Investigation on the Effects of Twin Tunnel Excavations Beneath a Road Underpass

Investigation on the Effects of Twin Tunnel Excavations Beneath a Road Underpass Eshagh Namazi Researcher, Department of Geotechnics and Transportation, Faculty of Civil Engineering, Universiti Teknologi Malaysia (Formerly a Tunnel Engineer at Bamrah Construction); email: eshagh.namazi@gmail.com Hisham Mohamad Lecturer, Department of Geotechnics and Transportation, Faculty of Civil Engineering, Universiti Teknologi Malaysia ; email: mhisham@utm.my Mohammad Ehsan Jorat Researcher, Department of Geotechnics and Transportation, Faculty of Civil Engineering, Universiti Teknologi Malaysia; email: mohammadehsanjorat@yahoo.com Mohsen Hajihassani Researcher, Department of Geotechnics and Transportation, Faculty of Civil Engineering, Universiti Teknologi Malaysia; email: mohsen_hajihassani@yahoo.com ABSTRACT Excavation of tunnels underneath cities often intrudes the existence of piled foundation and in severe cases, can cause damage to the overlying structures. As there are very limited published case studies concerning understanding of the interaction between piled structure and tunneling, there is a significant uncertainty regarding tunnel-pile interaction. In this paper, a case study of the effects of two subway tunnels on the contiguous pile walls which support a road underpass is investigated using three-dimensional Finite Element simulations. The interaction between the tunnels and piles is investigated with a special attention to the effect of tunnel face pressures. Through the numerical modelling and field data, it is shown with presence of the piles, the minimum pressure to support the tunnel face is less than minimum face pressure in the green field condition. Field experience indicates that excessive tunnel face pressure can cause temporary heave to the ground surface but also cause damage to the cutter head of tunnel boring machine. KEYWORDS: Pile Walls, Numerical Modeling, Surface Settlement, Face Pressure INTRODUCTION Construction of subway tunnels in the urban environment is a complex problem particularly when tunnels are excavated very close to existing structures supported with pile foundation system. The design and execution of these tunnels requires assessment of the impact of the tunnel-induced ground movement on the stability and integrity of existing piled foundations (Mohamed and Mattar, 2009; Cheng et al. 2007; Mroueh and Shahrour, 2002; Jacobsz et al. 2001; Leung et al. 2000; Chen et al. - 441 -

Vol. 16 [2011], Bund. D 442 1999; Vermeer and Bonnier, 1991). Several researchers have focused on the influence zone for 2D pile-soil-tunneling interaction based on case studies and numerical simulations (Lee and Bassett, 2007; Selemetas et al., 2006; and Kaalberg et al. 2006; Lee et al. 2007; Coutts and Wang, 2000). In these studies the effects of different parameters (e.g. distance of the pile from tunnel centre, position of the pile tip regarding to the horizontal tunnel axis, pile and tunnel diameter) on the interaction between pile and tunnel have been investigated to develop understanding of the interaction mechanism between tunnels and piles. In the first part of this paper, a case study of the effects of tunneling on the contiguous piles is presented. The 3D-finite element (PLAXIS 3D TUNNEL Package) was performed to investigate the effect of tunnel advancement on the contiguous pile with special attention to the most important parameter of excavation called face pressure. In the second part, parametric study of the effect of tunnel face pressure on the interaction between tunnel and piles is carried out. The last part represents the longitudinal settlement measured at the ground surface where the high face pressure was used in the tunneling operation. SITE DESCRIPTIONS The growth and expansion of Shiraz, a southern city of Iran, and increase in the number of vehicles and population led to construction of subway in order to overcome the transportation problems. The South eastern part of Line І of that subway with length of approximately 14 km consisted of a twin tunnel. The running tunnels were excavated by Earth Pressure Balance (EPB) of 6.88 m external diameter with tail-skin grouting. The tunnel linings were made from pre-fabricated reinforced concrete segments forming an internal tunnel diameter of 6 m. A particular interest of Line Ι project between stations Zirgozar and Zand Cross was the constructions of tunnels below an existing Zand Underground Motorway (Zand Underpass). Figure 1 shows the longitudinal section of Shiraz tunnel between the stations, under the Zand Underpass. The distance between the stations is 1215 m whereas the underpass length is about 607 m. Next to the Zand Underpass, cars must pass downward and upward ramps of 6.5% with lengths of 135 m. The twin tunnel underneath, on the other hand, were excavated with a generally more gentle slope of 1.9% running from the stations to their deepest point of approximately 16 m along the Underpass. The spacing of the two tunnel centre-lines next to the Underpass is equivalent to two tunnel diameters. Figure 1: longitudinal section of the tunnels under the Zand Underpass The Underpass was formed by two contiguous pile walls and a roof slab which was connected to the walls by pin connection. The roof slab was 0.8m thick. The contiguous pile walls formed of many piles with diameter of 1.2m and spacing of 0.1m. The filling between - 442 -

