Professional Documents
Culture Documents
CIVIL ENGINEERING
KEYWORDS Abstract The Greater Cairo metro-Line 3, the major project of underground structure in Cairo
Tunnels; city, Egypt, is currently under constructed. Ground movement is expected during the construction
Surface settlement; with tunneling boring machine as Cairo metro tunnel passes through sand soil.
Numerical modeling and In the present study, finite element model is used to model tunnel system performance based on
analysis; the case study. An elasto-plastic constitutive model is adopted to represent the soil behavior sur-
Displacement; rounding the tunnel. The effects are expressed in terms of surface displacement and soil stress
Design; change caused by tunneling. The subsoil stresses undergo three phases of change. At these phases,
Deformations
the loading steps of the tunnel construction are predicted using the 2-D finite element analysis.
Ground movement and construction influence are obtained by the numerical model. A compar-
ison is made between the computed tunnel performance and the observed behavior. The compari-
son reveals a good agreement between the calculated and the observed values.
2013 Ain Shams University. Production and hosting by Elsevier B.V.
All rights reserved.
2090-4479 2013 Ain Shams University. Production and hosting by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.asej.2013.04.002
710 S.A. Mazek, H.A. Almannaei
Darabi et al. [9] presented an appropriate model to predict settlement and vertical displacement at different locations
the behavior of the tunnel in Tehran No. 3 subway line. They and levels around tunnel system. The constitutive model for
employed empirical methods to determine the variation of ra- this analysis utilizes elasto-plastic materials. A yielding func-
dial displacements along the longitudinal direction of a tunnel. tion of the Mohr–Coulomb type and a plastic potential func-
They also determined the tunnel deformation using numerical tion of the Drucker–Prager type are employed. A linear
analyses. Elsayed [10] used 2-D finite element analysis to model constitutive model is employed to represent the tunnel liner.
two phases of tunneling process. First phase, the excavation Model boundaries and volume losses (VL) are discussed to
phase, was responsible to determine the pre-lining rock mass understand the performance of the metro tunnel. The associ-
deformations and the reduced in situ stresses. Second phase, ated stress changes in soil are studied and presented based
the interaction phase, modeled the compatibility of the rock- on different loading steps of the tunnel construction. The re-
lining system. The deformations and stresses of the rock-lining sults obtained by the 2-D FEA are compared with those ob-
system and the final rock mass pressure acting on the lining tained by the field measurement to assess the accuracy of the
were determined. Cavalaro et al. [11] analyzed the influence 2-D finite element model. There is a good agreement between
of the contact deficiencies between the segmented tunnel lining readings obtained by both the FEA and the filed data.
during the construction of tunnels and the consequent damage
procedure with TBM. They used the finite element analysis to 2. Greater Cairo metro-Line 3 (case study)
simulate the contact deficiencies.
Kivi et al. [12] investigated settlement control of large span The Greater Cairo metro-Line 3 passes through a wide range
underground station in Tehran metro using 3-D finite element of soil condition [20]. Line 3 moves underground from Abb-
analysis. They discussed the impact of central beam column asia station to Attaba station. Line 3 is 4.3 km long. Fig. 1
(CBC) on the rigidity of the supporting tunnel system. Liu shows the plan of the Greater Cairo metro-Line 3 [20]. The
et al. [13] discussed the ground movement property caused internal diameter of tunnel liner is 8.30 m, as shown in
by shield tunneling and expanding construction. Ground Fig. 2. Based on the project document prepared by the NAT
movement and construction influence were obtained by [20], the soil profile is analyzed and discussed to use soil prop-
numerical model. Wang et al. [14] used finite element analysis erties at the 2-D finite element model. Line 3 tunnel was exca-
to predict surface settlement above tunnel in clay till. The influ- vated under the central region of Cairo city. Line 3 was built
ence of drainage condition on surface settlement was investi- on alluvial deposits through the Nile valley.
