Damage of New Sanyi Railway Tunnel During the 1999 Chi-Chi Earthquake

Transcription

1 Damage of New Sanyi Railway Tunnel During the 999 Chi-Chi Earthquake Chih-Chieh Lu and Jin-Hung Hwang Ph.D. student, Department of Civil Engineering, National Central University, No.00, Jhongda Rd., Jhongli City, Taoyuan County 00, Taiwan(R.O.C.), Tel.: #46, Professor, Department of Civil Engineering, National Central University, No.00, Jhongda Rd., Jhongli City, Taoyuan County 00, Taiwan(R.O.C.), Tel.: #46, ABSTRACT: Historically, the damaged cases of tunnels during earthquake are rare due to their good earthquake-resistant performances. However, the tunnel may still be damaged owing to the violent squeeze of surrounding ground when subjected to an extreme strong shaking. On September 999, the new Sanyi railway tunnel in central Taiwan was seriously damaged by the Chi-Chi earthquake. At that time, the tunnel has been in operation for about one year. The damaged conditions of the tunnel were carefully investigated and well-documented. This paper summarizes the background information related to the geology, strong ground motions, design and construction of the tunnel and presents the damage patterns of the tunnel. Based on the compiled data, the possible reasons why the damages only occurred at some special locations have been proposed. Numerical analyses have been conducted to analyze one of the typical damaged cross-sections in order to explore the damage mechanism of the tunnel. The analyzed damage pattern agrees very well with field condition. The numerical results show that the bad shape of tunnel, the imperfect backfill grouting, and the secondary lining without any steel reinforcement were main factors that caused the tunnel seriously damaged during the earthquake. Keywords: new Sanyi railway tunnel, Chi-Chi earthquake, numerical analysis, damage investigation. INTRODUCTION Most modern tunnels in the world were constructed by NATM (New Australian Tunneling Method). The NATM is a design concept using flexible primary support, such as rock bolt, shotcrete, wire-mesh and light steel rib to take part of rock pressure from geostatic stress during tunnel excavation, and let the rock surrounding the tunnel also take the other part of rock pressure during its deforming process. By doing that, the design of tunnel support is more economic than traditional American Steel Support Method (ASSM) which assumes the rock pressure is fully taken by support. Thus, when the interaction of rock and tunnel support reaches a stabilized state of equilibrium, the secondary lining installed after that state will not bear any rock load theoretically in NATM concept. In general, constructing second lining is only for esthetic demand. However, when a large earthquake occurs, surrounding ground will shake and squeeze the tunnel, an extra load will apply on the second

2 lining. If this extra load is not considered at the beginning of second lining design, the non-reinforced lining will be damaged and even collapse during strong earthquake. For instance, in the 999 Chi-Chi earthquake, a total of fifty tunnels were reported to have been damaged. Among them, twenty-six were slightly damaged, eleven were moderately damaged, and thirteen were severely damaged (Wang et al., 000). After the earthquake, seismic capacity of the second lining became an important topic to tunnel engineers in Taiwan, a seismically active area. New Sanyi railway tunnel is the most seriously damaged tunnel in the 999 Chi-Chi earthquake. It is located in the western foothills of central Taiwan, which passes through Sanyi fault and Shihliufen fault zones, and other weak ground. The tunnel was completed and opened to vehicle traffic in 998 for replacing old Sanyi railway tunnel built in 908. This 76m long mountain tunnel was designed and excavated with NATM, Its overburden ranges from 0m to 0m. Unfortunately, after operating just about one year, the tunnel was shaken by the 999 Chi-Chi earthquake and severely damaged. The railway traffic was interrupted several months after this catastrophic earthquake. There are eight main damaged sections and various types of damage conditions were observed. Some damaged conditions are shown in Photo. -4. Based on the field investigation, a number of damaging factors were identified, which include excessive earthquake shaking, bad geometry of tunnel, imperfect backfilling, non-reinforced second lining, and geological weak zone. Since well-documented literatures of damaged tunnels are rare in the past, this paper summarizes and arranged the background data of new Sanyi railway tunnel, identifies its failure mechanism to provide a valuable data base for tunnel engineering profession. Moreover, this paper uses the modified cross section racking deformation (MCSRD) method proposed by the authors (Hwang and Lu, 007) to analyze the failure mechanism of the damaged zone at Sta. 6K+00, which was the most seriously damaged place. The simulation agrees well with the field damage pattern observed in 999 Chi-Chi earthquake. Thus, the MCSRD might provide a simple and effective numerical tool for analyzing seismic capacity of tunnel second lining to tunnel engineers. Photo. Spalling of lining extended to the opening of side wall (6K+00) (Wang et al., 00). Photo. Spalling of crown (64K+740) (Hsu and Weng, 000). Photo. Longitudinal cracks of crown (64K+77) (Hsu and Weng, 000). Photo. 4 Inclined cracks of side wall (6K+80) (Hsu and Weng, 000).. SEISMIC GROUND MOTIONS On September 999, a strong quake with a magnitude of 7. on the Richter scale occurred near the town of Chi-Chi. A large earthquake like this had never been experienced in Taiwan for the past 00 years. The maximum ground accelerations measured by the strong motion seismographic stations in the Nantou-Wufeng area