Vol. 16 [2011], Bund. D 443 the piles comprised of cement and bentonite mixture. Figure 2 shows the geometry of the structure and the position of the pile walls regarding to the tunnels. The mechanical properties of the Underpass partitions are given in table 1. 8 m 5.2 m Made ground Clayey sand 4 m 3.2 m 3.6 m 13.6 m 1.2 m Inorganic Silt 6.7 m 7 m Clayey sand 3 m Inorganic Silt 1.8 m R=3.44 m Clayey sand Figure 2: Description of geological conditions of the site GROUND CONDITION The site investigation includes three boreholes close to the area (Bamrah Construction, 2004). The sequence of strata identified from these boreholes is summarized in Figure 3. The ground profile consisted of made ground at the top, and the next clayey sand overlying the intermittent layers of clayey sand and inorganic silt. The tunnels were excavated in the clay and inorganic silt. Table 1 summarizes the characteristics of the soil used for the analyses. The water table was taken as approximately 8m below the ground surface, i.e. within the inorganic silt. Table 1: Mechanical properties of encountered materials (After Bamrah Construction, 2004) GROUP γ unsat (kn/m 3 ) γ sat (kn/m 3 ) E(MPa) υ C (kpa) Φ(o) Ψ(o) R int MADE 16 19 51.5 0.3 20 25 0 0.8 GROUND CLAYEY 17.7 22.8 88.3 0.25 24.5 29 0 0.6 SAND INORGANIC SILT 16.9 20.9 30 0.25 10 36 0 0.7 SEGMENT EA=30000(MN/m) EI=225(MN/m/m) ROOF SLAB E=23(GPa) Thickness = 0.8 (m) CONTIGUOUS PILE E=19.23(GPa) Diameter =1.4 (m) - 443 -

Vol. 16 [2011], Bund. D 444 NUMERICAL MODELING In order to investigate the tunnel advancement on the contiguous piles, 3D-Numerical modelling was performed using a commercial Finite Element program, i.e. PLAXIS 3D Tunnel (Brinkgreve and Broere, 2004). This software provides the flexible features to model the details of tunnel construction in soils. The finite element mesh used in numerical modelling is presented in Figure 3. The model is 100 m wide, 30 m deep and 70 m long. The geometrical boundaries considered here was found to be far enough from the tunnels axis in order to minimise the influence of boundaries on the tunnelling model. The model includes 4130 elements and 12389 nodes. The soils were modelled using 15-noded wedge elements, whereas 8-node plate elements represented the tunnel lining (Figure 3). To simulate the soilstructure interaction, a 16-node interface element was used. The water table is assumed to produce the hydrostatic initial pore water pressure. An elastic-plastic soil model using the Mohr-coulomb failure criterion is adapted in this study. Because the soils are prevalently fine-grained and relatively low permeable, the analyses were carried out in undrained condition. In general, the process of tunnel construction under the Underpass was modelled in two steps (for more information for simulation of tunnelling, see Potts and Zdravkovic, 2001). First, the initial conditions were set up for the model before excavation of the tunnels. It was achieved by specifying the distribution of effective vertical and horizontal stress (using coefficient of earth pressure at rest, K0=0.5) and pore water pressure. The initial conditions were completed with simulating the underpass structure. In this stage, the vehicles loads were calculated and applied to the model. After establishing the initial conditions, the analyses continued with modeling excavation of the first tunnel. The tunnel excavation process was done through a step-by-step method in 16 phases. In each phase, the excavation process consists of: (i) excavation of the soil, (ii) application of pore water pressure, (iii) support pressure at the tunnel face to prevent active failure at the face, (iv) installation of the tunnel lining and finally (v) the grouting of the gap between the soil and the newly installed lining. The second tunnel excavation was modeled after the completion of the first tunnel in which the same manner of step-by-step method is applied. Figure 3: Three dimensional finite element model of the tunnels under the Underpass - 444 -

