Mixed-face Ground Research Articles

Real-time unsupervised monitoring of earth pressure balance shield-induced sinkholes in mixed-face ground conditions via convolutional variational autoencoders

This study introduces a real-time unsupervised monitoring framework for monitoring sinkhole formation events during earth pressure balance (EPB) shield tunneling operations. A feature extractor (FE) is constructed by coupling variational Autoencoders structure with convolutional neural network layers (VAE-CNN) to manage the complexity of EPB operational data, including non-linearity and temporal dependencies. The monitoring framework consists of two main phases: offline modeling and online monitoring. In the offline modeling phase, an FE model is trained using data-intensive techniques to define a subspace characterizing the behavior of multivariate data without sinkhole formations. The squared prediction error (SPE) statistics and the control limits are computed for detection. During the online monitoring phase, unseen EPB data is propagated to generate SPE values and determine sinkhole events based on whether these values surpass the control limit. Sensor validity index violation counts were used to isolate the most influential variables, while the results demonstrated the superiority of the proposed VAE-CNN method, achieving a 100% detection rate and a 0.9% false alarm rate. The influential variables identified include cutter resolutions per minute, jack speed, screw pressure, torque, and cutter seal components. The monitoring system shows great potential for early warnings during EPB operations to mitigate sinkhole formation risks.

Random forest-based predictors for driving forces of earth pressure balance (EPB) shield tunnel boring machine (TBM)

During the design and excavation stage of the shield tunnel, the prediction of driving forces is pivotal, especially in complex strata. The accuracy of theoretical models for driving loads calculation is insufficient, owing to the academic hypothesis and parameter setting deviation. In this regard, a load prediction approach of an earth pressure balance (EPB) shield tunnel boring machine (TBM) based on a machine learning method (i.e., random forest, RF) was developed. Geological conditions and shield operational data were chosen as features to build the prediction models. The models were then applied to the thrust and torque prediction of an EPB shield TBM in a soil-rock mixed-face ground. The application effect of the RF-based method was evaluated by the coefficient of determination (R2) and the root mean square error (RMSE). Then we discussed the transferability of the built RF model for other EPB shield projects. The results showed that the load prediction capacity of the approach based on the RF algorithm was more accurate than that of theoretical models. Variables importance analysis took the decrease of the mean square error (MSE) as an evaluation indicator. The cover depth of the tunnel ring, the foam quantity, the advance rate, and the rotation speed of the screw conveyor were the top four most influential features for the RF-based thrust and torque prediction model. Overall, this study provides a valuable reference for the safe and efficient EPB shield tunnel excavation in the soil-rock mixed-face ground.

Prediction of Subsidence during TBM Operation in Mixed-Face Ground Conditions from Realtime Monitoring Data

The prediction of settlement during tunneling presents multiple challenges, as such settlement is governed by not only the local geology but also construction methods and practices, such as tunnel boring machine (TBM). To avoid undesirable settlement, engineers must predict the settlement under given conditions. The widely used methods are analytical solutions, empirical solutions, and numerical solutions. Analytical or empirical solutions, however, have limitations, which cannot incorporate the major causes of subsidence, such as unexpected geological conditions and TBM operational issues, among which cutterhead pressure and thrust force-related factors are the most influential. In settlement prediction, to utilize the machine data of TBM, two phases of long short-term memory (LSTM) models are devised. The first LSTM model is designed to capture the features affecting surface settlement. The second model is for the prediction of subsidence against the extracted features. One thing to note is that predicted subsidence is the evolution of settlement along TBM drive rather than its maximum value. The proposed deep-learning models are capable of predicting the subsidence of training and test sets with excellent accuracy, anticipating that it could be an effective tool for real-world tunneling and other underground construction projects.

Study of Rock-Cutting Process by Disc Cutters in Mixed Ground based on Three-dimensional Particle Flow Model

With the increasing number of long tunnelling and urban subway constructions, mixed-face ground conditions are frequently encountered. Rock fragmentation mechanism under disc cutter cutting in TBM tunneling through the mixed-face ground is complex and can lead to engineering difficulties. During TBM tunneling in mixed-face ground with soft rock in upper layer and hard rock in the lower layer, reduction of the advance rate and reduced rotational speed of cutter head occur compared with homogeneous ground. As a result, the muck in the working chamber cannot be replaced timely, leading to the formation of mud cake. Additionally, the disc cutters cannot rotate normally and are worn eccentrically and severely. Finally, the cutters collide with hard rock periodically at the interface between soft and hard rock, thus being subject to a huge impact load, even overload on some cutters, resulting in chipping of the cutter ring and damage to the cutter holder. This paper presents numerical analysis of the disc cutter cutting process considering the difference of rock-cutting behaviors of disc cutters in the mixed-face ground with the aid of PFC3D code. Based on the forces imposed on the disc cutter and rock crack propagation, TBM tunneling in the mixed-face ground is investigated. The decrease of the mean rolling force of the disc cutter causes rotation hindering in the disc cutter in soft rock stratum leading to flat cutter wear. The gap of the normal force between the soft rock and hard rock generates the overturning moment of the cutter head, which causes the eccentricity and vibration of the cutter head.

Prediction Model of TBM Disc Cutter Wear During Tunnelling in Heterogeneous Ground

When shield tunnelling is constructed in complex geological conditions using a tunnel boring machine, the disc cutter in the cutterhead easily wears to the failure state, particularly when the ground conditions are heterogeneous. This paper summarises the failure modes of the disc cutter in heterogeneous ground conditions into three categories, based on the observed wear data from field: (1) uniform disc cutter wear, (2) non-uniform disc cutter wear, and (3) breakage of cutter ring. Subsequently, the stress state of a disc cutter in the heterogeneous ground was analysed and the effective factors were investigated. The relationships between friction energy during cutting, working status of the machine and the characteristics of the geological conditions were evaluated. Based on the stress analysis and friction energy, a prediction model was proposed. The proposed model was applied to two field case studies: pertaining to uniform and mixed-face ground conditions, for which the empirical coefficient k for energy transfer was also determined. The preliminary results from this research indicated that the proposed model was valid for both homogeneous and heterogeneous ground conditions. Further case studies provided by co-operators are expected to improve the effectiveness of the proposed model.

Physical model tests of the surrounding rock deformation and fracture mechanism of mixed-face ground under TBM tunneling

To investigate the surrounding rock deformation and fracture mechanism of mixed-face ground under tunnel boring machine (TBM) tunnelling, tests of a mixed-face ground model with TBM tunnelling were conducted by using a self-developed 3D large-scale simulation test machine and TBM excavation equipment. A fibre Bragg grating sensor and a distributed optical fibre sensor have been installed along the micro-pressure cell and are used to monitor the variation of stress, strain and displacement in the surrounding rocks while ‘uncoordinated deformation is observed at the interface’ during the excavation process. By comparing the experimental observation obtained in the soft and hard rocks located near the interface, a significant difference is observed in the deformation and stress measurement. The fractures observed in the shallow layer of the surrounding rock mass were mainly tensile fractures and tensile shear fractures, and pressure shear fractures were observed at a certain depth of the surrounding rock during TBM tunnelling. The results not only provide reference data on the mechanical behaviour of mixed surrounding rocks during tunnelling by TBM but also can guide construction units to prevent and control the disaster of TBM in tunnelling through mixed-face ground.

