Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2023 Feb 7;8(7):6361–6375. doi: 10.1021/acsomega.2c06497

Experimental Study of the Fracture Network Expansion Mechanism and Three-Dimensional Reconstruction Characteristic of High-Rank Coal in Guizhou Province, China

Wei Liu †,*, Zihao Pan †, Xiong Zhang †,*, Yuanlong Wei ‡, Lingyun Zhao ‡, Xuanshi Zhu †, Youzhou Zhao §, Qiang Guo ∥
PMCID: PMC9947985  PMID: 36844508

Abstract

graphic file with name ao2c06497_0027.jpg

To better understand the fracture propagation characteristics and spatial distribution pattern of a high-rank coal reservoir during hydraulic fracturing, a true triaxial physical simulation device was used to conduct a hydraulic fracturing experiment on large-sized raw coal from Zhijin, Guizhou Province, China. Computed tomography technology was used to scan the three-dimensional morphology of the fracture network before and after fracturing, then AVIZO software was used to reconstruct the internal fractures of the coal sample, and fractal theory was used to quantify the fractures. The results show that (1) the sudden increase of the pump pressure curve and acoustic emission signal is an important identification feature of hydraulic fractures, and the in situ stress difference coefficient plays a leading role in the complexity of coal and rock fractures. (2) When a hydraulic fracture encounters a primary fracture in the process of expansion, the opening of the primary fracture, the penetration, bifurcation, and turning of the hydraulic fracture are the main reasons for the formation of complex fractures, and the existence of a large number of primary fractures is the basis for the formation of complex fractures. (3) The fracture shape of coal hydraulic fracturing can be divided into three categories: complex fracture, plane fracture + cross fracture, and inverted T-shaped fracture. The fracture shape is closely related to the original fracture shape. The research results of this paper provide strong theoretical and technical support for coalbed methane mining design such as Zhijin high-rank coal reservoirs.

1. Introduction

Historically, energy resources have evolved from high carbon to lower carbon fuels (from coal to oil to natural gas) and then to noncarbon (hydroelectric, geothermal, wind power, and solar).1 The vigorous development of coalbed methane (CBM) is an inevitable trend. At present, the United States, Australia, China, and Canada are the main countries in the world that have realized the industrial development of CBM.2−6 China’s coal measures strata are thick, widely distributed, and rich in coal measures gas resources. The exploration and development of CBM have attracted much attention.7 The proven geological reserves of CBM are about 634.5 × 109 m3, and the annual output of surface-extracted CBM is about 5.5 × 109 m3. The exploration rate and resource utilization rate of CBM are generally low.8−11 Guizhou Province is a big coal resource province in southern China, and the geological resources of Upper Permian CBM in Guizhou are 3.06 × 1012 m3, ranking second in China. The geological resources of Liupanshui, Zhina, and Qianbei coalfields are 2.83 × 1012 m3, accounting for 92.6% of the total resources in the province.12 Among them, the geological resources of Zhina coalfield are 7.00 × 1012 m3, accounting for 22.91% of the total geological resources in Guizhou Province. The geological conditions of CBM in Guizhou Province have the following characteristics: (1) low water-richness in Longtan Formation; (2) many gas-controlling structures and many coal seams; (3) high gas content in coal seams, high resource abundance, and high reservoir pressure and in situ stress; and (4) high CBM resources, great changes in coal rank, great changes in coal seam permeability, and great changes in vertical geological conditions.13 Therefore, hydraulic fracturing is needed to promote complex fractures in coal reservoirs. However, until now, the basic research on fracture propagation basis and fracture network characteristics of hydraulic fracturing in Guizhou coal reservoirs is still insufficient, and the theory and technology of CBM development in Guizhou Province still need to be improved.

A lot of research studies have been carried out on hydraulic fracture morphology and fracture propagation law of oil and gas reservoirs at home and abroad. Warpinski et al.14−16 discovered the phenomenon of simultaneous extension of main fracture and complex branch fracture during fracturing for the first time through field tests and put forward the concept of fracture extension zone. Mahrer17 indicated that the reservoir with primary fractures will form a complex fracture network after fracturing. Beugelsdijk et al.18 confirmed the existence of a fracture network during fracturing through laboratory tests. Through the hydraulic fracturing experiment of shale with primary fractures, Blanton19,20 pointed out that there may be three situations when hydraulic fractures encounter primary fractures: penetration, turning or penetration and turning. Fisher et al.21−24 monitored the reticular fractures formed during hydraulic fracturing of shale reservoirs by microseismic mapping. Cheng and Zhang25 used large-sized granite for hydraulic fracturing and studied the influence of the injection flow rate on characteristic parameters of different pressure curves. Mayerhofer et al.26,27 believed that increasing the volume of reservoir reconstruction and forming the maximum distribution of fractures and fractures are the key to the success of shale reservoir reconstruction. Different from shale and other single jointed rocks, coal has the characteristics of large deformation and well-developed cleat. During hydraulic fracturing, the hydraulic fracture forms are very complex, and the fracture propagation mechanism and fracture morphology description have always been a hot and difficult point in CBM reservoir research.28−30 Ai et al.31 studied the initiation and propagation characteristics of hydraulic fracture when the angle between cleat and maximum horizontal principal stress in coal face is different through the hydraulic fracturing test. Hou et al.32 carried out a series of hydraulic fracturing experiments on mechanical parameters and established the crack initiation criterion of horizontal well water in coal seam. Huang et al.33 observed the initiation and propagation direction of natural cracks in coal seam under different in situ stresses. Liang et al.34 studied the influence of in situ stress, tensile strength, and natural macroscopic crack on crack formation and propagation. Chen et al.35 studied the in situ stress distribution in the depth of 136–1244 m in western Guizhou and concluded that low horizontal stress anisotropy leads to a complex hydraulic fracture network.

In the aspect of hydraulic fracture morphology detection, acoustic emission (AE) technology and computed tomography (CT) scanning have been applied to hydraulic fracturing tests. Many scholars have obtained the spatial geometric shapes of fractures and porosity in samples through CT scanning technology.36−38 Ju et al.39 constructed heterogeneous sandstone and homogeneous sandstone models under hydraulic fracturing by using CT three-dimensional (3D) reconstruction and discussed the influence of the horizontal stress ratio on the spatial distribution and shape of fracturing fractures. Li et al.40 used high-frequency AE to monitor true triaxial hydraulic fracturing of coal. Zhang et al.41,42 used AE in the experiments of coal-bearing rocks and different types of coal to detect the generation of microcracks and the evolution of macroscopic and microscopic failures. Hao et al.43 obtained 3D reconstructed coal samples by CT scanning of bituminous coal with different water contents. Jiang et al.44 used CT to scan shale samples before and after fracturing, and the CT scanning images of internal fractures were reconstructed and visualized by 3D reconstruction.

At present, the evaluation of the reconstruction effect of the fracture network can be realized by characterizing the complexity of the fracture network after fracturing. The fractal theory not only reflects the distribution characteristics of cracks, but also quantitatively describes the complexity of cracks.45−47 Liu et al.48 introduced fractal theory to study the quantitative relationship between gas adsorption and pore characteristics of bituminous coal. Based on CT scanning and fractal theory, Li et al.49−51 characterized the fracture network of coal. Zhao et al.52 pointed out that the compressive strength of fractured rock mass decreases with the increase of fractal dimension.

It can be seen from the above research that the research of hydraulic fracturing mostly focuses on theoretical analysis, numerical simulation, and fracturing of similar materials. Most of them are aimed at shale reservoirs and hard coal reservoirs in northern China, but the research on the large-scale fracturing simulation experiment and fracture reconstruction of coal reservoirs in Guizhou Province is rare. Considering the complex geological conditions and special physical properties of Guizhou coal reservoirs, the existing research is difficult to be directly applied to guide the hydraulic fracturing and construction of Guizhou coal reservoirs. Therefore, it is urgent to carry out special research on fracture propagation and fracture network characteristics of Guizhou coal reservoirs.