Vol. 16 [2011], Bund. D 445 PREDICTION OF SURFACE SETTLEMENT TROUGH Prior to assessing the influence of piles walls on the settlement-induced by tunnelling, the green-field surface settlement trough obtained by numerical modelling is compared with empirical method and field data. The green-field settlement troughs are obtained in the same geology conditions as the Zand Underpass ones. Figure 4 compares the surface settlement profile after excavation both tunnels. There is an agreement between field data and numerical modelling in terms of magnitude and shape of profile except maximum settlement which is under predicted. Results calculated using empirical method which was expressed by Peck (1969) are also shown in Figure 4. The superposition principle is used to obtain the settlement trough due to excavation of both tunnels. The volume loss of 1.1% was used to calculate the settlement trough induced by each tunnel. Although the empirical method predicts maximum settlement more accurately than the numerical model, the settlement in the far field was over predicted. Figure 4: Surface Settlement Trough in the Green-field conditions Figure 5 shows the predicted surface settlement by numerical modelling where both of the tunnels have been completed, for green-field and actual conditions (with the presence of the Underpass). The existing of structure would normally modify the surface settlement owing to tunnelling excavation. In this case however, the two Finite Element (FE) ground surface settlement plots are almost identical to each other except at the point where the contiguous piles walls are located. Piles walls do not follow exactly green-field movement induced by tunnels at the piles location and soil movement surrounding the piles also altered due to presence of piles. This is due to the additional of displacement (settlement) caused by the piles. The displacement-induced by pile can be divided to the displacement caused by pile loading and the displacement by presence of pile without load. The displacement induced by loading is in accord to the green-field ground movement and displacement caused by presence of pile is in resistant to the green-field ground movement. Position of the pile tip regarding to the horizontal tunnel axis determines which one is dominance: pile loading or presence of the piles. In this example, because the tunnels are excavated exactly beneath the piles, loading increases the vertical effective stress and consequent ground movement beneath the pile tip. But in the different situation where the tunnel is excavated adjacent to the piles, the existing of the piles decreases the ground displacement induced by excavation (Ng et al., 2005; Huang - 445 -

Vol. 16 [2011], Bund. D 446 et al., 2009). Such phenomenon, as discussed in the previous section, is called the shielding effect. Surface Settlement (mm) 0-3 -6-9 -12-15 -18-21 -24 Transversal Coordinate (m) 0 10 20 30 40 50 60 70 Actual Condition Green-Field Condition Figure 5: Effect of contiguous piles on the surface settlement trough Figure 6 shows the soil displacement pattern around the Zand Underpass after excavation of the tunnels in FE. In general, before excavation of the tunnels the pile-soil system was in equilibrium. The surrounding soil applies an upward friction force (i.e. positive friction) to resist the downward displacement of the pile. When the tunnels are excavated, however, the equilibrium is disturbed and the soil moves to the tunnels boundaries. In this situation, the displacement of surrounding soil is larger than displacement of the piles and the downward fiction force (i.e. negative friction) exerts an additional load on the piles. The piles transfer their loads to the soil before excavation of the tunnel but after excavation of the tunnel the piles carry the load induced by soil displacement. Figure 6: Soil displacement pattern around the Zand Underpass in FE - 446 -

Vol. 16 [2011], Bund. D 447 TUNNEL FACE PRESSU URE The evaluation of the tunnel face-support pressure is a critical component in both the design and construction phases of Tunnel Boring Machines (TBM). In practice, face pressure is usually the primary control parameterr during excavation and is one of the most significant factors that have a direct effect on the magnitude of surface settlements. The higher face pressures are applied, the smaller surface settlementt will be observed. In cases of very high face pressure, the surface heave occurred. Today, there are several analytical and empirical methods to calculate the tunnel face pressure in the green-field conditions (without presence of surface structures) based on failure mechanism of tunnel face (Broere, 2001; Anognostou and Kovari 1996; Leca and Dormieux 1990; Atkinson and Potts 1977; Broms & Bennemark 1967). These methods are not able to consider the effects of existing structures on the face pressure. This limitation can be overcome by introduction of the finite element method. In this paper, in order to investigate the influence of the contiguous piles on the face pressure, the minimum pressure obtained from green-field conditions was compared with that of actual conditions (with the presence of the Underpass). The minimum face pressures in the models were obtained by decreasing the initial face pressure until the failure occurs in the tunnel face. The failure occurred when the face pressure remained constant in the infinite displacement (Figure 7). The initial face pressure in the green-field condition was calculated to be 150 KN/m2 from the analytical solution (Leca and Dormieux 1990). Figure 7 illustrates the minimum face pressure as fraction of initial pressure against soil displacement. It can be seen that the minimum face pressure of the tunnel support in the green-field is more than when the condition of structure exists. The underpass existencee declines the minimum face pressure from 68% to 65% of initial load. Such finding can be explained as follows. The contiguous piles wall in the Underpass act as shield and do not let the soil in front of the TBM face move freely. This shielding effect decreases the soil volume which moves to the face and consequently declines the face pressure. Figure 7: Face pressure related to face displacement at Greenfield and actual (with existence of Zand Underpass) - 447 -