Estimation of the cutterhead torque for earth pressure balance TBM under mixed-face conditions

In this paper, the cutterhead torque of the modified earth pressure balance (EPB) tunnel-boring machine (TBM) in mixed-face ground is analyzed. The theoretical model of the cutterhead torque of EPB TBM in single soft ground is generalized to the mixed-face ground. To determine cutting torque, a theoretical model for drag bits and disc cutters, which are suitable for mixed-face ground and nonhomogeneous distribution of drag bits on cutterhead, is established and the force formulas are derived. The components of cutterhead torque are investigated in detail. It is found from the results that the friction torques acting on the frontal surface, circumference and back surface of cutterhead, cutting torque, shearing torque acting on opening and agitating torque are important torque components and occupy a fairly large proportion. Moreover, in mixed-face ground, cutting torque induced by disc cutters is an important part, especially when hard rock has a greater proportion.

Two-Dimensional Numerical Simulation of Rock Fragmentation by TBM Cutting Tools in Mixed-Face Ground

In the study described in this paper, a general particle dynamics (GPD) code was developed to simulate rock fragmentation by tunnel-boring machine (TBM) cutters in mixed-face ground. In the GPD code, a novel algorithm of the contacts at the boundary between two different media was proposed. The mechanism of rock fragmentation by two disc cutters and the effects of cutter spacing on rock fragmentation in mixed conditions were investigated. With increasing cutter spacing, the propagation pattern of Hertzian cracks varied and an optimal cutter spacing was found. The mechanism of rock fragmentation by a disc cutter and a drag bit and the effects of the acting position of different TBM cutters on rock fragmentation in mixed conditions were studied. The optimal acting position of the disc cutter and the drag bit was found to benefit the effectiveness of TBM rock fragmentation.

Prediction of TBM performance in mixed-face ground conditions

A significant number of TBM performance prediction models in full-face soil or rock conditions are available in literature. However, very few prediction models exist for tunnels excavated in mixed-face ground conditions, mainly due to the complexity of the ground.Tunnelling in areas where two or more materials with significantly different geotechnical properties are simultaneously encountered on the tunnel face has often been described as one of the most challenging tunnelling scenarios. In such cases tunnel face instabilities are common and a thorough geotechnical assessment is necessary in order to select the most appropriate tunnelling technique.TBM performance for tunnels in mixed-face conditions was studied, based on the analysis of data from a tunnel project in Singapore excavated into Bukit Timah Granite. The areas considered are located within granite characterised by different weathering grades, which frequently results in mixed-face ground conditions represented by varying percentages of rock and soil simultaneously present at the tunnel face.A new TBM performance prediction index was proposed for mixed-face ground conditions, named as Mixed-face Penetration Index (MFPI). MFPI accounts for the total thrust force exerted by the TBM on the tunnel face (MN) and for the penetration rate (mm/rev). The index was correlated with: (a) the percentage of rock on the tunnel face, (b) the uniaxial compressive strength (UCS) of intact rock and (c) the rock quality designation (RQD) of rock mass, by using a weighted rock mass rating of the tunnel face (RMRm). Although no correlation was determined between MFPI and the geotechnical properties of the soil, its presence was considered in the TBM performance prediction model by assigning an RMR value to the soil portion equal to zero.MFPI and RMRm were correlated using regression analysis tools. The proposed relationship between these two indexes has been validated based on the recorded data of TBM performance in mixed-face ground along the excavated tunnel in Bukit Timah granite.

Prediction Model for TBM Disc Cutter Wear in Mixed-face Ground Condition

The disc cutter failures during shield machine tunnelling in mixed-face ground were summarised by classifying the wear data gathered from the construction of Guangzhou Metro Line 7. The stress state of disc cutter and different effective factors in mixed-face ground condition were analysed. A prediction model for disc cutter wear in mixed-face ground was proposed taking the working status of TBM and the characteristics of geological conditions into consideration. The TBM parameters and geological characteristics of the north tunnel in Guangzhou Metro Line 7 was used to determine the prediction model. With the prediction model, the disc cutter wear of the south tunnel was predicted. The effectiveness of the prediction model was verified by comparing the predicted value and the actual value of the south tunnel.