3 were as high as g. It caused significant damage in the nearby area in central Taiwan. Due to sufficient seismographic stations in that area, the acceleration history records of the earthquake were available. The ground acceleration values can refer to the nearby seismographic stations. The closest seismographic station with a distance of km to the tunnel is Jian-Jhong elementary school and was used in this analysis. Since the axis of tunnel is about South-North direction, the critical motion is in the East-West direction due to the racking action on the tunnel section. Therefore, Fig. shows the acceleration, velocity and displacement histories in EW direction at that station. Note that although the maximum ground acceleration is only 0.4g, the maximum velocity is 6.cm/sec, which is a more important motion index than acceleration in using MCSRD method to assess the seismic capacity of the tunnel. Acceleration (g) Velocity (cm/sec) Displacement (cm) Max. acceleration=0.4g Max. velocity=6.cm/sec Max. displacement=.cm time (sec) Fig. Seismic motion histories in East-West direction near the tunnel (after the Central Weather Bureau).. THE NEW SANYI RAILWAY TUNNEL. Geology and construction condition The new Sanyi railway tunnel passes through a series of small mountain ridges and terraces neighboring a small valley, as shown in the Digital Terrain Model (DTM) of Fig.. Its overburden depths range from 0m to 0m. The ground formations that the tunnel passes through are the Miocene Kuantaoshan Sandstone, the Shihliufeng Shale, the Toukeshan Gravel, and the river gravel terrace. The tunnel crosses over two fault zones, Sanyi fault and Shihliufen fault. Based on engineering characteristics, the rock mass along the tunnel can be classified into 6 grades, as shown in Table, and the geological profile accompanying with construction conditions during tunneling are shown in Fig.. In addition, the geological investigation, conducted by United Geotech (989), indicated that the ground formations can be roughly divided into eight kinds of formations. The physical and mechanical properties of these formations are summarized in Table. The tunnel cross section was designed for electric double-track railway system. To satisfy the demand of operation and maintenance, the refuge spaces were excavated on the side wall in 0m spacing for the small refuge and 00m spacing for the large one. The three types of cross-section, including the standard one, of tunnel are shown in Fig. 4. For ordinary sections, the tunneling adopted Drill and Blasting (D&B) method with bench excavation. In case of encountering some difficult ground condition, ground treatment or special excavation method were adopted. From the monitored records during the construction, the horizontal converge deformation ranged from -mm, and the settlement of crown ranged from -mm in most typical sections. The problems encountered during excavation included () roof spalling due to fractured rock, () rock mass sliding along the planes of cleavage and

