Damage-free coring technique for rock mass under high in-situ stresses

Journal of Rock Mechanics and Geotechnical Engineering. 2012, 4 (1): 44 53 Damage-free coring technique for rock mass under high in-situ stresses Peng Yan 1, 2, 3, Wenbo Lu 1, 2*, Ming Chen 1, 2, Zhigang Shan 3, Xiangrong Chen 3, Yong Zhou 3 1 State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan, 430072, China 2 Key Laboratory of Rock Mechanics in Hydraulic Structural Engineering of Ministry of Education, Wuhan University, Wuhan, 430072, China 3 HydroChina Huadong Engineering Corporation, Hangzhou, 310014, China Received 24 January 2011; received in revised form 16 December 2011; accepted 5 February 2012 Abstract: Rock sampling with traditional coring method would cause initial damage to rock samples induced by in-situ stress relief during coring. To solve this problem, a damage-free coring method is proposed in this paper. The proposed coring scheme is numerically modeled first, and then it is verified by comparative laboratory tests using rock samples both obtained by conventional coring method and the proposed damage-free coring method. The result indicates that the in-situ stresses in sampling area could be reduced by 30% 50% through drilling a certain number of destressing holes around the whole sampling area. The spacing between adjacent destressing holes is about 10 cm. The average uniaxial compressive strength (UCS) of rock samples obtained by the damage-free coring method in Jinping II hydropower station with overburden depth of 1 900 m is higher than that of samples obtained by the conventional coring method with the same depth by 5% 15% and an average of 8%. In addition, the effectiveness of damage-free coring method can also be verified by acoustic emission (AE) monitoring. The AE events monitored during uniaxial compression test of damage-free coring samples is fewer than that of conventional coring samples at the primarily loading phase. Key words: deep rock masses; damage-free coring; Jinping II hydropower station 1 Introduction At present, the issues of deep excavation or deep mining are frequently encountered (He et al., 2005; Zhou et al., 2005; Wang et al., 2006). Under high in-situ stresses, especially high horizontal stress, deep rock masses are always in a state of high compression, thus the excavation-induced responses would be severe, such as intensive rockbursts or other high stressinduced failures. The excavation damaged zone (EDZ) under high in-situ stresses would be much larger than that under low in-situ stresses (Martin, 1997; Backblom and Martin, 1999; Martino and Chandler, 2004; Read, 2004; He et al., 2005; Zhou et al., 2005; Wang et al., 2006). Deep rock sampling has to address the problem of coring damage or disturbance induced by in-situ stress Doi: 10.3724/SP.J.1235.2012.00044 *Corresponding author. Tel: +86-27-68772221; E-mail: wblu@whu.edu.cn Supported by the National Basic Research Program of China (973 Program) (2010CB732003) and the National Natural Science Foundation of China (51009013, 50909077) unloading. When a rock sample is cored from a deep borehole, it is removed from the original in-situ stress field to a state of destressing. Fig. 1 qualitatively shows the stress path during coring in a vertical well, based on numerical simulations performed by Dyke (1989). In Fig. 1, the point X is initially under a state of vertical and horizontal stresses. When a core is drilled, the vertical stress is gradually removed as the coring bit approaches from above. The horizontal stress is removed only after the point X gets inside the core barrel. Thus, a large deviatoric stress may occur during drill-out of the core from the strata. Such a stress may lead to yielding and failure of rocks (Holt et al., 2000). The differential strain curve analysis can be used to estimate in-situ stresses by measuring the distributions of microcracks in an oriented core (Strickland and Ren, 1980), and the principle of this method is that the stress relief in the core produces microcracks perpendicular to the principal stresses in proportion to the magnitude of each principal stress. Carlson and Wang (1986) showed that microcrack porosity in a rock core increases with the mean in-situ stress, which may indirectly verify the above method. Charlez et al.

Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 45 (a) Initial stress state of coring. which was conducted at the early stage of construction in shallow part of the transportation tunnel parallel to the diversion tunnels, and the maximum principal stress after regression at the largest overburden depth of the diversion tunnels could reach 72 MPa (Shan and Yan, 2010). The surrounding rocks of the transportation tunnels are mainly of marble with a saturated UCS of 65 90 MPa and a tensile strength of 3 6 MPa. Due to the complex geological conditions, large overburden depth and high in-situ stresses, intensive rockbursts induced by high in-situ stresses occurred frequently during excavation (Fig. 2). (b) Stress path during coring. Fig. 1 Illustration of stress relief during coring (Holt et al., 2000). (1986) proposed a rheological relaxation model to explain micro-fracturing induced by stress relief. Through laboratory tests, Holt et al. (1996, 1998), Kenter et al. (1997) and Brignoli et al. (1998) measured the material properties in virgin (simulated in-situ) conditions, and compared these to the properties of cores that have been unloaded to simulate coring and subsequently reloaded to in-situ conditions. The results indicated that the compression modulus was permanently reduced as a result of an unloading-reloading cycle, and the wave velocities were also permanently reduced. The unconfined strength was found to be strongly affected by the simulated coring. Chu (2009) also indicated that the rock strength may be underestimated by using damaged samples in laboratory tests, and furthermore, the coring damage may lead to a misunderstanding of rock mechanical characteristics. For deep rock project design, a reasonable estimation of coring damage extent of core samples or pre-established measures for reducing coring damage would be of great importance. Jinping II hydropower station, located on Jinping River bend of Yalong River in West China, is a long diversion-type hydropower station. The average length of the four diversion tunnels is around 16.7 km. These diversion tunnels pass through the Jinping Mountain and are generally with an overburden depth of 1 500 2 000 m, and the maximum depth is 2 525 m. The maximum in-situ stress measured in field is 46.1 MPa, (a) High stress-induced cracks. (b) Rockbursts. Fig. 2 Cracks and rockbursts induced by high in-situ stresses in Jinping II hydropower station. Relationship curves between the depth of V-shaped notches and the in-situ stress over the past 30 years are shown in Fig. 3 (Martin and Christiansson, 2009). In Fig. 3, D f is the depth of V-shaped notches from the center of circular tunnel, a is the radius of the circular tunnel, is the maximum boundary stress, and sm is the rock mass spalling strength. Except for the cases of a few failed zones induced by fracture rockburst, the depth of V-shaped notches induced by strain rockburst or slabbing is generally about 0.5 m in the drainage tunnel of Jinping II

46 Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 D f a Df/a Df 0.48 0.1 0.5 θθ a / θθ sm sm Strain (10 3 ) (a) Stress-strain curves. Fig. 3 Relationship curves between depth of V-shaped notches and in-situ stress (Martin and Christiansson, 2009). hydropower station during tunnel boring machine (TBM) excavation, with the overburden depth less than 1 800 m (Chu, 2009). Taking the drainage tunnel for example, the ratio of the maximum principal stress to the minimum one is considered to be 1.12, and the minimum principal stress is regarded as the weight of overlying rock masses (Fig. 3). For the tunnel with a radius of 3.5 m, calculation results show that the UCS 5 of the intact marble ( T 2y of Yantang group) is 150 180 MPa, which does not agree with laboratory test results. In the feasibility study stage of the project, laboratory tests (Chu, 2009) show that the average UCS of marble T 5 2y is about 125 MPa, and the maximum value is approximately 150 MPa. Although the differences may be caused by adaptability of the empirical formulae shown in Fig. 3, further study on the genuine strength of Jinping marbles is needed. In this paper, a damage-free coring method is proposed for deep rock masses, and numerical simulations and laboratory tests are also employed. 