting the most important features among all factors to reduce the size of the input variables. However, two important questions need to be solved: (1) how do feature selection methods affect the performance of machine learning models? and (2) which feature selection method is the most suitable for a given machine learning model? This paper aims to address these two questions by comparing the predictive performance of 13 feature selection-based machine learning (FS-ML) models and 5 ordinary machine learning models on LSA. First, five commonly used machine learning models (i.e., logistic regression, support vector machine, artificial neural network, Gaussian process and random forest) and six typical feature selection methods in the literature are adopted to constitute the proposed models. Then, fifteen conditioning factors are chosen as input variables and 1,017 landslides are used as recorded data. Next, feature selection methods are used to obtain the importance of the conditioning factors to create feature subsets, based on which 13 FS-ML models are constructed. For each of the machine learning models, a best optimized FS-ML model is selected according to the area under curve value. Finally, five optimal FS-ML models are obtained and applied to the LSA of the studied area. The predictive abilities of the FS-ML models on LSA are verified and compared through the receive operating characteristic curve and statistical indicators such as sensitivity, specificity and accuracy. The results showed that different feature selection methods have different effects on the performance of LSA machine learning models. FS-ML models generally outperform the ordinary machine learning models. The best FS-ML model is the recursive feature elimination (RFE) optimized RF, and RFE is an optimal method for feature selection.
The use of supporting fluids to stabilise excavations is a common technique adopted in the construction industry. Rapid detection of incipient collapse for deep excavations and timely decision making are crucial to ensure safety during construction. This paper explores a hybrid framework for forecasting the collapse of fluid−supported circular excavations by combining physics-based and data-driven modelling. Finite element limit analysis is first used to develop a numerical database of stability numbers for both unsupported and fluid−supported circular excavations. The parameters considered in the modelling include excavation geometry, soil strength profile and support fluid properties. A data-driven algorithm is used to ‘learn’ the numerical results to develop a fast ‘surrogate’ amenable for integration within real−time monitoring systems. By way of example, the proposed forecasting strategy is retrospectively applied to a recent field monitoring case history where the observational method is used to update the input parameters of the data-driven surrogate.
In the area of high sulfate concentration in China, the problem of sulfate corrosion in coal mine shaft is increasingly prominent. Currently, shaft repair methods are limited to diversion, interception, and backwall grouting. However, after conventional cement stabilization, shafts still must contend with poor durability and weak resistance to sulfate corrosion. To solve these problems, this study combined theoretical research, laboratory tests, and field tests to reveal the corrosion mechanism of shafts, explored novel anti-sulfate corrosion grout, and built a model for the migration of sulfate ions (SO42−) in strata and shafts. According to the results of this study, shaft corrosion was a process of continuous penetration towards the concrete interior following a cycle of concrete compaction, expansion, and cracking. Laboratory tests show that the specimens of ordinary Portland cement mortar with 20–30% fly ash not only have good long-term strength, but also have a significantly improved resistance capacity to sulfate solution. Based on the laboratory test results, the HSR42.5 mixed with 20% fly ash was used for wall grouting, and the water-cement ratio of the slurry was 1:1. After wall grouting, the water inflow was effectively controlled in the grouting area in the main shaft, auxiliary shaft, and air shaft, reducing from 18.5 m3/h, 20.9 m3/h, and 10.0 m3/h to 3.5 m3/h, 4.6 m3/h, and 3.2 m3/h, respectively. Moreover, after nearly three years of continuous monitoring, the water inflow did not show any significant increase in the shafts. Based on the migration law of SO42− in concrete, a shaft geological model and a salt solution migration model after grouting reinforcement under sulfate corrosion conditions were constructed. In addition, the analysis of the migration law of sulfate ion in the shaft and grouting reinforcement formation revealed that the service period of the grouting shaft can be extended for approximately 6–8 years.
An accurate estimation of the jacking forces likely to be experienced during microtunnelling is a key design concern for the design of pipe segments, the location of intermediate jacking stations and the efficacy of the pipe jacking project itself. This paper presents a Bayesian updating approach for the prediction of jacking forces during microtunnelling. The proposed framework is applied to two pipe jacking case histories completed in the UK including a 275 m drive in silt and silty sand and a 1237 m drive in mudstone. To benchmark the Bayesian predictions, a ‘classical’ optimisation technique, namely genetic algorithms, is also implemented. The results show that predictions of pipe jacking forces using the prior best estimate of model input parameters provide a significant over-prediction of the monitored jacking forces for both drives. This highlights the difficulty in capturing the complex geotechnical conditions during tunnelling within prescriptive design approaches and the importance of robust back-analysis techniques. Bayesian updating is also shown to be a very effective option where significant improvements in the mean predictions, and associated variance, of the total jacking force are obtained as more data is acquired from the drive.