Vol. 16 [2011], Bund. D 448 EFFECTS OF FACE PRESSURE ON THE SURFACE SETTLEMENT TROUG H The numerical parametric study has been done to assess the effect of tunnel face pressure on the surface settlement due to excavation of tunnels under the structure. The tunnels are excavated with face pressures of 115Kpa, 140Kpa and 215Kpa at the level of tunnel axis. Figure 8 shows the surface settlement trough after excavation of both tunnels related to the different face pressures. In general, a decrease of the face pressure causes larger stress release of the soil at the cutting face and it leads to increasing of the surface settlement (Kasper and Meschke, 2006) ). When the tunnels are excavated underneath the Underpass, the piles resist release of more soil stress at the face. In other words, when the face pressure decreases, the piles contribute to support the tunnel face and resist increase of more surface settlement. As the figure shows, when the face pressuree decreases around two times from 215Kpa to 115Kpa the maximum surface settlement increases just 1.4mm. In another way, at presence of the piles, large increasing of the face pressure causes small decreasing of the surface settlement. Figure 8: Surface settlement trough for different value of tunnel face pressure INFLUENCE OF HIGH FACE PRESSURE In the previous section by parametric study, we showed that the piles resist the upward force generated by the face pressure when controlling the surface settlement. This section represents the longitudinal settlement measured at the ground surface where the high face pressure was used in the tunnelling operation. Since there was an uncertainty in the effect of the Underpass structure on the face pressure, actual high face pressure of 210kPa was applied in the first 169m-part of the tunnel operation under the Underpass of the site. This was much higher than the predicted minimum face pressure mentioned in the previous sections. Figure 9 shows longitudinal settlements during excavation of the first tunnel for three TBM head positions. Clearly when the TBM advanced, the ground surface settlement increased. As the figure shows a slight temporary heave of the ground surface of approximately 1mm can be observed in front of the TBM. Although the high face pressure decreases the final settlement, this pressure can also cause damage to TBM cutter head as - 448 -

Vol. 16 [2011], Bund. D 449 indicated by the field observation (Bamrah Construction, 2010). Replacement of the cutter head had to be made which lead to further delays and interruption to the tunnelling operation. Surface Settlement (mm) 3 2 1 0-1 -2-3 -4-5 -6-7 -20-5 10 25 40 55 70 85 100 115 130 145 160 Figure 9: Field measurement of longitudinal surface settlement CONCLUSION A 3D numerical analysis with assistance of field data has been presented to study the effects of face pressure on the surface settlement induced by tunnelling under the piled walls. Numerical modelling results showed the minimum pressure to support the tunnel face is less than minimum face pressure in the green field condition. In fact, applying the green-field pressure to the tunnel face in the presence of piles is a conservative method. The parametric study illustrated that increasing the face pressure significantly does not help to reduce the final surface settlement significantly but only slight. Field observation of the Shiraz subway tunnels under the existing Zand underpass showed excessive tunnel face pressure causes temporary heave to the ground surface but also cause damage to the TBM cutter head. REFERENCES TBM position:+10.5m Longitudinal Coordinate (m) 1. Anagnostou, G. and Kovári, K.(1996). Face Stability Conditions with Earth-Pressure- Balanced Shields. Tunn. Undergr. Sp. Tech. 11(2), 165 173. 2. Atkinson, J.H., and Potts, D.M. (1977). Stability of a Shallow Circular Tunnel in Cohesionless Soil. Géotechnique 27(2),.203 215. 3. Bamrah Construction. (2004). Geotechnical engineering of Shiraz subway report. Shiraz Subway Construction Company, Shiraz, Iran (In Persian). 4. Bamrah Construction. (2010). Report of first TBM damage in Shiraz subway. Shiraz Subway Construction Company, Shiraz, Iran (In Persian). 5. Brinkgreve, R. B. J., and Broere, W. (2004). Plaxis 3D tunnel manual. Version 2. Online manual. 6. Broere, W.(2001). Tunnel Face Stability and New CPT Applications. Ph.D Thesis,Technical Univer-sity of Delft. www.library.tudelft.nl. 7. Broms, B.B., and Bennmark, H.(1967). Stability of Clay at Vertical Opening. ASCE Journal of the Soil Mechanics and Foundations Division. SM1, 71 94. - 449 -

