Read Manuscript Preparation for International Conference on Problematic Soils text version

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering, 28-30 May 2009, Near East University, Nicosia, North Cyprus

d

Geotechnical considerations on TBM tunneling in rock mass

Saffet Yaiz

Pamukkale University, Engineering Faculty, Geological Engineering Department, Denizli 20020 Turkey

KEYWORDS: Performance, rock mass, TBM tunneling ABSTRACT: The industry deals with underground excavation has always shown a major interest in the use of Tunnel Boring Machine (TBM) that is full-face tool to excavate rock, because of their demonstrated capabilities in attaining high rates of advance in tunnel construction. TBM tunneling may differ from other tunneling methods such as; drill and blast, due to the high level of machinerock interaction. Accurate estimates of machine performance are a crucial part of any mechanical tunneling. It is almost impossible for owners or contractors to make realistic evaluations of time and cost required for completing a project, without estimating the machine advancement. Intact rock properties, including uniaxial compressive strength, Brazilian tensile strength and brittleness of rock, and also rock mass properties such as orientation of discontinuities, with machine specifications could be used for investigating TBM performance in rock mass. This paper presents the results of some case histories in TBM tunneling, field and laboratory data processing together with recent improvement for estimating TBM performance in rock mass. 1 INTRODUCTION Since first TBM (Tunnel Boring Machine) was built by Robbins in late 1950, various researches have been conducted for predicting TBM performance that is essential for assessing time and cost to complete relevant project (Ozdemir, 1977; Rostami, 1997; Bruland, 1999; Yagiz, 2002; Yagiz et al. 2008). TBM performance is known to be influenced by a number of factors may be categorized as rock mass and machine parameters. However, affect of intact and mass rock properties on machine performance could be various in weight depend on rock mass condition and utilized machine specification in tunnel site. For example, as rock mass were fractured and faulted, the discontinuity properties could be more important than the strength of intact rock for estimating TBM penetration rate. In this paper, literature review and tunnel cases including the Queens Water tunnel designed to improve fresh water distribution in the City of New York, USA, the second tailrace tunnel of the Manapouri hydro project with about 1.6km long, an underground power station located in the Fiordland area of southwestern in New Zealand, Milyang hydro-tunnel project about 5.4km long in South Korea, and more Boston Harbor tunnel about 10 km long in Massachusetts, USA have been overviewed and the findings are discussed herein from the scope of geotechnical consideration on TBM tunneling in rock mass.

504

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering, 28-30 May 2009, Near East University, Nicosia, North Cyprus

