Laboratory model tests and field investigations of EPB shield machine tunnelling in soft ground in Shanghai

Tunnelling and Underground Space Technology 26 (2011) 1–14

Contents lists available at ScienceDirect

Tunnelling and Underground Space Technology journal homepage: www.elsevier.com/locate/tust

Laboratory model tests and field investigations of EPB shield machine tunnelling in soft ground in Shanghai Qianwei Xu a,b, Hehua Zhu a,⇑, Wenqi Ding a, Xiurun Ge c a

Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education, Tongji University, Shanghai, China Urban Rail Transit and Railway Engineering Department, Tongji University, Shanghai, China c School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiaotong University, Shanghai, China b

a r t i c l e

i n f o

Article history: Received 11 April 2010 Received in revised form 21 September 2010 Accepted 22 September 2010 Available online 20 October 2010 Keywords: Shield tunnelling Soft ground Model test Soil-shield machine system

a b s t r a c t In the last decades, many studies have been presented by different scholars on the environmental problems induced by the shield tunnelling in soft ground. But it mainly concentrated on ground surface settlement, tunnel surface stability and the influence to existing structures. Among them, little attention was paid to soil disturbance caused by the mismatch of machine’s operation parameters. In order to reveal this inherent relation, a series of laboratory model tests were carried out in the 1/16 scale for a field tunnel in practice where the tunnel had 6.4 m diameter. The tests to simulate earth pressure balance (EPB) shield tunnelling in soft ground were conducted with a microshield machine (0.4 m diameter). Measurements were carried out simultaneously in each test for total jack thrust force, cutting torque, chamber pressure, mucking ratio, ground surface displacement, and earth pressure. Based on the analysis of test results under different overburden ratio, cutterhead aperture ratio and screw rotation speed, the relationships between machine’s operation parameters themselves and its influence on disturbance to surrounding soil mass were discovered. Such discoveries were also verified by the field investigations. These results are useful for engineers and technicians to select suitable and serviceable machine operation parameters and reduce environmental influence at all stages of tunnel construction. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction In order to resolve the problems of increasing traffic congestion, reduced surface vacancy and huge pedestrian volume, Shanghai’s city planners are becoming more interested in using the underground space. According to the overall plan of Shanghai city, there are 21 lines planed in the urban rail transit network. Among them, 16 lines are subway tunnels which are all constructed with shield tunnelling method. Since Shanghai is located in the eastern coast of China, thick layer of soft clay is wildly distributed in the underground of it. When shield tunnelling in such soft ground, it will inevitably perturb the surrounding soil. Therefore, it is necessary to have a comprehensive understanding on the tunnelling induced problems. Practically, such problems may not be solved completely by theoretical research, but through field investigations and model tests sometimes. Based on the field investigation data, great progress in calculating surface settlement or lining stresses had been made in the past few decades (e.g. Peck, 1969; Rowe and Kack, 1983; Harris et al., 1994; Nakamura et al., 2003; Maeda and Kushiyama, 2005; Migliazza et al., 2009, etc.). But due to time consuming and cost ⇑ Corresponding author. Tel.: +86 021 65985014; fax: +86 021 65985140. E-mail address: [email protected] (H.H. Zhu). 0886-7798/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.tust.2010.09.005

expensive, the instrumented projects and field investigations are limited sometimes. For these reasons, the reduced physical model test has been an effective and economical way in studying the shield tunnelling problems. Physical modeling is concerned with replicating a process or a phenomenon in a reduced scale version of the prototype (Viswanadham, 2005). Recently, many tests have been conducted under 1g conditions or in a centrifuge to investigate the shield tunnelling behavior in soil (Meguid et al., 2008). For example, in order to verify the performance of the excavation mechanism of the DPLEX shield method, cutting face support, etc., an experimental shield having rectangular cross section was fabricated to excavate artificial grounds of fine sand, compacted sand, gravel, and gravel with cobbles. In the tests, the machine’s operation parameters were taken into account to keep the balance state of cutting face. But the relationship between the operation parameters was not investigated (Kashima et al., 1996). Nomoto et al. (1999) developed a 100 mm diameter miniature tunnel boring machine to simulate the complete process of shield tunnel construction from cutting to tail void formation in centrifuge. In his tests, only earth pressure around the tunnel and ground surface displacement were monitored and analyzed. Sharma et al. (2001) used polystyrene foam to be dissolved by organic solvent to simulate the progressive tunnel face advance and the gap of shield tail in a centrifuge. In

2

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

that test, the development of settlement trough was reproduced, but no model shield was used. Sterpi and Cividini (2004) conducted laboratory model tests to investigate the behavior up to failure of shallow tunnels excavated with EPB shield machine in strain softening cohesionless media. The tunnelling works were simulated through the gradual reduction of the air bag pressure, until failure of its face occurs. Lee and Bassett (2007) carried out two dimensional laboratory model test for the pile–soil-tunnelling interaction. The model tunnel device can reduce its diameter to simulate volume loss. As can be seen, a lot of research works on the shield tunnelling problems have been done by many pioneers, but it mainly concentrated on ground surface settlement, redistribution of soil stress, tunnel surface stability and the influence to existing structures. Among them, little attention was paid to the soil disturbance caused by the mismatch of machine’s operation parameters. Therefore, in order to reveal this inherent relation, the interval tunnel from Siping Road station to Quyang Road station of Shanghai Metro Line 8 was taken as reference prototype, then a series of laboratory model tests and field investigations were carried out accordingly. Based on the analysis of test results and field investigation data, some useful conclusions can be drawn. 2. Project overview 2.1. Engineering background Shanghai Metro Line M8 lies in the city central area. It crosses through six districts totally, i.e. Yangpu, Hongkou, Zhabei, Huangpu, Luwan District and Pudong New Area. Along the tunnel alignment, there are many residential areas. Therefore, the tunnel will have to pass beneath the urban areas with heavy traffic or dense underground pipelines as well as crowded building groups. The tunnel longitudinal gradient is in humped type or single slope type with maximum gradient of 3%, and the buried depth of tunnel axis is mainly from 7.06 to 31.125 m. The interval tunnel from Siping Road station to Quyang Road station is located in Yangpu district. It is an important part of Shanghai Rail Transit. As shown in Fig. 1, the interval tunnel starts from Siping Road station along West Dalian Road to Quyang Road station, about 870 m long, with its mileage from AK11 + 090 to AK11 + 960. The tunnel having outer diameter of 6.4 m is constructed with EPB shield machine. The elevation at tunnel bottom is from 16.0 or so to 11.0 m. During the construction, the shield machine mainly works under West Dalian Road. 2.2. Site location and topography Along West Dalian Road, there are dense residential areas, such as Yangpu Primary School, Hongwei Middle School, Yutian Village, Dalian Village, and Honglian Building. Most of them are multi-