In this paper, the typical high rank coal in Zhijin, Guizhou Province, China is taken as the research object, the underground raw coal blocks are selected and processed into large-sized cubic coal samples, and the physical simulation experiment of true triaxial horizontal well fracturing is carried out. Moreover, the influence of the in situ stress difference coefficient and primary fracture on the fracturing network of coal reservoirs is discussed. The coal samples were scanned by CT before and after the fracturing experiment. By comparing the 3D reconstruction model and fractal dimension characteristics of fractures before and after fracturing, the fracture morphology and propagation law were quantitatively analyzed. It provides theoretical and technical guidance for solving the demand of fractured coal reservoirs in Guizhou Province, China.

2. Materials and Methods

2.1. Experimental Equipment

The true triaxial hydraulic fracturing simulation experiment system adopted in this study can load different triaxial stresses on coal samples, simulating the stress state of coal in the formation. The fracturing simulation experimental system is mainly composed of a true triaxial servo loading system, a hydraulic fracturing pump pressure servo control system, an AE detection system, and other auxiliary devices. The overall structure is shown in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Route chart of the triaxial hydraulic fracturing simulation test of large-sized coal and rock.

In the process of hydraulic fracturing physical simulation, the true triaxial fracturing control system is used to inject hydraulic oil into the hydraulic bag, loading different directions, and magnitude of stress on the coal sample. Among them, two pairs of horizontally symmetrical hydraulic bladders respectively apply horizontal maximum principal stress σH and vertical principal stress σv to the coal sample. The horizontal minimum principal stress σh is applied when the hydraulic bag at the bottom and the pressure bearing plate at the top of the fracturing chamber are placed. Hydraulic bladders (bearing plates) in three directions can be independently loaded, and the pressure range is 0–70 MPa, as shown in Figure 2.

Figure 2.

Figure 2

Open in a new tab

True triaxial hydro-fracture simulation test device. (a) Appearance of fracturing equipment. (b) Inner structure.

The pump injection system includes a water tank, a fracturing pump, a dye container, and a fracturing pipeline. The fracturing pump can realize two injection modes of constant current and constant pressure, the flow control range is 0–80 mL/min, and the maximum pressure is 60 MPa. The dye container is filled with fracturing fluid with dye, which is used to track the fracture propagation.

The DS5 AE detection system is used for fracture propagation monitoring. AE signals were collected by the DISP AE monitoring device, and two AE probes and corresponding preamplifiers were arranged on each diagonal of four surfaces around the sample, with an amplifier of 40 dB.

2.2. Coal Sample Preparation

The coal used in this hydraulic fracturing physical simulation experiment sample is taken from Longtan Formation in Zhijin area, Guizhou Province, China. The Zhijin area is dominated by high-rank coal, with the coal depth ranging from 200–800 m, hard and brittle texture, some primary fissures, and a good gas-bearing property. It is one of the areas with the greatest experiment CBM development potential in Guizhou Province. Large-sized coal is taken from No.16 coal seam in Zhijin area, with a sampling depth of about 300 m. The coal is high rank coal with a shiny surface. The complete large coal of the working face is selected and transported to the ground and then packed tightly and sent to the processing plant.

Due to the limited size of the coal core taken from the Zhijin area, it is difficult to directly prepare the sample that meets the requirements of fracturing simulation size. Therefore, first, the large coal is cut to the maximum size by wire cutting, and the parts with missing edges and corners are repaired by cement mortar and then put into a special mold for curing, and a 200 mm × 200 mm × 200 mm standard cube sample is made. To minimize the strength difference between coal and cement mortar, a variety of cement mortar samples were prepared before the formal sample repair. It was finally found that the cement mortar prepared according to the mass ratio of ″quartz sand: 325 cement: coal powder = 7: 2: 1″ had similar uniaxial compressive strength and failure characteristics to Zhijin coal. Therefore, the cement mortar used to repair the sample is configured in this ratio. Considering that Zhijin coal seam is generally thin (minable coal seam is mostly 1–3 m), when repairing coal samples with cement mortar, the coal samples were tried to place in the middle position, which is used to observe the fracture propagation form at the interface between coal and cement mortar and provide reference for analyzing the fracture propagation at the interface between the roof and floor of coal seam.

The sample was cured for 28 days in a standard curing room with a temperature of (20 ± 2) °C and a relative humidity of over 95%. After curing, a hole was drilled with a depth of 130 mm and a diameter of 20 mm in the center of the sample to simulate the wellbore. The fracturing pipe (simulating field casing) is made of high-strength steel pipe. The designed fracturing pipe is 130 mm long, 14 mm in outer diameter, and 10 mm in inner diameter. Three rows of water outlets with a diameter of 6 mm are symmetrically opened in the range of 67 mm at the bottom end, and the internal threads are turned. The bottom end of the steel pipe is welded and sealed. The coal seam map, large-size raw coal, the map of samples with pouring concrete after wire cutting, schematic diagram of hole sealing samples, and wellbore sealing is shown in Figure 3, respectively. Then the outlet pipe with a length of 6 mm and external thread was connected to the outlet hole, and the exposed length is 3 mm. The fracturing pipe was inserted into the hole of the coal sample to simulate the perforation section around the casing, and the outlet pipe was blocked with soluble salt powder to prevent glue from infiltrating. Epoxy resin is used to fill and bond the gap between the outer wall of the fracturing pipe and the wellbore, simulating the cement ring between the wellbore and the surrounding rock of the wellbore. The schematic diagram of hole sealing sample and wellbore sealing is shown in Figure 4. After similar materials and epoxy resin are fully bonded with raw coal, hydraulic fracturing experiments were carried out.

Figure 3.

Figure 3

Open in a new tab

Coal rock standard sample.

Figure 4.

Figure 4

Open in a new tab

Photo of fracturing steel pipe and schematic diagram of wellbore sealing.

2.3. Experimental Method

The hydraulic fracturing experiments in this study mainly consider the influence of primary fracture development and in situ stress difference coefficient on the expansion and extension of hydraulic fracture morphology. (1) Primary fractures: They will affect the direction, deflection, and bifurcation of fractures in the process of extension and also affect the final fracture volume density. The development of primary fractures in Zhijin coal seams is different; therefore, the influence on the final fracturing effect needs to be explored and identified, which is used to provide a basis for selecting layers and fracturing fluid dosage in the same coalfield. (2) In situ stress difference coefficient: It will affect the extension direction of the main fracture and the final transformation scale. Considering the complex structure and large change of in situ stress in Zhijin coalfield, the in situ stress difference coefficient is also taken as the analysis object. To simulate fracturing in coal seam, the directions of all shafts are parallel to coal bed bedding (simulating horizontal well fracturing). To better fit the site, the fracturing fluid used in this experiment is all 1% KCl solution (close to the site), and KCl solution has an obvious expansion inhibition effect.

The geological structure of the Guizhou coal reservoir is complex, and the in situ stress state is changeable. The in situ stress difference coefficient is an important factor affecting the initiation and expansion of the fracture network in coal reservoirs. In situ stress difference coefficient k is defined as eq 1:53

2.3. 1

where σH and σh are the horizontal maximum principal stress and the horizontal minimum principal stress, respectively, MPa.

This paper provides an experimental method for simulating hydraulic fracturing of horizontal wells. The experimental parameters are shown in Table 1, and the loading mode is shown in Figure 5, where X is the front-back direction of the experiment equipment (vertical direction of the stratum), the loading stress is the maximum principal stress σV and perpendicular to the bedding plane, Y and Z are the horizontal direction of the stratum, and Y is also the left–right direction of the experimenting machine, and the loading stress is σH. Z is the up-down direction of the experimenting machine, and the loading stress is σh.