4 join, and () rock softening owing to ground water leakage, The problematic locations are also indicated in Fig.. Table Rock mass classification of the new Sanyi railway tunnel (after Hsu and Weng, 000). Classification Rating Description Stable I Sandstone or alternations of sandstone and silty sand. Some local zones have the conditions Slight fracture Moderate fracture or swelling High fracture or swelling Weak and non-cohesive II III IV a IV b V a V b of leakage or seepage but would not affect the strength of rock mass. Alternations of sandstone and shale, which intercalate thin shale. Degree of fracture is slight with some local severe fracture. If there is not suitable support, wedge failure would occur. Alternations of sandstone and silty sand, which intercalate thin shale. Degree of fracture increases obviously and usually coincides with fault zones. Unfavorable orientations of discontinuity are existent. Thin alternations of sandstone and shale. Strength of rock mass obviously decreases due to high fractured zones and faults. Quantity of water leakage is not serious. The geological characteristic is similar to IV a, but the excavated surface would be unstable due to a great quantity of water leakage and wide fault zone. Lateritic gravel. It is a different type overburden, a great part of which contains large pebble and gravel intercalated sandy and silty soil. The geological characteristic is similar to V a, but the excavated surface would be unstable due to a great quantity of water leakage and wide fault zone. Table The geological parameters of the formations (after United Geotech, 989). Formation Unit weight G B Vp Vs type (t/m φ o c ) (GPa) (GPa) (m/sec) (m/sec) (kpa) Description A Sandstone B Sandstone intercalated siltstone or shale C Siltstone or Shale intercalated fine sandstone D Siltstone or Shale E Fault gouge and mudstone F Weak bonding sandstone TK - a river gravel terrace or lateritic gravel FK Toukeshan gravel a Lack of experimental data SANYI SUNSEN Liyutan Reservoir Shihkang Dam Seismographic station New railway 8 9 Old railway New Sanyi railway tunnel FONGYUAN Note: N = The number of old Sanyi railway tunnel Sun Yat-sen Freeway Fig. The geographical location of the tunnel and the surrounding topography (after Hwang and Lu, 007) 4

5 Elevation (m) Mileage Ground formation Rock mass classification K 6K 6K 64K 6K IIIIII IV III a V a IV a III IV a IV a III III IV a III IV a III IV a III IV a III IIIIV a IV IV a a a 0 00m S:/000 66K 67K IV a III IV a III III III IV a III IV a IVa IV III IV a III V a IV a Construction hazard a Seismic damage section, PS:a. Construction roof spalling after excavation due to accident fractured rock rock mass sliding along cleavage and joint the softening of rock mass owing to encountering water leakage 4 Legend of ground formation Gravel Sanyi fault Sandstone intercalated fine shale Sandstone and alternations of sandstone and shale 6 7 Sandstone and sandstone intercalated fine shale Alternations of sandstone and shale Fig. Geological profile and construction conditions along the new Sanyi railway tunnel (after United Geotech, 989). 8 Sandstone Shale Thick sandstone intercalated fine shale Loessial gravel C L Y R R M Sanyi A Taichung A M +.70 M R R Shotcrete Inner lining C L Y Shotcrete Inner lining Waterproof C L Y Shotcrete Inner lining Waterproof Y ± Standard cross section (non-invert) X -.70 LC eastbound LC westbound ± X -.70 CL eastbound +.80 CL westbound 00 ±0.000 X LC Waterproof layer Shotcrete Non-invert section Invert section Non-invert section Invert section R R M Sanyi Taichung M A Y A M ±0.000 X Inner lining ± R= 0 Cross section of large refuge hole R= ± Cross section of standard refuge hole Standard cross section (invert) Fig. 4 The cross sections of the main tunnels and the tunnels with refuge spaces (after Hsu and Weng, 000). Table The basic information of eight damaged sections along the new Sanyi railway tunnel after the 999 Chi-Chi earthquake (after Wang et al., 00). Sec Damage Overburden Geological Rock mass Construction Concrete Location. types a Opening (m) condition classification hazard condition Approx. Sanyi Large Local void,4-6 IV 6k+00 Fault a - refuge existing Approx. Sanyi Cave-in and Small,4-4 IV 6k+60 Fault a Good collapse refuge Sanyi Small 6k+7-40,,4 - IV Fault a - Void existing refuge Approx. Small Local void 4,4 0 III - 64k+740 refuge existing 64k+78-80, III - - Good 6 64k IV a - - Void existing 7 6k ,, 0-0 III Squeezing and support damage - Void existing Approx. Fractured Small 8,4 IV 6k+800 zone a - refuge Void existing a Damage types: () longitudinal cracks; () transverse cracks; () inclined cracks; (4) cracks nearby the opening.