storey buildings, except that Honglian Building and Quyang residential area are high-rise buildings. Along the tunnel line, the ground surface level is about +3.57 to +3.99 m. Generally, the working site belongs to coastal plain landform type topography. 2.3. Soil composition and characteristics As revealed by engineering geological investigation report, the stratum from surface to 40.45 m deep can be divided into six layers in terms of its genetic type. Among them, layer s and layer x can be further divided into several sublayers according to the engineering properties. In the proposed interval, the average ground surface elevation is +3.70 m; the elevation of tunnel floor is 10.789 to 17.236 m. The embedded depth of tunnel floor presents the trend of large in the middle and small at two ends, as shown in Fig. 2. As the stratum was cut by ancient river firstly, and then the grey silt of littoral-estuarine facies deposited, such as layer s31 and layer s32, which leads to the miss of grey silty clay, i.e. layer t. But the other layers under layer u distribute evenly. The characteristics of the different soil layers are described as below. Layer r is miscellaneous fill, about 1.0–2.3 m thick. The superficial coat of this layer is mainly concrete floor, under which is the ballast mixed with few cohesive soil. In the central, the main component is rock block and broken brick mixed with cohesive soil. In the lower part, it is mainly filled with cohesive soil mingled with scree and cinder. Layer s1 is brown to grey yellow silty clay. The layer surface elevation is 1.46–2.75 m, generally 2.1 m. It is thin, with mean value of 1.2 m. Layer s31 is grey clayey silt mixed with cohesive soil. It is slightly dense. The layer surface elevation is 0.47–1.29 m, generally 0.9 m. The thickness varies greatly from 1.1 to 10.1 m, generally 4.9 m. Its specific penetration resistance (Ps) is about 1.72– 2.77 MPa and the standard penetration numbers (N) is about 3–7. Layer s32 is grey sandy silt, slightly dense to middle dense. The layer surface fluctuates obviously, and its elevation varies from 9.45 m to 0.32 m, generally 4.0 m. The layer thickness is from 2.8 to 11.8 m, generally 7.5 m. Its Ps value is about 3.60–7.74 MPa and the N value is about 3–14. Layer u is grey silt clay. Its surface elevation is about 13.38 to 8.92 m, generally 10.7 m. The layer is about 4.0 m thick and its Ps value is 0.72–0.78 MPa. Layer v1 is grey silty clay, soft plastic. It distributes throughout the entire tunnel interval. The layer surface elevation is about 15.55 to 13.08 m, generally 14.6 m. Its thickness is about 5.40–7.75 m, generally 6.5 m. Layer w is silty clay, dark green to grey yellow, evenly distributed. The surface elevation is 21.88 to 20.35 m, normally 21 m and the thickness is about 3.20–5.30 m, generally 4.3 m. Its Ps value is about 2.05–2.71 MPa.

West Dalian Road

Road bao Tian

ad n Ro Miyu

Quya ng R

oad

We

Fig. 1. Plan layout of the interval tunnel.

st

Da

lia

nR

oa

d

g

n pi Si

ad Ro

3

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

6 4 2 0 -2 -4 -6 -8 -10 -12 -14 -16 -18 -20 -22 -24 -26 -28 1

-30 -32 -34

t 2

-36

Fig. 2. Longitudinal geological profile of the interval tunnel.

Layer x1 is straw yellow to grey-green clayey silt. It distributes evenly in middle dense to dense state. The surface elevation is about 25.67 to 24.85 m, generally 25.3 m. Its thickness is about 4.50–9.70 m, normally 8.0 m. Its Ps value is about 7.45– 9.94 MPa and the N value is about 18–31. Layer xt is grey-green silty clay located below layer x1. It is thin, generally 0.7–2.6 m thick. Its Ps value is about 2.39–3.70 MPa. Layer x2 is grey silt sand, in middle dense to dense state. Its surface elevation is about 35.02 to 32.64 m, generally 34.0 m. Its Ps value is about 9.05–15.49 MPa and the N value is about 29–34. 2.4. Soil mechanical properties According to geological investigation report, the test results of soil mechanical properties are listed in Table 1. Since layer r is miscellaneous fill, it is not suitable for sampling test, and therefore is not reflected in Table 1. In this table, the cohesion and friction angle were measured by consolidated quick shear test, and the compression modulus was measured by standard consolidation test. 2.5. Hydrogeological conditions and soil permeability The groundwater of the working site can be divided into two types of phreatic water and confined water. The groundwater in the shallow layer is phreatic water with buried depth of 0.5– 0.9 m, which is mainly buried in layer s3. It is mainly supplied by the meteoric water and surface water. The confined water is

buried in layer x1 which is the first confined aquifer in Shanghai. The permeability coefficient of various soil layers can be found in Table 1. 2.6. Overview of construction method During the tunnelling process, the shield machine mainly crosses through grey sandy silt (s32), grey silt clay (u) and grey silty clay (v1). Two EPB shield machines are used in this project. Fig. 3 shows the front view of the shield cutterhead. It was equipped with 129 knives totally, with head aperture ratio of 40%. Detailed shield specification can be found in Table 2. 3. Design of model test 3.1. Similarity relations of the model test Shield tunnelling is essentially an interaction process between soil and shield machine. Therefore, when model test is used to study the shield tunnelling problem, it is necessary to consider soil and shield machine as one system, i.e. the ‘soil-shield machine system’, and to determine which variables should be taken into account in this system, by which we can establish a link from prototype to model test. Theoretically speaking, if the system variables are considered more exhaustively, then we can grasp the system nature more precisely. But if more detailed variables are considered in the system, it will not only increase the experiment difficulty but also make the

Table 1 Soil mechanical properties.

s 1 s 31 s 32

u v 1

w x 1 x t x 2

Water content (%)

Void ratio

Soil weight (kN/m3)

Cohesion (kPa)

Friction angle (°)

Compression modulus (MPa)

Permeability coefficient (cm/s)

34.6 32.9 31.6 52.2 35.5 24.7 26.7 26.2 26.8

0.99 0.94 0.90 1.46 1.02 0.72 0.78 0.77 0.79

18.1 18.2 18.3 16.6 17.9 19.4 18.9 18.9 18.7

15 11 6.0 12 17 38 9.0 30 4.0

21.0 24.5 49.5 9.50 17.5 21.5 30.5 23.5 33.0

4.75 7.66 8.22 2.15 3.74 7.99 12.04 10.40 13.51

6.38  106 1.90  104 1.09  103 5.20  107 6.02  106 – – – –

4

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

There are totally 13 system variables, but the number of independent fundamental physical quantity is three, that is, force, length and time. According to the Buckingham theorem, there should be 10 dimensionless factors Pi, as shown in Table 4. From Table 4, the dimensionless factors Pi can be expressed as:

T ; EH3 qV 2 P6 ¼ ; E

P1 ¼

P ; EH2 n H P7 ¼ 1 ; V

P2 ¼

P3 ¼

r E

;

c E

P8 ¼ ;

P4 ¼

d ; H

P 9 ¼ u;

P5 ¼

D ; H

P1 0 ¼ n

ð1Þ

If the model is required similar to the prototype, there should be:

ðPi Þp ¼ ðPi Þm ; ði ¼ 1; 2; . . . ; 10Þ

Fig. 3. EPB shield machine.