Table 1. Relevant Parameters of the True Triaxial Hydraulic Fracturing Simulation Test Scheme for Coal Samples.

specimen no. triaxial stress /MPa
      
  σV σH σh in situ stress difference coefficient k the displacement of fracturing fluid mL min–1 types of fracturing fluids
ZJ-1 5.60 5.58 4.46 0.25 45 1%KCl
ZJ-8 5.60 5.58 4.46 0.25 45
ZJ-5 5.60 5.58 4.29 0.30 45
ZJ-9 5.60 5.58 4.29 0.30 45
ZJ-10 5.60 5.58 4.13 0.35 45
ZJ-11 5.60 5.58 4.13 0.35 45
Open in a new tab

Figure 5.

Figure 5

Open in a new tab

Schematic diagram of stress loading and naming diagram of each surface.

Large hydraulic fracturing equipment is used to simulate the 3D stress state of real stratum, and the hydraulic fracturing pump servo system accurately controls the displacement of fracturing fluid. The DS5 AE detector with eight channels can detect the information of fracture initiation and propagation in the hydraulic fracturing process in real time. Before and after the experiment, the coal samples were scanned by CT, and the spatial forms of primary fractures and hydraulic fractures were reconstructed. The specific steps are as follows:

  • (1)

    The sample is put into the fracturing chamber after its orientation is calibrated, and two AE probes and corresponding preamplifiers are arranged on each diagonal of four surfaces around the sample, and bond them with a coupling agent. Before the experiment, check whether there is a signal in AE.

  • (2)

    Loading 3D stress is started until the set value, so as to simulate the real in situ stress state of coal.

  • (3)

    Hydraulic fracturing pump pressure servo control system and AE detection system are started to collect pump pressure information and AE signals in real time.

  • (4)

    After the pump pressure is stable and a stable fracture channel appears in the sample, the servo control system of hydraulic fracturing pump pressure is turned off first; after the pump pressure drops and stabilizes, the data acquisition is stopped, and the AE system is turned off; finally, the 3D stress of the hydraulic fracturing device is unloaded to zero.

  • (5)

    The sample is taken out, photos of six surfaces of the sample are taken, the fractures on the surface of the sample are observed, and the fracture form from its appearance is preliminarily recognized.

  • (6)

    After the fracturing experiment, the second CT scan is performed on the sample, and it is compared with the CT scan results before fracturing, so as to preliminarily determine the fracture propagation law. Then, combined with the hydraulic pump pressure–time curve and AE detection information, the fracture initiation, propagation, and spatial morphology are comprehensively analyzed, and the 3D reconstruction of the sample fracture is carried out to discuss the formation mechanism of Zhijin coal hydraulic fracturing fracture.

3. Results

3.1. Analysis Combined with AE, Pump Pressure Curve, and Sample Cutting

3.1.1. Result Analysis of the ZJ-8 Sample

The ZJ-8 sample is relatively complete, with few primary fractures on its surface, and the in situ stress difference coefficient k = 0.25. At the beginning of the experiment, 3D stress is applied to the sample in X, Y, and Z directions until σV = 5.60 MPa, σH = 5.58 MPa, and σh = 4.46 MPa. After the pressurization system is stable, the pumping system was started and the fracturing fluid was injected at a constant flow rate of 45 mL/min. The curve of pump pressure-AE ringing-time of sample Z8 is shown in Figure 6. With the continuous injection of fracturing fluid, the drilling hole is gradually filled with fracturing fluid, and the pressure at the bottom of the drilling hole rises slowly, and the whole process lasts for 25.00 s. At this time, the wall of the drilling hole breaks and fractures under the action of water pressure, and the pump pressure is 0.62 MPa, but the AE ringing information is not obvious. After that, the pump pressure began to rise rapidly. At 69.08 s, a bump appeared on the curve, and the pump pressure was 14.53 MPa, which caused the corresponding number of AE rings to surge, further causing fractures. Then the pump pressure rose slowly, which indicated that at this time, at least one main fracture channel had been formed in the sample, and some fracturing fluid was filtered out from this channel. At 98.50 s, the peak point of pump pressure is reached, and the pump pressure is 16.00 MPa, which indicates that the hydraulic fracture has steadily expanded and extended in the coal seam, until the fracture finally extends to the surface of the sample, and the experiment is stopped after the fracturing fluid flows out from the surface of the sample stably.

Figure 6.

Figure 6

Open in a new tab

Pump pressure-AE ringing-time diagram of sample ZJ-8.

It can be seen from Figure 7 that obvious hydraulic fractures can be seen on the surface of sample ZJ-8. On the Z1 plane, the hydraulic fracture mainly starts from the hole wall perpendicular to the horizontal minimum principal (σh) direction and extends along the horizontal maximum principal stress (σH) direction. During the expansion process, some deflections occur, and the extension direction keeps an included angle of 15° with the horizontal maximum principal stress. On the Y1 plane, it is observed that there is a certain angle between the extension direction of hydraulic fracture and the primary fracture, which is about 30°, and the angle between the fracture and the primary fracture is 75–90°. It can be seen from the Y2 plane that two hydraulic fractures (approximately perpendicular to σh) are formed. Under the influence of the horizontal maximum principal stress, they finally intersect and connect with the weak primary bedding plane to form a stable fracture channel, which is also consistent with the phenomenon of three bumps in the pump pressure curve of the ZJ-8 sample. As the maximum principal stress (σV = 5.60 MPa) and the second principal stress (σH = 5.58 Mpa) of the ZJ-8 sample are almost equal, their influences on fracture propagation are similar, so fractures in both directions occur.

Figure 7.

Figure 7

Open in a new tab

Surface fracture morphology of sample ZJ-8 after fracturing (Gray-bedding, red-hydraulic fracture, and blue-initial fracture).

3.1.2. Result Analysis of the ZJ-9 Sample

The surface of sample ZJ-9 has a relatively large number of primary fractures, and its integrity is not as good as that of sample ZJ-8. The in situ stress difference coefficient k = 0.30. At the beginning of the experiment, 3D stress is applied to the sample in the directions of X, Y, and Z until σV = 5.60 MPa, σH = 5.58 Mpa, and σh = 4.29 MPa. After the pressurization system is stable, the pumping system is started, and the flow rate of 45 mL/min is still controlled in constant current mode to inject fracturing fluid. As shown in Figure 8, with the continuous injection of fracturing fluid, the pump pressure rises slowly at the initial stage, and the pump pressure is only 0.24 MPa at 7.53 s. However, after that, the pump pressure increased rapidly, and with the continuous increase of the number of AE rings, the sample was fractured, and the turning point appeared at 46.89 s, and the pump pressure was 2.50 MPa. After that, in the process of 47–283 s, the pump pressure increased monotonously and was similar to the level, without great fluctuation. This shows that at this stage, the injected amount of fracturing fluid and the discharged amount of fracturing fluid along the fracture reach a dynamic balance, and no new fracture needs water pressure energy expansion. However, at 283.62 s, the pump pressure reached the peak value of 2.65 MPa, and there was a big fluctuation, which indicated that under the peak pump pressure, new fractures had been produced and quickly penetrated with the original fractures. On the whole, the pump pressure of the ZJ-9 sample is low, and the AE ringing signal is active, which indicates that the internal fracture is obvious, suggesting that there is a strong interaction between hydraulic fracture and primary fracture. Therefore, the effect of primary fracture on compressive fracture network and reconstruction is worthy of attention.

Figure 8.

Figure 8

Open in a new tab

Pumping pressure-AE-ringing-time diagram of sample ZJ-9.