6 Panorama view mileage Panorama view mileage Crown Crown Left Right Left Right K K+80 Panorama view mileage Panorama view mileage Left Crown Right Left Crown Right K+70 6K+400 6K+60 64K K+740 6K+90 6K K K+70 6K+80 64K+880 6K+40 64K K+780 6K+40 6K+70 6K K K+770 6K+40 6K+00.6m 7.m 7.m.6m 64K+80 64K+840.6m 7.m 7.m.6m 64K+760 6K+400.6m 7.m 7.m.6m.6m Legend Refuge Construction joint Cracks Spalling of concrete 0 Estimating value of the cracks?width (mm) Measured line of ground penertration radar The zone with bad compaction and completeness of concrete Fig. The mapping results of GPR at the vault of the crown (after Hsu and Weng, 000).. Field investigation of the tunnel after the Chi-Chi earthquake In order to understand the damaged conditions of second linings caused by the earthquake, a non-destructive inspection technique, ground penetration radar (GPR), was used in combination with visual inspection to map the -D failure conditions. The mapping results of which are shown in Fig.. It was observed that there are eight main damaged sections, and the relevant information of these sections is summarized in Table. Based on the field investigation, the following damaging factors were identified (Hsu and Weng, 000), which include () excessive earthquake loading: the design of second lining did not consider such a large earthquake loading ; () bad geometry of tunnel cross section: the small and large refuges was symmetrically installed on side walls of the tunnel every 0m and 00m, respectively. It was found that the cross sections with refuges were most easily damaged due to the irregular geometry which will cause stress concentration at the corner or intersection when subjected to ground shaking; () imperfect backfill: since the new Sanyi railway tunnel was constructed for electric double-track railway system, the electric leakage should be avoided. Therefore, the complete circumferential waterproof membrane was installed after constructing the primary support to prevent the leakage of ground water. This causes an interface between the first and second linings and might generate some voids during concreting the second lining. If the voids are not perfectly backfilled, the second lining will not tightly bond with the surrounding ground which will cause the spalling of concrete of crown and sidewall during strong shaking; (4) non-reinforced second lining: the 7.m 7.m.6m 6K+90 6