Accordingly, the similarity relations of soil-shield system can be deduced as bellows:

Table 2 Shield machine performance parameters.

C T ¼ C E C 3l ;

Shield

2

External diameter Length Electric motor

6400 mm 6450 mm 1  45 kW

Cutter External diameter Rotation speed Torque Electric motor

6320 mm 0–0.8 rpm 473 tm 2  37 W

Screw conveyor External diameter Rotation speed Screw pitch Torque Electric motor

702 mm 0–15 rpm 600 mm 4600 kg fm 2  37 kW

CqCv ¼ CE ;

C P ¼ C E C 2l ;

Cr ¼ CE ;

Cd ¼ CH ;

CH ¼ CD;

C n1 C l ¼ C v ;

Cc ¼ CE;

C u ¼ 1;

Cf ¼ 1

ð3Þ

where C denotes the similarity constant between the prototype and the model, while the subscript is the corresponding parameter involved in the system, for instance, Cl is the geometric similarity ratio, CT is the torque similarity ratio, and the rest symbols may be deduced by analogy. 3.2. Preparation of model ground

experimental precise difficult to control. Therefore, according to the working mechanism of EPB shield machine and its disturbance mechanism to stratum, only 13 variables are selected in this system. (1) Soil properties: cohesion (c), internal friction angle (u), compression modulus (E) and soil density (q). (2) System properties: thickness of covering soil (H), shield diameter (D), cutterhead rotation speed (n1) and cutterhead aperture ratio (f). (3) Dependent variables: advance rate (v), total jack thrust force (P) and cutting torque (T), soil stress (r) and soil deformation (d). When the physical quantities affecting the physical phenomena are known, but the functional relationship between these physical quantities has not been known, then dimensional analysis method can be used to study the physical phenomena. As for general engineering structures, the dimension of its physical quantities can be derived by three fundamental dimensions, that is, force, length and time (FLT system). In accordance with the FLT system, the dimensional matrix of the soil-shield system variables is shown in Table 3. In each column, the digit represents the power of corresponding fundamental dimension.

Table 3 Dimensional matrix of the soil-shield system variables.

F L T

ð2Þ

H

V

E

T

P

r

d

D

q

n1

c

u

n

0 1 0

0 1 1

1 2 0

1 1 0

1 0 0

1 2 0

0 1 0

0 1 0

1 4 2

0 0 1

1 2 0

0 0 0

0 0 0

Since it is the typical soft stratum in Shanghai area, the grey silt clay, i.e. layer u, is taken as reference prototype stratum. For better understand the properties of this soil layer, it was re-sampled and tested again in laboratory. The test results can be found in Table 5. They are a little different form that shown in Table 1. In this table, both the cohesion and friction angle were all measured by quick shear test. In this test, the selected geometric similarity ratio is Cl = 16, the density similarity ratio is Cq = 1, and the strength similarity ratio is Cr = 16. That means the model ground should have a close density to that of the prototype, and its mechanical strength should be scaled down accordingly. In fact, it is very difficult to find such low mechanical strength material to simulate soft soil. Thus, in most cases, we just try to find the material close to the specified requirements. Based on a large number of laboratory tests, the following fabricating scheme is determined, that is, Barite powder:fly ash:iron powder:clay powder:fine silt: kaolin:engine oil:water = 150:25:200:100:250:50:85:160. The mechanical properties of model soil can also be found in Table 5. Only under the specified density can the closest mechanical properties be obtained. In order to reach the specified density, on the one hand, the model soil is filled into the box layer by layer

Table 4 Dimensional factors Pi.

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10

H

v

E

T

P

r

d

D

q

n1

c

u

n

3 2 0 1 1 0 1 0 0 0

0 0 0 0 0 2 1 0 0 0

1 1 1 0 0 1 0 1 0 0

1 0 0 0 0 0 0 0 0 0

0 1 0 0 0 0 0 0 0 0

0 0 1 0 0 0 0 0 0 0

0 0 0 1 0 0 0 0 0 0

0 0 0 0 1 0 0 0 0 0

0 0 0 0 0 1 0 0 0 0

0 0 0 0 0 0 1 0 0 0

0 0 0 0 0 0 0 1 0 0

0 0 0 0 0 0 0 0 1 0

0 0 0 0 0 0 0 0 0 1

5

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

of 420 mm was provided in one of the sidewalls, which allows model shield machine to thrust into it. As shown in Fig. 5, the model shield machine is an EPB one with diameter of 400 mm. But its length is 2 m, and is not scaled down accordingly. In order to keep earth pressure balance state at the cutting face, a valve is installed at the end of model shield to control soil dumping volume; on the other hand, four earth pressure gauges are embedded in the chamber wall so as to control chamber pressure value. For the purpose of improving the excavated material flowability, a bentonite injection hole is also arranged in the chamber wall. From Fig. 6 we can see that there are three kinds of cutting tools equipped on the cutterhead, namely, center cutter, scraper and ripper, which are all scaled down according to the prototype cutter size. The model shield machine is driven by electric control system. The cutterhead is rotated by hydraulic control system, and its rotation speed can be adjusted manually.

Table 5 Soil mechanical properties. Stratum name

Density (kg/m3)

Uniaxial compressive strength (kPa)

Cohesion (kPa)

Friction angle (°)

Prototype Model

1730 1850

31.4 3.87

22.0 1.88

10.5 11.3

with thickness of 5 cm; on the other hand, the filled soil layer is then vibrated by the vibrator installed on the box sidewall. 3.3. Test devices The test devices is shown in Fig. 4, it includes three parts, that is, soil tank, model shield machine and driving system. The soil tank is made of a steel frame, with inside dimensions of 2400  1200 mm in plan and 2400 mm in height. A prepunched hole with diameter

Operation control console

PC computer

switch cabinet A B displacement meter

motor drive box

300 300 300

300

D2

x

jack oil tank D2

D3

D4

earth pressure cell

soil tank

A3 800

jack

C

D5

support

B3 800

800

0 y

A5

500

D1

tunneling direction

A2 A4

A6

315

C screw motor

A1

A3

A7

sliding track

muck outlet

A 400

z

0 x

B

400

400

D1

D2 D3 D4 D5 D6

A-A(B-B)

400

400

400

D7

D8 D9 D10 D11 D12

D13

C-C Fig. 4. Schematic of the model test (unit: mm).

earth pressure gauge

earth pressure gauge

outer shield shell grout hole inner shield shell

Fig. 5. Sketch of model shield machine (unit: mm).