As shown in Figure 9, it can be seen from the surface of Z1 that the hydraulic fracture starts from around the borehole and forms an angle of about 30° with the horizontal maximum principal stress. The specific reasons need to be deeply analyzed in combination with CT scanning. It can be seen from the Y1 and X2 planes that the hydraulic fracturing fractures are generated and expanded from the lower end of the sample and finally intersect and merge with the primary fractures until they reach the surface. It is found that the hydraulic fractures on both sides are roughly perpendicular to the second principal stress, which indicates that when there are primary fractures, the hydraulic fractures start and expand preferentially in the vertical direction of the primary fractures and the larger principal stress. In the process of expansion, they will be affected by the primary fractures, and then the expansion path will deflect. The final result is that the hydraulic fractures intersect and merge with the primary fractures, which is conducive to the formation of a crisscross spatial seam network.

Figure 9.

Figure 9

Open in a new tab

Surface fracture morphology and sketch of specimen ZJ-9 after hydraulic-fracturing.

3.1.3. Result Analysis of the ZJ-10 Sample

There are relatively few primary fractures on the surface of the ZJ-10 sample, and the integrity is close to that of the ZJ-8 sample. The in situ stress difference coefficient k = 0.35. At the beginning of the experiment, 3D stress is applied to the sample in X, Y, and Z directions until σV = 5.60 MPa, σH = 5.58 MPa, and σh = 4.13 MPa. After the pressurization system is stable, the pumping system is started and the fracturing fluid is injected at a constant flow rate of 45 mL/min. When the fracturing fluid slowly enters the borehole and gradually fills the borehole, the pump pressure quickly rises to 10.08 MPa and reaches its peak within 32.8 s. With the maximum number of AE rings, the pump pressure quickly drops to 8.09 MPa. In the next 120 s, the pump pressure does not change much and continuously drops monotonously, and there is no fluctuation (Figure 10).

Figure 10.

Figure 10

Open in a new tab

Pump pressure-AE ringing-time diagram of sample ZJ-10.

It can be seen from Figure 11 that hydraulic fractures can be observed on the surface of the ZJ-10 sample. On the Z1 plane, the hydraulic fracture starts from around the borehole, spreads along the vertical principal stress, and forms an angle of about 15 with the maximum principal stress. On the X2 plane, two parallel hydraulic fractures are produced and connected with the primary fractures along the bedding plane. Relatively speaking, the ZJ-10 sample has fewer fractures after fracturing, which may be due to its higher in situ stress difference coefficient (k = 0.35) and fewer primary fractures, which leads to a single propagation direction of hydraulic fractures and weak interaction between fractures and primary fractures.

Figure 11.

Figure 11

Open in a new tab

Surface fracture morphology of sample ZJ-10 after hydraulic-fracturing.

3.2. CT Scanning Results and Analysis

CT scanning is an important means to indirectly detect the internal structural features of sample. It is a common and effective means to study internal fractures, holes, and structural changes of nontransparent solid materials (rocks, concrete, and ceramic). Fracturing of coal is only based on the observation and analysis of its surface fracture morphology, but the research on fracture extension and fracture network characteristics in the sample is not sufficient. Therefore, with the help of CT scanning technology, this paper makes a visual and quantitative analysis of the fracture characteristics of coal samples before and after fracturing and provides guidance for exploring the fracture propagation law of coal under hydraulic fracturing and the mutual interference mechanism between hydraulic fractures and primary fractures.

The thermal-hydro-mechanical coupling CT experimental system of coal in Chongqing University is selected for the CT scanning experiment. The CT system consists of a CT workstation, a console, and an image processing system. Among them, there are 24 detector rows, and the maximum X-ray bulb current and maximum X-ray voltage of detector 16 fault are 345 mA and 130 kV. The maximum scanning time is 100 s, the scanning length is 1530 mm, the conventional application pitch is 0.4–2.0, the layer thickness is 0.6–10 mm, the reconstructed visual field is 5–50 cm, and the maximum IRS reconstruction speed is 20 images per second. 300 CT slices can be obtained in a single scan of the sample.

In this paper, the Gaussian filter in AVZIO is used to preprocess the CT scan image, and the CT image with less noise is obtained, as shown in Figure 12. The following CT images in this paper are all preprocessed images.

Figure 12.

Figure 12

Open in a new tab

Original CT section and filtered CT section. (a) Prefiltering. (b) After filtering.

All the samples used in this CT scanning research are Zhijin high rank coal, and the size after processing is 200 mm × 200 mm × 200 mm. The samples were scanned by CT before and after hydraulic fracturing. In the actual research, the density difference of heterogeneous materials is more concerned than the absolute value of density, so the concept of CT value is introduced in the research process eq 2:44

3.2. 2

where μs represents the decay coefficient of the material, and μw represents the decay coefficient of water.

The unit of the CT value is Hu (Hounsfield unit), and the CT value is proportional to the density of the material. It can be distinguished and identified in CT slices by using the principle that the corresponding CT values of the coal matrix, high-density minerals, and fractures in coal are different.

Figure 13 shows the CT slices (150 layers) of the ZJ-8 sample scanned from XY, XZ, and YZ before fracturing. The gray part in the picture is coal matrix, the white highlight part is a high-density mineral, and the gray part is cement mortar. Because the scanned sample is not absolutely homogeneous, the slice density of each scan is also different. Therefore, the black and white degrees of each CT slice are different, and the displayed fracture area is also different. Through CT scanning, the high density is white and the low density is black. So the more black parts, the larger the fracture area.

Figure 13.

Figure 13

Open in a new tab

CT sections of 150 layers with different directions before fracturing of sample ZJ-8.

Figure 14 shows the CT scanning slices of the ZJ-8 sample on the 90th, 120th, 150th, 170th, and 190th layers in XY direction before fracturing. From the scanning images, it can be seen that there are a few high-density minerals around the borehole and in the coal matrix, and there is a through primary fracture between the 90th and 190th layers.

Figure 14.

Figure 14

Open in a new tab

CT scanning section of sample ZJ-8 before fracturing.

Compared with Figure 15, it can be seen from the slice of the 150th floor of the figure that, due to the pressure-out effect, the hydraulic fracture starts from the last fracturing hole at first and then extends around. During the extension process, there are phenomena such as bifurcation and turning of the fracture, which continue to extend and finally intersect with the original fracture and penetrate the surface of the sample.

Figure 15.

Figure 15

Open in a new tab

CT scanning section of sample ZJ-8 after fracturing.

Only two-dimensional (2D) CT scanning slices cannot predict and judge the extension direction and spatial form of hydraulic fractures after hydraulic fracturing. Therefore, it is necessary to complete 3D reconstruction of the scanned 2D CT images in order to better analyze the spatial form of fractures.

3.3. 3D Reconstruction Model of Coal Fractures

Since the 1970s, CT scanning has become more and more mature as a nondestructive experiment technology and has been widely applied and developed all over the world. Because of the advantages of CT scanning technology, such as nondestructiveness and high precision, many researchers have done a lot of research on rock materials such as coal by using CT technology, including the internal structure of coal materials, the method of extracting internal fractures, the information parameters of internal fractures, the law of damage evolution, and the volume fractal dimension. However, most of them are only limited to the information on 2D CT slices, while the research on the spatial distribution of fractures in coal and rock under 3D reconstruction is not much involved. To accurately obtain the propagation and extension law of hydraulic fracture of high rank coal in Zhijin area, Guizhou Province, CT scanning was carried out before and after the fracturing experiment, and the 3D reconstruction and visualization of coal were studied by AVIZO software, so as to further understand the propagation law of hydraulic fracture of high rank coal in Zhijin area, Guizhou Province.

The preprocessed CT scan image is combined with AVIZO software, the fracture is segmented based on the gray level, the threshold value is adjusted, and the fracture information is selected by the Magic Wand function. As shown in Figure 16, the pink line pointed by the arrow is the fracture information.

Figure 16.

Figure 16

Open in a new tab

CT scan sections before and after treatment. (a) Before treatment. (b) After treatment.

The processed images are imported into AVIZO software, and the corresponding 3D model of fractures can be reconstructed by using relevant algorithms. The obtained model can truly reflect the spatial shape and distribution of internal fractures in coal, and it is also in good agreement with the surface fractures of real samples.

