ATS11-02113 ABSTRACT Geological evidences of collapse zones in TBM tunneling; a case study of Ghomroud water conveyance tunnel, IRAN Mahdi Zolfaghari, Ehsan Mokhtari, Massoud Morsali Sahel consultant engineers, No. 55, Ardakani street, Resalat highway,tehran, Iran. There are many factors such as equipments, management, personnel skills and ground condition that affected the TBM performance and mechanized excavation. The adverse geological condition that encountered in the tunnel is one of the most important parameters that affect the excavation process. Nature of the adverse geological conditions and fuzziness of them cause to decrease the accuracy of their prediction. It seems there are some evidences that can lead us to detect the problematic zones more exactly. To research the role of the geologic evidences in the collapse zones detection, the data gathered from a water conveyance tunnel excavated in central Iran were considered and analyzed. The rock formations along the tunnel path consist of metamorphic and sedimentary rocks aged from Jurassic to cretaceous. During the tunnel excavation the adverse geological conditions several times cause to collapse of tunnel and subsequently sticking of TBM. The parameters such as quartz content, fragment size and maximum fragment size of cuttings and amount of injected pea gravel behind the lining were monitored during the excavation, especially in collapse zones. The mentioned parameters have a variable rate along the tunnel path and these variations depend on the geologic condition. Quartz content of cutting materials in the collapse zones are higher than surrounding ground of these zones and the fragment size and maximum size of fragments in the collapse zones show an increasing trend relative to the normal condition of ground. Also, the injected pea gravel in collapses decreases in respect of other parts of tunnel. The results of this study show that the monitoring of variation in some geological parameters such as the amount of secondary minerals in cutting materials and the size of cutting fragments, also the amount of injected pea behind the lining of tunnel can help us to better prediction of collapse zones in the metamorphic rocks. KEY WORDS: TBM,collapsezone,quartzcontent,fragmentsize,cuttings. 1. INTRODUCTION Mechanical excavation especially excavation with TBM has many advantages over conventional drill and blast methods. These advantages include lower cost and higher advance rate than drill and blast excavation in most cases, improved safety, minimal ground disturbances, elimination of blast vibration, reduced ventilation requirements etc. There are many factors such as equipments, management, personnel skills and ground condition that affected the TBM performance and mechanized excavation. Ground condition or geology is one of the most important affecting factors in mechanized excavation. The effect of geologic condition on mechanized tunneling can be grouped into two main categories; the first one is the geologic condition that affects the machine choice and design. Such condition determined before the choice of machine and its designation. The second category consists of adverse geological condition that encountered during the tunnel excavation and this condition mostly is unexpected or accepted. This means the second group of geological condition has an adverse effect on excavation but we accepted the presence of this condition. In tunneling projects to determine the situations of these adverse condition, site investigation studies and surface geological surveys conducted before the beginning of excavation and geophysical studies and probe hole drilling performed during the excavation. Nature of the phenomena and fuzziness of the geological problems cause to decrease the accuracy of these determinations, so in real cases prediction of the adverse geologic condition is a rough approximation to what that happen. To overcome this problem and reduce the fuzziness it is necessary to monitor all the excavation process and ground condition during the work. It seems there are some evidences that can lead us to detect the problematic zones more exactly. Many researchers have studied geological parameters that affect the excavation and tunneling in difficult geological
condition. One of the first researches in this field has been carried out by Deere, D.U. (1981) who studied the effect of adverse geological condition on TBM tunneling. Lombardi G., Panciera A. (1997) showed the TBM tunneling problems in squeezing ground condition and Barla G. and Barla M. (1998) researched the tunneling in different adverse ground condition such as fault zones and squeezing grounds. Tseng Y.Y. et al (1998) also researched the mechanized tunneling in difficult ground. Barla G. (2000), Barton N. (2000), Shang Y. (2004), M. Sharifzadeh et al (2006), Mirmehrabi et al (2008), studied the effect of adverse geologic condition on TBM tunneling in recent years. In this paper, some geologic evidences before the entering the collapsible ground condition are surveyed and the relation between these evidences and occurred geologic hazards in the tunnel are researched. 2. GEOLOGICAL SETTING AND PROJECT DESCRIPTION In this research, the data gathered from a water conveyance tunnel excavated in central Iran were considered and analyzed. The Ghomroud water conveyance tunnel is one of the components of a water management system in central Iran (Figure 1). This involves a 36 km tunnel from the Dez River in Lorestan province to the Golpayegan dam reservoir in Esfahan province. The tunnel was divided into different parcels that two parts of the tunnel with about 18 km length excavated by Ghaem Construction Co., a subsidiary of Khatam Corp. This segment constructed using a 4.5 m diameter double shield TBM at a grade of 0.134% and finished with a concrete segmental lining to a diameter of 3.8 m. Figure 1- satellite photo from the tunnel path The tunnel is located tectonically in Sanandaj Sirjan belt in Iran plate. This zone consists of a series of Jurassic-cretaceous metamorphic and sedimentary rocks that has been formed during the clash of Arabian plate and central Iran plate. As a result of clash a wide trusted and folded belt has been formed called Sanandaj-Sirjan zone. The rock formations along the tunnel path consist of four main categories. These formations that age from Jurassic to cretaceous are as follows: I. Limestone formation: this formation consists of massive and thick bedded limestone and dolomite. II. III. Slate and shiest: foliated metamorphic shiest and slates embrace the most parts of tunnel path. This formation is faulted and oriented in different directions. Also the most of the schistose rocks contain organic components. Quartzite and quartz veins: because of the geologic setting of study area and presence of tectonic activities such as faulting and intrusion of plutonic rock in this zone many secondary quartz veins with igneous source have been injected into the discontinuities formed by faults. These veins can be
seen randomly in different parts of Jurassic formations. The maximum thickness of these veins reaches up to 80 m. Their strength reaches over 100 MPa. The origin of these veins is mainly from pegmatite formations in the area. Based on the surface geologic studies bore holes data and as built geologic maps, a geologic section of tunnel path was drawn that can be seen in figure 2. Table 1 collapse zones along the tunnel path NO. Chainage of collapse zones Lithology 1 2250 2265 Graphite shiest, shiest 2 2525 2545 Graphite shiest, shiest 3 3145 3215 Slate, shiest 4 4670 4730 Slate and shiest 5 5235 5490 Slate, shiest and Graphite shiest 6 5650 5700 Slate, shiest and Graphite shiest 7 6360 6400 Slate, shiest, sandstone As can be seen in figure 2 there are many faults that affect the tunnel. These structural defects cause to decrease of rock mass engineering properties and subsequently to take place tunnel collapses at different zones. The chainage of main collapse zones along the tunnel are as bellow: Figure 2- geological cross section of tunnel path 2. 1. Methods During the tunnel excavation several times the adverse geological condition such as fault zones and grounds with weak rock masses caused to collapse of tunnel roof and walls and subsequently sticking of TBM. Before the first and second collapse the presence of quartz veins in the parent rock was reported and quartz content of waste materials increased (figure 3&4). Such conditions in two collapses excite this idea in mind that there are some geological evidences of zones with collapse potential.
and rock mass condition. So the change of chips size is one of the important parameters that can be noticed in collapse zones. Figure 5 shows the chips size of the cuttings from the tunnel in metamorphic shiest and slates. Figure 3 quartz veins in rock mass at the first collapse of tunnel Figure 5-Cuttings of metamorphic shiest and slates from the tunnel Figure 4 Quartz chips taken from cuttings With this idea all of tunneling process before the entering the collapse zone was reviewed and geology related parameters such as waste materials, ground convergence and machine parameters have been checked. One of the most important parameters is the cutting type and size. The type of cuttings show the geological formation and rock type that embedded the tunnel and change in the rock cuttings means the change in host rock condition. Size of cuttings or chips size is the other important parameter. Chips are formed between two cutters. The size of chips is controlled mainly by the distance of cutters, the spacing and orientation of rock joints, rock strength and brittleness of rock (Cong and Zhao, 2009). During the excavation by means of TBM chips size changes depends on variation of joint spacing The machine depending parameters such as trust pressure or torque of cutter head also change in different geological zones but because of dependency of the machine parameters on the human decisions, in this research these parameters are not noticed. The pea gravel injection is one of the parameters that depend on ground condition. In squeezing ground and collapse zone that the ground be closer to the machine shields the pea gravel injection decrease in a considerable amount. The mentioned parameters were monitored during the excavation, especially in collapse zones. The normal cutting size based on the observations during the excavation is in the range between 2 to 10 centimeters. In two first tunnel collapses the percent of abnormal chips size with length less than 2 cm increased in matrix of waste materials. Also the maximum size of chips in these situations changes from the normal condition. Figure 6A shows the variations of percent of fragments and chips with size less than 2 cm. this figure also shows the position of collapse zones. The maximum chips size variation along the tunnel path has been shown in figure 6B. The presence of quartz vein in collapse zones and increase in quartz content of cuttings has been shown in figure 6C. Also the variation of the amount of injected pea gravel behind the tunnel lining can be seen in figure 7.
Figure 6 variation of geological evidence for collapse zone determination along the tunnel path: A- Percent of fragments with length less than 2 cm. B- Maximum sizes of fragments. C- Quartz content (%) of cuttings.