2 Damage-free coring technique 2.1 Coring damage of Jinping marble The laboratory tests conducted by Chu (2009) indicated that the average UCS of white marble (T 2b ) in Jinping II hydropower station was around 113 MPa, and the maximum and minimum values were 144 and 98 MPa, respectively. The results are basically consistent with those obtained by Hohai University. The typical stress-strain curves and corresponding AE counts at each loading level given by Chu (2009) are shown in Fig. 4. The axial stress-strain curves in Fig. 4(a) show a concave profile, indicating a closure of existing cracks right after loading. However, a large number of AE (b) AE counts. (c) Failure of specimen B11. Fig. 4 Stress-strain curves, AE counts and failure of specimen B11 under loading (Chu, 2009). events have been recorded in the meantime, as shown in Fig. 4(b). It indicates that a large number of new cracks have been developed under a rather low stress level, which is 30% of UCS, below the crack initialization strength. As the recorded AE events are mobilized when stress level is lower than the crack initialization strength, the AE events seem not to indicate an initial cracking. Instead, the propagation of existing damages is likely to account for the AE. In other words, significant damage exists in the specimen prior to loading (Shan et al., 2010). AE usually results from cracking in tension. The significant AE under low stress level suggests that the damage propagates sub-vertically in such an axial loading condition. The failure surface of this specimen agrees well with such suggestion. As shown in Fig. 4(c), the failure surface appears sub-vertically, or simply the existing damages are mostly oriented more or less in parallel with the specimen axis. Existing damage in specimen A45 can also be identified with the laboratory test results, as shown in Fig. 5. Similarly, damage propagation can be revealed by AE events. However, the profiles of both stress-

Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 47 2 Strain (10 3 ) (a) Stress-strain curve. Strain (10 6 ) Fig. 6 Stress-strain curve and AE count under loading using specimen obtained at shallow overburden condition (Zhao et al., 2006). Shear failure (b) AE counts. (c) Failure of specimen A45. Fig. 5 Stress-strain curves, AE counts and failure of specimen A45 under loading (Shan et al., 2010). axial strain curve and stress-volumetric strain curve for this specimen are quite different from those in Fig. 4. It suggests that most existing cracks are sub-horizontally oriented to favoring a closure response under axial loading. In addition to this difference, the failure surface penetrates through the specimen in an inclined manner, which indicates a shearing failure mechanism. Such mechanism is consistent with the presumption that sub-horizontal cracks are dominated as they have a little contribution to specimen failure under vertical loading (Shan et al., 2010). Fig. 6 shows the AE curve using rock specimen obtained in shallow overburden condition (Zhao et al., 2006), and it is significantly different from specimens B11 and A45. The significant difference indicates again that there would be a certain degree of initial damage in the core of Jinping marble, which may lead to an underestimation of the UCS of Jinping marble. 2.2 Scheme of damage-free coring All the laboratory test results have identified the presence of existing damages inside 14 specimens (Chu, 2009). The existing damage could result from core drilling or tectonic movement in geological history. A numerical model with 3DEC was built to investigate stress changes inside the core during drilling. The model geometry and interpreted stress are illustrated in Fig. 7 (Shan et al., 2010). Three points A, B and C locate at 0.8R, 0.5R and 0.2R (R is the radius of the rock core) in the rock core in Fig. 7(a), respectively. The stress states of the three points corresponding to different buried depths or in-situ stress levels are shown in Fig. 7(b). Once the stress state points locate in the area above the initiation strength line, the core damage will occur. The size of borehole in the model was set to be 76 mm in diameter, which is of a typical case in practice. The UCS of intact rock is considered as 120 MPa, referring to the laboratory test results in Figs. 4 and 5. (a) Calculation model. Rock core Drill part

48 Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 Stress increasing regions A B C Stress (MPa) 2.0000E+02 1.8000E+02 1.6000E+02 1.4000E+02 1.2000E+02 1.0000E+02 8.0000E+01 6.0000E+01 4.0000E+01 2.0000E+01 9.7557E 01 (a) Circle. (b) Square. Fig. 8 Concepts of damage-free coring technique (Chu, 2009). Maximum principal stress, 1 (MPa) (b) Stress distribution at depth of 1 500 m. Minimum principal stress, 3 (MPa) (c) Stress states of key points A, B and C. Fig. 7 3DEC simulations of stress path and potential damage caused by coring (Shan et al., 2010). Four simulations were carried out under four in-situ stress levels, equivalent to the depths of 800, 1 000, 1 500 and 2 000 m, respectively. The simulation results suggested that the stress was lower than crack initialization strength when drilling at the depth of 800 m from surface, as shown in Fig. 7. Core damage is thus not expected due to drilling. For the other three schemes examined, drilling can possibly cause core damage as the stress in the core can exceed crack initialization strength. In this model, the excavationinduced damage area could be 70% of the core cross-section when drilling at a depth of 2 000 m from surface. However, the central area of the core seems to remain intact. As drilling-induced damage propagates from boundary to centre, it seems that a solution to drill larger holes and extract the central column of core is needed. This approach is expected to work effectively when damage is not significant. In Fig. 8, another concept could be named as damage-free coring technique, which seems applicable to complex conditions. The idea is implemented through drilling a number of holes to release high stress in the area to be sampled (Chu, 2009). 