The filter cake on excavation face would be destructed periodically by cutting tools during slurry shield tunneling. The broken filter cake has a risk for face stability. The influence of cutting tools on filter cake was studied in this paper. A slurry-soil interaction model based on a multiphase flow theory which considers the solid particles and fluid in slurry was developed. The whole process of slurry penetration and filter cake formation on the excavation face of slurry shield can be described by this model. The motion state of cutting tools can be combined with this model and the effect of cutting tools on the slurry-soil interaction and pressure transfer mechanism was analyzed. Subsequently, the comparative calculations were presented to discuss the influence of tunneling parameters and the design of cutting wheel on filter cake formation during slurry shield tunnelling. The results indicate that the total area of filter cake on excavation face increases with the decreasing of revolutions per minute of cutting wheel and shield machine advance rate. The area ratio of filter cake on the center of excavation face always larger than other zone due to the lesser amount of cutting tools within one track in the center of cutting wheel. The results can provide a better understanding of how to set the shield tunneling parameters and design the layout of cutting tools for the stability of tunnel face.
Cracking and crushing were observed in the first tunnel lining of the Kaidagu double-arch tunnel (China), excavated in weathered surrounding rock. This concrete damage occurred during excavation of the second tunnel. It was necessary to monitor the continuous strain distribution along the entire cross section of the lining, so distributed fibre optic monitoring was carried out for nearly 1 year to study the deformation and stress of the damaged lining of the first tunnel. Optical fibres were attached to the inner surface of the concrete secondary lining and on opposite sides of the temporary arch support. Crack development was detected on the concrete lining, and the strain distribution indicated that vertical ovalisation and oblique ovalisation deformation was generated on the basis of the original deformation. Overall compression deformation was observed as this deformation mode developed over time. Ignoring the effects of axial force, the radial displacement of the concrete lining caused by the bending moment was calculated from the fibre optic strain data. The results were compared with those obtained using total stations, and obvious errors were found. According to the curved beam theory, a more accurate inversion analysis scheme based on the fibre optic strain data was proposed to determine the deformation mode of arch structures. The combined effect of the axial force and the bending moment was considered for the first time. This scheme was used to calculate the displacements, forces and loads of arch structures and was verified by finite-element simulations. The proposed method was applied to analyse the temporary arch support. The calculation results were consistent with those obtained using total stations and conformed with the tunnel deformation mode. The results and the proposed method are useful in explaining the subsequent deformation mode of the damaged lining in the first tunnel and provide valuable information for post-failure reinforcement.
Microtunnelling is an increasingly popular means of locating utilities below ground. The ability to predict the total jacking force requirements during a drive is highly desirable for anomaly detection, to ensure the available thrust is not exceeded, and to prevent damage to the pipe string and/or launch shaft. However, prediction of the total jacking force is complicated by site geology, the use of a lubricated overcut, work stoppages, tunnel boring machine driving style and pipe misalignment. This paper introduces a probabilistic observational approach for forecasting jacking forces during microtunnelling. Gaussian process regression is adopted for this purpose which allows forecasts to be performed within a probabilistic framework. The proposed approach is applied to two recent UK microtunnelling monitoring projects and the forecasts are appraised through comparisons to predictions determined using design methods currently applied in industry. The results show that the proposed framework provides excellent forecasts of the monitored field data and highlights a significant opportunity to complement existing prescriptive design methods with probabilistic forecasting techniques.
Detecting sudden changes in geological conditions (e.g., karst cavern and fault zone) during tunnelling is a complex task. These changes can cause shield machines to jam or even induce geo-hazards such as water ingress and surface subsidence. Tunnelling parameters that relate closely to the surrounding geology have proliferated in recent years and present a substantial opportunity for the application of data-driven artificial intelligent (AI) techniques that can infer patterns from data without reference to known, or labelled, outcomes. This study explores the potential for support vector machines (SVM) to identify changes in soil type during tunnelling towards reducing the possibility of jamming and geo-hazard development. All tunnelling data were pre-processed to convert time series data into feature-based sub-series. A selection of the most popular parameter optimisation algorithms was explored to improve the accuracy of the AI predictions. Their relative merits were evaluated through comparisons with a recent pipejacking case history undertaken in gravel and clayey gravel soils. The results highlight an exciting potential for the use of optimisation algorithm-based SVMs to identify changes in soil conditions during pipejacking.
