Abstract
Granite is a common brittle rock material in underground surrounding rock, such as Yebatan Hydropower Station. The underground cavern of Yebatan Hydropower Station has characteristics of large span, high ground stress, fault development, and abundant groundwater. Long-term stability of surrounding rock is a key factor for the project safe operation. Especially for the surrounding rock of large underground cavern, its long-term rheological stability must be considered. In this paper, taking the surrounding rock of a large underground cavern in Yebatan Hydropower Station as research object, creep mechanical test of the combined axial pressure and water pressure under normal water level and dead water level is carried out, and long-term damage mechanism of rock under different water confining pressures is revealed. Based on rock mechanical properties, calculation and analysis of surrounding rock stability under different support time are carried out, and optimal support time is also discussed.
1. Introduction
Yebatan Hydropower Station is located on the main stream of Jinsha River in Baiyu County, Sichuan Province and Gongjue County, Tibet. Surrounding rock of underground cavern is quartz diorite, with high overall strength and good tunnel forming conditions. However, faults are relatively developed, in situ stress is high, and groundwater is abundant, having a great impact on surrounding rock stability. In addition, the rock of Yebatan project has the following remarkable characteristics: (1) intact rock has high strength, but weak weathering resistance; (2) Yebatan slope rock intersects with the riverbed at a large angle, and structural fissures are developed, facilitating for river water to penetrate into the underground cavern; (3) excavation scale of underground cavern is huge and stress redistribution is obvious, resulting in obvious stress concentration in the rocks in some areas. These characteristics make Yebatan underground cavern rock vulnerable to a combined action of in situ stress and groundwater, which may have an adverse impact on the stability of surrounding rock.
In the past, related research on surrounding rock stability was mostly aimed at instantaneous mechanical properties of the rock mass, such as conventional uniaxial compression or triaxial compression tests, in order to obtain relevant mechanical parameters and carry out subsequent excavation, support, and reinforcement work. Zhang Yi studied the properties of soft surrounding rock of tunnel expansive soil through triaxial strength test [1]; Guo et al. obtained mechanical properties of weathered granite through triaxial compression test, which provided theoretical guidance for grouting reinforcement and lining of tunnel surrounding rock [2]; Xi and Zhao studied the mechanical properties of surrounding rock under high temperature and high pressure through laboratory tests, which provided a theoretical basis for the maintenance of drilling surrounding rock stability [3]; Shan et al. studied mechanical properties of quartz mica schist through uniaxial and triaxial tests, providing theoretical guidance for surrounding rock classification [4]; Takahashi et al. studied deformation and strength characteristics of Kimachi sandstone under confined compression and extension test conditions [5].
However, for large underground caverns with a long service life, engineering problems such as long-term stability and safety caused by rheology are increasingly prominent, and the long-term stability of surrounding rock cannot be ignored, which has achieved a consensus in the academic and engineering circles. Wang et al. used FLAC3D to study influence of roadway confining pressure on rock creep rate and revealed influence law of confining pressure on surrounding rock rheological properties [6]; Wang et al. compared creep deformation of tunnel surrounding rock before and after lining under different buried depths by numerical simulation method [7]; Tao et al. studied rheological law of weak surrounding rock under high geostress and put forward a principle of engineering control for rheology of weak surrounding rock [8]. It can be seen that previous studies did not reflect influence of groundwater on the long-term rheological properties of rocks. In recent years, more and more scholars pay attention to water-rock interaction and analyze effect of water on the mechanical properties of rocks. Fu et al. studied evolution characteristics of sandstone mechanical parameters under different water-rock interaction and obtained that uniaxial compressive strength and elastic modulus of sandstone decreased with increase of water-rock interaction times [9]; Niu et al. studied damage law of mechanical properties for altered rock under water-rock interaction and introduced a concept of damage rate [10]; Cao and Chen studied propagation law of subcritical cracks in rock blocks under natural and saturated conditions [11]. However, they did not conduct relevant research on rocks mechanical properties under coupling action of axial pressure and hydraulic pressure in a real water pressure environment. The presence of pressurized water may alter the microstructure inside the material. The macroscopic mechanical response and damage model of materials are closely related to the microstructure [12].
Based on this, with the help of advanced test equipment, this paper takes the Yebatan underground surrounding rock as research object, carries out creep mechanical test for rock under normal pool level and dead water level in real water environment, and reveals long-term damage mechanism of rock under different water confining pressures. On the other hand, supporting time is also a key factor to maintain surrounding rock stability. According to the rock indoor mechanical test, this paper selects reasonable rock mechanical parameters, combines the on-site geological conditions, and establishes a three-dimensional numerical model to calculate the typical area, in order to provide targeted theoretical guidance for the on-site construction.