Vol. 16 [2011], Bund. D 450 8. Coutts, D.R. and Wang, J. (20000. Monitoring of reinforced concrete piles under horizontal and vertical loads due to tunnelling. In: Tunnels and underground structures by Zhao, Shirlaw and Krishna, Singapore, 541-546. 9. Chen, L. T., Poulos, H. G. and Loganathan, N. (1999). Pile responses caused by tunnelling. Journal of Geotechnical and Geoenvironmental Engineering, 125 (3), 207-215. 10. Cheng, C. Y., Dasari, G. R., Chow, Y. K., and Leung, C. F. (2007). Finite element analysis of tunnel-soil-pile interaction using displacementcontrolled model. Tunnelling and Underground Space Technology, 22(4), 450-466. 11. Huang, M., Li, C. Z. Z. (2009). A simplified analysis method for the influence of tunneling on grouped piles. Tunnelling and Underground Space Technology,24 (4), 410-422. 12. Jacobsz, S. W., Standing, J. R., Mair, R. J., Soga, K., Hagiwara, T. & Sugiyama, T. (2001). Tunnelling effects on driven piles. Proc. Int. Conf. Response of buildings to excavation induced ground movements, London. 13. Kaalberg, F.J., Von Tol, A.F., Bosch, J.W. (2006). Dutch research on the impact of shield tunnelling on pile foundations. Proceeding of 5th International symposium at Amesterdam, 2005: Geotechnical Aspects of Underground Construction in soft Ground. 14. Kasper, T., Meschke, G. (2006). On the influence of face pressure, grouting pressure and TBM design in soft ground tunnelling. TunnellingandUndergroundSpaceTechnology, 21, 160-171. 15. Leca, E., and Dormieux, L. (1990). Upper and lower bound solutions for the face stability of shallow circular tunnels in frictional material. Ge otechnique, 40(4), 581 606. 16. Lee Y, J., Bassett, R. H. (2007). Influence zones for 2D pile soil tunnelling interaction based on model test and numerical analysis. Tunnelling and Underground Space Technology, 22, 325 342. 17. Leung, C. F., Chow, Y. K., and Shen, R. F. (2002). Behavior of pile subject to excavationinduced soil movement. J. Geotech. Geoenviron. Eng., 126(11), 947 954. 18. Mohamed A. Meguid., Mattar, j. (2009). Investigation of Tunnel-Soil-Pile Interaction in Cohesive Soils. Journal of Geotechnical and Geoenvironmental Engineering, 135 (7), 973-979. 19. Mroueh, H., and Shahrour, I. (2002). Three-dimensional finite element analysis of the interaction between tunneling and pile foundations. International journal for numerical and analytical methods in geomechanics, 26, 217-230. 20. Ng, C.W.W., Chan, S.H., Lam, S.Y. (2005). Centrifuge and numerical modeling of shielding effects on piles in consolidating soil. In: Proceedings of 2nd China Japan Geotechnical Symposium, Tongji University Press, Shanghai, 7 19. 21. Peck, R.B. (1969). Deep excavations and tunnelling in soft ground. Proc. 7th Int. Conf.on Soil Mechanics and Foundation Engineering, Mexico City, State of the Art Volume, 225-290. 22. Potts, D.M., and Zdravkovic, L. (2001). Finite element analysis in geotechnical engineering:application. Thomas Telford, London, UK 23. Vermeer, P.A., and Bonnier, P.G. (1991). Pile settlements due to tunnelling. Proc. 10thEuropean Conf. on Soil Mechanics and Foundation Engineering, Florence, Balkema, Vol. 2, p. 869-872. 2011 ejge - 450 -