By observing the surface fracture characteristics and CT scanning images of Z8-Z10 before and after coal fracturing, and combining with the analysis of fracture three-dimensional reconstruction model, it can be seen that the fracture extension and extension forms after coal fracturing are as follows (Figures 17−19):

Figure 17.

Figure 17

Open in a new tab

3D reconstruction models of Sample ZJ-8 at different angles after fracturing. (a) 3D reconstruction model of different angles before fracturing. (b) 3D reconstruction model of different angles after fracturing.

Figure 19.

Figure 19

Open in a new tab

3D reconstruction models of sample ZJ-10 at different angles after fracturing. (a) 3D reconstruction model of different angles before fracturing. (b) 3D reconstruction model of different angles after fracturing.

Figure 18.

Figure 18

Open in a new tab

3D reconstruction models of Sample ZJ-9 at different angles after fracturing. (a) 3D reconstruction model of different angles before fracturing. (b) 3D reconstruction model of different angles after fracturing.

3.3.1. Complex Fractures (for Example, ZJ-8)

The fracture before fracturing is shown in Figure 17a. In the middle of the borehole, there is a fracture running through the sample with a high right and a low left curve, and there is a parallel fracture above the inside of the sample. After fracturing, as shown in Figure 20, first, the fractures in the upper part of the original sample are compacted and annihilated, and then a main hydraulic fracture perpendicular to the horizontal minimum principal stress (σh) and extending along the direction of the maximum principal stress (σH) is formed at the bottom of the borehole. Because there are few high-density minerals in the ZJ-8 sample, the fracture can turn and branch to the direction perpendicular to the maximum principal stress (σV) without restriction during the propagation process, and finally it will intersect with the primary fracture to form crisscross fracture patterns. This is consistent with the phenomenon that the pump pressure curve of the ZJ-8 sample keeps rising during fracture propagation. The reason for this kind of fracture may be related to its in situ stress and the integrity of the sample. For ZJ-8 sample, the in situ stress difference coefficient is only 0.25, and the sample itself is relatively complete. Strictly speaking, both control the direction of fracture propagation, so the vertical direction of the two directions (both ∥σH) has a certain scale of fractures, which leads to the appearance of complex fractures.

Figure 20.

Figure 20

Open in a new tab

Fracture model diagram of ZJ-8 after hydraulic fracturing.

3.3.2. Plane Fracture + Cross Fracture

As shown in Figure 21, the primary fractures of ZJ-9 run through the X1, X2, and Y2 surfaces, and the fracture surface is relatively flat. There is an X-shaped fracture on the Y1 surface crossing the X1 surface. After fracturing a main hydraulic fracture perpendicular to the horizontal minimum principal stress and extending along the vertical and horizontal maximum principal stress directions formed at the bottom of the borehole, the fracture runs through the X1 and X2 planes. The other single hydraulic fracture starts around the upper part of the borehole and forms an angle of about 30° with the vertical principal stress, which is combined with the original fracture at the upper part of the sample and runs through the Z1 plane. This is consistent with the monotonic increase of surface fracture and pump pressure curve after the second rupture point in Figure 8. It should be noted that although the main fracture is a plane fracture, there are still many secondary hydraulic fractures in this sample, that is, due to the existence of primary fractures, the interaction between the main hydraulic fractures and the primary fractures is caused, resulting in the occurrence of secondary fractures penetrating through the hydraulic fractures and the primary fractures (⊥ σh) locally, thus realizing the ″cross-penetration″ between the hydraulic fractures and the primary fractures, which is still beneficial to the formation of complex volume fractures. Therefore, it is also very important to judge the development of primary fractures in coal seam. Its existence and its relationship with the main control stress will have an important impact on the expansion direction of main fractures and the development degree of intersection fractures.

Figure 21.

Figure 21

Open in a new tab

Fracture model diagram of ZJ-9 after hydraulic fracturing.

3.3.3. Inverted T-Shaped Fracture

It can be seen from Figure 22 that the fracture first started near the bottom row of hydraulic holes in the well, and the fracture was perpendicular to the horizontal maximum principal stress and expanded along the vertical direction of the large principal stress, forming a microarc hydraulic fracture penetrating through the surfaces of X1 and X2. In the middle of the sample, there is an inverted ″T″-shaped fracture with its branch extending along the horizontal maximum principal stress, which is formed near the X2 surface of the first and second rows of fracturing holes. However, as shown in Figure 19, there are high-density minerals in the sample Y1 and Y2 with two parallel faces, which directly affect the expansion and extension of the inverted T-shaped hydraulic fracture.

Figure 22.

Figure 22

Open in a new tab

Fracture model diagram of ZJ-10 after hydraulic fracturing.

3.4. Quantitative Characterization Analysis of Fractures

Since Mandelbrot first proposed the concept of fractal in 1975, a large number of scholars have applied fractal theory to the field of rock. In the field of rock, the most intuitive box counting dimension is usually used to describe the pore structure characteristics and fracture distribution characteristics of rock. Fractal theory can accurately and quantitatively reflect the complexity, roughness, orientation, and opening degree of rock fractures. In this paper, 2D slices are obtained by CT scanning before and after the sample, and 3D fractures are reconstructed by AVIZO software. Then the fractal dimension is calculated by the cube covering method, which can truly reflect the complexity of fractures.

A digital image of a fracture is shown in Figure 23a and consists of a series of pixels, each of which corresponds to a corresponding gray scale, and the space is divided into a cube Figure 23b of pixels. In this paper, the dichotomy in eq 3 is used to construct δk.54

3.4. 3

where k ⊂ N, C is the equal score of the image.

Figure 23.

Figure 23

Open in a new tab

3D fracture image and fractal computing grid. (a) 3D fracture diagram. (b) Box diagram of the 3D covering method.

The number of cubes (Nk) required to cover the fracture space varies with the size of the fracture space. The fracture image is obtained by binarizing the image. In a binary image, the pixel values only include 0 and 1, where 1 represents a fracture. Therefore, as long as the pixels contained in the cube coverage area are greater than 0, it can be counted as the number of boxes (Nk). The whole fracture space was traversed to get the number of cubes covering the whole fracture area after C equal division. For any value, the corresponding value can be calculated. From this, (1/Ck – 1, Nk) a series of data can be obtained. In the double logarithm coordinate, (1/Ck – 1, Nk) a straight line can reflect the fractal characteristics, and its slope is the fractal dimension of fractures. The straight line equation is eq 4(55)

3.4. 4

where D is the fractal dimension.

Based on the fractal calculation method and CT 3D reconstruction method, the fractal dimension of coal before and after hydraulic fracturing can be calculated, and then the fracture volume can be quantitatively calculated by using the Global Analyze module in AVIZO. The fractal dimension and volume of fractures before and after fracturing are shown in Table 2.

Table 2. Fractal Dimension and Volume of Fractures before and after Hydro-Fracturing.

  before fracturing
 after fracturing
specimen no. volume/mm3 fractal dimension/D volume/mm3 fractal dimension/D
ZJ-8 22,440.0 1.848 81,534.0 2.041
ZJ-9 45,225.9 1.928 95,024.2 2.025
ZJ-10 31,422.4 1.868 68,365.5 1.965
Open in a new tab

The fractal dimension of the original fracture before fracturing is 1.848–1.928, and that of the fracture after fracture is 1.965–2.041, both of which are larger than the fractal dimension of the fracture grid of the same sample before fracturing. Among them, the fractal dimension and fracture volume of the ZJ-9 sample are the largest, which indicates that the primary fractures in the ZJ-9 sample are the most developed.