6

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

influence on shield tunnelling. Therefore, only those variables which greatly influence tunnelling effect were selected, such as cover to depth ratio (H/D), cutterhead aperture ratio (n) and cutterhead rotation speed (n). Table 6 shows the details of the model tests. In each test, cutterhead rotation speed in the first half part and the second half part is different. 4. Model test results and discussion 4.1. Comparison of frontal face thrust force

Fig. 6. Model cutterhead.

3.4. Data acquisition system The shield advance rate, cutterhead rotation speed, total jack thrust force and cutting torque can be displayed on the PC computer automatically. As shown in Fig. 4, the variation of earth pressure around the shield tube can be recorded by two groups of earth pressure gauges arranged in sections A–A and B–B. Longitudinal and transverse ground surface settlements were measured by 13 displacement transducers, which can be found in section C–C. In order to facilitate later analysis on test datum, the Cartesian coordinate system is also shown in Fig. 4. 3.5. Design of test content

Fig. 7 shows the loads produced during the shield tunnelling process. Under the condition of model test, total jack thrust force is the sum of all resistances, it includes the frontal face thrust force (Fp) and the soil-tube friction force (Ff). Among them, frontal face thrust force is the sum of all external forces acted on the cutterhead along the tunnel axis, while friction force is mainly caused by vertical earth pressure (Pv), horizontal earth pressure (Ph) and the weight of shield machine itself (Pg). According to the static equilibrium principles, the total jack thrust force can be depicted as following formulas (Pellet-Beaucour and Kastner, 2002).

P ¼ Fp þ Ff Ff ¼

Z

Table 6 Details of model tests. Test serial number

Cover to depth ratio (H/D)

Aperture ratio (%)

Currerhead rotation speed (rpm)

No. 1

2

36

2 3

No. 2

3

36

4 5

No. 3

4

36

2 4

No. 4

4

54

3 6

L

f  p  D  dl

ð5Þ

0

f ¼l

As there are many system variables involved in the tunnelling process, it is impossible to change every variable to study its

ð4Þ



Pg Pv þ Ph þ pD 2

 ð6Þ

where f is the frictional stress of unit area, L is the shield tube length entered into soil tank, and l is the friction coefficient between soil and shield tube. As for prototype shield machine, its total jack thrust force may vary slightly but will not increase with the tunnelling length directly. However, as mentioned before, the model shield length is not scaled down in accordance with the prototype, and then, with the increase of shield tunnelling length, the contact area between soil and tube will also increase, which results in the rise of the total jack thrust force gradually. In order to better understand the variation law of jack thrust force in model test, the soil-tube friction force should be subtracted from the total jack thrust force. That means the variation law of frontal face thrust force in model test can be used to present that of total jack thrust force in prototype. Fig. 8 shows the frontal face thrust force curves when cutting face arrives at different position in No. 1, No. 2 and No. 3 test. These

Pv Pv Ff Chamber wall

P Fp

Ph

Ph

Screw conveyor Cutterhead

Pv+Pg

Pv+Pg

(a)

(b) Fig. 7. Loads during tunnelling process.

7

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

2.6

H/D=2, ξ=36% H/D=3, ξ=36% H/D=4, ξ=36%

12

H/D=2, ξ=36% H/D=3, ξ=36% H/D=4, ξ=36%

2.4 2.2

Cutting torque (kNm)

Frontal face thrust force (kN)

14

10 8 6

2.0 1.8 1.6 1.4 1.2 1.0

4 0.8

2

0.6

20

40

60

80

100

120

140

160

180

0.4

Position of cutting face in z-axis (cm) Fig. 8. Frontal face thrust force under the condition of different cover to depth ratios.

20

40

60

80

100

120

140

160

180

Position of cutting face in z-axis (cm) Fig. 10. Cutting torque under the condition of different cover to depth ratios.

2.6

H/D=4, ξ=36% H/D=4, ξ=54%

H/D=4, ξ=36% H/D=4, ξ=54%

2.4 12

2.2

Cutting torque (kNm)

Frontal face thrust force (kN)

14

0

10

8

6

2.0 1.8 1.6 1.4 1.2 1.0

4

20

40

60

80

100

120

140

160

180

Position of cutting face in z-axis (cm) Fig. 9. Frontal face thrust force under the condition of different cutterhead aperture ratios.

0.8 0.6

0

20

40

60

80

100

120

140

160

180

Position of cutting face in z-axis (cm) Fig. 11. Cutting torque under the condition of different cutterhead aperture ratios.

three tests were carried out under the condition of different cover to depth ratios (H/D = 2, 3, 4) but with the same cutterhead aperture ratio (n = 36%). We can see that, the larger the buried depth is, the larger the frontal face thrust force is needed. In Fig. 9, the frontal face thrust force curves in No. 3 and No. 4 test were presented. These two tests were carried out under the condition of different cutterhead aperture ratios (n = 36%, 54%) but with the same cover to depth ratio (H/D = 4). It can be seen that, the larger the cutterhead aperture ratio is, the smaller the corresponding frontal face thrust force is. In addition, in both Figs. 9 and 10, the frontal face thrust force presents a trend of rapid growth at the initial stage and then tends to be relatively stable. The rapid growth of the curves at the initial stage may be influenced by the infilling process of the soil chamber. 4.2. Comparison of cutting torque The cutting torque magnitude represents the difficulty degree of cutting soil by the cutterhead. Fig. 10 shows the test results of cutting torque under different cover to depth ratios (H/D = 2, 3, 4) but with the same cutterhead aperture ratio (n = 36%). Fig. 11 gives the cutting torque under different cutterhead aperture ratios (n = 36%, 54%) but with the same cover to depth ratio (H/D = 4). It can be seen that, the larger the buried depth is, the greater the cutting torque is required; secondly, the larger the cutterhead aperture ratio is, the smaller the cutting torque is needed. Addi-

tionally, the cutting torque value also tends to be stable after the rapid growth in the initial phase, which is consistent to what Figs. 8 and 9 have reflected. 4.3. Frontal face thrust force and cutting torque From Figs. 8–11, we can find that both frontal face thrust force and cutting torque has similar variation law under different buried depths or different cutterhead aperture ratios. Therefore, a close relationship may be existed between them. As revealed by Figs. 12 and 13, the relationship between frontal face thrust force and cutting torque presents multilevel linear characteristics. What is more, if cutting torque is in a certain range, the line slope will increase with cover to depth ratio but decrease with the cutterhead aperture ratio. A very interesting phenomenon is that, when the cutting torque reaches a certain higher value, the slope of the broken line will have a sudden change, which means that both frontal face thrust force and cutting torque will enter into a new linear relationship. It also indicates that there must be many multistage torque limit points existed during the tunnelling process. 4.4. Frontal face thrust force, cutting torque and penetration per revolution The ratio of advance rate (v) to cutterhead rotation speed (n1) is called penetration per revolution (PRev). It is often used to evaluate

