Feasibility evaluation for excavation of Naghshe Jahan Square subway station by underground methods

Morteza Abdi Cherlo, Hamid Hashemolhosseini, Masoud Cheraghi, Saeed Mahdevari

DepartmentofMiningEngineering,IsfahanUniversityofTechnology,Isfahan,Iran

Feasibility evaluation for excavation of Naghshe Jahan Square subway station by underground methods

Morteza Abdi Cherlo∗, Hamid Hashemolhosseini, Masoud Cheraghi, Saeed Mahdevari

DepartmentofMiningEngineering,IsfahanUniversityofTechnology,Isfahan,Iran

A R T I C L E I N F O

Articlehistory:

Received 10 May 2013

Received in revised form 3 August 2013

Accepted 1 September 2013

Subway station

Sequential excavation method

Compressed air

Ground freezing

Large-diameter curved pipe roofing method

In recent years, in reaction to the increasing usage of urban areas, the excavation of underground spaces has been developed. One of the most challenging issues encountered by engineers is the construction of subway stations as large underground spaces at shallow depth with soft surrounding soils. In this paper, Naghshe Jahan Square subway station located in Isfahan, Iran, has been simulated by geomechanical finite difference method (FDM). This station is located under important historical structures. Therefore, the ground displacement and surface settlement induced by the excavation of the subway station should be strictly controlled. Many of such problems are affected by selected excavation method. For these reasons, different underground excavation methods associated with construction have been studied. In this study, sequential excavation method and large-diameter curved pipe roofing method are used and the numerical results of the two methods are compared. The presence of groundwater table obliges us to choose special techniques for the stability of the ground around the subway station during construction; hence compressed air and ground freezing techniques are utilized in the simulations of the subway station. Finally, after choosing appropriate support systems, the large-diameter curved pipe roofing method with 1.5 m spacing between curved pipes is proposed.

© 2013 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting by Elsevier B.V. All rights reserved.

1. Introduction

In recent years, different excavation methods have been used in tunnel excavation, such as sequential excavation method (Gomes et al., 2009), cut and cover method, enlargement shield tunneling method (Kunihiko and Kenichi, 2006), multi-face shield construction method (Kunihiko and Kenichi, 2006), microtunneled arch method (Lunardi, 1990), concrete arch pre-supporting system (Sadaghiani and Dadizadeh, 2010) and large-diameter curved pipe roofing method (IDI, 2008). In this study, at the first stage, the sequential excavation method as a commonly used method in Iran and the large-diameter curved pipe roofing method as a new method at present are introduced and then the excavation of Naghshe Jahan Square subway station by taking advantage of numerical modeling is analyzed. Finally, the feasibility of the subway station excavation with regard to numerical results is evaluated.

2. Excavation methods

2.1.Sequentialexcavationmethod

In this method, after the gradual excavation of underground space and a permissible deformation, a suitable support system must be installed. Performance procedure in this method is related to the soil condition around the underground space. According to the soil homogeneity, underground space could be excavated partially or completely (Tonon, 2010). One of the excavation methods that is effective in reducing the settlement of ground surface is the sequential excavation method with middle columns introduced by Irshad and Helfin (1988).

2.2.Large-diametercurvedpiperoofingmethod

The large-diameter curved pipe roofing method was introduced in Japan in 2008. This method was initiated for the construction of the Tomigaya Entrance/Exit Tunnel project of the Tokyo Metropolitan Expressway. In this method, before the excavation of underground structures, large-diameter steel curved pipes were placed as freezing equipment from inside of the pre-excavatedtunnel to create frozen soil as a pre-supporting system so that the inside of the pre-supporting system could be excavated by mountain tunneling method. Fig. 1 shows the inside of a shielded enclosure and equipment for the construction of underground structures by large-diameter curved pipe roofing method, such as (1) the departure entrance; (2) the lower frame; (3) the bottom thrusting equipment; (4) large-diameter curve boring machine; (5) the lower and upper parts of the curved pipe roof segment; and (6) the end-point entrance room of the ramp shield (Kunihiko and Kenichi, 2006).

Fig. 1. Construction image inside enclosure (Kunihiko and Kenichi, 2006).

3. Different methods to prevent water entrance into underground space

There are some methods for the excavation of urban underground space in soft ground beneath water table, such as dewatering, chemical consolidation and grouting, freezing, closed face tunneling machines and compressed air. All of these methods have some advantages and disadvantages. The compressed air method is still frequently used, and it is important that engineers understand this method’s effect upon both labor force and environment in order to decrease its incompatible effects.