7 tunnel was designed using NATM concept, so the second lining was not reinforced with steel rebar. Therefore, it is weak in resisting shear force and bending moment. When a strong earthquake happens, large shear force and bending moment will be induced in the second lining due to dynamic squeezing of the surrounding ground. Thus, the lining will be easily damaged, even collapse; () geological weak zones: most of the eight damaged sections are at geological weak zones. For Sta. 6K , it is located at Sanyi fault and its overburden is about -6m. There were cave in and collapse of supporting system during construction and the deformation of crown was about 8-44cm. For Sta. 64K , the ground formation is sandstone intercalated with fine-grained shale. Its mass classification is III. The deformation of crown was about.-cm and there was a collapse of support system happened nearby. For Sta. 6K , the ground formation is alternations of sandstone and shale, and its rock mass classification is III-IV. The deformation of crown was about 40cm. There were lots of emergent ground treatments at 6K and 6K NUMERICAL ANALYSIS Among all damaged sections, the most seriously damaged one is located at the Sta. 6K+00 (Section ). It would therefore be selected to be analyzed in detail by the modified cross section racking deformation (MCSRD) method, to understand the failure mechanism of the lining when subjected to ground squeezing. This method imposes a prescribed seismic shear strain on the boundaries of a ground domain with a tunnel built in. Then, observe the mechanical interaction of the tunnel and surrounding ground through numerical steps. The MCSRD method had been used to assess the seismic capacity of old Sanyi railway tunnel, and gave a good prediction of seismic performance of the tunnel by comparison with its field performances in the 9 Hsinchu-Taichung and 999 Chi-Chi earthquakes (Hwang and Lu, 007). The procedures of the MCSRD method can be summarized as below and Fig. 6.. Set up a numerical model with a suitable mesh domain, and initialize in-situ geostatic stresses, and construct a tunnel in the domain.. Apply a small seismic shear strain with a slow shear strain rate on the boundaries of the analyzed domain.. Check whether or not the internal forces, such as the moment, shear and axial forces on the lining, exceed the nominal lining member s strength curves of P u -M u and P u -V u. If some segments of lining reach the limit state, that means the lining will begin cracking and lose its strength. It is very difficult to trace the after cracking behavior of the non-reinforced concrete with temperature steel bar. However, in order to conduct pushover analysis, it is necessary to set up the capacity curves of each lining segment first. Thus, this research assumes the segment s strength will reduce to 0% of the original strength to simulate the behavior of non-reinforced concrete after cracking. If more experimental data are added, the assumption can be modified accordingly. 4. Repeat step to until the applied seismic shear strain reaches to the design value, which is calculated by the following equation: 7

8 v γ des = () Vs Where: γ des is the design seismic shear strain, which is the maximum shear strain of the site induced by earthquake; v is the peak ground velocity (PGV) of the site during the earthquake; V s is the average shear wave velocity of the ground.. Keep the analyzed records on each loading stage and observe the failure mechanism of the lining. 4. Numerical modeling The geological parameters and the grid mesh of the analyzed model are shown in Fig. 7. The boundaries were located at m distant from the center line of the tunnel. The shear strain was imposed on the whole boundaries with a sufficient slow rate to make the shear strain distribution uniform in the whole analyzed domain. In the numerical model, the ground formations were modeled by solid elements with Mohr-Coulomb failure criteria. Based on the in-situ drilling data, the geological parameters are of the formation type C shown in table. The second lining was modeled by piece-wise beam elements with three different interface conditions between lining and surrounding ground. The parameters of second lining and three interface conditions, which simulate three different qualities of backfill, are shown in Table 4. Based on the measured PGV in E-W direction, the total imposed shear strain on the boundaries was 0.079% according to equation (), and the applied shear strain rate was as slow as about. 0 - m/sec. No Start Set up numerical model Apply seismic shear strain incrementally on the boundaries Check if the lining yields or not Yes Reduce the yielding lining strength to 0% of the original Check if the input shear strain reaches to design value Yes End No Fig. 6 Flow chart of the MCSRD analysis procedure. Table 4 The parameters of lining and interfaces Case Thickness(m) f c (kg/cm ) E(GPa) Moment inertia(0-4 m 4 ) Tensional bond of interface(kpa) - a a No interface between lining and surrounding ground (No slip) 0m Applied shear strain γ =.4tf/m c = 60kPa o φ =. V = 484m/sec s Tunnel 0m Fig. 7 The geological parameters and grid mesh of the model. 8