d

2 BACK GROUND Prediction of TBM performance depends on not only intact and mass rock properties but also site condition and machine specification in a tunnel. Therefore, before the project commenced, tunnel site condition, excavated rock properties and utilized TBM characteristic should be known to make better project in short time and low cost without unexpected problems. Various studies have been conducted to investigate the relationship between cutting force and rock properties (Ozdemir, 1977; Rostami and Ozdemir, 1993; Rostami, 1997). The most popular intact rock properties are uniaxial compressive strength (UCS) and Brazilian indirect tensile strength (BTS) to measure cutting forces and TBM performance estimation. Rock brittleness is also an important factor that could be obtained via various tests as index but have not gained universal recognation yet. As rock fracture toughness has been used for estimation of cutting forces by Sanio ( 1985), Tarkoy (1975) utilized total hardness of rock for TBM performance studies in the field, but not directly cutting force estimation. Punch penetration or indentation test has also been conducted to obtain rock brittleness to use for estimating TBM performance (Yagiz and Ozdemir, 2001; Yagiz, 2002, 2008). Several formulas have been offered for estimation of the cutting forces from rock properties and cutting geometry (Ozdemir, 1977; Lindquist, 1984; Sanio, 1985; Rostami, 1997). Colorado School of Mines (CSM) has developed a semi-theoretical model, based on the cutting forces on individual cutter (Ozdemir, 1977). Rostami and Ozdemir (1993) improved this model theoretically by estimating cutting force as a function of uniaxial compressive strength and Brazilian tensile strength of rock. Further, Yagiz (2002) introduced the Modified CSM model, giving more precise estimation result in fractured rock mass condition, by inserting rock brittleness and rock mass properties (i.e., distance between planes of weakness, alpha angle that is the angle between the plane of weakness and tunnel driven direction) into the existing CSM model. The last set, so called TBM prognosis, models including Tarkoy (1975), Nelson and O'Rourke (1983), and Norwegian model (NTNU), are based on field observation of TBM performance in various tunneling projects (Blindheim, 1979). The most commonly used model is the Norwegian hard rock TBM prognosis model developed by Norwegian University of Science and Technology (NTNU) at Trondheim (Lislerud, 1988; Bruland, 1999). The NTNU model uses rock mass characteristics and joint information in conjunction with indices representing rock drillability and abrasivity. 3 PROJECT DESCRIPTIONS The Queens Water Tunnel # 3 is intended to improve fresh water distribution throughout the City of New York (Figure 1). Upon completion, it will allow for the maintenance of two existing tunnels that have been operating since 1917 and 1936 and will be an important connecting link for operation of the New York City water tunnel system. About 8km long concrete-lined pressure tunnel through the hard metamorphic rocks of the Appalachian mountain belt was excavated beneath Brooklyn and Queens at an average depth of 200m below sea level in west-central Queens County by utilizing an open type Robbins machine (Model 235-282) that specifications are given in Table 1. The machine bored through hard jointed formations of varying rock types, including biotite-hornblende gneiss intermixed with granite gneiss, amphibolite, pegmatite, and biotite schist (Yagiz, 2002; Mergurian and Ozdemir 2003). Intact and mass rock properties along the tunnel were investigated and the average value of measured rock properties is given in Table 2. Besides Queens Tunnel project, about 2 km long Manapouri hydro tunnel project that was excavated along the calcslicate, meta dolorite, meta andesite, paragneiss, granitic gneiss type of rock mass in New Zealand; Milyang Tunnel project about 5.4km long was excavated along the igneous rock mass (i.e., granite and andesite) to deliver clean water from Milyang dam to Yangsan area through 2.6m-diameter hydro-tunnel in South Korea (Kim 2004); and more Boston Harbor tunnel was excavated in variety of argillite rock mass to conceive designs and constructs the facilities to clean up the harbor in Massachusetts, USA was also overviewed and the field and

505

Geotechnical consideration on TBM tunneling in rock mass Yaiz S.

laboratory database were used to obtain the relationship between TBM and rock mass conditions. However, due to restriction of space herein, only the Queens Water Tunnel was introduced in detail as an example project on TBM tunneling in rock mass.

Figure 1. Location of Queens Tunnel Project in New York City, USA (Yagiz, 2002)

Table 1. Average rock properties for the Queens Water Tunnel Project Rock no 1 2 3 4 5 Rock Type Rhyodacite dike Granitoid gneiss Diorite dikes Gneiss Garnet mica gneiss UCS (MPa) 151 158 161 137 148 BTS (MPa) 8.9 9.3 9.9 9.4 9.7 BI (kN/mm) 34 34 43 35 33 Fs (m) 0.1 1.02 0.56 1.11 1.1 Alpha (degree) 42.5 46.1 28.3 45.8 46.7

Table 2. Specifications of the Queens Tunnel Boring Machine (Model 235­282) Machine Diameter Diameter Range Cutters Number of Disc Cutters Recommended Individual Cutter Load Cutterhead Max. Operating Cutterhead Thrust Cutterhead Power Cutterhead Speed Cutterhead Torque Thrust Cylinder stroke TBM Weight (approx.) 7.06 m 6.5m-8.5m Series 19, 0.482m 50 300 kN 1750 tons 4220 hp 8.3 RPM 1335 short tons 1.83m 640 tons

506

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering, 28-30 May 2009, Near East University, Nicosia, North Cyprus

d

4 AFFECTS OF ROCK PROPERTIES ON TBM PERFORMANCE TBM parameters including thrust and power are keys to provide sufficient amount of forces and torque to support the excavation operation. Machine thrust should provide the enough force to efficiently penetrate the tools into the rock surface. Besides machine specification and some constrain (haulage, support requirement and groundwater handling etc) encountered in the field, intact and mass rock properties are also must be well understood. Otherwise, time and cost increment for the project is inevitable. In the following parts, there is some intact and mass rock properties were discussed from the scope of TBM performance evaluation. 4.1. Intact Rock Properties The most of developed models and equations utilizes the UCS of rock as one of the input parameters for estimating TBM performance; however, mechanical cutting predictions relying only on the UCS of intact rock may provide inaccurate results (Cigla et al. 2001; Yagiz, 2006). Brazilian tensile strength (BTS) is another common rock property that is used in making boreability predictions together with the UCS. The BTS is generally intended to provide an indication of rock toughness from a viewpoint of crack propagation between adjacent cutter paths. Further, rock toughness/brittleness also has an effect on boreability. In general, rock cutting efficiency of any mechanical tool improves with increasing brittleness exhibited by the rock formation. Thus, brittleness is a highly desirable feature of rock from a boreability standpoint. Punch penetration index test can be carried out to estimate combination of rock brittleness and toughness. Usefulness and producers of the test were discussed in some detail by different researchers (Dollinger et al 1998; Cigla et al 2001; Yagiz, 2002, 2009). As using the CSM Model or the modified model for predicting TBM performance, UCS, BTS and rock brittleness (BI) of intact rock must be measured in the laboratory as input variable together with TBM parameters such as torque, power, thrust, cutter geometry. After known rock and machine properties entered into the model, the model would be run to estimate TBM penetration rate for given intact and mass rock properties and also TBM specification. Typical TBM performance curve obtained from the CSM Model for given project are demonstrated in Figure 3. Further, the Modified CSM model uses rock brittleness as input variable to estimate TBM performance. TBM penetration rate increases with increasing rock brittleness (Figure 4). 4.2. Rock Mass Properties Rock mass condition has enormous affects on TBM performance in the field. Fracture properties, orientation, spacing between the plane of weakness, fault, shear zones and any other observed geological structures should be investigated and quantified in the field to obtain enough rock mass data for performance estimation. Spacing between plane of weakness or fractures (Fs) and the alpha angle ( that is the angle plane of weakness make with tunnel axis, have affect on machine advancement. Quantification of the discontinuity properties could be obtained with the equation of alpha angle developed by the NTNU. As the alpha angle in 50-60 degree, TBM reaches maximum performance resulting of the Modified CSM model (Figure 5). In general, TBM performance increases with decreasing Fs; however as the Fs lower than 20cm, TBM performance decreases due to increment of down time for tunnel support and maintenances so forth (Figure 6). Thus, maximum penetration rate are obtained as Fs ranges from 20 to 40cm via the Modified CSM Model.

507

Geotechnical consideration on TBM tunneling in rock mass Yaiz S.

9 8

3,000

2,500 7

CH. Power (kW) / Thrust (ton)

6 5

2,000

ROP (m/hr)

1,500 4 3 2 1 0 0 50 100 150 200 250 300 Uniaxial Compressive Strength (MPa) ROP (m/hr) Cutterhead Power (kW) Thrust (ton) 500 1,000

0

Figure 3. TBM performance curve generated from the CSM Model (Rostami, 1997; Yagiz 2002)

4.0 3.5 3.0

ROP (m/hr)

2.5 2.0 1.5 1.0 0.5 0.0 0 5 10 15 20 25 30 35 40 45 50 Rock brittleness index (kN/mm)

Figure 4. The relationship between the ROP and rock brittleness (Yagiz, 2002, 2008)

508

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering, 28-30 May 2009, Near East University, Nicosia, North Cyprus

d

2.5 2.0

ROP (m/h)

1.5 1.0 0.5 0.0 0 10 20 30 40 50 60 70 80 90 Alpha Angle (degree)

Figure 5. The relationship between the ROP and the alpha angle (Yagiz, 2002, 2008)

2.5 2.0

ROP (m/hr)

1.5 1.0 0.5 0.0 0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

Spacing between fractures (m)

Figure 6. The relationship between the ROP and spacing between fractures (Yagiz, 2002, 2008)

509

Geotechnical consideration on TBM tunneling in rock mass Yaiz S.