3.1.Compressedair

Compressed air in shallow tunneling is used to prevent the flow of water into underground structures and nowadays, especially in urban areas, it is very useful because the usage of compressed air impedes the decrease of groundwater level and obstructs the advent of pollutant materials into underground space. One of the notorious diseases related to work in places with high pressure is decompression illness and also there are many pernicious in fluences on workers’ health while they are working under compressed air (Kindwall, 1988; Lamont, 2000). In this condition, the compressed air is applied uniformly in underground structure perimeter and it is equal to the water pressure in underground space invert. The pressure applied to the model is different between the compressed air pressure and water pressure, and the remaining pressure is applied in underground space and its perimeter that temporary support system was installed (Fig. 2) (Kirkland, 1984). After the installation of final lining, the compressed air pressure was removed. In this paper, the magnitude of the pressure for the excavation of the subway station in sequential excavation method is 1.026 bar (1 bar = 100 kPa) that is less than the critical value (2.4–2.7 bar) (Bickel and Kuesel, 1982).

3.2.Groundfreezing

In ground freezing process, by converting in situ pore water into ice, the soil particles link together like the cement in concrete. According to the strength and impermeability of the frozen soil mass, using ground freezing especially in tunneling under the foundation of buildings is very significant in reducing ground surface settlement and the movement of building foundations. Brine or liquid nitrogen is usually used for ground freezing. The strength characteristic of frozen soil has been widely used in the design and construction of underground activities (Habetaand and Schäfers, 2006). Regarding different researches, the Mohr–Coulomb yield criterion is suitable for frozen soil when the confining pressure is less than 4 MPa, and it is not suitable when the confining pressure is higher than 4 MPa. A non-linear Mohr–Coulomb criterion was suggested by Yang et al. (2010), in which the generalized friction angle and cohesion are regarded as the functions of confining pressure. This criterion was proposed based on the experimental results. For determining the bulk modulus and shear modulus of frozen soil, the curves of relation between them and confining pressure were used (Lai et al., 2009; Yang et al., 2010).

Fig. 2. The different pressures and the remaining pressure applied to the station.

4. Site specification

Naghshe Jahan Square as an important historical construction in Isfahan, Iran, has the dimension of 507 m × 158 m. The diameter of twin subway tunnels excavated by shield tunneling method is 7 m and the dimension of subway station for the two excavation methods is about 24 m in width and 102 m in length. The groundwater level is 10 m lower than the ground surface and the distance between the station crown and the ground surface is about 11 m. The cross-sections of the subway station and soil layers for different excavation methods are shown in Fig. 3.

5. Numerical modeling

The two models in numerical analysis have the dimension of 160 m in width (x-direction), 76 m in height (y-direction), and 80 m in length (z-direction). The number of meshes with 0.5 m in length (z-direction) is 160 (Fig. 4). The segments used in TBM tunneling have the thickness of 0.3 m and the length of 1.5 m along the axis of tunnel. The model boundaries are fixed inx-direction andydirection, and also the bottom boundary of the model is fixed inzdirection, but the top boundary of the model is free and could move alongzaxis. The ratio of horizontal stress to vertical stress (K) is 0.6. Traffic load is about 20 kPa which is applied over ground surface and the groundwater table is applied in the model. Table 1 shows the soil layers parameters surrounding Naghshe Jahan Square subway station.

Fig. 3. The cross-sections of the subway station excavated by (a) sequential excavation method and (b) large-diameter curved pipe roofing method.

Fig. 4. The first model (sequential excavation method).

5.1.Sequentialexcavationmethod

After the modeling of stationary situation in the ground, the simulation of different stages in the model was started. The different stages of excavation and the stages of temporary and final support installation are shown respectively in Figs. 5 and 6. The sequence of excavation is according to drift number so that the distance between different drifts in Fig. 5 is kept until the completion of excavation. We should consider that the twin tunnels were completely excavated by TBM before the excavation of the subway station. The drifts No. 3 were excavated and the segments of the subway tunnels were removed simultaneously, and the magnitude of advance step is equal to segment length along the tunnel axis (1.5 m) in the drifts No. 3. Considering the great influence of advance step on ground surface settlement, ultimately the advance step in drifts No. 4 is chosen to be equal to 1 m. The sequence of excavation and the support systems applied in the model are shown in Fig. 7.

Table 1 Soil parameters around the subway station (Zamin Fanavaran Engineering Co., 2007).

Fig. 5. The sequence of excavation with various steps in different drifts.

Fig. 6. The sequence of excavation and the installation of support systems in different stages.

5.1.1.Theselectionandsimulationofsupportsystems

For supporting drifts, three types of supports were used that are the combination of shotcrete, lattice girder and welded wiremesh. Table 2 shows the characteristics of lattice girders, the parameters in which are illustrated in Fig. 8; and Table 3 shows the different supports used for supporting drifts. One of the key factors isthe stability of drifts faces in different stages; considering the instability of the drifts faces, the performance of shotcrete with 5 cm thickness was proposed. For the modeling of lattice girder and shotcrete, the equivalent section approach introduced by Carranza-Torres and Diederichs (2009) was used. In this approach, the equivalent thickness and Young’s modulus could be calculated according to the following equations:

Fig. 7. (a) The sequence of excavation applied in the model and (b) the support systems simulated in different stages.

Table 2 The characteristics and types of lattice girders.

Fig. 8. Regarding lattice girders cross-section, in Table 2, A is the area of four bars. Parameters B, H, S and D are respectively the lattice girders width, lattice girders height, bar size and stiffener diameter. W and J refer to the first and second moment of area, and the weight of lattice girders is also shown in Table 2.

Table 3 The different support systems used in the drifts.

Fig. 9. Vertical stress induced by the excavation of the subway station (unit: Pa).

whereteqandEeqare the equivalent thickness and the equivalent Young’s modulus, respectively;KiandDi(i= 1, 2) are the flexibility and compressibility coef ficients related to steel set and shotcrete, respectively;nis the number of lattice girders used inbmeter length of shotcrete (Hoek et al., 2008; Carranza-Torres and Diederichs, 2009); andi= 1 (steel set) and 2 (shotcrete).

To calculateKandD, the following equations were proposed by Carranza-Torres and Diederichs (2009):

whereEi,Aiand νiare the Young’s modulus, cross-sectional area and Poisson’s ratio, respectively.

5.1.2.Supportsystemdesign

One of the important factors in the support of underground space is the magnitude of bending moment and thrust force derived from support systems during excavation. In sequential excavation method, columns play a determining role in the enduring of the load induced by the subway station excavation. The columns used to support the station have the cross-section of 1.5 m × 1.4 m and involves 50 bar with 30 mm in diameter. According to Fig. 9, the maximum stress appears in the columns, showing the importance of using columns in this method of excavation. According to ACI 318-89 (ACI Committee 318, 1989), the maximum stress induced in the columns is less than the maximum endurable stress in the columns (safety factorFs= 1.77).

For the final lining of the station, considering the control of bending moment and thrust force, ultimately, the reinforced concrete with 40 cm thickness was selected (Fig. 10). According to numerical results, the maximum bending moment and thrust force in the final lining (Fig. 11) are shown in Fig. 12 (Fs= 1.6).

Fig. 10. The cross-section of the final lining used in the station.

Fig. 11. Bending moment and thrust force induced in final lining. (a) Bending moment (unit: N m) and (b) thrust force (unit: N).

Fig. 12. Thrust force–bending moment interaction diagram (ACI Committee 318, 1989).

5.1.3.Obtainedresults

The vertical displacement extracted from numerical analysis in sequential excavation method for different support system types is shown in Fig. 13. In this method, the displacement is constant after 60 m advance, consequently, because of eliminating the effect of underground excavation face, only the 60 m of the station has been simulated. The results related to ground surface settlement are shown in Figs. 14 and 15. About ground surface settlement, Skempton and MacDonald (Murthy, 2002) suggested that the maximum angular distortion must be lower than 0.002 and also Skempton announced that the maximum settlement must be less than 7.6 cm, and according to the 1955 Soviet code the allowable value of settlement is 8 cm (Murthy, 2002).

Fig. 13. Vertical displacement after the excavation of the station (unit: m).

Fig. 14. Ground surface settlement in the cross-section of the station.

Fig. 15. Ground surface settlement in the cross-section at different excavation steps (support type 3).

5.2.Large-diametercurvedpiperoo fingmethod

Soil freezing before the excavation of underground structures, in addition to preventing water flow into underground space, has many in fluences on increasing soil stability. This in fluence is obvious in numerical results (Fig. 16). For the simulating of frozen area with about 2 m in thickness, the mechanical parameters of frozen soil are necessary and these parameters were extracted from diagrams announced by Lai et al. (2009) and Yang et al. (2010). The vertical stress in the ground surrounding the station is about 0.2 MPa and the water content of the soil around the station is 15%. The mechanical parameters extracted from diagrams are shown in Table 4. The excavation sequence and installation stages of temporary and final support systems are shown in Fig. 17, and Fig. 18shows the frozen soil and the support elements used for simulation of the subway station.

Table 4 The mechanical parameters of the frozen soil.

Fig. 16. The slide of soil between curved pipes and the stability of soil in the frozen part.

5.2.1.Supportdesign

After the installation of curved pipes and ground freezing, the subway station space has been excavated in the whole length of the station. After the excavation of the station space and temporary support, the process of segment cutting and the installation of final lining and columns started. The advance step for the performance of final lining and columns is 5 m (the spacing between two columns in a row along the station). The circular columns used for the final support of the station is 70 cm in diameter, according to ACI318-89 (ACI Committee 318, 1989), the columns are capable of tolerating 10,919 kN thrust force. In this instance, the maximum thrust force induced in columns is less than the permissible value (Fs= 2.1) (Fig. 19).

According to Eq. (5), the magnitude of stress induced by bending moment and thrust force (Fig. 20) in curved steel pipes with 60 cm in diameter and 3.5 cm in thickness (outer diameter is 60 cm, and inner diameter is 53 cm) is 289 MPa, which is less than the permissible value that pipes can tolerate (0.66fy):

Fig. 17. (a) The sequence of excavation and installation of support systems and (b) final lining and columns construction.

Fig. 18. (a) The excavation of model after the simulation of frozen soil and (b) the support elements installed in the model.

Fig. 19. The columns cross-section used for the final support of the station and the thrust force induced in columns (unit: N).

Fig. 20. Bending moment and thrust force induced by the space excavation (the spacing between the pipes is 1.5 m). (a) Thrust force (unit: N) and (b) bending moment (unit: N m).

Fig. 21. The characteristics (bulk modulus) of initial soil applied in the model after the end of ground freezing stage and the performance of final lining (unit: Pa).

where (fb)max,FmaxandMmaxare the maximum stress, the maximum thrust force and the maximum bending moment, respectively; andC,AandIare the pipe radius, pipe cross-section area and pipe moment of inertia, respectively.

Before the end of ground freezing stage, the final support of the station is performed. In order to model the final support system, instead of frozen soil characteristics, the characteristics of initial soil are applied in the model (Fig. 21), and the water pressure is applied on the final support system which is reinforced concrete with 40 cm in thickness. In this condition, the maximum bending moment and thrust force induced in support system (Fig. 22) are also in allowable limits (Fs= 1.6).

5.2.2.Modelingresults

For evaluation of pipes spacing, three different situations have been modeled (the spacing between the pipes is 1.5 m, 2 m and 2.5 m). Fig. 23 shows the displacement induced by the excavation of the station. The ground surface settlement for different spacing between the pipes at various advance steps is shown in Figs. 24 and 25. In this method, the whole part of the station was simulated because the displacement during the advance steps was not fixed before 80 m excavation.

Fig. 22. Thrust force and bending moment induced in final support system (the distance between the pipes is 1.5 m). (a) Thrust force (unit: N) and (b) bending moment (unit: N m).

Fig. 23. The displacement of the ground surrounding the station after excavation (unit: m).

Fig. 24. Ground surface settlement in cross-section for different pipe spacing.

Fig. 25. Ground surface settlement in cross-section at different advance steps.

Fig. 26. Ground surface settlement for different values of elastic modulus (1.3Ei, Eiand 0.7Ei, Eiis the initial elastic modulus).

6. Sensitive analysis

In this section, considering the effect of different parameters and quantities on displacement and ground surface settlement and also the influences of parameters on the simulated models, the effects of different parameters such as cohesion, Poisson’s ratio, elastic modulus, friction angle and the ratio of horizontal stress and vertical stress (K) have been evaluated. The results show that the enhancing of the cohesion, Poisson’s ratio, elastic modulus, friction angle andKvalue will decrease the ground surface settlement, and with the decreasing of these parameters the ground surface settlement will increase. The changes in settlement are more sensitive to the decrease of elastic modulus and Poisson’s ratio. The maximum changes evaluated for different parameters are aboutKvalue, especially the decrease ofKvalue. For example, the ground settlement for different values of elastic modulus is shown in Fig. 26.

7. Conclusions

Considering the sensitive location of Naghshe Jahan Square as a prominent historical structure in Isfahan, the evaluation of ground surface settlement is inevitable. According to the results, the maximum angular distortion on the ground surface in large-diameter curved pipe roofing method for different spacing between the pipes, 1.5 m, 2 m and 2.5 m, are respectively 0.00145, 0.00208 and 0.00258. While using the pipes with 1.5 m spacing, the maximum angular distortion is less than the permissible values suggested by Skempton and MacDonald (the maximum angular distortion is 0.002) (Murthy, 2002). Therefore, considering the results derived from this study, large-diameter curved pipe roofing method using pipes with 1.5 m spacing is recommended as an excavation method with high safety factor for the excavation of the station. Concerning sequential excavation method, the maximum angular distortion using support type 3 is 0.00362 which is more than permissible value according to Skempton and MacDonald suggestion (Murthy, 2002). Nevertheless, the maximum settlement in this method is about 7 cm (support type 3) which is in allowable limit regarding allowable values proposed by Skempton (7.6 cm) and the Soviet code (8 cm). Consequently, this method could totally be considered as an excavation method with low danger. The settlement value in 30 m distance from the station center in cross-section is zero, then if the station is located further than this distance from historical structures, this method of excavation would be considered with high safety factor.

ACI Committee 318. Building code requirements for reinforced concrete (ACI318-89) and commentary (318R-89). Detroit: American Concrete Institute; 1989.

Bickel JO, Kuesel TR. Tunnel engineering handbook. New York: Van Nostrand Reinhold Co.; 1982.

Carranza-Torres C, Diederichs M. Mechanical analysis of circular liners with particular reference to composite supports. For example, liners consisting of shotcrete and steel sets. Tunnelling and Underground Space Technology 2009;24(5):506–32.

Gomes ARA, Mercado C, Houska M, Maia K. Metro Santiago—underground works of the new line 5 to Maipu. In: Safety Tunneling for the City and Environment, ITAAITES World Tunnel Congress 2009 and the 35th ITA-AITES General Assembly; 2009.

Habetaand H, Schäfers P. Application of ground freezing for underground construction in soft ground. In: Proceedings of the 5th International Symposium TC28. London: Taylor & Francis; 2006.

Hoek E, Carranza-Torres C, Diederichs M, Corkum B. Integration of geotechnical and structural design in tunnelling. In: The 56th Annual Geotechnical Engineering Conference. Minneapolis: University of Minnesota; 2008.

Infrastructure Development Institute (IDI). Japanese infrastructure newsletter. IDI Quarterly 2008;39(45):2–5.

Irshad M, Helfin LH. Soft-ground NATM tunnel designs for the Washington, D.C. Metro. Tunnelling and Underground Space Technology 1988;3(4):385–92.

Kindwall EP. Medical aspects of compressed air tunnelling: background and present state of the art. Tunnelling and Underground Space Technology 1988;3(2):169–73.

Kirkland CJ. Design and construction of urban tunnels-compressed air. In: Advances in Tunnelling Technology and Subsurface Use. Oxford: Pergamon Press Ltd.; 1984. p. 191–4.

Kunihiko T, Kenichi Y. The state of affairs of large diameter shield tunnel method for subway and the recent trend of shield technology in Japan. In: International Symposium on Underground Excavation and Tunnelling; 2006. p. 57–66.

Lai Y, Jin L, Chang X. Yield criterion and elasto-plastic damage constitutive model for frozen sandy soil. International Journal of Plasticity 2009;25(6):1177–205.

Lamont DR. Tunnels under pressure—mitigating the human response. Tunnelling and Underground Space Technology 2000;15(4):485–90.

Lunardi P. The cellular arch method: technical solution for the construction of the Milan railway’s Veneziastation. Tunnelling and Underground Space Technology 1990;5(4):351–6.

Murthy VNS. Geotechnical engineering. Boca Raton: CRC Press; 2002.

Sadaghiani MH, Dadizadeh S. Study on the effect of a new construction method for a large span metro underground station in Tabriz-Iran. Tunnelling and Underground Space Technology 2010;25(1):63–9.

Tonon F. Sequential excavation NATM and ADECO: what they have in common and how they differ. Tunnelling and Underground Space Technology 2010;25(3):245–65.

Yang Y, Lai Y, Chang X. Laboratory and theoretical investigations on the deformation and strength behaviours of artificial frozen soil. Cold Regions Science and Technology 2010;64(1):39–45.

Zamin Fanavaran Engineering Co.. The geological report of Takhti subway station; 2007.

∗Corresponding author. Tel.: +98 9360463911.

E-mail address: mortezaabdi90@gmail.com (M.A. Cherlo).

Peer review under responsibility of Institute of Rock and Soil Mechanics, Chinese Academy of Sciences.

1674-7755 © 2013 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting by Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.jrmge.2013.09.003