9 4. Results and Discussion The internal forces of the tunnel lining in different cases are shows in Fig. 8. The result of case is not shown owing to the final collapse of the tunnel lining. It can be seen that the internal forces of lining are quite small at the beginning (before earthquake shaking). That is compatible with the concept of NATM, which assume the second lining does not take any forces from geostatic stresses. However, when seismic shear strain was applied to the boundaries of model, forcing the ground motion to squeeze the tunnel, the internal forces of lining largely increase, especially at the corners of refuge space. The computed results seem to be similar whether the interface between lining and surrounding ground was considered or not. Bending moment Axial force Shear force Initial condition After being applied shear strain Lining condition Lining condition (without interface) (with interface) Max. value=0.7kn-m Max. value=7.6kn-m Max. value=8.kn-m Max. value=0.kn Max. value=48.kn Max. value=48.8kn Max. value=4.0kn Max. value=64.6kn Max. value=607.kn Fig. 8 The internal forces of lining before and after ground squeezing due to earthquake. n : Yielding point : Location of yielding point n : Yielding sequence of the segment in the lining r : Shear strain as the corresponding point yields v : Ground velocity as the corresponding point yields r=0.006%, v=8.9cm/s r=0.0068%, v=0.cm/s r=0.0066%, v=9.87cm/s 4 r= %, v=0cm/s Fig Collpase (b) Case r=0.009%, v=8.8cm/s 6 r=0.006%, v=7.8cm/s r=0.00%, v=.cm/s 4 r=0.0084%, v=.cm/s r=0.00%, v=7.77cm/s (a) Case r=0.008%, v=7.99cm/s r=0.007%, v=8.48cm/s 4 r=0.008%, v=.7cm/s r=0.007%, v=.9cm/s r=0.007%, v=7.98cm/s The yielding damage mechanism of the lining. (c) Case Through observation in the process of numerical simulation, the yielding mechanism of second lining is that one of the lining segments yields first due to suffering too much internal forces and then loses its load-carrying capacity, and then the extra forces have to be shared by the other non-yielding lining elements, which will reach the state of yielding in turn, and finally the lining system will collapse. Fig. 9 shows the development of yielding mechanism along the tunnel lining in three different cases. For Case and, both yielding states occur at the left-upper and right-lower corners of the refuge owing to the rightward seismic shear strain. Nonetheless, the development of yielding states of Case is faster than Case, which demonstrates the allowance of relative slip between the tunnel lining and 9

10 surrounding ground slows down the development of yielding state in the same loading conditions. In this case, when the upper corners of refuge yield, the crown of tunnel can only sustain its weight by its small tensional bond with the upper ground. Since the bonding force is weak, the crown finally collapses as shown in Fig. 0. This indicates imperfect backfilling between the lining and the surrounding ground is harmful. Fig. 0 The collapse of the tunnel lining in Case.. CONCLUSION AND SUGGESTION. This paper summarizes and documents the background information and the damage pattern of the new Sanyi railway tunnel, which is the most valuable case of the seriously damaged tunnels during the 999 Chi-Chi earthquake in Taiwan.. The major damaging factors to the tunnel are strong ground shaking, bad geometry of tunnel section, imperfect backfilling, non-reinforced second lining, and weak geological conditions.. The proposed MCSRD method can simulate the harmful effects of all the above major damaging factors, and the numerical analysis of the section at Sta. 6K+00 can identify its failure mechanism and agree well with its performance during the Chi-Chi earthquake. 4. The second lining is suggested to be reinforced with steel bars to resist earthquake loading, especially in seismic area. 6. REFERENCES Central Weather Bureau: Hsu, L.P. and Weng, S.L. (000). The geological treatment for railway tunnel after seismic damage a case study of Sanyi no. railway tunnel. Treatment Technology of Engineering Geology on Tunnel. -. (in Chinese) Hwang, J.H. and Lu, C.C. (007). Seismic capacity assessment of old Sanyi railway tunnels. Tunnelling and Underground Space Technology. (4): United Geotech (989). Drilling and testing report of route changing section of Sanyi no. runnel route changing and double track project. Technical Report, United Geotech, Taipei. (in Chinese) Wang, W.L., Wang, T.T., Su, J.J., Lin, C.H., and Huang, T.H. (000). The seismic hazard and the rehabilitation of the tunnels in central Taiwan after Chi-Chi. Sino-Geotechnics. 8: (in Chinese) Wang, W.L., Wang, T.T., Su, J.J., Lin, C.H., and Huang, T.H. (00). Assessment of damage in mountain tunnels due to the Taiwan Chi-Chi earthquake. Tunneling and Underground Space Technology. 6: -0. 0