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

the difficulty degree of shield tunnelling in soil. For example, if the soil mechanical strength is high, it is difficult for shield tunnelling and in return the corresponding PRev value will be low. In most practical engineering, the PRev has a close relationship with total jack thrust force and cutting torque. As the reason mentioned before, we only chose the frontal face thrust force, cutting torque and PRev in the model test to analyze here. As shown in Figs. 14 and 15, both frontal face thrust force and cutting torque present a form of exponential relation with PRev under different conditions of buried depth or cutterhead aperture ratio. We can see that, if the PRev keeps invariant, the higher the overburden ratio is or the smaller the aperture ratio is, the larger the frontal face thrust force and the cutting torque will be needed. On the other hand, with the increase of PRev, the cutting torque grows slowly but tends to be stable quickly, which means cutting torque is easier to reach its rated value.

Cutting torque (kNm)

1.8 y4 = 0.5815x - 2.8065

1.6

1.2 1.0 0.8

y4 = 0.1631x + 0.1024

4

6

2.2

1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 3.2

1.6

0.0

2.0 1.8 1.6

1.2

0.8 0.6

y2 = 1.0338x - 7.3687

0.4

y1 = 1.0093x - 4.6646

0.2

9

10

11

12

13

Frontal face thrust force (kN) Fig. 12. Cutting torque vs. frontal face thrust force under the condition of different overburden ratios.

3.2

4.8

6.4

8.0

9.6

11.2

12.8

Frontal face thrust force (kN)

In a sense, the shield advance rate can reflect theoretic volume of soil excavated by cutterhead per unit time, and the screw rotation speed represents the mucking capacity. Therefore, the ratio of screw rotation speed to advance rate (n2/v) may be regarded as an alternative form of mucking ratio. As shown in Fig. 5, both the cutterhead

y3 = 1.0868x - 8.8527

8

14

4.6. Advance rate, screw rotation speed and chamber pressure

1.0

7

12

Fig. 14. Frontal face thrust force, cutting torque and PRev under the condition of different overburden ratios.

1.5

6

1.6

Cutting torque (kNm)

y1 = 0.0174x + 1.669

5

10

thrust force (H/D=2, ξ =36%) thrust force (H/D=3, ξ =36%) thrust force (H/D=4, ξ =36%) torque (H/D=2, ξ =36%) torque (H/D=3, ξ =36%) torque (H/D=4, ξ =36%)

Penetration per revolution (cm/rev)

2.0

1.4

0.6

8

Frontal face thrust force (kN)

1.8

0.9

y3 = 1.0868x - 8.8527

torque limit point

1.4

y2 = 0.0648x + 1.2598

1.2

y3 = 0.1737x + 0.1008

torque limit point

2.0

y3 = 0.1737x + 0.1008

H/D=2, ξ=36% H/D=3, ξ=36% H/D=4, ξ=36%

torque limit point

2.2

3.2

1.6

0.0

thrust force (H/D=4, ξ =36%) thrust force (H/D=4, ξ =54%) torque (H/D=4, ξ =36%) torque (H/D=4, ξ =54%)

Penetration per revolution (cm/rev)

According to the working mechanism of EPB shield machine, In order to keep the earth pressure balance state at the cutting face, the weight of the soil imported into soil chamber should be equal to that exported by screw conveyor. Therefore, the cutterhead advance rate (v) and the screw rotation speed (n2) are two decisive factors affecting the earth pressure balance state. When these two parameters do not match each other, it will lead to the weight imbalance between the imported and the exported soil, which means the existence of over excavation or under excavation phenomena. It is this imbalance that causes the variation of chamber pressure. The ratio of actual exported soil weight to theoretical imported soil weight is defined as mucking ratio, and is often used to evaluate this imbalance. In the tests, the actual exported soil was weighed whenever the shield machine advances a certain distance. Fig. 16 shows the chamber pressure vs. mucking ratio in No. 1 test. As can be seen, once the mucking ratio is greater than 100%, that is, the mucking speed is too fast, so the chamber pressure will drop down; on the contrary, when the ratio is lower than 100%, the soil in chamber will be squeezed, which results in the rise of chamber pressure. Figs. 17 and 18 gives the chamber pressure vs. mucking ratio under different conditions of overburden ratio and cutterhead aperture ratio. We can see that the chamber pressure decrease in the form of negative exponential function with the increase of mucking ratio. Additionally, if the chamber pressure is kept constantly, the larger the cover to depth ratio or the smaller the cutterhead aperture ratio, the greater the mucking ratio is needed.

2.1

y4 = 0.1414x + 0.658

Fig. 13. Cutting torque vs. frontal face thrust force under the condition of different head aperture ratio.

4.5. Mucking ratio and chamber pressure

2.4

H/D=4, ξ=36% H/D=4, ξ=54%

2.4

Cutting torque (kNm)

8

1.6

Cutting torque (kNm)

3.2

4.8

6.4

8.0

9.6

11.2

12.8

Frontal face thrust force (kN)

Fig. 15. Frontal face thrust force, cutting torque and PRev under the condition of different cutterhead aperture ratios.