PyOD is an open-source Python toolbox for performing scalable outlier detection on multivariate data. Uniquely, it provides access to a wide range of outlier detection algorithms, including established outlier ensembles and more recent neural network-based approaches, under a single, well-documented API designed for use by both practitioners and researchers. With robustness and scalability in mind, best practices such as unit testing, continuous integration, code coverage, maintainability checks, interactive examples and parallelization are emphasized as core components in the toolbox's development. PyOD is compatible with both Python 2 and 3 and can be installed through Python Package Index (PyPI) or https://github.com/yzhao062/pyod.
In recent years, there has been an increased resort to microtunnelling/pipe-jacking as a means of constructing underground conduits (for water, sewage, gas and other utilities) to avoid on-street disruption in urban areas. In this paper, technical details of two 1200 mm internal diameter microtunnels in silty sand totalling 550 m in length are discussed; the microtunnels were constructed by Ward and Burke Construction Ltd. as part of the Blackpool South Strategy project. A general overview of the tunnelling process is provided, including the separation plant, jacking facilities and the bentonite supply process. The results show that the lubrication system was very effective at maintaining low skin friction, and that the pipe string was almost fully buoyant for the majority of the drive. Stoppages were shown to have a significant but transient effect on the jacking force; high jacking forces upon resumption of jacking after a stoppage return to ‘baseline’ levels after the length of one pipe diameter. Machine deviations did not appear to play a major role in increasing jacking forces for this particular project.
There are several well-established jacking force models available for determining the jacking loads. However, their ability to characterise the tunnel bore conditions is limited. A simple approach to characterise the tunnel bore conditions is proposed and applied to a case study where four sewer pipelines of the Shulin district sewer network in Taipei County, Taiwan were constructed to verify its validity. In this paper, four jacking force models are reviewed. Based upon the given soil properties and pipe dimensions as well as the pipe buried depth, the normal contact pressure (σ’) in each jacking force model and the measured frictional stress (τ) in each baseline section are utilised for back-analysis of the frictional coefficient (μavg). The μavg values outside the range of 0.1-0.3 recommended for lubricated drives can be attributed to the increasing pipe friction resulting from excessive pipe deviation or ground closure or due to the gravel formation not being long enough to establish lower face resistance or total jacking load. JMTA (Japan Microtunnelling Association) has indicated a further potential use in assessment of the interface performance during pipe-jacking works.
Microtunnelling is an important trenchless construction technique that is used to successfully install essential utility pipelines in increasingly congested urban centres around the world. An important consideration for a microtunnelling project is the magnitude of the jacking force that will be required to advance the microtunnelling shield and the string of product pipes from the starting shaft to the receiving shaft. Frictional resistances along the surface of the pipeline have a major contribution to the total jacking force. This paper considers the frictional resistance mechanism involved in advancing concrete pipes through a coarse-grained soil and describes laboratory testing carried out with the aim of physically modelling the process. Comparisons are made with case histories from microtunnelling projects recently completed in coarse-grained soils. Recommendations are made on predicting likely jacking forces in advance of future projects.
The installation of underground trunk sewer lines in the Tuang formation of Kuching City, Malaysia, used trenchless technology in the form of the pipe-jacking method. The evaluation of pipe-jacking forces mainly involves empirical models developed for soils, with rather limited considerations for drives through weathered rock. Therefore, a novel approach is proposed to evaluate strength parameters by reconstituting and subsequently shearing scalped tunneling rock spoils in the direct shear apparatus. The direct shear results are then applied to a well-established pipe-jacking force model, which considers arching theory. The outcomes indicate that the backanalyzed frictional coefficients μavg are not only reliable but also related to their surrounding geologies because of soil-structure interaction. This study also highlights the significance of lubrication and effect of rock arching in assessing jacking forces. The successful characterization of reconstituted tunneling rock spoils in this paper has shown potential use in assessing jacking forces during microtunneling works.
|| Read More: http://ascelibrary.org/doi/abs/10.1061/%28ASCE%29GT.1943-5606.0001348 ||
Supplementary Material: http://ascelibrary.org/doi/suppl/10.1061/%28ASCE%29GT.1943-5606.0001348
Anomalies are data points that are few and different. As a result of these properties, we show that, anomalies are susceptible to a mechanism called isolation. This article proposes a method called Isolation Forest (iForest), which detects anomalies purely based on the concept of isolation without employing any distance or density measure---fundamentally different from all existing methods.