Theoretical model of the equivalent elastic modulus of a cobblestone–soil matrix for TBM tunneling

Cobblestone–soil mixed ground is a composite comprising cobblestones surrounded by soil. It is typical mixed-face ground encountered during tunnel boring machine (TBM) tunneling, and it may result in cutter wear, jamming of the roller cutterhead, poor TBM performance and cost overruns. The present paper investigates the deformation problem of cobblestone–soil mixed-face ground during TBM excavation. The ground under study is composed of two components (soil matrix and cobblestones) usually firmly bonded together at the interface, and can be regarded as a continuum. Previous studies have proposed many theoretical models for a composite material with two components. Representative models include the parallel model, series model, and effective medium theory model. Nonetheless, these models are limited by their assumptions and preconditions. In the present study, under an assumption of uniform strain, analytical solutions were derived for the equivalent elastic modulus while the cobblestone is assumed to be perfectly spherical or ellipsoidal. Triaxial compression tests were carried out to validate the analytical solutions. The equivalent elastic modulus derived from the triaxial experiments and theoretical models matched rather closely. The analytical solutions are helpful in clarifying the deformation of such ground and enhancing TBM performance.

TBM tunneling in mixed-face ground: Problems and solutions

Mixed-face ground encountered in Tunnel Boring Machine (TBM) tunneling presents great challenges and may trigger potential hazards without warning. A detailed understanding of such unfavorable conditions is therefore critical to a successful bored tunnel. In this paper, we firstly present a brief review of the definition, classification and the factors related to mixed-face conditions. Secondly, for a better understanding of this topic, we investigate the main difficulties and problems involved in TBM tunnelling under mixed-face ground with detailed cases. Thirdly, from the viewpoint of rock-machine interaction, we give some suggestions on the corresponding mitigation measurements from three categories: (i) selection of TBM type and modification of TBM, (ii) condition of ground and (iii) optimization of TBM operation.

Deformation characteristics under variable stiffness for the propelling mechanism of EPB shield machines in mixed ground

This study aims to investigate the deformation characteristics under variable stiffness for the propelling mechanism of Earth pressure balance shield machines in mixed ground. In terms of hyperstatic compatibility, a mechanical model for the driving mechanism in composite strata is first constructed. Based on the method of regarding spatial deformation difference as measuring effect on rectifying deviation in thrust system, the deforming difference trunk for the propelling mechanism is subsequently discovered in mixed face ground. Applying deformation difference trunk to the thrust mechanism of a tunneling machine in engineering applications, the characteristics of deformation in thrust mechanism that affect the thrust system correction are finally discussed. Results could provide a theoretical foundation and support for the design of a new pushing mechanism in tunneling machines.

Case studies of TBM tunneling performance in rock–soil interface mixed ground

This paper investigates the performance of tunnel boring machines (TBMs) in rock–soil mixed-face ground based on TBM tunneling projects in Singapore. Currently several methods are available to estimate TBM tunneling performance in homogenous rock or soil. However, the existing models cannot be effectively applied to predict TBM penetration rate in mixed ground. The tunnels in this study were excavated in adverse mixed-face ground conditions. The geological profiles and the TBM operational parameters are compiled and analyzed. The influence of different geological face compositions on the performance of the TBMs is studied. The statistical analysis shows that there is a possible correlation between the mixed-face ground characteristics and the TBM advancement. Different approaches are used to find a reliable model. Finally, a method is proposed to predict the TBM performance in mixed-face ground for project planning and optimization.

The Practice of Risk Identification and Assessment of EPB Tunnel in Mixed Face Ground

Earth Pressure Balance shield machines are used to construct metro tunnel in Shenzhen mixed ground. The ground condition is full of water and the diversity of two grounds in juncture is drastic. The risk mechanism of this ground is studied, and the main risks are found out. The main risks are assessed and classified employing the method of P=R×C, then the risk classification is determined. These studies ensure the projects in Shenzhen metro advance successfully.