9

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

Shield tunnelling process will inevitably break the original stress equilibrium state in soil, which can be manifested by the variation of earth pressure increment, i.e. the difference between the measured earth pressure and the earth pressure at rest. To illustrate it, the variation of earth pressure increment on A–A section in No. 1 test is selected to analyze in the following paragraphs. As shown in Fig. 3, the three gauges of A2, A4 and A6 are arranged at the same height above the tunnel. Fig. 20 shows the variation of earth pressure increment measured by these three gauges. we can see that the earth pressure increment value rises rapidly before the arrival of cutterhead; it reaches the maximum when cutterhead about 10 cm before A–A section; once cutterhead moves away, the pressure value will descend and tend to be stable finally, but it is still higher than its initial value. Another phenomenon is that, in the whole process, the earth pressure increment just above the tunnel axis is higher than that on both sides of it. Furthermore, before the arrival of cutterhead, the earth pressure increment values of these three gauges are relatively close; but when the cutterhead had passed away, the increment value of A4 drops relatively small as compared with that of A2 and A6. It also indicates that the soil disturbance just above the tunnel axis is relatively large. The variation of earth pressure increment measured by the three gauges of A1, A2 and A3 on the same vertical line is shown in Fig. 21. Its special variation law lies in two aspects. Firstly, the deeper the buried depth is, the larger the corresponding increment value is, which also indicates that the soil disturbance is larger in the deep. Secondly, the increment values of these three points do not reach the maximum at the same time, i.e. the gauge with shallow buried depth is sooner to reach the maximal increment value. As shown in Fig. 22, mucking ratio also affects earth pressure increment greatly, that is, when mucking ratio exceeds 100%, the earth pressure increment will descend accordingly; but if the efficiency is less than 100%, the increment value will increase immediately.

-0.0166x

H/D=2 H/D=3 H/D=4

y = 0.2694e 2 R = 0.8849

Chamber pressure (MPa)

4.7. Mucking ratio and earth pressure increment

0.15

0.12

-0.0146x

y = 0.1617e 2 R = 0.8095

0.09

0.06

0.03

-0.0242x

y = 0.1468e 2 R = 0.8757

0.00 20

40

60

80

100

120

140

160

Mucking ratio (%) Fig. 17. Chamber pressure vs. mucking ratio under different overburden ratios.

0.16 0.14

Chamber pressure (MPa)

and the screw conveyor use the same bearing in our tests, so they have the same rotation speed. Fig. 19 shows the chamber pressure vs. the ratio of screw rotation speed to advance rate (n2/v) in No. 1 test. It can be seen that the chamber pressure also reduces in a negative exponent form with the increase of n2/v value.

0.12 -0.0166x

y = 0.2694e 2 R = 0.8849

0.10 0.08 0.06 0.04

-0.0222x

y = 0.2155e 2 R = 0.8547

0.02 0.00 0

20

40

60

80

100

120

140

160

Mucking ratio (%) Fig. 18. Chamber pressure vs. mucking ratio under different cutterhead aperture condition.

0.05

4.8. Ground surface displacement As mentioned before, if the mucking ratio is low, it leads to the extrusion of the soil in front of cutterhead and further results in the 1.6

1.6 chamber pressure mucking ratio

1.4 1.2

1.0

1.0

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0 20

40

60

80

100

120

140

160

180

Position of cutting face in z-axis (cm) Fig. 16. Chamber pressure vs. mucking ratio in No. 1 test.

0.0

0.03

0.02

0.01

0.00

5

1.2

Chamber pressure ( 10 Pa)

Mucking ratio ( 100%)

1.4

Chamber pressure (MPa)

0.04

0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20

Screw rotation speed/advance speed (r/mm) Fig. 19. Chamber pressure vs. ratio of screw rotation speed to advance rate in No. 1 test.

upheaval of ground surface. On the contrary, if the mucking ratio is high, then chamber pressure deceases accordingly, which results in subsiding of ground surface. Fig. 23 shows the ground surface displacement at point D7 vs. mucking ratio in No. 1 test. It can be found that, when mucking ratio is less than 100%, the ground surface will upheave; but when it exceeds 100%, the surface presents a downward trend immediately.

10

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

A2, x=-300mm A4, x=0mm A6, x=+300mm

0.016 0.012

0.008 0.004

A-A section

0.000 -80

-60

-40

-20

0

20

40

60

80

100

Relative position of cutting face to A-A section (cm) Fig. 20. Variation of earth pressure increment at the same level in A–A section in No. 1 test.

Earth pressure increment (MPa)

A1, y=-150mm A2, y=-650mm A3, y=-1000mm

5. Field test results and discussion Since the earth pressure was not measured, it is not analyzed in the following parts. Moreover, the exported soil weight was not

18

1.6

Fig. 21. Variation of earth pressure increment at different depth in A–A section in No. 1 test.

Fig. 24 shows the longitudinal displacement curves of ground surface above tunnel axis when cutterhead arrives at different positions. We can see that, with the advance of cutterhead, longitudinal displacement curve moves forward in form of wave transmission. Fig. 25 shows the transverse surface displacement curves at A–A and B–B section when the cutterhead just arrives below A–A section. Since B–B section is far from the cutterhead, the deformation value in this section is relatively small.

Earth pressure increment muck discharging efficiency

1.5

0.030 0.025

1.0 0.020 0.015

0.5

0.010 0.0 40

60

80

100

120

140

160

14

1.2

12 1.0 10 0.8 8 0.6 6 0.4

4

0.2

2 0

40

60

80

100

120

140

160

180

0.0

Position of cutting face in z-axis (cm) Fig. 23. Ground surface displacement of point D7 vs. mucking ratio in No. 1 test.

18

Muck discharging efficiency ( 100%)

2.0 0.035

Displacement of D7 point (mm)

Relative position of cutting face to A-A section (cm)

1.4

Muck discharging efficiency ( 100%)

ground surface displacement muck discharging efficiency

16

A-A section

Earth pressure increment (MPa)

Fig. 26 shows the ground surface displacement at point D7 vs. mucking ratio in No. 2 test. Since the mucking ratio is less than 75% at the early stage of the tunnelling process, ground surface upheaves obviously; afterwards, if it reaches or exceeds 100%, then ground surface begins to subside accordingly. Fig. 27 shows the longitudinal surface displacement curves when cutterhead arrives at different positions. It can be seen that, if the mucking ratio is consistently greater than 100%, the longitudinal subsidence trough would deepen and move forwards with the advance of cutterhead. Fig. 28 shows the transverse surface displacement curves at A–A and B–B cross section when the cutterhead just arrives below A–A section. As can be seen, the displacement value at A–A section is greater than that at B–B section. Since the mucking ratio is depended on shield advance rate and screw rotation speed, then the ground surface displacement is also affected by these two factors. Figs. 29 and 30 shows the ground surface displacement of point D7 vs. the n2/v value in No. 1 and No. 2 test respectively. They are somewhat similar to that reflected in Figs. 23 and 26.

180

Ground surface displacement (mm)

Earth pressure increment (MPa)

0.020

16

7

z=40cm z=60cm z=80cm z=100cm z=120cm

4

14 12 10

1

8

10

6 4 13

2 0 20

40

60

80

100 120 140 160 180 200 220

Relative position in the z axis (cm)

Position of cutting face in z-axis (cm) Fig. 22. Variation of earth pressure increment and mucking ratio in No. 1 test.

Fig. 24. Longitudinal displacement curve of ground surface when cutterhead arrives at different position in No. 1 test.