2.3 Optimization of damage-free coring scheme According to the damage-free coring technique, the diameter of destressing holes was determined as 110 mm by consulting with the drilling engineers, so as to reduce the number of destressing holes outside. The clear spacing between adjacent coring holes is at least 20 cm, while the spacing between the coring hole and the surrounding destressing hole is at least 30 cm. The layouts of the destressing holes basically can be of two types, i.e. square and circle (Fig. 9). The length of the side of square stress relief area is about 90 cm, and the diameter of the circular stress relief area is around 100 cm. 1 006 400 900 (a) Square arrangement. 400 900 (b) Circular arrangement. Fig. 9 Arrangement types of destressing holes (unit: mm). After the above parameters are determined, the spacing of adjacent destressing holes is the only unknown, and the effect of damage-free coring mostly depends on the selection of the spacing. In order to choose a reasonable spacing for damage-free coring, 76 1 006 76 110 110

Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 49 the stress relief effects of five different intervals on adjacent destressing holes and two kinds of destressing hole distribution types are simulated by using FLAC 3D. The corresponding calculation parameters are shown in Table 1. Table 1 Calculation parameters of different destressing holes. Diameter of Spacing between adjacent Layout of Number of Case destressing destressing holes (mm) destressing destressing No. hole holes holes Center-to-center Wall (mm) spacing thickness 1 Circle 110 12 260.6 150.6 2 Circle 110 14 223.3 113.3 3 Circle 110 16 195.3 85.3 4 Square 110 16 225.3 115.3 5 Square 110 20 180 70 1 (MPa) 3 (MPa) The parameters of surrounding rocks adopted in calculation refer to those of the surrounding rocks of 5 class III of Yantang group marble ( T 2y ) in Jinping diversion tunnels, whose geological strength index (GSI) is 60, and the unconfined UCS is 110 MPa. The in-situ stress fields are: 1 x 0.83 MPa, 2 y 43.93 MPa, 3 z 38.53 MPa, and the directions of the maximum and intermediate principal stresses are both horizontal, while the direction of the minimum principal stress is vertical (Chu, 2009). Numerical model is shown in Fig. 10, in which the dimensions are 3 m 3 m 5 m (x y z), and the diameter of the holes is 110 mm. The bottom surface of the model is fixed. Taking case 2 for example, calculation results are shown in Fig. 11. z y x Fig. 10 Numerical model of damage-free coring. B z O (a) Distribution of the maximum principal stress. y A x y C z D x The principal stress (MPa) (b) Stress paths of points A, B and C. Destressing holes O 100 D 80 Stress relief area 1 In-situ stress area 60 2 40 3 20 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 r (m) (c) Distribution of the principal stresses on cross-section O D. Fig. 11 Simulation results of case 2. The distribution of the maximum principal stress in the sampling area after drilling of the destressing holes in turn is shown in Fig. 11(a). It can be observed that the stress concentration appears in the vicinity of destressing holes. The curves shown in Fig. 11(b) are the stress paths of points A, B and C during destressing holes drilling. The stress level of point C exceeds the empirical envelope of crack initiation strength, and it means that the rock around point C may be damaged during drilling process. But the stress states of points A and B are both always lower than the empirical envelope of crack initiation strength, so no cracks occur there. Fig. 11(c) shows the distribution of the principal stresses along the cross-section O-D. It could be observed that the maximum principal stress reduces from 60 to 30 MPa after stress relief, and the difference between the maximum and minimum principal stresses reduces from 20 to around 10 MPa. The stress state has been changed significantly. Fig. 12 indicates the stress relief rates of above five cases, of which three cases with circular arrangement of destressing holes and two with square arrangement, as shown in Figs. 12(a) and (b), respectively.

