A Review of the Effects of Tunneling on Adjacent Piles Mustafa Hasan Abdullah PhD student, Department of Civil and Structural Engineering, University Kebangsaan Malaysia, Malaysia e-mail: [email protected]

Mohd Raihan Taha Professor and head, Department of Civil and Structural Engineering, University Kebangsaan Malaysia, Malaysia e-mail: [email protected]

ABSTRACT The excavation processes of tunnels has caused stresses and deformation in the surrounding soil, then they would transferred to the adjacent structured and specifically to the nearby foundations. In addition, because most of the foundations of buildings in urban areas are piles, it is necessary to study the effect of these operations on piles to evaluate the displacements and stresses acted on the piles, because they may cause a serious risk to the structures they are supporting. Various studies which included numerical or/and experimental solutions were conducted by many researchers to describe this phenomena. Numerical solutions were based on using various constitutive models (e.g. Mohr-Coulomb, elasto-plastic, elastic, displacement control model, Hardening Soil model and Soft-soil model), while the experimental approaches were based on photo-elasticity, photogrammetric technique, small scaling model, and centrifuge tests. This paper presents and discusses a review of these studies. Furthermore comparisons between them will be presented to show the most suitable approach.

KEYWORDS:

Tunnel-soil-pile interaction, tunneling effects on piles, experimental methods in tunneling, numerical analysis of tunneling..

INTRODUCTION Many big cities all over the world are using tunnels, either for transportation, sewerage or telecommunication purposes. Due to the increase of using these tunnels in urban area and because of the route of these tunnels are close to structures, it is important to study the tunneling effects on these structures. Tunneling affects the foundation of structures and becomes more significant for piled foundations. This is because the tunnels generally located in deep levels, so they will be - 2739 -

Vol. 18 [2013], Bund. N

2740

neighboring to these deep foundations (piles). It is also important to study the influence of tunneling operations on the adjacent piles in terms of additional stresses and displacements of these piles in order to control the dangerous settlements of these piles. Many researchers studied this interaction. Some of them focused on numerical simulations (e.g. Mroueh and Shahrour, 2003, Surjadinata et al., 2006, Huang et al., 2009) where their analyses were based on the tunnelpile interaction and they applied various constitutive models of the soil. In addition they studied the geometry effect and whether the foundation is single pile or grouped piles in 2D or 3D. Other researchers used the experimental methods in studying the behavior of piles due to tunneling (e.g. Standing and Leung, 2005, Lee and Yoo, 2006, Lee and Bassett, 2007, Meguid and Mattar, 2009) to investigate many factors which may limit the effects of tunnel excavation processes on nearby piles, and to provide realistic results for verification against numerical methods results. Most of these studies were conducted in the laboratory by using specific techniques (e.g. photo-elasticity technique, photogrammetric technique, small scale testing model, etc.). This paper summarizes selected studies dealing with the effects of tunneling on nearby piles, including numerical methods and laboratory model tests. In addition, this paper will discuss the different approaches in describing this effect.

COMPUTATIONAL METHODS Analytical Methods The analytical methods are usually used to obtain the deformational characteristics of zones between a tunnel and adjacent piles. This approach may be divided into closed-form analytical solutions and numerical methods.

Mindlin’s Model Huang et al. (2006, 2009) presented two-stage analysis with finite difference method for studying the behavior of pile groups subjected to tunneling-induced ground movement in homogeneous and non-homogeneous soils. They assumed Loganathan-Poulos analytical expression based on a Winkler model to calculate the response of a single pile. Tunneling effect of passive pile groups due to pile-soil-pile interaction was investigated. They used the RandolphWroth logarithmic attenuation function for the vertical response and Mindlin’s solution for the lateral response to simulate the pile-pile interaction. Responses of group piles due to tunneling were obtained by the superposition principle. Their method was checked through comparisons with results obtained by the boundary element method program, centrifuge test data and field measurements. The comparison showed good agreements between these studies especially for the non-linear analysis effects as shown in Figures: 1, 2 and 3. As shown in Figure 1, the pile near the ground surface has a settlement more than the free-field ground settlement at the position of the pile that may cause a tension force for that portion of the pile.

Vol. 18 [2013], Bund. N

2741



Figure 1: Comparison of induced pile displacement and free-field soil movements: (a) vertical displacement, (b) lateral displacement (Huang et al., 2009)



Figure 2: Comparison of induced pile: (a) axial force and (b) bending moment (Huang et al., 2009)

Vol. 18 [2013], Bund. N

2742

Figure 3: Comparison of induced pile: (a) axial force and (b) bending moment for test 1 (Huang et al., 2009) Mindlin’s solutions were also conducted by Kitiyodom et al. (2006). They studied the deformation analysis of piled raft foundations subjected to ground movements induced by tunnelling three-dimensional simplified analytical method. They also assumed the flexible raft as thin plates, the piles as elastic beams, and the soil as interactive springs, while the interaction between these members was modelled according to Mindlin’s solutions for vertical and lateral forces. The results were compared with the results of boundary element method and with FLAC3D and show good agreements.

Numerical Models The numerical simulation for any geotechnical problem depends mainly on the accuracy of the constitutive model used to simulate the stress-strain relationship of the soil. In addition, several constitutive soil models have been developed, wherein most of them are formed from rigorous mathematical equations. Many researchers developed and applied the different constitutive soil models generally in tunneling and specifically in tunnel-soil-pile interaction. In principle, their aim was to simulate and to describe the behavior of the soil in this interaction as it occurs in the field. In spite of the large numbers of constitutive soil models, the researchers were focusing on Mohr-Columb, elasto-plastic, elastic, displacement control model, hardening soil model and soft-soil model).

Mohr-Coulomb Model Mohr-Coulomb Model is a first-order model. The stress-strain relationship is linear in the elastic zone, while it is curved in the plastic zone (see Figure 4). The model contains five intercept parameters (e.g. Young’s modulus, E and Poisson’s ratio,  the effective friction angle, ϕ’, effective cohesion, c’, and dilatancy angle, ψ). E and  parameters defined the failure criteria while ψ described the flow rule.

Vol. 18 [2013], Bund. N

2743

Figure 4: The assumptions of Mohr-Coulomb model (Ti et al., 2009) Mroueh and Shahrour (2003) studied the interaction between tunneling in soft soils and adjacent structures, using three-dimensional finite element method. They suggested the MohrCoulomb model to describe the behavior of the soil. They concluded that tunneling-induced forces largely depend on the presence of adjacent structures. Xiang et al. (2008) investigated the effect of tunneling in clayey and sandy soil on urban piled overpass structures of Beijing metro station. Their study involved empirical, theoretical and numerical (using Mohr-Coulomb Model) predictions and in situ monitoring data. From in situ data, they found that the post-underpinning pile foundation showed a large amount of additional settlement prior to behave like the nearby non-underpinned pile foundation, for the same tunneling influences (Figure 5). They concluded that the surcharge effect from an existing structure on the ground increases the effect of tunneling on ground deformation. Also they recommended that the general control procedure should consist of assessing the superstructure capacity, prediction of tunneling and dewatering-induced ground surface and pile foundation settlements, establishment of criteria for distinctively constraining ground surface, pile foundation settlements, and execution of reinforcement measures for metro tunneling in close range to an existing urban overpass supported on pile foundations.

Figure 5: Temporal evolvement of the foundation settlements (Xiang et al., 2008)

Vol. 18 [2013], Bund. N

2744

Recently, Vahdatirad et al. (2010) adopted this model in studying the effect of boring the tunnel of Tabriz, Iran on an underground commercial center which is located on the tunnel passage. The soil type that had been excavated is silty sand. They presented the risk level assessment for settlements of a commercial center foundation due to tunneling. Empirical and numerical methods using Mohr-Coulomb model were used to obtain the value of the underground commercial center structure settlement. The results showed that the risk category of building settlement would be type A, that any tunneling process in this section needed a special monitoring system and consolidation measures before passing the TBM. For the same tunnel case Edalat et al (2010) studied the procedure of choosing TBM method in construction of this tunnel using Multi Criteria Analysis. Their investigations deal with technical, economical and environmental factors that affect the TBM method type. They concluded that the Earth Pressure Balance method (EPB) is the best choice for tunnel excavations in urban area.

Elastic Model Surjadinata et al. (2006) studied the 3D analysis of tunneling effects on adjacent piles (Figure 6 shows the definition of the problem). The finite- and boundary-element (FAB) methods was used to describe the tunnel-soil-pile interaction. The finite-element method was used to predict the free-field ground movements, while the response of an embedded pile to these ground movements were predicted by the boundary-element method. The results showed a good agreement between predictions of the pile response obtained by the FAB method and a 3D finiteelement analysis (Figures: 7 and 8).

Figure 6: Problem definition (Surjadinata et al., 2006)

Vol. 18 [2013], Bund. N

2745

Figure 7: Comparison of FE and BE predictions of lateral pile displacements (Surjadinata et al., 2006)

Figure 8: Comparison of FE and BE predictions of longitudinal pile displacements (Surjadinata et al., 2006) Hou et al. (2007) investigated the behavior and deformation of Shanghai South Railway Station excavation. They simulated the three-dimensional finite-element analysis using ABAQUS program for an oversize deep excavation in soft deposits. Numerical analyses based on elastic modeling of the soil were conducted to study the anisotropic stiffness effects using isotropic and anisotropic soil stiffness parameters. A comparison was made between the obtained wall deflection and ground movements with the observed data in the field and showed an agreement between them (see Figure 9). They concluded that the soil stiffness anisotropy had an important effect on the accuracy of prediction of the diaphragm wall deflection and the ground movement around excavation for the oversize excavation in soft deposits.

Vol. 18 [2013], Bund. N

2746

Figure 9: Measured and calculated wall deflection and soil lateral displacement (Hou et al., 2007) Elasto-plastic Model One of the main difficulties of a constitutive relationship is to describe the stress-strain behavior of soil beyond failure. Some relationships include the elastic and plastic behavior. Using the yield locus located in a shear stress – normal stress space this behavior can be obtained. When the stress state of a soil is located inside the yield locus, it is regarded as elastic and subjected to recoverable deformation. If it is located outside the yield locus, plastic or irrecoverable deformation of soil occurs. The elasto-plastic model distinguishes between the recoverable and irrecoverable deformations, thus helps to understand the stress strain behavior of soil during loading and unloading. Sung et al. (2004) investigated the behaviors of existing pile raft and group pile due to tunnel excavation using two-dimensional finite element method and elasto-plastic model. They concluded that the inclination of group pile and pile raft depend on the distance between the pile tip and the excavation block and it increased with the decreasing this distance. The vertical load of the pile near the excavation was decreased with the tunnel excavation, while the vertical load of the rear pile was increased with the tunnel excavation (see Figure 10). Also they found that the surface settlement were dependent on the magnitude of the load of the existing structure.

Vol. 18 [2013], Bund. N

2747

Figure 10: Percentage of the total load in the piled raft (Sung et al., 2004) Tunneling effects on shotcrete lining and rock bolts was studied by Liu et al. (2008) using this constitutive model. ABAQUS and TUNNEL3D were used in the analysis through full threedimensional (3D) finite element calculations coupled with the tunneling procedure, the interaction between the shotcrete lining and rock mass, the interaction between the rock bolts and rock mass, and the elasto-plastic behavior of the rock mass, the shotcrete lining and the rock bolts. They concluded that tunneling was affecting the existing support system of the adjacent tunnel especially when the tunnel face passed the existing support system and this effect became smaller when the face was far from it.

Displacement Control Model (DCM) Many simulations of tunneling problems were conducted by applying forces to the nodes on the tunnel boundary. These simulations are called Force Controlled Models (FCM). However, these models may lead to a wide surface settlement trough and far field settlements than field or centrifuge test data (Cheng et al., 2007). Meanwhile, there is another approach to simulate these problems based on the displacement convergence pattern around a deforming tunnel boundary. Sagaseta (1987) proposed an analytical solution for this model, in which he assumed that the soil around the unsupported tunnel converges inwards in a radial pattern towards a point on the tunnel vertical line of symmetry. Figure 11 shows the various assumptions for DCM.

Vol. 18 [2013], Bund. N

2748

Figure 11: Proposed displacement mechanism around excavated tunnel (Cheng et al., 2007) Cheng et al. (2004, 2007) performed a finite element analysis of tunnel-soil-pile interaction using this model. The soil type was clay. The results showed that for the case of a single floating pile, induced bending moments were generally negligible beyond a pile horizontal offset from tunnel centre greater than 2 tunnel diameters, and the cracking moment was easily exceeded with small tunnel volume loss for a pile horizontal offset from tunnel centre ≤ 1 tunnel diameter. The analysis showed a good agreement with field results (see Figure 12, 13).

Figure 12: Location of pile tip relative to tunnel axis level and zone of large displacements (Cheng et al., 2004)

Vol. 18 [2013], Bund. N

2749

Figure 13: Computed pile and soil (far field) (a) horizontal and (b) vertical displacements (Cheng et al., 2007) Recently Yang et al. (2011) applied this model to study the effects of tunneling on nearby piles using 3D finite element analysis. The results of the model were compared with those of centrifuge test and a set of parametric studies. The comparisons showed a good agreement between the results. They concluded that the tunneling-induced pile internal force and deformation depends on the distance between the pile and the tunnel, the pile length to tunnel depth ratio and the volume loss. They observed that there were two different zones which were divided by a 45° line projected from the tunnel springline. The pile was subjected to tensile force and large settlement in the zone of influence, while the outer zone of influence, drag load and small settlement were induced. They showed that the effects of tunneling on a pile group were smaller than on a single pile in the same location the rear of the group.

Hardening Soil Model The Hardening Soil model is considered as a second order model for soils. It includes friction hardening and cap hardening. The former is used to model the plastic shear strain in deviatoric loading, while the later is used to model the plastic volumetric strain in primary compression. The model characteristics include stress dependent stiffness according to a power law (m), plastic straining due to primary deviatoric loading (Eref50), plastic straining due to primary compression (Eref oed), elastic unloading/reloading input parameters (Eref ur , νur) and failure criterion according to the Mohr-Coulomb model( c, φ and ψ) (Ti et al. , 2009 ). Lee et al. (2009) and Mica et al. (2009) investigated the tunneling effects on the nearby piles and structures and used this constitutive model in their analyses. Lee et al. (2009) used the finite element program, Plaxis-GiD to model a tunneling case history in Singapore. The analysis included the effect of tunneling on adjacent piles in three dimensions. They used the Hardening Soil-Small model to predict the behavior of sandy silty clay soil. The stress relaxation method was used to model excavations for the tunnels at steady state settlement conditions. The analysis results agreed with those of measured ground surface settlements and subsurface horizontal displacements (see Figure 14).

Vol. 18 [2013], Bund. N

2750

Figure 14: Settlements after tunnel passage (Lee et al., 2009) Mica et al. (2009) investigated the interaction influence of the municipal tunnel, Czech on the behavior of the retaining wall and vice versa. They analyzed the cases using FEM to examine the influence of position of the existing underground structure on the deformation mode and internal forces of the retaining structure respectively (Figure 15). Three constitutive soil model i.e. MohrCoulomb model, Hardening Soil model and Soft-soil model, are used in the analysis to describe the behavior of the over consolidated Miocene clays. They found that the excavation had a major influence on bending moments of tunnel lining and it was increasing (up to 120%) with the depth increasing of the tunnel (Table 1). Also they concluded that for excavation pits of smaller dimensions, the influence of the tunnel would be more significant due to longitudinal stiffness of the tunnel, which couldn’t be taken into account in a typical plane strain analysis.

Figure 15: Geometry of the cross-section used for tunnel analysis (cross-section) (Mica et al., 2009)

Vol. 18 [2013], Bund. N

2751

Table 1: Change of diaphragm wall bending moments (Mica et al., 2009) Position 5 6 7 8 9 10 11 12 13 14 15 16

HS model intrados extrados -12.43 -6.56 -4.95 -4.17 -2.88 -4.83 -5.02 -4.57 -0.41 -1.99 -3.45 -3.84

14.22 10.68 10.8 9.92 4.8 9.57 9.75 9.88 7.47 10.17 10.28 9.91

SS model intrados extrados -12.65 -5.18 -2.25 -1.35 -1.22 -4.77 -3.37 -2.34 -6.68 -0.46 -1.83 -2.07

3.39 -2.12 -2.86 -2.35 -9.48 -2.96 -3.22 -2.83 -6.68 -0.28 -0.61 -2.32

MC model intrados extrados -6.23 -2.52 -1.22 -0.8 -3.4 -3.31 -1.64 -0.94 3.96 0.37 -0.57 -0.66

0.76 0.02 -0.28 -0.36 -2.31 -1.17 -0.68 -0.5 -0.24 -0.97 -0.71 -0.57

Experimental Methods Although the development in computational methods produces analytical and numerical studies, tunneling researchers prefer to support their studies by experiments and model tests. The experimental methods have played an important role in tunneling effects studies. Several tests were conducted to investigate the pile behavior due to nearby tunnel excavation, to understand this phenomenon.

Photo-Elasticity Technique Standing and Leung (2005) investigated the behavior of the effects of piling on a tunnel using the method of photo elasticity. They used an apparatus with the various photo-elasticity components to simulate the tunnel and stress patterns developed in the granular assembly surrounding it during the piling process (see Figure 16 and 17). The results showed that for piles having an offset from the centre of the tunnel inclusion less than a diameter of the inclusion, there was considerable transfer of stress from the pile toe to the tunnel periphery. This was indicated by lines of bright light joining the base of the pile to the inclusion, representing highly loaded chains of particles (Figure 18). The number of lines increased as the penetration depth of the pile increased, and reached a maximum at a specific depth of penetration of the pile depending on its lateral position.

Vol. 18 [2013], Bund. N

2752

Figure 16: Photo-elasticity components and test chamber (Holister, 1967)

Vertical loading by means of actuator Pyrex particles saturated with paraffin Solid perspex inclusion representing tunnel Tap for draining paraffin

Photo-elasticity chamber

V

y

Model pile of perspex

H Tap for gravity feeding paraffin

Figure 17: Schematic diagram showing test set up for studying the effect of piling on a nearby tunnel (Standing and Leung, 2005)

Figure 18: Test 2 photo-elastic image at H = 45mm, y = -8mm (Standing and Leung, 2005)

Vol. 18 [2013], Bund. N

2753

Photogrammetric Technique Photogrammetric technique was used by Lee and Yoo (2006) and Lee and Bassett (2007). Lee and Yoo (2006) studied a 2D behavior of a model tunnel adjacent to a row of piles during tunnel excavation. The pile-soil-tunnel model was built from a multi-sized aluminum rods mixture (Figures: 19 and 20). A photogrammetric technique adopted the analysis of the tunnel center movements with respect to volume loss. They concluded that the pile tip locations influence the tunnel movement, and the model tunnel may be shifted when a line of loaded piles located on close to the tunnel perimeter (see Figure 21).

Figure 19: 2D model pile-soil-tunnel interaction test (Lee and You, 2006)

Figure 20: Realistic image of test (Lee and You, 2006)

Vol. 18 [2013], Bund. N

2754

Figure 21: Center point movements of the single model tunnel (Lee and You, 2006) In Lee and Bassett (2007) study, the tunnel-soil-pile interaction was evaluated using both two-dimensional laboratory model test and numerical analysis. They used aluminum rods for the testing model because the aluminum can be regarded as the frictional granular material. They measured the displacements of reflective markers using a close range photogrammetric technique. The results of the finite element analysis showed a good agreement with the result of model test. They concluded that the pile axial forces were greatly influenced by the location of pile tip from the tunnel centre line. Also they showed that the influence zones were dependent on the location of pile tip, 2D volume loss, soil strength, pile working load, pile size, dilation effect of the granular material, and tunnel size (See Figure 22).

Vol. 18 [2013], Bund. N

2755

Figure 22: Postulated influence zones for the pile–soil-tunneling interaction (Lee and Bassett, 2007) Small Scale Testing Model Meguid and Mattar (2009) designed an experimental setup to investigate the effect of existing pile foundation on the stresses developing in a flexible lining system of tunnels installed in soft cohesive soils. They built a small scale testing model to simulate the process of tunnel excavation and lining installation in the close vicinity of preinstalled model piles, which the later were symmetrically installed at separation distances of 0.7D, 2D, and 2.7D from tunnel perimeter, and the stresses of the lining were measured in each stage of the test. The model consisted from a galvanized steel pipe with an outer diameter of 152 mm, and an inner diameter of 150 mm, while the casing pipe was approximately 405 mm in length (Figures 23, 24). The excavation process of the tunnel opening was done by using a piston head of 12 mm diameter threaded hole diameter connected to hydraulic jack (Figure 25). Model piles consisted of 30 piles 25 mm in diameter, and composed of steel bars located symmetrically in three rows of five piles on both sides of the tunnel. They were fixed in the horizontal direction at the top and bottom of the box by a metal grid and by a perforated wooden plate, respectively. The results showed a continuing decrease in the lining load when the piles were located within a distance of one tunnel diameter from the tunnel lining (see Figure 26).

Vol. 18 [2013], Bund. N

2756

Figure 23: Pile group arrangement (Meguid and Mattar , 2009)

Figure 24: Inner surface of lining with strain gauges (Meguid and Mattar , 2009)

Figure 25: Hydraulic jack mounting with threaded rod (Meguid and Mattar, 2009)

Vol. 18 [2013], Bund. N

2757

Figure 26: Experimental results (Meguid and Mattar, 2009)

Centrifuge Model Test Ong et al. (2007) studied the effect of tunneling in clay on the adjacent piles. They used both the centrifuge model tests and the finite element analysis to evaluate the free-field soil movements due to tunneling and the effects of soil movements on a free-head single pile located nearby. Figure (27) shows the centrifuge model setup which had been used in the study. It consists of stainless steel alloy box soil container with internal dimensions of 525mm in length, 200mm in width, while the height is 490mm. The model tunnel was made of stainless steel, while the piles were square aluminum tubes. The bending moment along a pile and the axial forces along the other were measured by using strain gauges attached along the pile shafts (see Figure 28). In addition they developed a technique to simulate the inward tunnel deformation at the tunnel springline. This technique was done by dissolving the over-shaped high density polystyrene foam placed outside a model tunnel lining using an organic solvent, acetone, while the centrifuge was in-flight at 100 g. Also they made a discharge outlet to permit the drainage of acetone into the tunnel cavity.

Vol. 18 [2013], Bund. N

2758

Figure 27: Centrifuge model setup (Ong et al., 2007)

Figure 28: The model pile (Ong et al., 2007) The results showed that the calculated ground settlements increased with time after tunnel excavation, but the magnitude of this increase was smaller than that observed in the centrifuge test (see Figure 29). This might be due to the reconsolidation of the clay, with the dissipation of pore water pressure caused by tunneling and may lead to produce a wide settlement trough with time after tunnel excavation. As shown Figure 30, it is clear that the pile has experienced large settlement which increased with time.

Vol. 18 [2013], Bund. N

2759

Figure 29: Comparison of surface settlement troughs (Ong et al., 2007)

Figure 30: Induced pile settlement (Ong et al., 2007)

Summary and Conclusions Various modeling methods of tunnel-soil-pile interaction are the most important factor of design tunnels. These different methods which based on different approaches (e.g. analytical, numerical and experimental) provide also different results of the behavior of piles due to tunneling. The closed-form solutions for the analytical method feature simple steps of calculation, especially when using the boundary element methods. Nevertheless, they do not give good approximations for sophisticated problems and cases. In another hand, the numerical approaches supported by finite element method play an essential role in tunneling analysis generally, and tunneling effects on adjacent piles specifically. The various concepts which based on the constitutive models (e.g. Mohr-Columb, elasto-plastic, elastic, displacement control model, Hardening Soil model and Soft-soil model) understand the behavior of these problems and applications. Meanwhile, those researchers developed these constitutive models to gain nearly realistic results compared with in situ data.

Vol. 18 [2013], Bund. N

2760

Table 2: Advantages and disadvantages of the methods Method

Advantages

Disadvantages

Analytical Describe both vertical and

Mindlin’s Model

horizontal ground deformations within few steps of calculation

Their results are not accurate as those for numerical

Numerical Models

Mohr-Colommb Model

Simple and applicable to three-dimensional stress space model

Doesn’t include the influence of rotation of principal stress axes which has the effect in tunnel analysis where liquefaction is possible

Elastic Model

Gives reasonable results for small loading cases

Doesn’t predict realistic results for unloading effects

Helps to understand the stress strain behavior of soil during loading and unloading by distinguishes between the recoverable and irrecoverable deformations

has large numbers of parameters may reach 15, and some of them need special lab tests to get them

Elasto-plastic Model

Displacement Control Model

Hardening Soil Model

Predicts realistic subsurface movements Its two types of hardening (e.g. friction hardening and cap hardening) give an accurate description for problems involving a reduction of mean effective stress and mobilization of shear strength

Producing surface settlement troughs wider than the field observations The Failure in this model is defined by means of Mohr-Coulomb failure criterion

Experimental Methods Photo-Elasticity Technique

Gives a roughly indication about stresses transfer from tunnel to piles

Lack to deformation and creep observations

Photogrammetric Technique

Analyze the tunnel center movements with respect to volume loss

Does not simulate 2D models

Small Scale Testing Model

Centrifuge Model Test

Well simulations of the tunnel advance operations

Conducts the simulation in semi-ideal environment and gives real results for both short and long terms

Good results have been taken when piles are symmetrically arranged around the tunnel

Does not support the creep effects of the soil

Vol. 18 [2013], Bund. N

2761

Finally, the experimental tests were adopted by many researchers and be the verification references for others. The essence of this approach is how to build an environment that resembles the real one. These methods were based on photo-elasticity, photogrammetric technique, small scaling model, or centrifuge tests. Despite of all these approaches and methods of analyzing the interaction of the tunnel-soil-pile, but they complement each other and make a continuum chain of assumptions and theories to describe this phenomena. Meanwhile, these methods have advantages and disadvantages, should take into account when using them. Table 2 shows these advantages and disadvantages.

REFERENCES 1. Antoine, M. M. and J. M. Bradford (1982) “Parameters for Describing Soil Detachment due to Single-Water Impact,” Sad Sam Soc. Am. J., 46, 36-840. 2.

Cheng, C.Y., Dasari, G.R., Leung C.F., Chow, Y.K. and Rosser, H.B. (2004) “3D Numerical Study of Tunnel-Soil-Pile Interaction”, Proceedings of World Tunnel Congress and 13th ITA Assembly, Singapore.

3.

Chen, L.T.and Poulos, H.G. (1996) “Behaviour of piles adjacent to tunnelling operations”, Research Report - University of Sydney, School of Civil and Mining Engineering, R728, 155.

4.

Edalat, K., Vahdatirad, M.J., Ghodrat, H., Firouzian, S. and Barari, A. (2010) “Choosing TBM for Tabriz subway using multi criteria method”, Journal of Civil Engineering and Management, 16(4), 531-539.

5.

Holister, G.S. (1967) “Experimental stress analysis: principles and methods”, Cambridge University Press, UK.

6.

Hou, Y.M., Wang, J.H. and Zhang, L.L. (2007) “Three-Dimensional Numerical Modeling of a Deep Excavation Adjacent to Shanghai Metro Tunnels”, Computational Science – ICCS, 7th International Conference, Beijing, China, Proceedings, Part III, 1164–1171.

7.

Huang, M., Zhang, C. and Li, Z. (2009) “A simplified analysis method for the influence of tunneling on grouped piles”, Tunneling and Underground Space Technology, 24, 410-422.

8.

Huang, M.S., Zhang, H.B. and Lu, R.X. (2006) “Analysis of pile responses caused by tunneling”, Yantu Lixue/ Rock and Soil Mechanics 27(8), 1379-1383.

9.

Kitiyodom, P., Matsumoto, T. and Kawaguchi, K. (2006) “Analyses of pile foundations subjected to ground movements induced by tunnelling”, Geotechnical Aspects of Underground Construction in Soft Ground, 601-606.

10. Lee, Y.J. and Bassett, R.H. (2007) “Influence zones for 2D pile–soil-tunnelling interaction based on model test and numerical analysis”, Tunnelling and Underground Space Technology 22, 325-342. 11. Lee, S.W., Cheang, W., Swolfs, W. and Brinkgreve, R. (2009) “Plaxis-GiD Modeling of Tunnel-Pile Interaction”, 2nd international conference on computational methods in Tunneling, Ruhr University Bochum, Aedificatio Publishers , 1045-1052. 12. Lee, Y. and Yoo, C. (2006), “Behaviour of a Bored Tunnel adjacent to a Line of Loaded Piles”, Tunneling and Underground Space Technology 21, Issues 3-4, page 370. 13. Liu, H.Y., Small, J.C. and Carter, J.P. (2008), “Full 3D modelling for effects of tunnelling on existing support systems in the Sydney region”, Tunnelling and Underground Space Technology 23, 399-420.

Vol. 18 [2013], Bund. N

2762

14. Mica, L., Racansky, V. and Krasny, O. (2009) “Analysis of Interaction of Retaining Walls with Underground Structures in Clay”, 2nd international conference on computational methods in Tunneling, Ruhr University Bochum, Aedificatio Publishers, 233-240. 15. Meguid, M.A. and Mattar, J. (2009) “Investigation of Tunnel-Soil-Pile Interaction in Cohesive Soils”, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 973979. 16. Mroueh, H. and Shahrour, I. (2003) “A full 3-D finite element analysis of tunneling–adjacent structures interaction”, Computers and Geotechnics 30, 245–253. 17. Ong, C.W., Leung, C.F., Yong, K.Y. and Chow, Y.K. (2007) “Performance of pile due to tunneling-induced soil movements”, Underground Space – the 4th Dimension of Metropolises, London, 619-624. 18. Sagaseta, C. (1987) “Analysis of undrained soil deformation due to ground loss”, Geotechnique, 37 (3), 301–320. 19. Standing, J.R. and Leung, W.Y.M.T (2005) “Investigating stresses around tunnels and piles using photo-elasticity techniques”, Geotechnical Aspects of Underground Construction is Soft Ground, Proceedings of the 5th International Symposium TC28, Amsterdam, Netherland. 20. Sung, E., Shahin, H.M., Yamamoto, M., Nakai, T. and Hinokio, M. (2004) “Behavior of existing piled raft foundation due to tunnel excavation”, Proc. of the 16th Central Japan Geotechnical Symposium, Nagoya, Japan, 13-22. 21. Surjadinata, J., Hull, T.S., Carter, J.P. and Poulos, H.G. (2006) “Combined Finite- and Boundary-Element Analysis of the Effects of Tunneling on Single Piles”, International Journal of Geomechanics, 374-377. 22. Ti, K.S., Huat, B.B.K., Noorzaei, J., Jaafar, M.S. and Sew, G.S. (2009) “A Review of Basic Soil Constitutive Models for Geotechnical Application”, EJGE 16, 1- 18. 23. Vahdatirad, M.J., Ghodrat, H., Firouzian, S. and Barari, A. (2010) “Analysis of an underground structure settlement risk due to tunneling-A case study from Tabriz, Iran”, Songklanakarin J. Sci. Technol 32 (2), 145-152. 24. Xiang, Y., He, S., Cui, Z. and Ma, S. (2005) “A subsurface ‘‘drift and pile’’ protection scheme for the construction of a shallow metro tunnel”, Tunnelling and Underground Space Technology 20, 1–5. 25. Xiang, Y., Jiang, Z. and He, H. (2008) “Assessment and control of metro-construction induced settlement of a pile-supported urban overpass”. Tunnelling and Underground Space Technology, 23, 300-307. 26. Yang, X.J., Deng, F.H., Wu, J.J., Liu, J. and Wang, F.Q. (2009) “Response of carrying capacity of piles induced by adjacent metro tunneling”, Mining Science and Technology, 19,176–181. 27. Yang, M., Sun, Q., Li, W. and Ma, K. (2011) “Finite element analysis of tunnel–soil–pile interaction using displacement controlled model”, Journal of Central South University of Technology, 18 (3), 909-916.

© 2013 ejge