11

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

that cutting torque increases gradually with the increase of cutterhead rotation speed. In addition, the larger the thrust force is, the greater the corresponding torque is needed.

Ground surface displacement (mm)

12 D4

A-A section B-B section

10

D3

5.2. Total thrust force and cutting torque

8

D5

Fig. 32 shows the relationship between total thrust force and cutting torque during a certain period of the tunnelling process. As can be seen, a piecewise linear relationship is existed between the thrust force and the cutting torque. It is consistent with the laws derived from the model tests, as shown in Figs. 12 and 13.

6 D2

4

D6

2 0 -50

D10

D9

D8

-40

-30

D11

-20

-10

0

10

20

5.3. Total jack thrust force, cutting torque and PRev

D12

30

40

50

Relative position to the tunnel axis (cm) Fig. 25. Transverse displacement of ground surface at A–A and B–B section in No. 1 test.

The relationship between total jack thrust force, cutting torque and PRev is presented in Fig. 33. It is consistent with the test results revealed by Figs. 14 and 15. 5.4. Advance rate, screw rotation speed and chamber pressure

1.25

1.25

1.00

1.00

0.75

0.75

0.50

0.50

0.25

0.25

0.00

0.00

-0.25

-0.25

-0.50

-0.50 20

40

60

80

100

120

140

160

Fig. 34 shows the chamber pressure vs. the ratio of screw rotation speed to advance rate (n2/v). It can be seen that the chamber pressure reduces in negative exponent form with the increase of n2/v value. It is also consistent with the test results shown in Fig. 19.

Distance from tunnel axis (cm) -50 0.00

Ground surface displacement (mm)

1.50 ground surface displacement muck discharging efficiency

Muck discharging efficiency ( 100%)

Ground surface displacement (mm)

1.50

180

Position of cutting face in z-axis (cm) Fig. 26. Ground surface displacement of point D7 vs. mucking ratio in No. 2 test.

recorded, so it will not be discussed also here. The ground conditions and shield operation parameters can be found in Section 2.

-40

-30

-20

-10

0

Fig. 31 gives the relationship between cutting torque and cutterhead rotation speed under different thrust forces. It can be seen

20

30

D8

40

50

D12

-0.02 D9 D2

D11

D10

-0.04

D6

D5

-0.06

A-A section B-B section

D3 D4

-0.08

5.1. Cutting torque and cutterhead rotation speed

10

Fig. 28. Transverse displacement of ground surface at A–A and B–B section in No. 2 test.

Ground surface displacement (mm)

4

0.04

7 10

13

0.00 -0.04 -0.08 z=40cm z=60cm z=80cm z=100cm z=120cm

-0.12 -0.16 -0.20 -0.24 -0.28 20

ground surface displaement screw rotation speed to advance rate

16

1.4

14 12

1.2

10 1.0

8 6

0.8

4 2

0.6

0

Screw rotation speed to advance rate (r/cm)

1

Displacement of D7 point (mm)

18 0.08

-2 40

60

80

100 120 140 160 180 200 220

Relative position in the z axis (cm) Fig. 27. Longitudinal displacement curve of ground surface when cutterhead arrives at different position in No. 2 test.

40

60

80

100

120

140

160

180

Position of cutting face in z-axis (cm) Fig. 29. Ground surface displacement of point D7 vs. ratio of screw rotation speed to advance rate in No. 1 test.

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

5.5. Ground surface displacement In order to study the soil disturbance caused by tunnel construction, a series of displacement observation points are arranged

Displacement of D7 point (mm)

ground surface displaement screw rotation speed to advance rate

3

3

2

2

1

1

0

0

-1 20

40

60

80

100

120

140

160

Screw rotation speed to advance rate (r/cm)

4

4

along and perpendicular to the tunnel axis, as shown in Fig. 35. Along the tunnel axis, every five rings there is an observation point arranged at the selected intervals, that is to say, the points numbered as SX1–SX157 are located on the uplink, while the points numbered as XX1–XX155 are on the downlink. From Siping Road station and Quyang Road station, there are total 17 groups of observation sections are arranged to monitor transverse surface displacement. These sections are numbered as JSX-1 to JSX-9, and each section is composed of nine observation points.

90 80

Penetration per revolution

12

-1 180

Position of cutting face in z-axis (cm) Fig. 30. Ground surface displacement of point D7 vs. ratio of screw rotation speed to advance rate in No. 2 test.

jack thust force cutting torque

70 60 50 40 30 20 10

6000 4000 2000

0

0

2000 4000 6000 8000 10000 12000 14000

Cutting torque (kNm) 6400

Fig. 33. Jack thrust force, cutting torque and PRev.

P=10741kN P=10876kN P=12694kN

5600 0.352 0.350

4800

4000

3200

0.50

0.55

0.60

0.65

0.70

0.75

0.80

0.85

Chamber pressure (MPa)

Cutting torque (kNm)

Jacke thrust force (kN)

0.348 0.346 0.344 0.342 0.340 0.338

Cutterhead rotation speed (rpm)

0.336

Fig. 31. Cutting torque and cutterhead rotation speed.

4.2

4.3

4.4

4.5

4.6

4.7

4.8

4.9

Screw conveyor rotation speed/advance rate (r/mm) Fig. 34. Chamber pressure vs. ratio of screw rotation speed to advance rate.

14000

Cutting torque (kNm)

12000

3m

10000

2m

9m

2m

JS17-9 JS17-8 JS17-7

JS17-6

JS17-5

2m JS17-4

2m

3m

JS17-3 JS17-2 JS17-1

8000 SX-2

XX-2

SX-1

XX-1

6000

3m JS1-9

4000

JS1-8

9m

2m 2m JS1-7

JS1-6

JS1-5

2m JS1-4

2m

JS1-3

3m

JS1-2

11000 12000 13000 14000 15000 16000 17000 18000

Thrust force (kN) Fig. 32. Total thrust force and cutting torque.

uplink

downlink

Fig. 35. Monitoring profiles of ground surface displacement.

JS1-1

13

4

0.200

2

0.195

0

0.190

-2

0.185

-4

-20

-10

0

10

0.180 30

20

Relative position to JS3 section (m) -20

-10

0

10

20

30

40

50

0

Ground surface displacement (mm)

ground surface displacenment rotation speed/advance rate

Screw conveyor rotation speed/advance rate

Maximum displacement of ground surface (mm)

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

-10 -20 cutting face

-30 -40 -50 -60

Relative position to JS3 section (m) Fig. 39. Longitudinal ground surface displacement (subsidence). Fig. 36. Maximum ground surface displacement before cutting face vs. n2/v value.