gated. Wang et al. [15] applied finite element analysis to
investigate the effect of tunneling induced ground movement
3. Soil and metro tunnel interaction
on buried pipelines. Modeling of geotechnical properties and
tunneling procedure is a sophisticated numerical problem
[8,16–19]. The analyzed project area lies within the alluvial plain, which
The Greater Cairo metro tunnel-Line 3 is presented and covers the major area of the lowland portion of the Nile valley
discussed in the present study as a case study. Line 3 has been in Cairo vicinity [20]. Five distinct soil layers are encountered
constructed since 2011. 2-D finite elements analysis (FEA) is through the case study [20].
used to understand performance of tunnel system based on a Site investigations along Line 3 alignment indicate that the
case study. The 2-D FEA is also adopted to estimate surface soil profile consists of a surficial fill layer ranging from 2 to 4 m
Figure 1 Plan of the Greater Cairo metro (Line 3) (after NAT, 2009).
Finite element model of Cairo metro tunnel-Line 3 performance 711
Figure 2 Cross section of the Greater Cairo metro (Line 3) (after NAT, 2009).
in thickness. The fill is followed by upper sand layer. The is not as local construction as it is simulated in the model,
upper sand layer is varied from dense to very dense poorly which is conservative. The computed normal forces and bend-
graded silty sand with a thickness varying from 2.1 m to ing moment values must comply with the strength of the 50-cm
6.6 m. Underneath the upper sand layer, the soil is very dense thick reinforced segments and the 30-cm thick joints between
sand. This sand layer is called middle sand below the upper the segments. The tunnel liner is assumed to behave in a linear
sand layer. The thickness of the middle sand layer is varied manner in the 2-D FEA. The characteristic of the tunnel liner
from 5.5 m to 10 m. Beneath the middle sand layer, there is is tabulated in Table 2.
a very stiff silty clay layer with thickness of the layer which Effective stress is used in the finite element analysis, as the
is varied from 1.5 m to 4.4 m. Below the silty clay layer, the soil metro tunnel is located in the lower sand layer. The variation
layer is sand soil. This sand layer is called lower sand layer. of soil modulus (Es) with confining pressure is related to effective
The lower sand layer is very dense gravely sand. The lower pressure based on Janbu’s empirical equation [21] as presented
sand layer is extended down to bedrock. The ground water ta- in Eq. (1). The different soil parameters (m, n) are selected to
ble varies between two meters to four meters from the ground simulate the behavior of different sand soil types [22,23].
surface. The main soil parameters required to model the per- n
r3
formance of the case study are analyzed by the authors and Es ¼ mPa ð1Þ
Pa
presented in Table 1 based on the soil report prepared by the
National Authority for Tunnels (NAT) [20]. In which, modulus number (m) and exponent number (n)
Fig. 3 shows the cross section of soil parameters. The tunnel are both pure numbers, Pa is the atmospheric pressure ex-
liner is composed of 50-cm thick segments. The segments joints pressed in appropriate units, and r3 is an effective confining
are never aligned along the tunnel, and the thickness reduction pressure.
Figure 3 Cross section of soil parameters along the Greater Cairo metro-Line 3 tunnel.
The horizontal plane at the bottom of the mesh represented numerical model should reflect the behavior of the metro tun-
a rigid bedrock layer, and the movement at this plane is re- nel in the field. The 2-D finite element mesh models soil block
strained in all directions. The vertical boundaries of the 2-D fi- width and depth in x and y directions, respectively, as shown in
nite element model are restrained by roller supports to prevent Fig. 4. The soil media are modeled and simulated using an
a movement normal to the boundaries. The movement at the elasto-plastic constitutive model. A yielding function of the
upper horizontal plane is free to simulate a free ground sur- Mohr–Coulomb types and a plastic function of the Drucker–
face. The 2-D finite element mesh is shown in Fig. 4. Prager type are adopted in this study [26]. At the plain strain
The finite element analysis is carried out to simulate the condition for soil media, the Drucker–Prager model can be
construction of Line 3. The stress changes in surrounding soil used to simulate the soil behavior at the 2-D FEA [26].