5 DISCUSSIONS The Colorado School of Mines Model could be very useful to estimate cutting force to penetrate rock mass, consequently to estimate TBM performance for given TBM specification and rock properties. However, as rock mass have fractures and discontinuities that makes rock weaker than expected, the model may not give precise result, therefore, it is modified (Yagiz, 2002). Even though more than one project were investigated and the database developed for the model modification, only Queens Tunnel project were given as an example tunnel case herein. Two performance prediction equations were presented in this study. One of the equation (1) deals with modification of the CSM Model, and another one is independent equation (2) as function of rock properties (UCS, BI, alpha, Fs). Both developed equation and modified model work more precise to estimate TBM performance in comparison with previous CSM model. Concluding remark is that first equation (1) that is linear regression equation including basic CSM model result is relatively more realistic to be used for given tunnel project. Second equation (Yagiz, 2008) were developed by using the database obtained from one case that is 8-km long hard rock tunnel as the CSM model are one of the cited and used model in the USA and European Countries. Thus, Modified CSM Model equation together with the Basic CSM Model result should be used for TBM tunnelling. For given five different rock mass condition, prediction result of the CSM, Modified CSM and developed equation by Yagiz (2008) are given in Table 3.

Table 3. Rate of penetration obtained from the Models and actual TBM in the field Rock no 1 2 3 4 5 ROPField (m/hour) 2.42 2.02 2.35 2.05 1.99 ROPCSM (m/hr) 4.07 3.86 3.71 4.31 4.00 ROPM-CSM (m/hr) 2.27 2.06 2.30 2.11 2.03 Yagiz (2008) 2.32 2.11 2.37 2.18 2.09

ROP

M CSM

0.272 0.437 log( ) 0.225 Fs 0.027 BI 0.097 ROP

CSM

(1) (2)

ROP ( m / h) 1.093 0.437 log( ) 0.003 UCS 0.029 BI 0.219 Fs

In these equations, the ROP refers to rate of penetration in meter per hour; alpha ( refers to the angle between planes of weakness to tunnel driven direction in degree; Fs is refers to spacing between the plane of weakness or fracture in meter; BI is in kN/mm as brittleness index, the UCS is uniaxial compressive strength in MPa; and ROPCSM is basic penetration rate obtained from the CSM Model is in m/hr, likewise ROPM-CSM for the Modified CSM Model. As seen that Modified CSM Model gives better agreement with actual TBM performance as well as given equation (2). So, if tunnel contractor and an engineer do not have access to the CSM model, they may use given equation (2) for their purposes, but not preferable. 6 CONCLUSIONS In present, all model development effort may contribute to the development of more precise TBM performance models. Since every models have different input machine and rock parameters to estimate machine performance, variation of their results obtained from each model are inevitable. Further, development needs to make improvement on the accuracy of the existing models. Research groups dealing with machine-rock interaction and performance prediction around the world carry on improving their models by expanding the database and developing new theories.

510

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering, 28-30 May 2009, Near East University, Nicosia, North Cyprus

d

REFERENCES

Blindheim, O. T. (1979). Boreabilty predictions for tunneling. PhD Thesis. Department of Geological Engineering. The Norwegian Institute of Technology. 406p. Bruland, A. (1999). Hard rock tunnel boring: Advance rate and cutter wear. Trondheim: Norwegian Institute of Technology (NTNU), Trondheim, Norway. Cigla, M., Yagiz, S., Ozdemir, L. (2001). Application of tunnel boring machines in underground mining development. 17th Int. Mining Congress & Exhibition of Turkey, 155-164 Ankara TR. Dollinger, G. L., Handewith, H. J., Breeds, C. D. (1999). Use of punch test for estimating TBM Performance. Tunneling Underground Space Technol. 13: 403 408 Kim T. (2004). Development of a fuzzy logic based utilization predictor model for hard rock tunnel boring machines. PhD Thesis, Depart of Mining Engineering Colorado School of Mines, 254p. Lislerud, A. (1988). Hard rock tunnel boring: Prognosis and Costs. Tunneling and Underground Space Technology, 3(1); 9-17. Lindqvist, P. (1984). Stress fields and subsurface crack propagation of single and multiple rock indentation and disc cutting. Rock Mechanics and Rock Engineering, 17, pp. 97-112 Merguerian, C., Ozdemir, L. (2003). Rock mass properties and hard rock TBM penetration rate investigations, Queens Tunnel Complex, NYC Water Tunnel #3, Stage 2. In: Robinson, R.A. and Marquardt, J.M. (Eds.) Proceedings of RETC., 1019-1036. Nelson, P.P., O'Rourke, T.D. (1983). Tunnel boring machine performance in sedimentary rocks, Report to Goldberg-Zoino Associates of New York, P.C., by School of Civil and Environmental of Civil Engineering. Cornell University. p.438, Ithaca, NY. Ozdemir, L. (1977). Development of theoretical equations for predicting tunnel borability, Ph.D., Thesis, T1969, Colorado School of mines, Golden, Co Rostami, J. (1997). Development of a force estimation model for rock fragmentation with disc cutters through theoretical modeling and physical measurement of crushed zone pressure. PhD. Thesis, Department of Mining Engineering, Colorado School of Mines, T 5008. Rostami, J., Ozdemir, L. (1993). A new model for performance prediction of hard rock TBMs, Proceedings of RETC, Boston MA, June 13 17,1993 Sanio, H.P. (1985). Prediction of the performance of disc cutters in anisotropic rocks" International Journal of Rock Mechanic and Mining Science & Geomechanic abstracts 22(3); 153 161. Tarkoy, P.J. (1975). Rock hardness index properties and geotechnical parameters for predicting tunnel boring machine performance. PhD. Thesis, University of Illinois, IL, 326p. Yagiz, S., Ozdemir, L. (2001). Geotechnical parameters influencing the TBM performance in various rocks. Program with abstracts, 44th Annual Meeting of Association of Engineering Geologists, Technical Session 10. Engineering geology for construction practices, MO, USA Yagiz, S. (2002). Development of rock fracture and brittleness indices to quantify the effects of rock mass features and toughness in the CSM model basic penetration for hard rock tunneling machines" PhD Thesis, Dept of Mining Engineering, Colorado School of Mines, T 5605. 289p. Yagiz, S. (2006). A model for prediction of tunnel boring machine performance. Substructures and underground space. Engineering geology for tomorrow's cities. In: The 10th IAEG Congress, Paper No. 383, The Geological Society of London, Nottingham, UK, 10p. in DVD. Yagiz, S. (2008). Utilizing rock mass properties for predicting TBM performance in hard rock condition, Tunneling and Underground Space Technology, 23(2008) 326-339 Yagiz, S., Rostami, J., Ozdemir, L. (2008). Recommended rock testing methods for predicting TBM performance: focus on the CSM and NTNU Models. In: Proceedings of the ISRM International Symposium 2008, 5th ASRM. Majdi A, Ghazvinian A. (Eds), pp1523-1530. November 24-26, 2008. Tehran IRAN Yagiz, S. (2009). Assessment of brittleness using rock strength and density with punch penetration test. Tunneling and Underground Space Technology. 24(1);66-74.

511

Information

Manuscript Preparation for International Conference on Problematic Soils

8 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

1219606


You might also be interested in

BETA
The Current State of Disc Cutter Design and Development Directions
An engineer who meets the following requirements may apply for registration with the Board:
text
ROADHEADERS PERFORMANCE OPTIMIZATION