According to the comparative quantitative analysis in Figure 24, the fractal dimension of the ZJ-8 sample before and after fracturing changed the most, with a change rate of 10.43%, and the corresponding volume also changed the most, with a change rate of 263.34%. The complexity of fracture morphology is positively correlated with fractal dimension, and the higher the complexity of fracture, the greater the fractal dimension. Therefore, it is further shown that the ZJ-8 sample has the largest fractal dimension and the most complex fracture morphology after fracturing, which is very consistent with the reconstructed 3D fracture model.

Figure 24.

Figure 24

Open in a new tab

Relationship between fractal dimension and volume of fracture and its change rate.

4. Analysis of Fracture Propagation Form of Horizontal Wells in Coal Seam

There are a lot of primary fractures in coal reservoirs, which will affect the propagation of hydraulic fractures. Generally speaking, when a hydraulic fracture encounters a primary fracture, the hydraulic fracture will eventually evolve into six modes as shown in Figure 25.

Figure 25.

Figure 25

Open in a new tab

Interaction between hydraulic fractures and natural fractures.

Mode 1: The hydraulic fracture extends along the direction of the maximum principal stress before it meets the primary fracture. Mode 2: Once the hydraulic fracture meets the primary fracture, under certain conditions, the hydraulic fracture will turn and merge into the primary fracture, becoming a part of the primary fracture, and the primary fracture will continue to expand instead of the hydraulic fracture. This is because the pump pressure is not large enough at this time, the primary fracture absorbs the energy from fracturing fluid, and the hydraulic fracture does not have enough energy to overcome the normal stress and friction on the wall of the primary fracture. Mode 3: When the hydraulic fracture encounters other primary fractures or weak structural planes when it expands in the primary fracture, it is easy for the hydraulic fracture to diverge, and the original primary fracture expansion is replaced by the optimal orientation expansion.

Mode 4: When the hydraulic fracture meets the primary fracture, the hydraulic fracture does not turn in the initial period, but the diameter of the hydraulic fracture continues to expand through the primary fracture. When the fracturing fluid in the fracture has enough energy and the energy of forward expansion is greater than the energy of the primary fracture opening, the behavior of Mode 5 will occur at this time. Mode 6: When the hydraulic fracture continues to expand, it encounters other primary fractures, etc. At this time, it will form the existence of multiple fractures. After hydraulic fracture passes through the primary fracture, it continues to expand, and at the same time, the primary fracture continues to open and expand. The two processes alternate repeatedly to form a complex fracture network.

The key of hydraulic fracturing of Zhijin coal in Guizhou lies in the complexity and extension distance of fractures. Combined with the fractal dimension and volume changes of fractures before and after fracturing, it shows that the in situ stress difference coefficient controls the complexity of the coal hydraulic fracturing network. Through 3D reconstruction observation, it is considered that a small in situ stress difference may form a complex fracture network in a local area, but a high stress difference can make the extension distance of the main fracture longer, so the stress difference is not as small as possible.

In practical engineering, there are many factors that affect the fracture morphology and fracture network of hydraulic fracturing in coal reservoirs, such as the in situ stress difference coefficient, mechanical properties of coal matrix and bedding, distribution of primary fracture system and microfracture development degree,56,57 and construction parameters such as fracturing fluid viscosity, pumping displacement, and construction time.58 Among them, the geological condition of coal reservoir is the key to hydraulic fracturing, and the high rank coal fractures in Zhijin, Guizhou Province are developed, which is conducive to the formation of complex fractures. As shown in Table 3, this paper only analyzes the influence of in situ stress difference coefficient and primary fracture on fracture morphology after fracturing, and other influencing factors need to be further discussed.

Table 3. Description of Coal Rock Fracture Morphology under Different In Situ Stress Difference Coefficients.

specimen no. the displacement of fracturing fluid (mL/min) in situ stress difference coefficient fracture morphology description
ZJ-8 45 0.25 Crack initiates in direction parallel to σH and σV, extends in direction perpendicular to σh communicates through the primary crack to form complex crack.
ZJ-9 45 0.30 The plane fracture initiates and extends from the bottom of the well in the direction parallel to σV, σH causes the local deflection of the main fracture, and the secondary cross fracture connects the main fracture with the primary fracture.
ZJ-10 45 0.35 The fracture starts from the middle of the well in direction of σH, and connects with the primary fracture to form an inverted T-shaped fracture.
Open in a new tab

5. Conclusions

  • (1)

    Different coal samples have different internal components and fractures before and after fracturing, so they can be visually characterized according to different CT values corresponding to coal matrix, high-density minerals, and fractures. The model obtained by 3D reconstruction based on 2D CT slices of coal and rock can truly reflect the spatial distribution of fractures, and it is also consistent with the surface fractures of samples.

  • (2)

    Most hydraulic fractures start from the bottom of the well along the direction of the maximum horizontal main force, and different fracture forms are formed in the process of expansion: complex fracture, plane fracture + cross fracture, and inverted T-shaped fracture. When the fracture extends, the pump pressure curve and AE signal rise continuously, which is also the obvious characteristic of hydraulic fracturing fracture.

  • (3)

    The change of fracture volume and fractal dimension of coal lies in the complexity of fracture formation and the extension distance of fracture. The smaller the difference coefficient of in situ stress, the more favorable it is to form complex fractures. The high-density minerals in the sample also have an important influence on fracture propagation.

  • (4)

    Based on the interaction between hydraulic fracture and primary fracture of coal, the propagation mechanism of fracture in horizontal well of coal is preliminarily explored. It is considered that the opening of primary fracture, penetration, bifurcation, and turning of hydraulic fracture are the main reasons for the formation of complex fractures, and the existence of a large number of primary fractures is the basis for the formation of complex fractures.

  • (5)

    For the hydraulic fracturing of Zhijin coal reservoir in Guizhou Province, the coal reservoir with developed fractures, a certain number of primary fractures, and few high-density minerals should be optimized, which is very important for efficient exploitation of coalbed methane and improvement of coalbed methane productivity in Zhijin area, Guizhou Province.

Acknowledgments

This study was supported by Guizhou Provincial Geological Exploration Fund (208-9912-JBN-UTS0)and 52000021MGQSE7S7K6PRP and the Graduate Research and Innovation Foundation of Chongqing, China (Grant No.CYB20023) and the Open Fund of the State Key Laboratory of Water Resource Protection and Utilization in Coal Mining Grant number GJNY-18-73.10, and Guizhou Provincial Fund Project (Grant Nos. (2022)ZD001) and by the Fundamental Research Funds for the Central Universities Grant number 2021CDJQY-030.

The authors declare no competing financial interest.