As a result, iForest is able to exploit subsampling (i) to achieve a low linear time-complexity and a small memory-requirement and (ii) to deal with the effects of swamping and masking effectively. Our empirical evaluation shows that iForest outperforms ORCA, one-class SVM, LOF and Random Forests in terms of AUC, processing time, and it is robust against masking and swamping effects. iForest also works well in high dimensional problems containing a large number of irrelevant attributes, and when anomalies are not available in training sample.
In this paper, we offer a gentle introduction to Gaussian processes for time-series data analysis. The conceptual framework of Bayesian modelling for time-series data is discussed and the foundations of Bayesian non-parametric modelling presented for Gaussian processes. We discuss how domain knowledge influences design of the Gaussian process models and provide case examples to highlight the approaches.
For many KDD applications, such as detecting criminal activities in E-commerce, finding the rare instances or the outliers, can be more interesting than finding the common patterns. Existing work in outlier detection regards being an outlier as a binary property. In this paper, we contend that for many scenarios, it is more meaningful to assign to each object a degree of being an outlier. This degree is called the local outlier factor (LOF) of an object. It is local in that the degree depends on how isolated the object is with respect to the surrounding neighborhood. We give a detailed formal analysis showing that LOF enjoys many desirable properties. Using real-world datasets, we demonstrate that LOF can be used to find outliers which appear to be meaningful, but can otherwise not be identified with existing approaches. Finally, a careful performance evaluation of our algorithm confirms we show that our approach of finding local outliers can be practical.
Suppose you are given some data set drawn from an underlying probability distribution P and you want to estimate a “simple” subset S of input space such that the probability that a test point drawn from P lies outside of S equals some a priori specified value between 0 and 1.
We propose a method to approach this problem by trying to estimate a function f that is positive on S and negative on the complement. The functional form of f is given by a kernel expansion in terms of a potentially small subset of the training data; it is regularized by controlling the length of the weight vector in an associated feature space. The expansion coefficients are found by solving a quadratic programming problem, which we do by carrying out sequential optimization over pairs of input patterns. We also provide a theoretical analysis of the statistical performance of our algorithm.
The algorithm is a natural extension of the support vector algorithm to the case of unlabeled data.
Scikit-learn is a Python module integrating a wide range of state-of-the-art
machine learning algorithms for medium-scale supervised and unsupervised
problems. This package focuses on bringing machine learning to non-specialists
using a general-purpose high-level language. Emphasis is put on ease of use,
performance, documentation, and API consistency. It has minimal dependencies
and is distributed under the simplified BSD license, encouraging its use in
both academic and commercial settings. Source code, binaries, and documentation
can be downloaded from http://scikit-learn.sourceforge.net.
In recent years, there has been an increased resort to microtunnelling/pipe-jacking as a means of constructing underground conduits (for water, sewage, gas, and other utilities) to avoid on-street disruption in urban areas. In this paper, technical details of two 1 200 mm internal diameter microtunnels in silty sand totalling over 550 m in length are discussed. While average skin friction values are extremely low for both drives suggesting effective lubrication practice, differences in normalised bentonite volumes appear to be responsible for differences in skin friction. Full or near full buoyancy of the pipeline has been demonstrated for the majority of the drive. The frictional stress increase after a stoppage is shown to depend on stoppage duration but also on the normalised lubriation volume. Interpretation of data in the manner presented in the paper is an important means of assimilating experience of microtunnelling in different ground conditions.
This case report presents two pedestrian passages underneath an urban expressway with a spacing of 0.5 m, which were constructed using rectangular pipe jacking technology in Nanjing, China. To better understand the field performance and the geoenvironmental impacts of rectangular pipe jacking technology, real-time monitoring was conducted with the installation of measurement devices at 43 measurement points. These points were set up to measure lateral earth pressure, excess pore water pressure, soil lateral displacement, and ground surface settlement. By analyzing field measurement data and operational parameters, this study documents operational parameters throughout the jacking process of two successively constructed tunnels and presents geoenvironmental impacts of the rectangular pipe jacking technology. In this process, the study finds that large ground settlements around the jacking shaft were due to the strong soil-carrying effect of the rectangular pipe jacking technology.
Generally, when there are only a few boreholes present along a tunnel design alignment, geological understanding of the worksite may not be adequate and the ability to optimise the tunnelling parameters is limited. This lack of boreholes will cause an increased potential of geo-hazards during tunnelling works. This study proposes an alternative method to determine the major and other components of ground under such circumstances. Five factors, cutter wheel torque, sieve residue, flow rate of feedline, pressure in the feed and discharge lines and density of bentonite slurry, are adopted for determining the major and other ground components. Comparisons of the soil types based upon the results of grading and Atterberg limits tests on the spoil and soil samples, respectively, and those resulting from the proposed method indicate good consistency. The proposed method provides an opportunity for establishing a more comprehensive geological structure for refining the tunnelling parameters, reducing the potential of geo-hazards associated with the inappropriate tunnelling parameters.
This paper presents a case study on the jacking load during the installation of Gongbei tunnel pipe roof using microtunneling. The pipe roof, consisting of 36 pieces of 1,620-mm-diameter steel pipes along a curved alignment, provides the preexcavation support for the Gongbei tunnel, which has a cross-sectional area of 338 m² and a length of 255 m. The design jacking load was compared with the field measurement, and the difference was investigated. The equation for estimating the face penetration thrust during the slurry type microtunneling was recommended, and a formulation for predicting the total jacking load of the segmented steel pipe along the curved alignment was developed on the basis of the force equilibrium analysis. The unit side frictional resistance and the frictional coefficient of the steel pipes were derived on the basis of the statistical analysis of the field data. The findings from this paper can provide guidance for the design of the jacking load during microtunneling operations for the installation of steel utility pipelines and pipe roofs.
This study investigates the influencing factors that affect the jacking loads during slurry pipe-jacking works at four drives in the Shulin district sewer network in Taipei County, Taiwan, with lengths varying from 73 to 126 m. The main factors which affect the jacking loads during tunnelling may include (1) overcut annulus and volume of injected lubricant, (2) work stoppages, (3) geology, and (4) misalignment. In the four pipe-jacking drives, the jacking forces are represented using the baseline technique. The pipe-jacking results show that the local variations (increasing or decreasing) of jacking force are ascribed to the varying face resistance due to driving between coarse soil and fine soil governed sand or gravel deposit or driving into and away from a buried wooden log. The increase in the jacking loads could also be due to the increasing friction resistance resulting from the pipe deviation being greater than a threshold value of 60 mm. Excessive injected volumes of lubricant result in very low pipe frictions incurred during pipe-jacking of the four drives and are reflected through the back-analysed μavg values which vary from 0.02 to 0.09. The jacking load increases due to either overnight stoppages or short breaks are more pronounced in poorly graded gravel or sand deposit than in clayey gravel or clayey sand deposit.
The purpose of this paper is to introduce some simple theoetical models of pipe-soil interaction during pipe jacking, and relate these to observations made in the field. Ground conditions, construction techniques and the degree of site control all influence the resistance to jacking of pipes, but if an appropriate model is chosen it should be possible to predict jacking forces with a reasonable degree of accuracy. Deviations of the pipeline from a straight line increase the jacking resistance. A new analysis, based on observations from the field monitoring, provides an explanation for the measured increases in pipeline resistance for pipes jacked through a stable bore; it highlights the important factors and emphasizes the need for careful control of pipeline alignment. Explanations are also sought for the apparently frictional behaviour in terms of total stress at the pipe-soil interface in firm and stiff cohesive soils. Time effects are shown to be important in high plasticity clays.
This paper presents a case history of the successful application of observational method to instruction microtunneling with successive pipe-jacking. The microtunneling project is the construction of four parallel pipes under Guan River in Jiangsu, China. Four parallel tunnels with external diameter of 4,160 mm and horizontally spaced at 4.8 m apart were jacked over 450 m in cemented silty clay and sand by two slurry-balance microtunnel boring machines (MTBM) at a depth of 4.6 m under the river bed. Since the overburden soil is very thin, proper control of tunneling operations was of utmost importance for maintaining the stability of the river bed. In order to optimize the operation parameters prior to construction under the river bed, a field trial was conducted, which included measurement of ground surface settlement, subsurface settlement, and lateral displacement of the subsurface soils, as well as excess pore water pressure and earth pressure. The relationship between ground response and construction operation parameters is summarized. Appropriate operation parameters were applied during tunneling under Guan River. Although tunneling and pipe jacking under the river was successfully carried out, a difficulty was encountered when the MTBM reached the opposite bank and large settlements were observed. This paper discusses the technical issues faced and lessons learnt from interpretation of the monitoring data collected.
There is currently no published guidance on the excavatability of Irish rock for microtunnelling applications. In this paper, new data and experiences of microtunnelling through rock (using a Herrenknecht AVN slurry shield machine with a rock head) at five Irish sites are presented and interpreted. The rock type is limestone at three of the sites, with mudstone/conglomerate and sandstone at the other two sites. The jacking forces are separated into face and friction (between the concrete pipe and rock) components. Useful relationships have been established between the excavatabilityindex andthe uniaxial compressive strength and brittleness index of rock.Cutter head wear is discussed in the context of rotational distance travelled and rock strength. In addition, the suitability of a number of prediction models documented in the literature for the prediction of microtunnel boring machine performance in Irish rock is examined.
Pipe ramming is a cost-effective trenchless pipe installation method in which percussive blows generated by a pneumatically or hydraulically powered encased piston rammer are used to advance a pipe or culvert through the ground. To evaluate the feasibility of a pipe ramming installation, engineers must be able to reliably predict the pipe drivability and installation stresses. Assessment of the drivability of the pipe and selection of the optimal hammer for pipe ramming installation requires that the static and dynamic soil resistance to ramming at the pipe face and along the casing be reliably estimated. However, pipe ramming-specific models are not currently available, and engineers often resort to the existing traditional pipe-jacking and microtunneling models for static soil resistance computations. This paper describes the results of four full-scale pipes rammed in the field and the corresponding static soil resistance to ramming in granular soils. A companion paper addresses dynamic soil resistance and pipe drivability. The accuracy of the existing pipe jacking and microtunneling-based static soil resistance models is evaluated herein and found to provide unsatisfactory estimates of the face and casing resistance. New semiempirical pipe ramming-specific models are proposed based on the field observations and are found to produce good estimates of static soil resistance for use in pipe drivability evaluations. (C) 2014 American Society of Civil Engineers.
Abstract: STL is a filtering procedure for decomposing a time series into trend , seasonal , and remainder components. STL has a simple design that consists of a sequence of applications of the loess smoother; the simplicity allows analysis of the properties of the procedure and ...
Pipe jacking is a trenchless technology method of installing pipes under existing facilities such as roads and railroads. Predicting jacking forces is important for planning, design, and construction phases of these types of projects. The jacking forces dictates shaft or pit locations, thrust block or backstop design, jacking equipment, use of intermediate jacking stations, and pipe bearing capacity. Accurate estimation of jacking forces depends on several site and project parameters, such as soil and site conditions, lubrication, size of overcut and steering corrections. Excessive jacking forces can damage the pipe, instable the thrust block, and may stop project progress. There are different methods of calculating jacking loads presented by researchers and industry organizations. By using different models, a discrepancy can be observed in results, which make precise estimation questionable. It can be concluded that analytical and empirical models are based on certain assumptions and field data, and more research is required. For example, ASCE 27 recommends using experimental values to calculate frictional forces and do not consider the face pressures. Other researchers have considered detailed analysis of soil conditions at the face and recommended including project specific conditions such as pipe depth and bore stability. This paper presents an analysis of literature and provides a framework for design engineers to refer to applicable guidelines for planning pipe jacking operations. A conceptual case study is provided to illustrate the differences.
This paper outlines methods for estimating the jacking forces associated with different types of microtunnelling operations. These methods have been developed using probably the most extensive database of microtunnelling jacking forces assembled to date. These data were collected by questionnaire in Japan as part of an initiative by the International Society for Trenchless Technology (ISTT) and coordinated by the Japanese Society for Trenchless Technology (JSTT). Methods for predicting jacking force are produced for slurry, auger and push-in type microtunnelling operations. Separate methods are suggested for these techniques as it was found that the jacking force is sensitive to the method of installation. The methods can also take into account the soil type found on a particular project. However, sensitivity of the measured jacking force to other factors, such as soil strength and depth of installation, were not included in these predictive methods as no discernible relationships could be established due to the variability in the data.Examples from two case histories, one involving a 1.0 m nominal diameter slurry microtunnelling machine in dense silty sand and the second involving a 500 mm nominal diameter microtunnelling machine in sand and gravel, are presented which use the equations proposed in this paper. The results from these examples show that the predicted jacking forces are comparable to those measured in the field.This paper therefore presents practical and reliable methods of predicting jacking forces associated with microtunnelling projects.
This paper is intended to describe an Italian case study of a microtunnelling project where the boring machine got stuck during jacking of a 760mm pipe in a limestone formation. Reference is made to rock mass characterisation, including site investigations and laboratory tests. The machine’s performance is compared to prediction. To allow for a better understanding of the conditions which led to the unexpectedly high jacking forces, continuum and discontinuum numerical analyses have been used. These analyses are shown to be essential at the design stage, when dealing with microtunnelling in a rock mass, in order to obtain a good prediction of the jacking forces.
STL is a filtering procedure for decomposing a time series into trend, seasonal, and remainder components. STL has a simple design that consists of a sequence of applications of the loess smoother; the simplicity allows analysis of the properties of the procedure and allows fast computation, even for very long time series and large amounts of trend and seasonal smoothing. Other features of STL are specification of amounts of seasonal and trend smoothing that range, in a nearly continuous way, from a very small amount of smoothing to a very large amount; robust estimates of the trend and seasonal components that are not distorted by aberrant behavior in the data; specification of the period of the seasonal component to any integer multiple of the time sampling interval greater than one; and the ability to decompose time series with missing values.
Friction forces usually constitute the main component of jacking loads. As a result of their increase with jacking length, it is these forces which limit the drive length. Therefore, it seems important to be able to quantify them accurately. The field monitorings, carried out as a part of the French National Project ‘Microtunnels’, have shown the effects of parameters such as lubrication, stoppages, deviation and overbreak on friction force values. Frictional stress, deduced from field monitoring, is compared with empirical results extracted from the literature and with the results of most frequently used calculation models.
Detecting outliers in a large set of data objects is a major data mining task aiming at finding different mechanisms responsible for different groups of objects in a data set. All existing approaches, however, are based on an assessment of distances (sometimes in- directly by assuming certain distributions) in the full-dimensional Euclidean data space. In high-dimensional data, these approaches are bound to deteriorate due to the notorious "curse of dimension- ality". In this paper, we propose a novel approach named ABOD (Angle-Based Outlier Detection) and some variants assessing the variance in the angles between the difference vectors of a point to the other points. This way, the effects of the "curse of dimensional- ity" are alleviated compared to purely distance-based approaches. A main advantage of our new approach is that our method does not rely on any parameter selection influencing the quality of the achieved ranking. In a thorough experimental evaluation, we com- pare ABOD to the well-established distance-based method LOF for various artificial and a real world data set and show ABOD to per- form especially well on high-dimensional data. Categories and Subject Descriptors
The functions and effects of 624 lubrication in pipe jacking
- G W E Milligan
- M A Marshall
Milligan, G. W. E., and M. A. Marshall. 1998. "The functions and effects of
624
lubrication in pipe jacking." In Vol. 2 of Tunnels and Metropolises,
625
edited by A. A. Negro, Jr. and A. A. Ferreira, 739-744. Rotterdam,
626
Netherlands: A.A. Balkema.
- Eng
Eng. Geol. Environ. 78 (4): 2933-2942. https://doi.org/10.1007/s10064
594
-018-1288-4.
- Eng.-Geotech
- Eng
Eng.-Geotech. Eng. 170 (4): 322-334. https://doi.org/10.1680/jgeen.16
598
.00117.
Frictional resistance of jacked concrete pipes at full scale
- P Norris
- G W E Milligan
Norris, P., and G. W. E. Milligan. 1992. "Frictional resistance of jacked
631
concrete pipes at full scale." In Vol. 92 of Proc., Int. Conf. on Trenchless
632
- Soc
- London
Soc. London, Ser. A 371 (1984): 20110550. https://doi.org/10.1098/rsta
664
.2011.0550.
665
- Eng
environ. Eng. 145 (2): 05018004. https://doi.org/10.1061/(ASCE)GT
684
.1943-5606.0002006.
685