Distance to the line center (m) -15

Relative position to JS3 section (m) 0

-80

-60

-40

-20

0

20

-10

-20

cutting face

-30

-40

Transverse ground surface displacement (mm)

Ground surface displacement (mm)

10

-10

-5

0

5

10

15

0 -5 -10 downlink

uplink

-15 -20 -25 -30 -35

Fig. 37. Longitudinal ground surface displacement (upheaval).

Transverse ground surface displacement (mm)

Fig. 40. Transverse ground surface displacement (subsidence).

2.0

Fig. 37 shows the longitudinal displacement curve of ground surface when cutting face is 20 m away before JS3 section, and its n2/v value is 0.189. Meanwhile, the corresponding transverse surface deformation shape at JS3 is displayed in Fig. 38. As can be seen, due to low n2/v value, the ground surface in front of the cutterhead was uplifted in arch shape. The deformation value of ground surface on downlink is a little larger than that on uplink. Figs. 39 and 40 show the longitudinal and transverse deformation of ground surface when the cutting face had past 25 m away from JS3 section. At this time, due to high n2/v value, a settlement trough appears at the ground surface before cutting face. Similarly, the subsidence value of ground surface on uplink is smaller than that on downlink.

1.8 1.6 1.4 1.2 1.0 0.8 0.6 downlink

uplink

0.4 0.2 -15

-10

-5

0

5

10

15

Distance to the line center (m) Fig. 38. Transverse ground surface displacement (upheaval).

In a sense, the ratio of screw rotation speed to advance rate (n2/v) represents mucking ratio. Fig. 36 shows the maximum ground surface displacement before cutting face vs. the n2/v value when cutting face arrives at different position to JS3 section. It can be seen that, when the n2/v value is high, the ground surface will subside; on the contrary, the surface will be uplifted. It will be illustrated in detail in the following paragraphs.

6. Conclusions In order to reveal the inherent relation between soil disturbance and the variation of shield machine operation parameters, a series of model tests to simulate EPB shield machine tunnelling in soft ground and field investigation were carried out simultaneously. Based on the analyses of test results and field monitoring data, the following conclusions can be drawn: (1) Both overburden ratio and cutterhead aperture ratio have great influence on frontal face thrust force and cutting torque, i.e. for a given penetration per revolution, the larger

14

Q. Xu et al. / Tunnelling and Underground Space Technology 26 (2011) 1–14

(2)

(3)

(4)

(5)

the overburden ratio is or the smaller the cutterhead aperture ratio is, the larger the thrust force and the cutting torque are needed. What is more, both frontal face thrust force and cutting torque presents a form of exponential relationship with PRev. The relationship between frontal face thrust force and cutting torque is piecewise linear. The linear slope is also influenced by the overburden ratio and cutterhead aperture ratio. As for EPB shield machine, chamber pressure is mainly affected by mucking ratio. Chamber pressure decreases in the form of negative exponential function with the increase of mucking ratio, which is also influenced by overburden ratio and cutterhead aperture ratio. Shield tunnelling process can cause earth pressure increment inside the soil body, and the generated earth pressure increment is relatively larger at the position which is close to the shield machine. Furthermore, this pressure increment is also affected by mucking ratio. Variation of machine’s operation parameters, such as screw rotation speed (n2) and advance rate (v), can influence mucking ratio greatly, which leads to variation of chamber pressure, and further results in ground surface displacement. That is, if the n2/v value decreases, the chamber pressure will increase accordingly, which ultimately results in the ground surface upheaving in arch shape before the cutting face; on the contrary, ground surface will subside in trough shape due to low n/v value. Therefore, during the tunnelling process, ground surface deformation information should be fed back to the tunnelling face to guide machine’s operation parameters immediately, which does good to reduce disturbance to surrounding environment.

Acknowledgements This research work is sponsored by the National High Technology Research and Development Program of China (40672184), the National Science Foundation for Post Doctoral Scientists of China,

and the Open Foundation of the Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education (Tongji University) of China. References Harris, DI., Mair, RJ., Love, JP., Taylor, RN., Henderson, TO., 1994. Observations of ground and structure movements for compensation grouting during tunnel construction at Waterloo station. Geotechnique 44 (4), 691–713. Kashima, Y., Kondo, N., Inoue, M., 1996. Development and application of the DPLEX shield method: results of experiments using shield and segment models and application of the method in tunnel construction [J]. Tunnelling and Underground Space Technology 11 (1), 45–50. Lee, Y.J., Bassett, R.H., 2007. Influence zones for 2D pile–soil-tunnelling interaction based on model test and numerical analysis [J]. Tunnelling and Underground Space Technology, 22:325–342. Maeda, M., Kushiyama, K., 2005. Use of compact shield tunneling method in urban underground construction. Tunnelling and Underground Space Technology 20, 159–166. Meguid, MA., Saada, O., Nunes, MA., et al., 2008. Physical modeling of tunnels in soft ground: a review [J]. Tunnelling and Underground Space Technology 23, 185– 198. Migliazza, M., Chiorboli, M., Giani, GP., 2009. Comparison of analytical method, 3D finite element model with experimental subsidence measurements resulting from the extension of the Milan underground [J]. Computers and Geotechnics (36), 113–124. Nakamura, H., Kubota, T., Furukawa, M., et al., 2003. Unified construction of running track tunnel and crossover tunnel for subway by rectangular shape double track cross-section shield machine. Tunnelling and Underground Space Technology 18, 253–262. Nomoto, T., Imamura, S., Hagiwara, T., et al., 1999. Shield tunnel construction in centrifuge [J]. Journal of Geotechnical and Geoenvironmental Engineering 125 (4), 289–300. Peck, R.B., 1969. Deep excavations and tunneling in soft ground. In: Proceedings of the 7th International Conference on Soil Mechanics and Foundation Engineering, Mexico City, pp. 225–290. Pellet-Beaucour, A.L., Kastner, R., 2002. Experimental and analytical study of friction forces during micro-tunneling operations [J]. Tunnelling and Underground Space Technology 17, 83–97. Rowe, RK., Kack, GJ., 1983. A theoretical examination of the settlements induced by tunneling: from cases history. Canadian Geotechnical Journal 20 (2), 299–314. Sharma, JS., Bolton, MD., Boyle, RE., 2001. A new technique for simulation of tunnel excavation in a centrifuge [J]. Geotechnical Testing Journal 24 (4), 343–349. Sterpi, D., Cividini, A., 2004. A physical and numerical investigation on the stability of shallow tunnels in strain softening media [J]. Rock Mechanical and Rock Engineering 37 (4), 277–298. Viswanadham, B.V.S., 2005. Centrifuge modeling of geotechnical structures [J]. Chemical Business, 59–65.