2. Project Overview
Yebatan Hydropower Station is located on the main stream of Jinsha River in Baiyu County, Sichuan Province and Gongjue County, Tibet. It is the 7th level of the 13-level development plan for upper reaches in Jinsha River. The underground powerhouse has a horizontal depth of 270∼480 m and a vertical depth of 240∼460 m. The longitudinal axis of powerhouse is N50°W. The size of main powerhouse is 268 m × 28.5 m × 67.1 m, with huge scale. The lithology of surrounding rock is quartz diorite (δο43). The underground cavern is mainly controlled by three secondary structural planes. In addition, there are several tertiary structural planes. The maximum measured in situ stress is 37.57 MPa, belonging to the high to extremely high stress area. So underground cavern has the characteristics of large span, high ground stress, fault development, and abundant groundwater, as shown in Figure 1.
(a)
(b)
(c)
(d)
3. Creep Test under Combined Action of Axial Pressure and Water Pressure
3.1. Test Equipment and Test Methods
Rock samples used in this test are taken from the main powerhouse of Yebatan Hydropower Station, and the lithology is granite. Drilling and blasting method is adopted for powerhouse excavation on site, so we can take samples from the large and complete rock blocks dropped by blasting. The cylindrical specimens that meet the test requirements are taken out from the rock blocks with a core-taking machine. Conventional physical and mechanical parameters of rock must be mastered before the creep test under the combined action of axial pressure and water pressure. Therefore, it is necessary to carry out conventional uniaxial compression test of rock. There are six samples numbered Y-1, Y-2, Y-3, Y-4, Y-5, and Y-6. Among them, Y-1, Y-2, and Y-3 will be soaked to saturation to carry out the test and Y-4, Y-5 and Y-6 will be kept in a natural state to carry out the test, as shown in Figure 2. MTS equipment is used to carry out experimental research, as shown in Figure 3. According to the requirements of the International Society for Rock Mechanics (ISRM), the uniaxial compression test of granite is carried out.
(a)
(b)
Physical parameters of granite in natural and saturated states are obtained, as shown in Table 1.
The stress-strain curves are obtained, as shown in Figure 4.
Note: the axial extensometer of the Y-2 specimen was loose during the test, and the axial strain could not be accurately measured, so the stress-strain curve of the Y-2 specimen was not given here.
The conventional mechanical parameters of granite in saturated and natural state are obtained, as shown in Table 2.
In the uniaxial compression test, saturated strength of granite is reduced by 7.3% and elastic modulus is reduced by 6.4% compared with the natural strength, which has a certain softening effect in water. At the same time, it can be seen from the stress-strain curve that, except for Y-4 specimen, when specimen reaches peak strength and brittle failure occurs, the volumetric strain is positive, indicating that specimen does not have capacity expansion at this stage, there is no obvious slip dislocation in the rock, and brittle characteristics are quite obvious. Failure mode of sample is shown in Figure 5.
As can be seen from the above figure, failure modes of each specimen show such a law: failure occurs at the upper and lower ends of the specimen, and failure mode is dominated by longitudinal tensile splitting failure, and specimen does not show obvious shear failure. According to the above test mechanical parameters, combined with the equipment parameters, creep specimen’s size is further determined, which is set as a cylindrical specimen of φ25 ∗ 50 mm. The initial axial load is 30 kN, and the water confining pressure needs to be determined according to the water level on site.
The water confining pressure is external water pressure in the actual project, and its value is obtained from “NB/T 10391-2020 Design Specification for Hydraulic Tunnels” [13], namely,where is external water pressure acting on the outer surface of the cave, kN/m2; is reduction coefficient of external water pressure; is bulk density of water (using 9.81 kN/m3), kN/m3; is acting head, m.
The normal pool level of Yebatan dam is 2889.00 m. Under special circumstances such as extremely dry years, on the premise of safe operation of the power station, water level can be reduced to 2855.00 m; that is, the dead water level is 2855.00 m. Check flood level of the dam is 2891.22 m. For main powerhouse, the elevation at the top arch is 2729.00 m, and elevation at the lowermost section is 2659.20 m. According to the check flood level and dead water level of Yebatan, combining formula (1) and Table 3 (using βe 0.2), the maximum water confining pressure is 0.50 MPa and the minimum water confining pressure is 0.25 MPa in the test. Schematic diagram of calculation is shown in Figure 6.
At this time, the maximum water confining pressure is exactly twice the minimum water confining pressure. In order to give consideration to the accuracy and timeliness of the test, step loading is adopted. According to the conventional uniaxial compressive strength of the specimen, the initial axial compressive load is determined to be 30 kN. If the specimen has creep failure, the test is over. If the deformation becomes stable, the next-level load is applied, and the load gradient is 5 kN until creep failure occurs. The loaded experimental scheme is shown in Figure 7.
The equipment used in this test is a microcomputer-controlled multichannel rock rheological test system under the combined action of axial pressure and water pressure, as shown in Figure 8.
For the equipment, axial pressure range is 10∼200 kN; water confining pressure range is 0∼10 MPa; axial displacement range is 0∼25 mm; and the measurement accuracy is ±0.2 μm. The rock rheological system truly realizes a complete contact between pressure water and rock sample. Stress diagram of rock sample is shown in Figure 9.
3.2. Creep Test Results under Different Water Confining Pressures
Creep test curve of rock under different water confining pressures is shown in Figure 10.
(a)
(b)
As can be seen from Figure 10, the rheological curve of granite shows the following laws:(1)For rheological curve of rock when the water confining pressure is 0.25 MPa, in the first two stages of creep, rheological curve can be divided into initial elastic stage, creep attenuation stage, and creep stability stage. In the last stage of creep, there is a creep acceleration stage, which leads to specimen final failure; at this time, the axial load of the specimen is 40 kN. For the rheological curve of rock with the confining pressure of 0.5 MPa, there are all stages including initial elastic stage, creep attenuation stage, creep stability stage, and creep acceleration stage. When the specimen is damaged, the axial load is 30 kN.(2)When the water confining pressure is 0.25 MPa, the rock is finally destroyed after three stages of loading. When the water confining pressure is 0.5 MPa, rock failure occurs after a single loading.(3)Water confining pressure has a significant effect on rock deformation after rheological failure. When the water confining pressure is 0.25 MPa, the strain of the specimen is about 0.00175. When the water confining pressure is 0.5 MPa, the strain of the specimen is about 0.00275.(4)The rheological failure time of rock under low water confining pressure is significantly longer than that under high water confining pressure, which is almost one order of magnitude different.
Based on the above points, it can be inferred that water confining pressure has an obvious control effect on rock rheology when rock is placed in real water environment. There are natural micropores and microcracks in rock. When rock is placed in pressure water, pressure water will penetrate into the micropores and microcracks. There are two possible effects:(1)Pressurized water has a confining pressure effect similar to that of original rock stress, which restrains micropores and microcracks and makes them difficult to open and expand, thus improving the strength of rock and enhancing the relevant mechanical parameters of rock. The higher the water confining pressure is, the more obvious the strengthening effect is.(2)The pressure water will form a seepage field in the rock, which makes the original micropores and microcracks forced to open and develop, the damage range continues to expand, and the relevant mechanical parameters of the rock are weakened. The higher the water confining pressure is, the more obvious the weakening effect is.
It can be seen from the final results that, under the action of water-rock coupling, compared with the water confining pressure of 0.25 MPa, when the water confining pressure is 0.5 MPa, the final deformation and failure time of specimen is shorter, the strain is larger and the axial load is smaller, so pressure water has a weakening effect on mechanical properties of rock. Therefore, the action (2) is consistent with actual situation of underground surrounding rock of the cavern. Therefore, for the surrounding rock of Yebatan underground cavern, groundwater is a significant adverse factor; especially when the water pressure is high, it will have an adverse effect on the long-term stability of rock. Pumping and drainage measures are important measures to ensure the stability of surrounding rock. Based on this, a total of three layers of antiseepage drainage corridors are designed for the underground caverns of Yebatan Hydropower Station, which effectively ensured the long-term stability of cavern group, as shown in Figure 11.
4. Analysis of Supporting Time of Surrounding Rock
The mechanical properties of surrounding rock of underground caverns have been studied in detail, including instantaneous mechanical properties and long-term mechanical properties under different water environments, thus deepening the understanding of surrounding rock stability from internal mechanism. On the other hand, external action is also an important factor to ensure surrounding rock stability, such as supporting time and construction technology. Based on the rock mechanical properties revealed in the previous research and combined with the geological conditions of the underground caverns in Yebatan, this paper studies the supporting time of prestressed bolt for surrounding rock, in order to provide theoretical guidance for surrounding rock stability under similar background.
4.1. Model Establishment
In this paper, 3DEC software is used to study prestressed bolt support timing. A tunnel excavation model is established according to geological data, relevant mechanical parameters of rock, and design scheme, as shown in Figure 12. The numerical model adopts displacement boundary conditions; that is, normal displacement constraints are imposed on the front, back, left, and right sides of the mountain model, the bottom of the model is fixed, and the top of the model is set as a free boundary. Considering the accuracy and time consumption of the calculation results, different meshing parameters are adopted for different parts of the model; that is, the plant is a main research part of the model, so meshing division is relatively small, and the cell mesh size is 1 m; for the rest of the mountain model except the plant, meshing division is relatively large, and the cell mesh size is 10 m.
(a)
(b)
4.2. Numerical Simulation Test Scheme
In practical projects, when a wool hole excavation is completed, support will always lag behind the working face a certain distance. In order to better get the influence of support time on surrounding rock stability, this paper plans to calculate and analyze the surrounding rock stability under different distances (10 m, 20 m, 40 m, and 80 m) of prestressed anchor behind the working face in the first two layers of the powerhouse, so as to provide a certain theoretical basis for support time selection. The parameters of prestressed bolt are L = 9m, φ = 32 mm, and T = 120 kN. And the support spacing is 1.2 m × 1.2 m. Schematic diagram of different distances of prestressed anchor lag working face is shown in Figure 13.
(a)
(b)
(c)
(d)
Surrounding rock deformation is an important reference factor to judge whether it is stable. Therefore, the maximum deformation of surrounding rock and unit sections 1#∼6# are taken as standards to evaluate the supporting effect in this paper. Unit sections 1#∼6# are shown in Figure 14.
4.3. Numerical Simulation Results
The deformation nephogram of surrounding rock under different support lag distances is obtained, taking the prestressed anchor lag working face 40 m as an example, as shown in Figure 15. When the prestressed anchor lags behind the working face by 10 m, 20 m, and 80 m, there is a similar nephogram diagram, so it will not be given here.
The maximum displacement mostly occurs at the downstream side arch waist and the upper part of the side wall, and the value is between 1.0 cm and 3.6 cm. Under different support lag distances, the maximum displacements of surrounding rock and unit 1–6# section are shown in Table 4.
Relationship between the maximum displacement at different positions and the distance of bolt support lagging working face is shown in Figure 16.
(a)
(b)
Taking the surrounding rock displacement when the support lags behind the working face for 10 m as the reference value, the displacement growth percentage under other spacing can be obtained, as shown in Table 5.
It can be seen from Table 5 that when the support lag distance changes from 10 m to 20 m, the overall maximum displacement of surrounding rock increases by only 1.24%, and the maximum displacement of 1#∼6# unit section increases by 10.56%, 11.04%, 10.59%, 10.84%, 11.26%, and 12.56%, respectively. On the whole, the increase proportion of displacement is small. When support lag distance increases from 20 m to 80 m, displacement growth ratio is obvious. For the stability control of surrounding rock, in theory, it is an ideal state when support closely follows the working face. However, considering the construction process on site, the support always needs to lag the working face by a certain distance. Therefore, from the perspective of the displacement growth ratio, it can be considered that the lag distance is within 20 m, and displacement of surrounding rock will not change significantly. It is suggested that 20 m should be used as the maximum distance of support lag under the conditions on site.
5. Conclusion
(1)When the rock in the underground cavern of Yebatan Hydropower Station is under uniaxial compression, the failure part occurs at the upper and lower ends of the specimen, and the failure form is mainly longitudinal tensile splitting failure. Under the condition of no confining pressure and no fault, the specimen does not show obvious shear failure, and the rock has a certain softening effect in water.(2)According to creep test results of rock in real water environment, when confining pressure of water is high, the final deformation and failure time is short, the strain is large, and the axial load is small. Therefore, pressurized water has a weakening effect on the long-term mechanical properties of rock. Therefore, for the surrounding rock of Yebatan underground cavern, underground pressure water is a significant adverse factor, and pumping and drainage measures are important to ensure the stability of surrounding rock. A total of three layers of antiseepage drainage corridors are designed for the underground caverns.(3)For Yebatan Hydropower Station, the numerical simulation shows that when the distance between the prestressed anchor and the working face is controlled within 20 m, the displacement of surrounding rock will not change significantly.In addition, for water conservancy and hydropower projects, except for traditional rock materials and cement slurry materials, more and more green environment and ecofriendly materials also have broad application prospects [14]. The experimental methods proposed in this paper may be applied to the study on mechanical properties of these materials in the future.
Data Availability
All data used to support the findings of this study are included within the article.
Conflicts of Interest
The authors declare no conflicts of interest.