50 Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 Stress relief rate (%) 70 60 50 40 30 20 10 0 1 2 8 9 10 11 12 13 14 15 16 Spacing between adjacent destressing holes (cm) (a) Circular arrangement of destressing holes. 3 (a) Test site. 50 2 40 3 30 20 10 0 9 10 11 12 Spacing between adjacent destressing holes (cm) (b) Square arrangement of destressing holes. Fig. 12 Stress relief rates with two kinds of destressing hole arrangements. Stress relief rate (%) From Figs. 11 and 12, the following conclusions can be made: (1) Generally, if the spacing between adjacent destressing holes is smaller, the stress relief effect would be better. (2) With the same spacing of adjacent destressing holes, the stress relief effect of schemes using circular arrangement of destressing holes is better than that with square arrangement, and the circular arrangement of destressing holes needs fewer holes than square arrangement. (3) For all stress relief schemes mentioned above, it is noted that the stress change in rock would not exceed crack initiation strength within 30 cm from the center of stress relief area (within point B in Fig. 11), and the rock specimen obtained from this area may satisfy the requirements of damage-free coring test. 1 (b) Coring position. (c) Cross-section of coring area. Fig. 13 Test site of damage-free coring. 3 Laboratory test verification Considering the construction difficulty and other factors comprehensively, case 2 with the circular arrangement of destressing holes was chosen finally. Specifically, the diameter of the destressing holes is 110 mm, the clear spacing of adjacent destressing holes is 113.3 mm, and 14 destressing holes were arranged. 3.1 Implementation of damage-free coring The site of the damage-free coring test is located in the test branch tunnel 2-1 of Jinping II hydropower station, which is shown in Fig. 13. Fig. 14 shows the situation of field lofting of destressing holes, damage-free coring holes, and the conditions after coring completion. (a) Lofting of coring holes. (b) Situation after coring. Fig. 14 Lofting of coring holes and situation after coring. 3.2 Situation of core discing The situations of damage-free coring holes and cores

Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 51 obtained from destressing holes (routine coring) are shown in Figs. 15 and 16, respectively. There are still core discing phenomena in the cores obtained by damage-free coring technique, but the discing extent and the hole segment length of discing are much less than those of destressing holes. As shown in Fig. 15, the core discing appears partly in holes C and E in all five damage-free coring holes, marked by A E separately, while it appears continuously and approximately in each destressing hole (Fig. 16). It indicates that the purpose of the designed damage-free coring technique is roughly satisfied. However, due to the rock heterogeneity along coring holes, the stress relief induced damage during coring still occurs locally. It means that the spacing between adjacent destressing holes should be further reduced. 3.3 Uniaxial compression tests and AE monitoring Comparison of uniaxial compression tests between the traditional coring method and damage-free coring method is made, where the tests are conducted on the MTS815 servo-controlled rock pressure testing machine. The test equipment and layout of AE monitoring are shown in Fig. 17. The AE monitoring system is made by PAC Company, and the monitoring sensor model is of R61-AST type. 13.52 m (a) MTS815 servo-controlled rock pressure testing machine. (a) Damage-free core from hole C. 14.12 m 17.31 m AE sensor (b) Damage-free core from hole E. 18.26 m Fig. 15 Discing phenomena of damage-free cores. (a) Cores in the 5th box from destressing hole No. 2. (b) Cores in the 6th box from destressing hole No. 4. Fig. 16 Discing phenomena of cores obtained from destressing holes. (b) Layout of AE monitoring. Fig. 17 Test equipment and layout of AE monitoring. Part of test results is shown in Table 2. The specimens D25-2, D28, D30, D31 and D35 are damage-free cores, and the specimens Y3-2 and Y4-2 are obtained from destressing holes with traditional coring method. It can be seen that the UCSs of damage-free cores are higher than those of traditional cores by 5% 15%, averaging about 8%. The buried-depth of coring position of the rock specimens used in this test is about 1 900 m (Fig. 13), but the maximum buried-depth of tunnels in Jinping II hydropower station is about 2 500 m. Thus, it could be inferred that the coring-induced unloading damage in rock cores that are sampled from larger buried-depths or higher ground stress positions would be larger than those monitored in the laboratory tests. And the rock strength would be seriously underestimated in traditional coring tests.

52 Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 Specimen No. Table 2 Results of uniaxial compression tests. Depth of sampling (m) Initial Final Diameter of sample (cm) UCS (MPa) Improving rate to average UCS of traditional cores (%) D25-2 12.45 12.67 4.98 98.2 5.76 D28 13.72 13.87 4.93 107.6 15.89 D30 15.68 15.86 4.93 97.5 5.01 D31 16.1 16.23 4.97 99.8 7.49 D35 17.4 17.58 4.96 98.9 6.52 Y3-2 9.97 10.37 4.92 93.3 Y4-2 6.06 6.36 4.92 92.4 The typical axial stress-strain curves and AE monitoring results are shown in Fig. 18. It could be observed that the AE counts monitored at the early loading stage of damage-free core is few, and the AE events mainly appear before peak strength. But for the cores obtained from destressing holes by the traditional coring method, the AE events appear uniformly at the whole loading stage, which indicates that coring damage exists in the core obtained by the traditional coring method. It is similar to the results shown in Figs. 4 and 5. Axial strain (10 3 ) (a) Specimen D28. proposed method is optimized by numerical simulations. Then, the method is applied to the test tunnel 2-1 in Jinping II hydropower station, and damage-free cores and traditional cores are both obtained. At last, comparative laboratory tests using two kinds of specimens are carried out to test the effect of the proposed method. The main conclusions are drawn as follows: (1) The statistics, core discing and laboratory tests all indicate that the traditional coring method would cause damage to rock samples to some extent under high in-situ stresses, and the damage is mainly induced by stress relief during coring. (2) By employing the proposed coring method, the influence of in-situ stress relief could be significantly reduced by drilling destressing holes with proper spacing around the coring area. Numerical simulations show that, if the clear spacing between adjacent destressing holes is about 10 cm, the in-situ stresses could be reduced by 30% 50%. (3) The UCSs of damage-free samples obtained from the test tunnel 2-1 with an overburden depth of 1 900 m in Jinping II hydropower station are 5% 15% higher than those of traditional coring tests, averaging about 8%. The effect of proposed damage-free coring method could also be verified by the results that the monitored AE events of damage-free cores during early loading stage are fewer than those of traditional cores. The work is only a preliminary study on the coring damage under high in-situ stresses. In addition, the researches on the range and extent of drilling-induced damage to cores and the influence of coring damage on rock mechanical characteristics are still needed. References Backblom G, Martin C D. Recent experiments in hard rocks to study the excavation response: implications for the performance of a nuclear waste geological repository. Tunneling and Underground Space Technology, 1999, 14 (3): 377 394. Brignoli M, Fanuel P, Holt R M, Kenter C J. Effects on core quality of a Axial strain (10 3 ) (b) Specimen Y3-2. Fig. 18 Typical axial stress-strain curves and results of AE monitoring. 4 Conclusions In this study, a damage-free coring method is proposed in order to reduce the coring damage induced by stress relief during drilling, and the effect of the bias stress applied during coring. In: Proceedings of EUROCK 98 (Vol.1). [S.l.]: [s.n.], 1998: 301. Carlson S R, Wang H F. Microcrack porosity and in-situ stress in Illinois Borehole UPH-3. Journal of Geophysical Research, 1986, 91 (B10): 10 421 10 428. Charlez P, Hamamdjian C, Despax D. Is the microcracking of a rock a memory of its initial state of stress? In: Proceedings of International Symposium on Rock Stress and Rock Stress Measurements. Stockholm: [s.n.], 1986: 341 350. Chu Weijiang. Wall rock stabilization and safety evaluation of deep buried