due to tunneling are investigated to study the detailed soil The element size is chosen to be three meters along the out-
behavior around the metro tunnel. The stresses in the subsoil er boundary of the soil block. The element size is chosen to be
have undergone three phases of change. At these phases, the one meter along the boundary of the tunnel liner. The diameter
loading steps of the metro tunnel construction are simulated of metro tunnel is 9.15 m. Numerical analysis is adopted for
using the 2-D finite element analysis. the metro tunnel located at 23 m depth from the ground sur-
First, the initial principal stresses are computed with the ab- face, as shown in Fig. 1. The volume loss (VL) is the ratio of
sence of the metro tunnel. Second, the excavation of the metro the difference between excavated soil volume and tunnel vol-
tunnel is modeled by means of the finite element method. The ume over excavated soil volume. In this section, the parametric
excavation is simulated by the removal of those elements inside study assumes volume loss to be 3%. In the present study, vol-
the boundary of the metro tunnel surface to be exposed by the ume loss is studied and varied from 1.5% to 4.5% as discussed
excavation. The excavated tunnel boundary is free to move un- later in Section 6.
til the soil comes into contact with the metro tunnel liner The suitable geometric boundaries (model width and model
resulting from volume loss. The volume loss is considered in height) are studied to reflect performance of tunnel system.
this study. Third, the calculated changes in stresses are then Based on soil stress change along geometric model boundaries,
added to the initial principal stresses computed from the first the parametric study is conducted to determine the suitable
phase to determine the final principal stresses resulting from dimensions beyond which no changes in soil stress and vertical
the metro tunnel construction. Fig. 5 shows final vertical stress displacement are occurred. In this study, the model width is
around the metro tunnel system. varied from 40 m to 120 m.
The initial in situ stresses before tunneling are calculated and The calculated surface settlements of the metro tunnel
plotted in Fig. 6 at the right-hand side. The vertical stress against different model widths are shown in Fig. 7. The calcu-
change after tunneling is calculated and shown in Fig. 6 at the lated crown settlement of the metro tunnel-Line 3 versus dif-
left-hand side. The results show that as the soil depth beneath ferent model widths is also presented in Fig. 8. The
the invert of the metro tunnel exceeds three times the tunnel calculated invert heave of the metro tunnel against different
diameter, there is no change in the soil stress due to tunneling model widths is again presented in Fig. 9. Based on the finite
process, as shown in Fig. 6. Based on the proposed 2-D FEM, element analysis, the 100-m model width is chosen for the rest
the calculated vertical stress change in soil due to tunnel con- of 2-D analysis to reflect the performance of the metro tunnel
struction at 54 m below ground surface level is equal to zero va- system beyond which no change in vertical displacement and
lue. At 54 m below the ground surface level, there is no soil soil stress is occurred.
displacement occurred due to tunneling. The results also reveal
that the soil above the crown of the metro tunnel settles down- 6. Volume loss impact
ward as the final vertical stress moves downward. The soil under
the invert of the metro tunnel excavation heaves as the final The suitable volume loss is studied to reflect the performance
vertical stress moves upward. of the metro tunnel system-Line 3. The 100-m model width
5. Geometric model boundaries and 54-m model height are studied using the 2-D FEA based
on different volume losses. The volume loss (VL) is ranged
from (1.5% to 4.5%) [1]. In the presented study, the VL is var-
A parametric study is conducted to choose the suitable geo-
ied from 1.5% to 4.5%.
metric boundaries of the 2-D numerical model. The 2-D
Surface settlement due to construction of the metro tunnel
is calculated by the 2-D FEM. The calculated surface settle-
ment against different volume losses is shown in Fig. 10. The
results show that the increase in the volume loss leads to in-
crease the surface settlements due to tunneling. The volume
loss of 3% is chosen to practically reflect the performance of
the metro tunnel system-Line 3.
8. Conclusions
Acknowledgement
References
Figure 11 Comparison between measured and calculated settle-
ment of the Grater Cairo metro tunnel-Line 3. [1] El-Nahhass FM. Soft ground tunneling in Egypt. Geotechnical
challenges and expectations. J Tunnel Underground Space Tech-
nol 1999;14(3):245–56.
[2] Mazek SA, El-Tehawy EM. Impact of tunneling running side-by-
side to an existing tunnel on tunnel performance using non-linear
analysis. In: Proceeding of the 7th ICCAE. Cairo, Egypt; 2008.
[3] El-Nahhas FM. Spatial mode of ground subsidence above
advancing shielded tunnels. In: Proceeding of international
congress on large underground opening, Fireze, Italy, vol. 1;
1986.p. 720–5.
[4] Bicel JO, Kuesel TR. Tunnel engineering hand book. New York,
USA: Van Nostrand Reinhold Company; 1982.
[5] Lee CJ, Chiang KH, Kuo CM. Ground movement and tunnel
stability when tunneling in sandy ground. Proc J Chinese Inst Eng
2004;27(7):pp. 1021–1032.
[6] Mazek SA, Law KT, Lau DT. Effect of grouting on soil
reinforcing and tunnel deformation. In: International under-
ground infrastructure research conference, Waterloo University,
Kitchener, Canada; 2001a.
[7] Abu-Krisha A. Settlement control of CWO Sewer tunnel during
boring El-Azhar road tunnels in Cairo. In: Proceeding of the
international world to congress Milano; 2001. p. 3–9.
Figure 12 Calculated vertical displacement at different levels [8] Abu-Farsakh MY, Tumay MT, Fellow ASCE. Finite element
along centerline of the metro tunnel-Line 3. analysis of ground response due to tunnel excavation in soils. In:
Proceeding of the international journal for numerical and analyt-
ical methods in geomechanics engineering, AinShams University,
Egypt, vol. 23; 1999. p. 524–5.
metro tunnel moves down due to the soil stress change. The
[9] Darabi A, Ahangari Kaveh A, Noorzad A, Arab A. Subsidence
soil stress change above the crown tunnel pushes soil to move estimation utilizing various approaches – a case study: Tehran No. 3
downward. The soil under the invert of the metro tunnel subway line. Tunnel Underground Space Technol 2012;31:117–27.
heaves due to the soil stress change around tunnel excavation. [10] Elsayed A. Study of rock-lining interaction for circular tunnels
The soil stress change beneath the tunnel invert pushes soil to using finite element analysis. Proc Jordan J Civ Eng
move upward. 2011;5(1):50–64.
716 S.A. Mazek, H.A. Almannaei
[11] Cavalaro SHP, Blom CBM, Walraven JC, Aguado A. 14- Dr. Sherif Abdel Aziz Mazek is an Associate
Structure analysis of control deficiencies in segmented lining. Prof. of Civil Engineering Department, Mili-
Tunnel Underground Space Technol 2011;26, 651-127. tary Technical College, Cairo, Egypt. He was
[12] Kivi A Valizadeh, Sadaghiani MH, Ahmadi MM. Numerical born on October 7, 1968. He completed B.Sc.
modeling of ground settlement control large span underground in Civil Engineering on July 1991. His Accu-
metro station in Tehran metro using central beam column (CBC). mulated Evaluation was Excellent with Honor
Tunnel Underground Space Technol 2012;28:218–28. and final year evaluation was also Excellent.
[13] Liu Jiangfeng, Qi Taiyue, Wu Zhanrui. Analysis of ground After graduation, he was worked as an
movement due to metro tunnel station driven with enlarging Assistance Lecturer (1991–1998). He got a
shield tunnels under building and its parameter sensitivity Egyptian scholarship from Faculty of Engi-
analysis. Tunnel Underground Space Technol 2012;28:287–96. neering, Carleton University, Ottawa, Canada
[14] Wang Zhechao, Wong Ron CK, Li Shucai, Qiao Liping. Analysis (1999–2003) for Ph.D. Degree in Civil Engineering. His Affiliation was
of ground movement due to metro tunnel station driven with a faculty member at Military Technical College (2003–2010). He
enlarging shield tunnels under building and its parameter sensi- completed his M. Sc. on April 1997 (1993–1997). His M.Sc. Thesis
tivity analysis. Tunnel Underground Space Technol Title was ‘‘Underpinning of tunnel and underground structure.’’ He
2012;30:85–92. got Faculty Graduation of M. Sc. in Military Technical College, Cairo,
[15] Wang Yu, Shi Jaingwei, Ng Charles. Numerical modeling of Egypt. He got Ph. D. on March 2003 (1999–2003). His Ph.D. Thesis
tunneling effect on buried pipelines. Can Geotech J Title was ‘‘Soil structure interaction modeling and analysis of tunnel
2011;48:1125–37. structures.’’ And he got Faculty Graduation of Ph. D. in Faculty of
[16] Chehade FH, Shahrour I. Numerical analysis of the interaction Engineering, Carleton University, Ottawa, Canada. He joined as a
between twin-tunnels: influence of the relative position and Consultant Engineer on June 2007. He also finished a course, ‘‘Certi-
construction procedure. J Tunnel Underground Space Technol fication at Time Management,’’ from American University, Cairo. He
2008;23:210–4. teaches the following courses: (1) Reinforced Concrete, (2) Foundation
[17] Mazek SA, Law KT, Lau DT. 3-D Analysis on the performance Analysis and Design, (3) Tunnel and underground structure Engi-
of a grouted tunnel. Canadian geotechnical conference, Calgary, neering, (4) Finite element analysis, (5) Theory of structure. He
Canada, vol. 1. p. 111–9. supervised nine theses (two Ph.D. Theses and seven M.Sc. Theses). His
[18] Mazek SA. Impact of grouting on an existing tunnel performance scientific activities are as follows: (1) Member at Egyptian Society of
under passed by another tunnel. In: Proceeding of the 11th Tunnels-Cairo-Egypt (Scientific Society), (2) Member at Intentional
ICSGE, 17–19 May, Cairo, Egypt; 2005. Society of Tunnels (Scientific Society), (3) Members at Egyptian
[19] Vermeer PA, Moller SC. Numerical simulation of tunnel instal- Geotechnical Society (EGS), (4) Members at International Society of
lation. J Tunnel Underground Space Technol 2008;23:461–75. Soil Mechanics and Geotechnical Engineering (ISSMGE). He pub-
[20] National Authority for Tunnels (NAT). Project Document; 2009. lished 35 papers.
[21] Janbu N. Soil compressibility as determined by oedometer and
triaxial tests. In: European conference on soil mechanics of
foundation engineering, Wiesbaden, Germany, vol. 1; 1963. p. 19– Eng. Hassan Abdul Wahab Almannai is a civil
25. engineer at engineering department in Bahrain
[22] Duncan JM, Chang CY. Nonlinear analysis of stress and strain in and at Ismail Khonji Associates (Consultant
soils. J Soil Mech Found Div ASCE 1970;96(SM5), September. Engineers). He was born on December 26,
[23] Duncan JM, Byrne PM, Wong KS, Mabry P. Strength, stress- 1984. He completed B.Sc. in Civil Engineering
strain, and bulk modulus parameters for finite element analysis of in July 2007 from Military Technical College,
stresses and movements in soil masses. University of California, Cairo, Egypt. His accumulated evaluation was
Berkeley, CA. Report no. UCB/GT/80-01; 1980. good and final year evaluation was also good.
[24] Byrne PM, Cheung H, Yan L. Soil parameters for deformation He completed his M.Sc. in civil engineering in
analysis of sand masses. Can Geotech J 1987;24:366–76. September 2012 (2010– 2012). He also got
[25] COSMOS/M program. Structural research and analysis corpora- M.Sc. from Military Technical College, Cairo,
tion, Los Angles, California, USA; 2005. Egypt. His M.Sc. Thesis Title was ‘‘Modeling and analysis of under-
[26] Chen WF, Mizuno E. Nonlinear analysis in soil mechanics. ground reinforced concrete structure.
Netherlands: Elsevier Science Publishers B.V; 1990.