References

  1. Ma H.; Chen S.; Xue D.; Chen Y.; Chen Z. Outlook for the coal industry and new coal production technologies. Adv. Geo-Energy Res. 2021, 5, 119–120. 10.46690/ager.2021.02.01. [DOI] [Google Scholar]
  2. Vinson D. S.; Blair N. E.; Ritter D. J.; Martini A. M.; McIntosh J. C. Carbon mass balance, isotopic tracers of biogenic methane, and the role of acetate in coal beds: Powder River Basin (USA). Chem. Geol. 2019, 530, 119329 10.1016/j.chemgeo.2019.119329. [DOI] [Google Scholar]
  3. Tao S.; Pan Z.; Tang S.; Chen S. Current status and geological conditions for the applicability of CBM drilling technologies in China: A review. Int. J. Coal Geol. 2019, 202, 95–108. 10.1016/j.coal.2018.11.020. [DOI] [Google Scholar]
  4. Cheung K.; Sanei H.; Klassen P.; Mayer B.; Goodarzi F. Produced fluids and shallow groundwater in coalbed methane (CBM) producing regions of Alberta, Canada: Trace element and rare earth element geochemistry. Int. J. Coal Geol. 2009, 77, 338–349. 10.1016/j.coal.2008.07.012. [DOI] [Google Scholar]
  5. Jamieson M.; Elson M.; Carruthers R.; Ordens C. M. The contribution of citizen science in managing and monitoring groundwater systems impacted by coal seam gas production: an example from the Surat Basin in Australia’s Great Artesian Basin. Hydrogeol. J. 2020, 28, 439–459. 10.1007/s10040-019-02050-8. [DOI] [Google Scholar]
  6. Tao S.; Chen S.; Pan Z. Current status, challenges, and policy suggestions for coalbed methane industry development in China: A review. Energy Sci. Eng. 2019, 7, 1059–1074. 10.1002/ese3.358. [DOI] [Google Scholar]
  7. Li Y.; Chen J.; Tang S.; Zhang S.; Xi Z. Biogeochemical Assessment of the Coalbed Methane Source, Migration, and Fate: A Case Study of the Shizhuangnan Block, Southern Qinshui Basin. ACS Omega 2022, 7, 7715–7724. 10.1021/acsomega.1c06496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bi C.-Q.; Zhang J.-Q.; Shan Y.-S.; Hu Z.-F.; Wang F.-G.; Chi H.-P.; Tang Y.; Yuan Y.; Liu Y. R. Geological characteristics and co-exploration and co-production methods of Upper Permian Longtan coal measure gas in Yangmeishu Syncline, Western Guizhou Province, China. China Geol. 2020, 3, 38–51. 10.31035/cg2020020. [DOI] [Google Scholar]
  9. Qin Y.; Moore T. A.; Shen J.; Yang Z.; Shen Y.; Wang G. Resources and geology of coalbed methane in China: a review. Int. Geol. Rev. 2018, 60, 777–812. 10.1080/00206814.2017.1408034. [DOI] [Google Scholar]
  10. Lau H. C.; Li H.; Huang S. Challenges and Opportunities of Coalbed Methane Development in China. Energy Fuels 2017, 31, 4588–4602. 10.1021/acs.energyfuels.7b00656. [DOI] [Google Scholar]
  11. Jianchao H.; Zhiwei W.; Liu P. Current states of coalbed methane and its sustainability perspectives in China. Int. J. Energy Res. 2018, 42, 3454. 10.1002/er.4085. [DOI] [Google Scholar]
  12. Zhao F. P.; Sun Z.; Sang S. X.; Zhou X. Z.; Han Z. Q.; Zhao L. Y.; Wu Z. L.. Progress of coalbed methane exploration and development in Guizhou and the 14th Five-Year Plan development strategy. 2022, 42, 65–75.
  13. Di G.; Yong Q.; Tong-sheng Y. CBM geology and exploring-developing stratagem in Guizhou Province, China. Procedia Earth Planet. Sci. 2009, 1, 882–887. 10.1016/j.proeps.2009.09.137. [DOI] [Google Scholar]
  14. Warpinski N. R.; Lorenz J. C.; Branagan P. T.; Myal F. R.; Gall B. L. Examination of a Cored Hydraulic Fracture in a Deep Gas Well. SPE Prod. Facil. 1993, 8, 150–158. 10.2118/22876-PA. [DOI] [Google Scholar]
  15. Warpinski N. R.; Teufel L. W. Influence of Geologic Discontinuities on Hydraulic Fracture Propagation (includes associated papers 17011 and 17074). J. Petrol. Technol. 1987, 39, 209–220. 10.2118/13224-PA. [DOI] [Google Scholar]
  16. Warpinski N. R. Hydraulic Fracturing in Tight, Fissured Media. J. Petrol. Technol. 1991, 43, 146–209. 10.2118/20154-PA. [DOI] [Google Scholar]
  17. Mahrer K. D. A review and perspective on far-field hydraulic fracture geometry studies. J. Petrol. Sci. Eng. 1999, 24, 13–28. 10.1016/S0920-4105(99)00020-0. [DOI] [Google Scholar]
  18. Beugelsdijk L.; de Pater H.; Sato K., Experimental Hydraulic Fracture Propagation in a Multi-Fractured Medium. 2000.
  19. Blanton T., An Experimental Study of Interaction Between Hydraulically Induced and Pre-Existing Fractures. 1982.
  20. Blanton T. L. In Propagation of Hydraulically and Dynamically Induced Fractures in Naturally Fractured Reservoirs, SPE Unconventional Gas Technology Symposium, 1986.
  21. Fisher M. K.; Wright C. A.; Davidson B. M.; Steinsberger N. P.; Buckler W. S.; Goodwin A.; Fielder E. O. Integrating Fracture Mapping Technologies To Improve Stimulations in the Barnett Shale. SPE Prod. Facil. 2005, 20, 85–93. 10.2118/77441-PA. [DOI] [Google Scholar]
  22. Maxwell S. C.; Waltman C. K.; Warpinski N. R.; Mayerhofer M. J.; Boroumand N. Imaging Seismic Deformation Induced by Hydraulic Fracture Complexity. SPE Reserv. Eval. Eng. 2009, 12, 48–52. 10.2118/102801-PA. [DOI] [Google Scholar]
  23. Fisher M. K.; Heinze J. R.; Harris C. D.; Davidson B. M.; Wright C. A.; Dunn K. P. Optimizing horizontal completions in the Barnett shale with microseismic fracture mapping. J. Pet. Technol. 2005, 57, 41–42. 10.2118/0305-0041-JPT. [DOI] [Google Scholar]
  24. Urbancic T I M. S. C. , Society of Petroleum Engineers SPE/ISRM Rock Mechanics Conference - (2002.10.20–2002.10.23) Proceedings of SPE/ISRM Rock Mechanics Conference - Microseismic Imaging of Fracture Behavior in Naturally Fractured Reservoirs. 2000.
  25. Cheng Y.; Zhang Y. Hydraulic Fracturing Experiment Investigation for the Application of Geothermal Energy Extraction. ACS Omega 2020, 5, 8667–8686. 10.1021/acsomega.0c00172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mayerhofer M.; Lolon E.; Youngblood J.; Heinze J., Integration of Microseismic Fracture Mapping Results With Numerical Fracture Network Production Modeling in the Barnett Shale. 2006.
  27. Mayerhofer M. J. J.; Lolon E. P. P.; Warpinski N. R. R.; Cipolla C. L. L.; Walser D.; Rightmire C. M. M. What Is Stimulated Reservoir Volume?. SPE Prod. Oper. 2010, 25, 89–98. 10.2118/119890-PA. [DOI] [Google Scholar]
  28. Wang H.; Xia B.; Lu Y.; Gong T.; Zhang R. Study on the Propagation Laws of Hydrofractures Meeting a Faulted Structure in the Coal Seam. Energies 2017, 10, 654. 10.3390/en10050654. [DOI] [Google Scholar]
  29. Zhao X.; Liu S.; Sang S.; Pan Z.; Zhao W.; Yang Y.; Hu Q.; Yang Y. Characteristics and generation mechanisms of coal fines in coalbed methane wells in the southern Qinshui Basin, China. J. Nat. Gas Sci. Eng. 2016, 34, 849–863. 10.1016/j.jngse.2016.07.063. [DOI] [Google Scholar]
  30. Zou J.; Chen W.; Yuan J.; Yang D.; Yang J. 3-D numerical simulation of hydraulic fracturing in a CBM reservoir. J. Nat. Gas Sci. Eng. 2017, 37, 386–396. 10.1016/j.jngse.2016.11.004. [DOI] [Google Scholar]