Figure 7 variation of Injected pea gravel along the tunnel path 3. RESULTS Monitoring the mentioned parameters such as quartz content, fragment size and maximum size of cuttings, also injected pea gravel behind the tunnel lining show a variable rate along the tunnel path and these variations depend on the geologic condition. An overall look at the parameters variation along the tunnel path reveals that there are some relative differences in the amount these parameters in collapse zones and other parts of tunnel. Quartz content of cutting materials in the collapse zones are higher than surrounding ground of these zones and the fragment size and maximum size of fragments in the collapse zones show an increasing trend relative to the normal condition of ground. Also, the injected pea gravel behind the lining of tunnel in collapses decreases in respect of other parts of tunnel. The variations of the mentioned parameters together along the tunnel path have been shown in figure 8. Figure 8 The variation of quartz content, fragment size and maxium size os cuttings and injected pea gravel along the tunnel path
4. DISCUSSION AND CONCLUSIONS The increase in quartz content of cuttings can be seen almost in the entire collapse zones and this phenomenon is related to the secondary quartz that re-crystalizes in openings of fracture and fault zones. The source of such silica and other secondary minerals is the plutonic activity in the region. In the fractured and fault zones due to frequency of fractures and low quality of rock mass the collapse potential increase and in the most cases collapse and rock falls are unavoidable. In fractured and crushed zones (the zones by high potential of collapse), there are unsystematic discontinuities and fractures that form rock blocks in different sizes. During the excavation in such zones these rock blocks will be released in the cuttings and consequently the maximum size of fragments. By increasing the fracture density in collapse zones, the fragment size of cuttings mainly controlled by the spacing and orientation of the fractures and joint, more than cutter head characteristics. In this condition commonly the average of fragment size of cuttings increases. The amount of injected pea gravel behind the tunnel lining depends on many factors from operational condition to geological situation. The squeezing ground, falling blocks, karts, groundwater condition, and ground collapse are some of the geologic phenomenon that affects the pea gravel injection. Closing to the zones with collapse potential because of the very deformable rock mass in collapsible zones around the tunnel the ground come closer to the lining and the amount of injected pea gravel reduce in a considerable rate. When the ground is collapsed the free space behind the lining completely is filled with collapse materials. In such condition the pea gravel injection is almost impossible. The results of this study show that in different geological condition with monitoring the excavation process it's possible to predict the adverse geological condition using some geological and operational parameters. In similar geological condition to this tunnel, geological parameters such as secondary minerals in fractures or mineral veins (e.g. calcite and quartz), size of cuttings fragments can be used to determine the collapsible zones locations. [7] Mirmehrabi, H., Hassanpour, J., Morsali, M., Tarigh Azali, S., 2008. Experiences gained from gas and water inflow toward the tunnel, case study: Aspar anticline, Kermanshah, Iran. Proc. 5th Asian Rock Mechanics Symposium, Tehran, Iran, 1469-1476. [8] Shang, Y., Xue, J., Wang, S., Yang, Z., Yang, J., 2004. A case history of tunnel boring machine jamming in an interlayer shear zone at the yellow river diversion project in China. Engineering Geology, 71, 199-211. [9] Sharifzadeh, M., Hemmati Shaabani, A., 2006. TBM Tunneling in adverse rock mass with emphasis on TBM Jamming accident in Ghomroud water transfer tunnel. Proceedings of the International Conference of Eurock, London, 643-647. [10] Tseng, Y.Y., Wong, S.L., Chu, B., Wong, C.H., 1998. The Pinglin Mechanized Tunneling in difficult Ground. St 8th Congr. Of IAEG, Vancouver, Canada. 5. REFERENCES [1] Barla, G., Barla, M., 1998. Tunneling in difficult conditions. Int. Conf. on Hydro Power Development in Himalayas, Shimla (India), 20 22 April. [2] Barla, G., Pelizza, S., 2000. TBM tunneling in difficult ground conditions. In: GeoEng2000 An International Conference on Geotechnical & Geological Engineering, Melbourne, Australia. [3] Barton, N., 2000.TBM tunneling in jointed and faulted rock. Balkema. Netherlands, 173. [4] Gong, Q.M., Zhao, J., 2009. Development of a rock mass characteristics model for TBM penetration rate prediction. International Journal of Rock Mechanics & Mining Sciences. 46, 8 18. [5] Deere, D.U., 1981. Adverse geology and TBM tunneling problems. Proc. Rapid Excavation and Tunneling Conference (RETC), San Francisco, 574-85. [6] Lombardi, G., Panciera, A., 1997. Problems with TBM and linings in squeezing ground. Tunnel and Tunneling International, June 1997, 54-57.