Peng Yan et al. / J Rock Mech Geotech Eng. 2012, 4 (1): 44 53 53 tunnel. Hangzhou: HydroChina Huadong Engineering Corporation, 2009 (in Chinese). Dyke C G. Core discing: its potential as an indicator of principal in-situ stress directions. In: Maury V, Formaintraux D ed. Rock at Great Depth. [S.l.]: A. A. Balkema, 1989: 1 057. He Manchao, Xie Heping, Peng Suping, Jiang Yaodong. Study of rock mechanics in deep mining engineering. Chinese Journal of Rock Mechanics and Engineering, 2005, 24 (16): 2 803 2 813 (in Chinese). Holt R M, Brignoli M, Kenter C J, Meij R, Schutjens P M. From core compaction to reservoir compaction: correction for core damage effects. In: Proceedings of EUROCK 98 (Vol.2). [S.l.]: [s.n.], 1998: 311 320. Holt R M, Brignoli M, Kenter C J. Core quality: quantification of coring-induced rock alteration. International Journal of Rock Mechanics and Mining Sciences, 2000, 37 (6): 889 907. Holt R M, Fjar E, Furre A K. Laboratory simulation of the influence of earth stress changes on wave velocities. In: Fjar E ed. Seismic Anisotropy of Social Exploration Geophysics. [S.l.]: [s.n.], 1996: 180 202. Kenter C J, Brignoli M, Holt R M. CMS (constant mean stress) vs. UCS (unconfined strength) tests: a tool to reduce core damage effects. International Journal of Rock Mechanics and Mining Sciences, 1997, 34 (3/4): 129. Martin C D, Christiansson R. Estimating the potential for spalling around a deep nuclear waste repository in crystalline rock. International Journal of Rock Mechanics and Mining Sciences, 2009, 46 (2):219 228. Martin C D, Read R S, Martino J B. Observations of brittle failure around a circular test tunnel. International Journal of Rock Mechanics and Mining Sciences, 1997, 34 (7): 1 065 1 073. Martino J B, Chandler N A. Excavation-induced damage studies at the underground research laboratory. International Journal of Rock Mechanics and Mining Sciences, 2004, 41 (8): 1 413 1 426. Read R S. 20 years of excavation response studies at AECL s underground research laboratory. International Journal of Rock Mechanics and Mining Sciences, 2004, 41 (8): 1 251 1 275. Shan Z G, Wu X M, Yan P, Liu N, Zhu Y S, Zhu H C. Behavior of marble at Jinping II Project part 1: intact rock. In: Proceedings of the 44th US Rock Mechanics Symposium, the American Rock Mechanics Association. [S.l.]: [s.n.], 2010: 72. Shan Zhigang, Yan Peng. Management of rock bursts during excavation of the deep tunnels in Jinping II Hydropower Station. Bulletin of Engineering Geology and the Environment, 2010, 69 (3): 353 363. Strickland F G, Ren N K. Use of differential strain curve analysis in predicting in-situ stress state for deep wells. In: Proceedings of the 21st US Symposium on Rock Mechanics. Rolla: [s.n.], 1980: 523 532. Wang Mingyang, Zhou Zeping, Qian Qihu. Tectonic, deformation and failure problems of deep rock mass. Chinese Journal of Rock Mechanics and Engineering, 2006, 25 (3): 448 455 (in Chinese). Zhao Xingdong, Tang Chun an, Li Yuanhui, Yuan Ruifu, Zhang Jianyong. Study on the AE activity characters under uniaxial compression loading. Chinese Journal of Rock Mechanics and Engineering, 2006, 25 (Supp.2): 3 673 3 678 (in Chinese). Zhou Hongwei, Xie Heping, Zuo Jianping. Developments in researches on mechanical behaviors of rocks under the condition of high geostress in the depths. Advances in Mechanics, 2005, 35 (1): 91 99 (in Chinese).