  31. Ai C.; Li X.-X.; Zhang J.; Jia D.; Tan W.-J. Experimental investigation of propagation mechanisms and fracture morphology for coalbed methane reservoirs. Petrol. Sci. 2018, 15, 815–829. 10.1007/s12182-018-0252-z. [DOI] [Google Scholar]
  32. Hou B.; Chen M.; Wang Z.; Yuan J.; Liu M. Hydraulic fracture initiation theory for a horizontal well in a coal seam. Petrol. Sci. 2013, 10, 219–225. 10.1007/s12182-013-0270-9. [DOI] [Google Scholar]
  33. Huang S.; Liu D.; Yao Y.; Gan Q.; Cai Y.; Xu L. Natural fractures initiation and fracture type prediction in coal reservoir under different in-situ stresses during hydraulic fracturing. J. Nat. Gas Sci. Eng. 2017, 43, 69–80. 10.1016/j.jngse.2017.03.022. [DOI] [Google Scholar]
  34. Liang W.; Wu P.; Li Z. In Study on Hydraulic Fracturing of Large-Size Coal Mass Containing Natural Macro-Fracture, 50th U.S. Rock Mechanics/Geomechanics Symposium, 2016.
  35. Chen S.; Tang D.; Tao S.; Liu P.; Mathews J. P. Implications of the in situ stress distribution for coalbed methane zonation and hydraulic fracturing in multiple seams, western Guizhou, China. J. Petrol. Sci. Eng. 2021, 204, 108755 10.1016/j.petrol.2021.108755. [DOI] [Google Scholar]
  36. Guo T.; Zhang S.; Qu Z.; Zhou T.; Xiao Y.; Gao J. Experimental study of hydraulic fracturing for shale by stimulated reservoir volume. Fuel 2014, 128, 373–380. 10.1016/j.fuel.2014.03.029. [DOI] [Google Scholar]
  37. Karpyn Z. T.; Alajmi A.; Radaelli F.; Halleck P. M.; Grader A. S. X-ray CT and hydraulic evidence for a relationship between fracture conductivity and adjacent matrix porosity. Eng. Geol. 2009, 103, 139–145. 10.1016/j.enggeo.2008.06.017. [DOI] [Google Scholar]
  38. He P.; Pan L.; Lu Z.; Zhou J.; Meng C.; Yu H. Experimental Study to Quantify Fracture Propagation in Hydraulic Fracturing Treatment. ACS Omega 2022, 7, 27490–27502. 10.1021/acsomega.2c02517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ju Y.; Liu P.; Chen J.; Yang Y.; Ranjith P. G. CDEM-based analysis of the 3D initiation and propagation of hydrofracturing cracks in heterogeneous glutenites. J. Nat. Gas Sci. Eng. 2016, 35, 614–623. 10.1016/j.jngse.2016.09.011. [DOI] [Google Scholar]
  40. Li N.; Sun W.; Huang B.; Chen D.; Zhang S.; Yan M. Acoustic Emission Source Location Monitoring of Laboratory-Scale Hydraulic Fracturing of Coal Under True Triaxial Stress. Nat. Resour. Res. 2021, 30, 2297–2315. 10.1007/s11053-021-09821-9. [DOI] [Google Scholar]
  41. Zhang R.; Ai T.; Ren L.; Li G. Failure Characterization of Three Typical Coal-Bearing Formation Rocks Using Acoustic Emission Monitoring and X-ray Computed Tomography Techniques. Rock Mech. Rock Eng. 2019, 52, 1945–1958. 10.1007/s00603-018-1677-9. [DOI] [Google Scholar]
  42. Zhang Q.; Li X.; Gao J.; Jia S.; Ye X. Macro- and Meso-Damage Evolution Characteristics of Coal Using Acoustic Emission and Keuence Testing Technique. Nat. Resour. Res. 2022, 31, 517–534. 10.1007/s11053-021-10006-7. [DOI] [Google Scholar]
  43. Hao D.; Tu S.; Zhang C. Experimental Study on the Effect of Moisture Content on Bituminous Coal Porosity Based on 3D Reconstruction of Computerized Tomography. Nat. Resour. Res. 2020, 29, 1657–1673. 10.1007/s11053-019-09528-y. [DOI] [Google Scholar]
  44. Jiang C.; Niu B.; Yin G.; Zhang D.; Yu T.; Wang P. CT-based 3D reconstruction of the geometry and propagation of hydraulic fracturing in shale. J. Petrol. Sci. Eng. 2019, 179, 899–911. 10.1016/j.petrol.2019.04.103. [DOI] [Google Scholar]
  45. Cai J.; Wei W.; Hu X.; Liu R.; Wang J. Fractal characterization of dynamic fracture network extension in porous media. Fractals 2017, 25, 1750023 10.1142/S0218348X17500232. [DOI] [Google Scholar]
  46. Xia Y.; Cai J.; Perfect E.; Wei W.; Zhang Q.; Meng Q. Fractal dimension, lacunarity and succolarity analyses on CT images of reservoir rocks for permeability prediction. J. Hydrol. 2019, 579, 124198 10.1016/j.jhydrol.2019.124198. [DOI] [Google Scholar]
  47. Liu C.; Zhang L.; Li Y.; Fugui L.; Martyushev D.; Yang Y. Effects of microfractures on permeability in carbonate rocks based on digital core technology. Adv. Geo-Energy Res. 2022, 6, 86–90. 10.46690/ager.2022.01.07. [DOI] [Google Scholar]
  48. Liu X.; Kong X.; Nie B.; Song D.; He X.; Wang L. Pore Fractal Dimensions of Bituminous Coal Reservoirs in North China and Their Impact on Gas Adsorption Capacity. Nat. Resour. Res. 2021, 30, 4585–4596. 10.1007/s11053-021-09958-7. [DOI] [Google Scholar]
  49. Li G.; Zhang R.; Xu X. L.; Zhang Y. F. CT image reconstruction of coal rock three-dimensional fractures and body fractal dimension under triaxial compression test. Rock Soil Mech. 2015, 36, 1633–1642. 10.16285/j.rsm.2015.06.015. [DOI] [Google Scholar]
  50. Ma T.; Ma D.; Yang Y. Fractal Characteristics of Coal and Sandstone Failure under Different Unloading Confining Pressure Tests. Adv. Mater. Sci. Eng. 2020, 2020, 2185492 10.1155/2020/2185492. [DOI] [Google Scholar]
  51. Ju Y.; Xi C.; Zhang Y.; Mao L.; Gao F.; Xie H. Laboratory In Situ CT Observation of the Evolution of 3D Fracture Networks in Coal Subjected to Confining Pressures and Axial Compressive Loads: A Novel Approach. Rock Mech. Rock Eng. 2018, 51, 3361–3375. 10.1007/s00603-018-1459-4. [DOI] [Google Scholar]
  52. Zhao P.; Xie L.-Z.; Fan Z.-C.; Deng L.; Liu J. Mutual interference of layer plane and natural fracture in the failure behavior of shale and the mechanism investigation. Petrol. Sci. 2021, 18, 618–640. 10.1007/s12182-020-00510-5. [DOI] [Google Scholar]
  53. Teufel L. W.; Clark J. A. Hydraulic Fracture Propagation in Layered Rock: Experimental Studies of Fracture Containment. Soc. Petrol. Eng. J. 1984, 24, 19–32. 10.2118/9878-PA. [DOI] [Google Scholar]
  54. Mao L.; Lian X.; Hao L. The fractal calculation of 3D cracks based on digital volumetric images and its application. J. China Univ. Min. Technol. 2014, 43, 1134–1139. [Google Scholar]
  55. He Q.; He B.; Li F.; Shi A.; Chen J.; Xie L.; Ning W. Fractal Characterization of Complex Hydraulic Fractures in Oil Shale via Topology. Energies 2021, 14, 1123. 10.3390/en14041123. [DOI] [Google Scholar]
  56. Zhu B. C.; Tang S. H.; Yan Z. F.; Zhang J. Z. Effects of crustal stresses and natural fractures on fracture pressure of coal reservoirs. J. China Coal Soc. 2009, 34, 1199–1202. [Google Scholar]
  57. Warpinski N.; Clark J.; Schmidt R. Laboratory investigation on the effect of in-situ stress on hydraulic fracture containment. SPE J. 1982, 22, 333–340. 10.2118/9834-PA. [DOI] [Google Scholar]
  58. Soliman M.; East L.; Augustine J.. Fracturing Design Aimed at Enhancing Fracture Complexity. 2010, 2.

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES