Construction of a large-section long pedestrian underpass using pipe jacking in muddy silty clay: A case study

This paper presents a case study of constructing a large-section long pedestrian underpass using pipe jacking method in Nanjing, China. The underpass, having a width of 7 m and a height of 4.3 m, was jacked 94.5 m in muddy silty clay under a busy roadway with 6.2 m overburden soil, meanwhile it traverses above the existed shield metro tunnels with just 4.5 m from the underpass bottom to tunnel vault. This paper introduced the design and construction schemes of this project in detail. A pre-construction three dimensional numerical simulation was conducted to investigate the responses of the roadway and metro tunnels to pipe jacking construction. Based on the simulation results, the field monitoring program was proposed, and the tunnels deformation and ground settlement were constantly monitored. The field performances of the metro tunnels and roadway were analyzed according to the monitoring data. In the jacking process, the micro-underbreak method was adopted. In order to decrease the tunnels uplift and ground settlement, the actual volume of soil conveyed out from soil chamber to ground surface was kept 95–98% of theoretical soil volume cut by cutter head. In general, this project is completed successfully without taking any additional time and money-consuming deformation control measures. The ground traffic and underneath metro runs well during the whole construction process.     

Opening the excavation chamber of the large-diameter size slurry shield: A case study in Nanjing Yangtze River Tunnel in China

Shield tunnel construction in a dense strata often encounters malfunction of shield-tunneling machine or abrasion of cutters. Accessing to an excavation chamber under compressed air is a main method to repair and replace worn cutters. And many safety issues such as stability of the excavation face were involved. However, the face stability due to opening an excavation chamber was not fully studied. To overcome this shortcoming, face support scheme and stability analysis were presented in a case history of opening the pressure chamber for a large-diameter (up to 14.93 m) slurry shield tunnel constructed underneath Nanjing Yangtze River. Since most of the damaged cutters were distributed along the edge of cutting wheel, only top 3 m of tunnel face within the chamber needed to be supported by compressed air, and remaining area would also to be supported by slurry pressure. A series of simple primary laboratory tests were carried out to design an optimum slurries mixing scheme to support the tunnel face as accessing to the pressure chamber in the project. The face stability was analyzed in terms of the pressure equilibrium (i.e., internal and external pressures) as well as three-dimensional numerical analysis by adopting properties of soils and filter cakes from laboratory tests. By injecting lower density slurry into the sand to form a stable infiltration zone, followed by using higher density slurry to create a filter cake at tunnel face, compressed air-support system could ensure face stability during maintenance of cutter wheel. The success of applying the mixed slurry and compressed air-support scheme in this project is valuable to shield tunnel constructions in similar ground conditions.     

Sequential excavation,NATM and ADECO: What they have in common and how they differ

Rabcewicz, 1964, Rabcewicz, 1965 maintained that “tunnels should be driven full face whenever possible”. ADECO, which stands for “Analysis of Controlled Deformations in tunnels”, now allows us to fulfill Rabcewicz’s dream in any stress–strain condition. In order to achieve that dream and its consequent control over cost and schedule, however, NATM must be abandoned for the ADECO. The paper traces the history of the sequential excavation, NATM (as first conceived) and Analysis of Controlled Deformations (ADECO) with the aim of shedding light on the unavoidable use of sequential excavation in “soft ground”, and of highlighting advances in tunnel design and construction that have occurred in Europe after and as alternates to the NATM. The paper presents the basic concepts in the ADECO approach to design, construction and monitoring of tunnels together with some case histories, including: full face excavation for Cassia tunnel (face area > 230 m²) in sands and silts under 5 m cover below an archeological area in Rome, Italy; Tartaguille tunnel (face area > 140 m²) advanced full face in highly swelling and squeezing ground under 100 m cover where NATM led to catastrophic failure, France; and 80 km of tunnels (face area > 140 m²) advanced full face in highly squeezing/swelling ground under 500 m cover for the high-speed railway line between Bologne and Florence, Italy (turnkey contract).     

Geotechnical risk management approach for TBM tunnelling in squeezing ground conditions

Historically, attempts to use tunnel boring machines (TBMs) in Himalayan geology have been unsuccessful, particularly where weak rocks exist at the significant depths often required for hydroelectric hydraulic tunnels resulting in squeezing ground conditions. The use of segmental tunnel linings erected by shielded TBMs presents additional risk, such that the advantages of potentially high rates of advance using this form of construction have not previously been realised. Programme demands for the 330 MW Kishanganga Hydroelectric Project in India required that 15 km of the 23 km headrace tunnel be constructed using a double-shield TBM erecting a segmental lining. Preliminary studies suggested difficult ground due to squeezing conditions along the 1400 m deep tunnel through weak meta-sedimentary rocks. To allow planning and construction to commence, a risk management approach to design and construction was formulated with contingency procedures and criteria developed to allow the risks to the TBM and the lining to be managed effectively. Advanced numerical modelling included analysis of the tunnel with the ground represented by a Stress Hardening Elastic Viscous Plastic (SHELVIP) model to take account of time dependent loading. The Kishanganga tunnel represents the first segmentally lined TBM tunnel to be successfully constructed in the Himalaya. This paper describes the risk-mitigation approach, the special measures developed to address the risks, the numerical modelling and laboratory testing undertaken, and includes results from the segmental lining monitoring. Recognition of the risks, the development of an innovative methodology and the provision of the means by which geotechnical risk could be managed effectively during construction, gave confidence to all stakeholders to proceed with a method of construction that had not previously been implemented successfully in the Himalaya.     

Key protection techniques adopted and analysis of influence on adjacent buildings due to the Bund Tunnel construction

The Bund Tunnel is 14.27 m in diameter. It is the first application of super diameter earth pressure balanced shield (EPBS) in China. There are many historical buildings along the construction line, and the minimum horizontal distance from the building to the tunnel side varies from 1.7 m to 30 m. Considering the importance of these historical buildings and the complicated construction processes, it is essential to adopt effective protection techniques to ensure safety during the tunnel construction. Three kinds of protection techniques are presented in this paper. Firstly, underground cut-off wall built by bored piles is used to separate the buildings and tunnel when the minimum horizontal distance from the building to the tunnel side is less than 5 m. Secondly, the grouting reinforcement technique is adopted when the minimum clear distance is between 5 m and 10 m. Finally, if the minimum clear distance is larger than 10 m, the optimized construction parameters are selected to reduce the influence induced by the EPBS excavation. The deformations of some typical buildings are monitored. The results of this project will be a useful reference for similar future projects.     

Challenges and strategies involved in designing and constructing a 6 km immersed tunnel: A case study of the Hong Kong–Zhuhai–Macao Bridge

A 6 km immersed tunnel of the Hong Kong–Zhuhai–Macao Bridge (HZMB) has been designed and funded. Once completed in 2016, the HZMB Tunnel will break the record of the 5.8 km BART Tunnel in the United States, although it will soon be overtaken by the ∼20 km Fehmarn Tunnel between Denmark and Germany that is to be completed in 2020. Construction of the HZMB Tunnel was started in 2011 and more than ten elements thus far have been installed on the site. This paper presents details about the challenges and solutions for the design and construction of the HZMB Tunnel on a strategic level. Special features of the HZMB Tunnel include a long length of 6 km, a deep water depth of almost 45 m, and a thick backfill of 23 m. Challenges include severe marine environment, strict requirements for waterproofing, construction of sections connecting the tunnel with artificial islands, and tunnel stability after future excavation of fairway trenches. Moreover, the HZMB Tunnel is challenged by possible sand liquefaction in seismic events, conservation of white Chinese dolphin, and waterway dispersion during construction. Details about the strategies are given in order to improve the immersed tunnel design and construction methods.     

Ground response of closely spaced twin tunnels constructed in heavily overconsolidated soils

As part of the City of Edmonton’s light rail transit expansion, twin 6.5 m diameter oval shaped tunnels were constructed using conventional tunnelling methods. The geology of the site consists predominately of a hard, fissured cohesive till unit. The tunnel face construction was sequenced as top header, bench and invert excavations. At its narrowest, the pillar separating the twin tunnels was less than 1.5 m across or 0.23 tunnel diameters. Typically, the minimum pillar spacing required to reduce the interaction between twin, side by side tunnels is one tunnel diameter. Because the tunnel construction was within an urban environment, there was considerable concern with ground loss and excessive surface settlements. This study demonstrates that within similar materials, a pillar width of 0.5 tunnel diameters sufficiently reduces the tunnel interactions and minimize the risk of damage to nearby structures.     

Application of ground penetrating radar in grouting evaluation for shield tunnel construction

Uniformity and quality of the grout behind the lining segments have a great influence on the long-term settlement in shield tunnel construction in soft soil areas. In order to evaluate the effectiveness of the grouting treatment before the tunnel operation, a nondestructive testing method using ground penetrating radar (GPR) was proposed to detect the grout thickness behind the lining segments of metro lines in Shanghai, China. GPR has shown to be a viable approach due to the facts that: (1) the detecting objects (concrete segments, grout and soil) were in the depth of one meter or less; (2) dielectric parameters of all the materials can be obtained from the laboratory; (3) the contrasts in the dielectric properties among these three materials were large enough; (4) only the boundary between the grout and the soil needed to be found since the concrete segments had a known even thickness of 0.35 m. Three GPR frequencies 250 MHz, 500 MHz and 1 GHz were used in the field tests in Shanghai Metro line 9. The results showed that the 250 MHz GPR had a low resolution while the 1 GHz GPR had a shallow detecting depth. Frequency at 500 MHz showed the most promising results. These tests results demonstrated that nondestructive geophysics techniques such as GPR detection can be used to mitigate the risks of long-term ground settlement, a critical issue of shield tunnel construction in soft soil areas such as Shanghai.     

Experiences of TBM operation in gas bearing water condition – A case study in Iran

The Nowsud tunnel (Lot 2B) project is a 25.7 km length water conveyance tunnel which was bored with a 6.73 m diameter double shield TBM. The tunnel consisting of 6.0 m inside diameter which lined with 25 cm thick, 4 pieces honeycomb precast segments. The geology is consisted of limestone, limy shale, black shale and Shally limestone of Cretaceous Garu and Pliocene Gurpi formations. During heading of tunnel, it was encountered with CH₄ gas emission and H₂S bearing water inrushes, up to 890 l/s. A source which could not be plugged with grouting and the heading had to be stopped for several days and months. As far as the author is aware, this condition has not been seen in double shield TBM tunneling.This case study attempts to discuss firstly the quality and origin of poison gases and water ingress into the excavations. Then, the destructive effects of the poison gases and water ingress on different parts of TBM, concrete corrosion, considerable delay in tunnel progresses and negative impact on tunnel personals productivity have been evaluated.Finally, with respect to restricted space in TBM and back up, a executable solution methods have been pursued to abatement and prevention of the poison gases and water ingress into the excavations, while the construction is in progress by double shield TBM.     

Analyses of the grouting results for a section of the APSE tunnel at Äspö Hard Rock Laboratory

The grouting results for a tunnel at a depth of 450 m in crystalline rock at Äspö HRL were studied. The aims were to investigate whether the methodology used resulted in a successful grouting design and producing a sufficiently dry tunnel, and whether grout penetration and inflow into the finished tunnel corresponded to the predictions. An analysis was made of data from an original cored borehole, drilled before the tunnel was constructed and mapped thoroughly with regard to fractures and transmissivities. The predicted inflow into the tunnel was calculated and found to be four times higher than the measured inflow. The latter was 5 l/min along the 70 m tunnel, considered to be a good result at the current depth. New cored control boreholes were drilled along a section of the tunnel. The inflow positions and quantities in these holes, and the positions of grout found in the corresponding cores, were compared with the data from the original borehole. It was found that at the predicted positions of larger fractures, grout was observed and there was no inflow, showing that these had been successfully sealed. At the predicted positions of small fractures, no grout was visible in the cores, and small inflows showed that the grout had not sealed these fractures. The results indicated that cement-based grout successfully sealed fractures down to a hydraulic aperture of about 50 μm but not below 30 μm. This concurs with the initial design aimed at sealing fractures larger than 50 μm.     

Characteristics and mechanisms of large deformation in the Zhegu mountain tunnel on the Sichuan–Tibet highway

The Zhegu mountain tunnel is a typical long, deep-buried highway tunnel at a high altitude, subjected to low temperatures and high geostress. The tunnel is excavated in carbon phyllite and slate at depths of up to 1000 m below ground, which has resulted in extreme deformation, especially in a depth of 3 m from the tunnel perimeter. The maximum deformation was monitored to be 60 cm, with a maximum deformation speed of 39.3 mm/day. In addition, it took 60–120 days to complete 90% of the deformation. The deformation of the Zhegu mountain tunnel is characterized by serious subsidence of the arch, squeezing outwards of sidewalls, buckling failure of sidewalls and local collapse. The swelling of soft rock is found not to be a main factor of large deformation in the subject tunnel. Three mechanisms of large deformation are derived based on the characteristics and geological conditions, which are plastic flow of soft rock, shear sliding of wedges, and bending of thin-layered soft rock.     

Combined freeze-sealing and New Tubular Roof construction methods for seaside urban tunnel in soft ground

Construction of seaside urban tunnels is significantly difficult due to the densely adjacent buildings and underground facilities in conjunction with unfavourable geological conditions. For this, this paper investigates the construction methods used for Gongbei Tunnel, which is a typical seaside tunnel connecting Hong Kong-Zhuhai-Macau Bridge in China. The construction methods and stability of the undercutting section will be focused. In the tunnel section of interest, dozens of thick steel tubes are jacked into the soft soil, and freeze-sealing method is applied to form a thick water-proofing wall. In this study, the combined freezing and New Tubular Roof (NTR) method are simulated by thermo-mechanical coupling analysis. The temperature field obtained in the freezing process indicates that the thickness of frozen wall grows approx. 2.0 m after 50 days of freezing. Besides, the stability of the surrounding ground and supporting structures in the bench-cutting stage are also investigated. Then the thawing process is simulated and associated suggestions for post-grouting to prevent excessive thawing settlement are proposed. The numerical results show that the tunnel is stable and the influence of tunnel excavation on adjacent buildings is within the permissible range. It also shows that the designed construction methods can be used to adequately meet safety and stability standards when adopting the proposed construction and supporting system.     

Evaluation of common TBM performance prediction models based on field data from the second lot of Zagros water conveyance tunnel (ZWCT2)

Zagros water conveyance tunnel (ZWCT) is a 49 km tunnel designed for conveying 70 m³/s water from Sirvan River southward to Dashte Zahab plain in western Iran. This long tunnel has been divided in 3 Lots namely 1A, 1B, 2. By November 2014, about 22 km of the Lot 2 (with a total length of 26 km) has been excavated by two double shield TBMs from two southern and northern portals. The bored section of tunnel passed through different geological units of 3 main formations of Zagros mountain ranges which mainly consist of weak to moderately strong argillaceous-carbonate sedimentary rocks. In this paper, the operating and as-built geological data collected during construction phase of the Lot 2 of ZWCT project was used to compare the calculated machine performance by empirical methods such as the Hassanpour et al. (2011), Q_TBM, NTNU, Palmstrom, and theoretical model of Colorado School of Mines or CSM. The predicted penetration rates were then compared with the observed field performance of the machine and the variations of predicted rates were examined by statistical analysis. The results showed that the site-specific model, which was based on TBM performance in similar formations can provide estimates closer to actual machine performance.     

Upgrading the Arlberg tunnel to current safety standards

Austrian road tunnels within the Trans-European Road Network (TERN) must fulfil the requirements of the Directive 2004/54/EC (European Commission, 2004) not later than April 2019. This regulation has to be applied to all tunnels in the TERN with a length of more than 500 m, whether they are in operation, under construction or at design stage, and aims at ensuring a minimum level of safety for road users. One of the main features of this directive is the requirement for providing an egress possibility to a safe environment every 500 m throughout the whole tunnel.The Arlberg road tunnel has a length of some 15.5 km and is in operation for more than 35 years. It is a single tube tunnel operated with bi-directional traffic, but carries a quite low traffic volume. Hence, the construction of a second tube is not really cost effective. Currently the tunnel is equipped with a transversal ventilation system with remotely controlled smoke extraction dampers providing smoke extraction every 100 m. The maximum distance between egress possibilities to a save environment is some 1500 m. Due to the high costs of a construction of a second tube or a parallel running escape gallery, a novel solution was found. The existing fresh air duct will be used as safe escape way between the existing egress possibilities. This solution has big impacts on the ventilation system and on the requirements for thermal structure protection of the new egress ways, i.e. the fresh air duct. In order to overcome this problem, massive changes in the ventilation design have to be performed, accompanied by the installation of a high-pressure water-mist system for structure protection.     

The problem of human response to blast induced vibrations in tunnel construction and mitigation of vibration effects using cautious blasting in half-face blasting rounds

Blast induced ground vibrations generated by explosives in tunnel construction may cause structural damages in or close to urban areas. Therefore, the aim in blasting must be to suppress the vibration effects and mitigate the possible hazard on structures. But the psychological character of human response to vibrations involves highly subjective attitudes about what kind of environment is “acceptable” even if no structural damage is occurred. Therefore, we utilize the method of cautious blasting for half-faces that is environmentally friendly, and easy to utilize for tunnel construction. Small charges in this method are detonated sequentially to produce minimum side effects. This method is tested in a tunnel construction in Istanbul with numerous experimental shots. In these experiments, the duration and also quantity of explosives are carefully controlled. Regarding human response, better results are obtained with short durations (about 300 ms) compared to long durations (9000 or 480 ms). The quantity of maximum co-operating explosive charge decreases from 3.088 to 1.744 kg while the vibration levels defined as peak particle velocity (PPV) become more tolerable around 300 ms shot durations.     

Experimental study of the effectiveness of a water system in blocking fire-induced smoke and heat in reduced-scale tunnel tests

A water system, consisting of several water mist nozzles, has been installed in a reduced-scale tunnel. Its effectiveness in blocking fire-induced smoke and heat is tested, with and without longitudinal ventilation. A total of 14 fire tests have been carried out, with 250 ml methanol in an iron tray (25 cm × 20 cm) as fuel. Temperatures have been measured by 30 thermocouples, located upstream and downstream of the fire location. The aim is to assess the effectiveness of the water system in preventing smoke spread and in reducing the temperature in the tunnel. Interaction of the water with the fire is avoided. The impact of water pressure, ventilation velocity and nozzle arrangement on the effectiveness in smoke blocking and temperature reduction is discussed. The result confirms that the water system effectively reduces the temperatures and prevents smoke spreading in the absence of longitudinal ventilation. However, strong longitudinal ventilation (0.8 m/s ventilation velocity in the reduced-scale tunnel, corresponding to critical velocity in full-scale (1:10) tunnel) reduces the effectiveness in blocking the smoke spreading by the water system, although the temperature reduction downstream the water system remains in place. Higher water pressure makes the cooling effect stronger, because more and smaller water droplets are injected into the tunnel. For a given level of water pressure level, the impact of the nozzle row configuration is small in the tests.     

Load transfer mechanism in pile group due to single tunnel advancement in stiff clay

Construction of tunnels in urban cities may induce excessive settlement and tilting of nearby existing pile foundations. Various studies reported in the literature have investigated the tunnel–soil–pile interaction by means of field monitoring, centrifuge and numerical modelling. However, the load transfer mechanism between piles in a group, the induced settlement and the tilting of a pile group due to tunnel advancement has not been investigated systematically and is not well understood. This study conducts three-dimensional, coupled-consolidation finite element analyses to investigate tunnelling effects on an existing 2 × 2 pile group. The construction of a 6 m diameter (D) tunnel in saturated stiff clay is simulated. Responses of the pile group located at a clear distance of 2.1 m (0.35D) from a tunnel constructed at three different cover-to-diameter-of-tunnel ratios (C/D) of 1.5, 2.5 and 3.5 are investigated. The computed results are compared to published data based on field monitoring. It is found that the most critical stage for settlement, tilting and induced bending moment of pile group due to tunnelling is when the tunnel face is close to the pile group rather than at the end of tunnel excavation. The depth of the tunnel relative to the pile group has a vital influence on the settlement, tilting of pile group and the load transfer mechanism between piles in pile group induced by tunnel excavation. Tunnelling near the mid-depth of the pile group (i.e. C/D = 1.5) induces the largest bending moment in the piles, but the settlement and tilting of the pile group are relatively small. Based on a settlement criterion, apparent loss of capacity of the pile group is 14% and 23% for tunnels constructed at depths of C/D = 1.5 and at both C/D = 2.5 and 3.5, respectively. The largest load redistribution between the front and rear piles in the group and the largest tilting of the pile cap towards the tunnel occurs when tunnel excavated at C/D = 2.5.     

Ceiling-jet thickness and vertical distribution along flat-ceilinged horizontal tunnel with natural ventilation

A series of fire tests was conducted in a 10.0 m (L) × 0.75 m (W) × 0.45 m (H) model tunnel with a rectangular cross section, and detailed measurements were taken of the temperature and velocity within a quasi-steady state fire-driven ceiling-jet running along the centre of a ceiling.The ceiling-jet thickness was defined as the distance from the tunnel ceiling to the point where the temperature and/or velocity dropped to half of their maximums. Correlations to represent the variation in the ceiling-jet thickness along the tunnel axis were developed with the aid of a theoretical approach. The coefficients included in these correlations were determined based on the experimental results obtained. It was found that the ceiling-jet thickness derived from the temperature was 1.17 times greater than that from the velocity in the tranquil flow region.In the tranquil region, both the velocity and temperature showed top-hat distributions, with a bulging shape from the apex of the distribution towards the tunnel floor. A cubic function and coordinate transformation were applied to develop empirical formulae for the temperature and velocity distributions, which were represented by the dimensionless distance from the tunnel ceiling and dimensionless temperature rise and/or velocity at a given distance from the fire source. The correlation developed for the temperature distribution was compared with the results of large- and full-scale tunnel experiments, which verified its applicability.     

Velocity and temperature attenuation of a ceiling-jet along a horizontal tunnel with a flat ceiling and natural ventilation

A series of fire tests was conducted in a small-scale tunnel with dimensions of 10.0 m (L) × 0.75 m (W) × 0.45 m (H) and a rectangular cross-section. Detailed measurements of the velocity and temperature within a steady fire-driven ceiling-jet running along the centre of the ceiling were conducted.Referring to a theoretical derivation process described in the literature as a starting point, correlations representing the velocity and temperature attenuation along the tunnel axis were developed.The values of the coefficients included in the developed correlation for the velocity attenuation were measured using a particle image velocimetry system during the experiments conducted in the small-scale tunnel. The value of the Stanton number was determined by considering the ceiling-jet thickness, which was derived from the velocity distribution. The values of the coefficients included in the developed correlation for the temperature attenuation were also determined based on experimental results described in the literature, which were obtained in a large-scale tunnel constructed using good heat insulation properties.Through these correlations developed for the velocity and temperature attenuations along the tunnel axis, the variation in the Richardson number of the ceiling-jet based on the distance from the fire source position along the tunnel axis was examined, and the position where the ceiling-jet changed from a shooting flow to a tranquil flow was determined. The boundary positions between the shooting and tranquil flows were determined using correlations between the velocity and/or temperature attenuation, which were compared with the variation in the Richardson number along the tunnel axis to verify their appropriateness.     

Maximum surface settlement based classification of shallow tunnels in soft ground

Prediction of the maximum surface settlement due to shallow tunnelling in soft grounds is a valuable metrics in ensuring safe operations, particularly in urban areas. Although numerous researches have been devoted to this issue, due to the complexity and a large number of the effective parameters, no comprehensive solution to the problem is available. In this study, a shallow tunnel classification system (STCS), based on maximum settlement, is proposed. The STCS holds on the results of several tunnelling projects around the world. The classifier categorises a tunnel based on geometry, ground, and performance characteristics. A decision tree classification method, after training with 20 cases, was successful to predict the maximum settlement for 14 tunnelling projects. The maximum surface settlement predictions were in the form of assigning a class label to each tunnel. Four tunnel classes were defined as follow: (i) class A (maximum settlement < 9.9 mm), (ii) class “B” (10 ⩽ maximum settlement < 19.9 mm), (iii) class “C” (20 ⩽ maximum settlement < 29.9 mm), and (iv) class “D” (maximum settlement ⩾ 30 mm). The most explanatory independent variables were selected, by the STCS, as follow: tunnel depth, diameter, volume loss, and normalised volume loss. The proposed classification scheme can be employed as a decision making aid in settlement prediction/prevention in shallow tunnelling in soft grounds. The STCS is proposed as a supplemental tool to the observational methods, and it is not expected to be a stand-alone measure for settlement.