Study on Hydraulic Fracture Propagation in Mixed Fine-Grained Sedimentary Rocks and Practice of Volumetric Fracturing Stimulation Techniques

Abstract

Yingxiongling shale oil is considered a critical area for future crude oil production in the Qaidam Basin. However, the unique features of the Yingxiongling area, such as extraordinary thickness, hybrid sedimentary, and extensive reformation, are faced with several challenges, including an unclear understanding of the main controlling factors for hydraulic fracturing propagation, difficulties in selecting engineering sweet layers, and difficulties in optimizing the corresponding fracturing schemes, which restrict the effective development of production. This study focuses on mixed fine-grained sedimentary rocks, employing a high-resolution integrated three-dimensional geological-geomechanical model to simulate fracture propagation. By combining laboratory core experiments, a holistic investigation of the controlling factors was conducted, revealing that hydraulic fracture propagation in mixed fine-grained sedimentary rocks is mainly influenced by rock brittleness, natural fractures, stress, varying lithologies, and fracturing parameters. A comprehensive compressibility evaluation standard was established, considering brittleness, stress contrast, and natural fracture density, with weights of 0.3, 0.23, and 0.47. In light of the high brittleness, substantial interlayer stress differences, and localized developing natural microfractures in the Yingxiongling mixed fine-grained sedimentary rock reservoir, this study examined the influence of various construction parameters on the propagation of hydraulic fractures and optimized these parameters accordingly. Based on the practical application in the field, a “three-stage” stimulation strategy was proposed, which involves using high-viscosity fluid in the front to create the main fracture, low-viscosity fluid with sand-laden slugs to create volume fractures, and continuous high-viscosity fluid carried sand to maintain the conductivity of the fracture network. The resulting oil and gas seepage area corresponding to the stimulated reservoir volume (SRV) matched the actual well spacing of 500 m, achieving the effect of full utilization. The understanding of the controlling factors for fracture expansion, the compressibility evaluation standard, and the main process technology developed in this study effectively guide the optimization of transformation programs for mixed fine-grained sedimentary rocks.

1. Introduction

The Yingxiongling sag, located in the core of western Qaidam Basin, contains oil that is mainly concentrated in the upper member of the Lower Ganchaigou Formation, which dates to the Paleogene, and multiple stages of semi-deep lake–deep lake facies shale are developed. It is a unique plateau, thick-mountain-type continental shale oil resource. The current oil and gas exploration results have proven it has a very good resource prospect. The Yingxiongling mixed fine-grained sedimentary rock reservoir is exceptionally thick, with a vertical thickness reaching up to 1200 m. It exhibits high-frequency lithological mixing and variation, developing six principal lithofacies. Previous studies have identified three main sweet spots in the upper, middle, and lower sections, which are controlled by factors such as burial depth, hybrid, and structural conditions. The regional geomechanical characteristics are complex, with lithofacies, stratigraphy, and target bodies influencing the mechanical properties, thereby governing the propagation of hydraulic fractures. The current understanding of the primary controlling factors for hydraulic fracture propagation in the mixed fine-grained sedimentary rocks is still not clear, which limits the optimization of target layer selection and the improvement of transformation strategies [1,2].

With the development of computer technology, numerous scholars have extensively employed numerical simulation methods to study the propagation of hydraulic fractures. Tan Peng et al. established a fracture propagation model for heterogeneous sandstone and conglomerate formations based on the discrete element method, analyzing the influences of gravel content, particle size, distribution, horizontal stress difference, fracturing fluid viscosity, and flow rate on the hydraulic fracturing propagation behavior [3]. Magnus Wangen utilized regular finite element meshes at the reservoir scale, without any special meshing for the fractures. The critical fluid pressure model was applied to derive the exact solutions for the propagation of hydraulic fractures in impermeable and homogeneous rock [4]. Zeng Qingdong et al. employed the extended finite element method to simulate fracture propagation, with the fracture propagation criterion based on the J-integral. By comparing numerical and analytical results, they validated the proposed numerical model and investigated the effects of plastic deformation on the fracture propagation process [5]. Effective formation of volumetric fractures and ensuring the overall stimulated reservoir volume (SRV) is the basis for stable shale oil production. Research on the fracture propagation laws of mixed fine-grained sedimentary rock reservoirs is relatively scarce. Guo Delong and Shen Yinghao et al. utilized rock compression and microscopic fracture monitoring experiments to evaluate the fracture propagation laws of different lithofacies. They believed that laminated calcareous dolomitic shale and laminated dolomitic shale could form complex fractures, with lower compressive strength and more complex fractures formed during rupture. Li Guoxin et al., using core scanning logging combined with X-ray diffraction analysis, considered that the Yingxiongling mixed fine-grained sedimentary rock reservoir has a high content of brittle minerals and good compressibility [1]. Current studies indicate that the propagation laws of hydraulic fractures are mainly controlled by factors such as the brittleness of reservoir rocks, the difference in formation stress, the degree of development of natural fractures, and the types of lithofacies. The reformation of volumetric fractures can be controlled by altering fracturing techniques such as pumping rate, cluster spacing, proppant loading, and fluid pumping rate [6,7,8,9].

Therefore, to address the limited research on fracture propagation laws in the mixed fine-grained sedimentary rock reservoir and the insufficient understanding of the primary controlling factors for hydraulic fracture propagation in different target zones, this paper focuses on the characteristics of “extra-thick, hybrid sedimentary, and intensive tectonization” of the Yingxiongling mixed fine-grained sedimentary rock reservoir. Utilizing a refined 3D geological-geomechanical model as a basis, numerical simulation experiments of hydraulic fracturing in different boxes and target layers were conducted. Combined with laboratory rock mechanics experiments, this research systematically evaluated the controlling effects of rock brittleness, stress differences, the degree of natural fissure development, and lithofacies groups on hydraulic fracture propagation [10]. The impact of construction parameter variations on fracture propagation and the formation of volumetric fractures is clarified. Employing a multi-factor analysis method, the primary controlling factors and their associated weight coefficients affecting compressibility are identified, thereby elucidating the hydraulic fracture propagation laws in the mixed fine-grained sedimentary rock reservoir. An evaluation system for compressibility in the mixed fine-grained sedimentary rock reservoir is established, guiding the selection of the next target zones and optimizing fracturing process parameters.

2. Regional Geological and Geomechanical Characteristics Analysis

The Yingxiongling area is the Paleogene sedimentary center of the western Qaidam Depression. Throughout the E₃² sedimentary period, it presented as a broad and gentle large-scale depression, with significant development of source rocks. The shale oil in the Yingxiongling area has a thick oil window, high concentration density, and a wide planar distribution range. The vertical thickness reaches approximately 1200 m, and the favorable exploration area covers over 800 km². The shale oil system is developed under the background of a deep-water, saline, and hypoxic environment, forming a rich gray-black, laminated, and organic-rich shale oil system [11]. The shale sequence exhibits complex vertical lithologic variations, with multiple high-frequency, interbedded, and spiral-shaped patterns.

The well log data from a typical well in the area are shown in Figure 1. The natural gamma and resistivity data from the typical well suggest that the characteristics of the corresponding lithology are complex hybrid, mixed composition. The total organic carbon (TOC) content is relatively low, ranging from 0.4% to 2.7%, with an average of 1%, and 97% of the samples are less than 2%. The potential for hydrocarbon generation is generally between 1 and 40 mg/g, with an average of 14 mg/g. The shale oil system is developed from both type I and type II kerogen, derived from marine algae and bacteria, providing a material basis for improving productivity. The initial productivity is high, and the shale oil system exhibits a “two-tier” pattern of low-maturity to high-maturity hydrocarbons. Under the influence of arid to semi-arid environmental conditions, the entire lake basin undergoes overall salinization, resulting in high salinity (average 20‰), with the development of alkaline minerals such as gypsum and halite. The primary mineralogical components of the shale in the Yingxiongling area are carbonate, clay, and long-chain organic matter. The shale is currently classified into six lithological categories based on bedding thickness and mineral composition: laminated gray shale, laminated mudstone, laminated shale, laminated mudstone, laminated claystone, and laminated silty shale. The dominant lithological types are laminated and layered gray shale and mudstone, accounting for over 70% of the total [12].

The Yingxiongling mixed fine-grained sedimentary rock reservoir exhibits significant lateral and vertical heterogeneity, with pronounced stress disparities between the upper and middle sweet spots. The stress variations in the horizontal and vertical directions within the Yingxiongling mixed fine-grained sedimentary rock reservoir exhibit very significant differences, yet no fault-induced discontinuous deformation has been observed within the study area. Using the vertical stress distribution of well chai13 as an example, the overlying rock stress increases with depth, varying from 87.80 to 103.61 MPa, and the stress gradient ranges from 2.22 to 2.30. The primary horizontal stress also increases with depth, from 76.02 to 109.82 MPa. The minimum principal stress gradient ranges from 2.01 and 2.04, and the stress gradient at the middle sweet spot is the smallest. The maximum horizontal stress gradient ranges from 2.43–2.56; the value decreases with depth, peaking at the upper sweet spot and declining at the lower sweet spot. Stress differences vary from 17.72 to 18.79 MPa, with the smallest difference observed at the lower sweet spot and the greatest at the middle. The stress difference coefficient decreases with depth, ranging from 0.19 to 0.23, indicating a substantial overall stress variability. The pore pressure distribution spans from 61.77 to 78.42 MPa, with a gradient of 1.66 to 1.74, generally trending upward with depth. The pore pressure gradient is most pronounced at the lower sweet spot, while that of the upper sweet spot is slightly elevated, approaching the gradient observed at the middle sweet spot.

3. Methodology

3.1. Three-Dimensional Modeling

This paper is based on a refined three-dimensional geological-geomechanical model, combined with lab rock mechanics experiments and oilfield experience guidance, to carry out simulation work on the hydraulic fracture propagation of different targets, analyze the control factors of hydraulic fracture propagation, establish an evaluation standard for the compressibility index of mixed fine-grained sedimentary rock reservoir, optimize the fracturing construction parameters, and conduct research on the volume fracture network transformation of on-site platform wells [13].

Firstly, a three-dimensional geological model is constructed based on actual geological conditions, logging data, stratification data, and seismic data. Combining geological modeling and the stress field distribution in the Yingxiongling shale oil research area, a refined three-dimensional geological-geomechanical model is established, as shown in Figure 2. The grid accuracy in the plane reaches 10 m, and the grid accuracy in the vertical direction reaches 1 m, indicating high model accuracy.

Based on a refined three-dimensional geological-geomechanical model, this study conducts hydraulic fracturing simulations at various target locations within the mixed fine-grained sedimentary rock reservoir block to analyze key fracture parameters such as fracture length, width, height, and conductivity. An analysis of reservoir parameters is carried out, examining the impact of factors such as brittleness, stress differences, and natural fractures on the hydraulic fracturing process in the Yingxiongling mixed fine-grained sedimentary rocks. The average daily fluid production rate is utilized as a reference for the complexity of hydraulic fracture propagation [14]. A multi-factor analysis method is employed to calculate the weight of each influencing factor, thereby establishing a comprehensive evaluation system for compressibility. Considering the high brittleness, large stress differences, and developed microfractures in the mixed fine-grained sedimentary rock reservoir, an analysis of construction factors for hydraulic fracture propagation is carried out. The laws of hydraulic fracture propagation under different construction parameters are simulated, and relevant fracturing construction parameters are optimized to guide field operations. Finally, a field application analysis is conducted. A volume transformation simulation study is carried out on the YY2H platform, presenting the current methods of perforation and cluster selection in fracturing, as well as the actual fracturing construction technology. The feasibility and efficiency of releasing the production potential of the current Yingxiongling mixed fine-grained sedimentary rocks through the volume transformation technology are verified through the actual production analysis of platform wells.

3.2. Fracture Propagation Simulation Model Description

The numerical simulation study on hydraulic fracture propagation conducted in this paper primarily employs the complex fracture network extension model (UFM). This model considers the mechanical properties of reservoir rocks, the irregular morphology of fractures, and the interference effects between hydraulic fractures. It precisely predicts fracture distribution, geometric shape, and proppant distribution by utilizing the three-dimensional fracture height equation and the proppant settlement equation, respectively. Consequently, the UFM model can effectively handle the fully coupled problem of fluid flow and fracture elastic deformation within the fracture network when addressing the complex hydraulic fracture propagation issues in shale reservoirs, demonstrating strong applicability. The basic control equations include fluid flow equations, fluid continuity equations, boundary conditions, and fracture interaction criteria [15].

It is assumed that the flow of fracturing fluid in the fracture is laminar flow of incompressible Newtonian fluid in a flat plate, ignoring the influence of gravity, and that it should satisfy Poiseuille’s law (Equations (1) and (2)):

$${\displaystyle \frac{\partial p}{\partial s}}=-{\alpha }_0{\displaystyle \frac{1}{{\overline{w}}^{2n^{\prime }+1}}}{\displaystyle \frac{q}{H_{fl}}}{\left|{\displaystyle \frac{q}{H_{fl}}}\right|}^{n^{\prime }-1}$$

(1)

$${\alpha }_0={\displaystyle \frac{2K^{\prime }}{\varphi (n^{\prime }{)}^{n^{\prime }}}}\cdot {\left({\displaystyle \frac{4n^{\prime }+2}{n^{\prime }}}\right)}^{n^{\prime }};\varphi \left(n^{\prime }\right)={\displaystyle \frac{1}{H_{fl}}}{\int }_{H_{fl}}{\left({\displaystyle \frac{w\left(z\right)}{\overline{w}}}\right)}^{{\displaystyle \frac{2n^{\prime }+1}{n^{\prime }}}}dz$$

(2)

where $p$ is the fluid pressure in MPa; $n^{\prime }$ is the power-law index; and $K^{\prime }$ is the consistency index.

According to the principle of conservation of mass, the mass of fluid flowing into the rock within a certain period is equal to the sum of the increase in internal fluid mass and the mass of fluid flowing out, satisfying the local continuity equation (Equation (3)) and the global mass conservation equation (Equation (4)):

$${\displaystyle \frac{\partial q}{\partial s}}+{\displaystyle \frac{\partial \left(H_{fl}\overline{w}\right)}{\partial t}}+q_L=0;q_L={\displaystyle \frac{2H_L{C}_L}{\sqrt{t-{\tau }_0\left(s\right)}}}$$

(3)

$$\underset{0}{\overset{t}{\int }}Q\left(t\right)dt=\underset{0}{\overset{L\left(t\right)}{\int }}H\overline{w}ds+\underset{H_L}{\int }\underset{0}{\overset{L\left(t\right)}{\int }}\underset{0}{\overset{t}{\int }}{q}_{L}dtdsd{h}_L$$

(4)

where $q$ represents the flow rate within the hydraulic fracture, measured in m³/s; $s$ denotes the distance from any point within the fracture to the fracture tip, m; $q_L$ is the volume of fracturing fluid lost, m³/s; $H_{fl}$ is the height of the hydraulic fracture at the current location, m; $H_L$ is the height of the lost circulation area, m; $\overline{w}$ is the average width at 115 the cross-section of the fracture at position s = s(x, y); $C_L$ is the total loss coefficient; ${\tau }_0\left(s\right)$ is the time, s; $t$ is time, s; $Q\left(t\right)$ is the injection rate at time $t$, m³/s; $H$ is the height of the fracture, m; and $L\left(t\right)$ is the total length of all extended fractures at time $t$, m.

Assuming the rock is a homogeneous linear elastic material and the fractures are vertical, the fracture width is related to the fluid pressure and the minimum principal stress at the respective position. The profile of the fracture width can be determined analytically as follows:

$$w\left(x,y,z\right)=w\left(p\left(x,y\right),H,z\right)$$

(5)

At the fracture tip, the boundary conditions are satisfied as follows:

$$p={\sigma }_n,w=0,q=0$$

(6)

where ${\sigma }_n$ represents the normal stress.

Due to variations in fracture height, the governing equations also incorporate the height growth calculation method described by Kresse et al. [15] and the extension criterion in the fracture height direction (Equations (7) and (8)):

$$K_{Iu}=\sqrt{{\displaystyle \frac{\pi h}{2}}}\left[p_{cp}-{\sigma }_n+{\rho }_{f}g\left(h_{cp}-{\displaystyle \frac{3}{4}}h\right)\right]+\sqrt{{\displaystyle \frac{2}{\pi h}}}\sum _{i=1}^{n-1}({\sigma }_{i+1}-{\sigma }_i)[{\displaystyle \frac{h}{2}}arccos({\displaystyle \frac{h-2h_i}{h}})-\sqrt{h_i\left(h-h_i\right)}]$$

(7)

$$K_{Il}=\sqrt{{\displaystyle \frac{\pi h}{2}}}[p_{cp}-{\sigma }_n+{\rho }_{f}g(h_{cp}-{\displaystyle \frac{1}{4}}h)]+\sqrt{{\displaystyle \frac{2}{\pi h}}}\sum _{i=1}^{n-1}({\sigma }_{i+1}-{\sigma }_i)[{\displaystyle \frac{h}{2}}arccos({\displaystyle \frac{h-2h_i}{h}})-\sqrt{h_i\left(h-h_i\right)}]$$

(8)

where $K_{Iu}$ and $K_{Il}$ represent the stress intensity factors at the top and bottom of the vertical fracture, respectively; $h$ denotes the fracture height, (m); $h_{cp}$ is the height from the perforation location to the bottom of the fracture, m; $p_{cp}$ is the fluid pressure at the depth $h_{cp}$, MPa; ${\sigma }_n$ is the normal stress in megapascals, MPa; ${\rho }_f$ is the fluid density in kilograms per cubic meter, kg/m³; $i$ is the sequence number of the stratigraphic layer from the top of the fracture to the bottom; and $h_i$ is the height from the top of the $i$-th layer to the bottom of the fracture, m.

4. Analysis of Dominant Factors and Compressibility Assessment for Hydraulic Fracture Extension

Given the Yingxiongling mixed fine-grained sedimentary rock reservoir’s distinctive lateral and vertical heterogeneity, with frequent facies mixing, a combined numerical and physical modeling approach was deployed to examine the distinct roles of brittleness, differential stress, natural fractures, and varying lithologies in fracture propagation patterns [16]. By leveraging data from production test wells, the study employed daily average fluid production as a proxy for the complexity of hydraulic fracture growth, delving into the influence of these variables. Utilizing a multi-attribute analysis technique, the relative significance of each factor in influencing hydraulic fracture extension in the Yingxiongling shale oil was quantified, ultimately resulting in the development of a compressibility assessment standard for the reservoir.

4.1. Influence of Brittleness on Hydraulic Fracture Extension

Brittleness is a pivotal factor in the hydraulic fracture propagation mechanism, which can be classified into modulus brittleness and mineral brittleness. Modulus brittleness, determined by Young’s modulus and Poisson’s ratio, reflects a material’s resistance to deformation. Greater modulus brittleness signifies stronger tensile strength. The rock’s modulus brittleness is primarily governed by these parameters, as expressed in Equation (9):

$$B_{YB}={\displaystyle \frac{\left(E_s-E_{min}\right)/\left(E_{max}-E_{min}\right)+\left(v_{max}-v_s\right)/\left(v_{max}-v_{min}\right)}{2}}$$

(9)

where $B_{YB}$ is the dimensionless rock modulus brittleness index, $E_s$ is the static Young’s modulus, GPa. $E_{min}$ is the minimum Young’s modulus within the study area or well log segment, GPa. $E_{max}$ is the maximum Young’s modulus, GPa. $v_s$ is the static Poisson’s ratio, $v_{max}$ is the maximum Poisson’s ratio, and $v_{min}$ is the minimum Poisson’s ratio.

Rock mineral brittleness primarily represents the brittle mineral content in the rock, with the calculation formula shown in Equation (10):

$$B_M=X_{qua}+X_{dol}+X_{cal}+X_{fel}+X_{pyr}$$

(10)

where $B_M$ is the mineral brittleness index. $X_{qua}$ is the quartz mass fraction. $X_{dol}$ is the dolomite mass fraction. $X_{cal}$ is the calcite mass fraction. $X_{fel}$ is the feldspar mass fraction. $X_{pyr}$ is the pyrite mass fraction.

Rock brittleness comprehensively considers modulus brittleness and mineral brittleness, playing a significant role in hydraulic fracture extension and having important implications for compressibility assessments. The rock brittleness index is calculated using Equation (11):

$$B_I=0.6\times B_{YB}+0.4\times B_M$$

(11)

where $B_I$ is the brittleness index. $B_{YB}$ is the rock modulus brittleness index. $B_M$ is the mineral brittleness index.

The Yingxiongling mixed fine-grained sedimentary rock reservoir exhibits high overall brittleness, making it essential to study the impact of brittleness on hydraulic fracture extension. A single-well, one-dimensional vertical profile was constructed, with seven representative locations selected for full three-dimensional hydraulic fracturing simulations from the upper part of the 13-2 layer to the lower part of the 12-1 layer (Figure 3). The results clearly show the inhibition of fracture propagation in the upper part of the 13-2 layer due to lower brittleness index values. The 13-1 and 12-2 layers, along with the lower parts of the profile, display higher brittleness index values, facilitating fracture extension. In contrast, the lower part of the 12-2 layer shows lower brittleness index values, restricting fracture extension upward.

By compiling the brittleness data from test wells in the work area and their corresponding average daily fluid production, the data were processed and subjected to correlation analysis (Figure 4). The fitting results indicate that there is a favorable positive correlation between reservoir brittleness and average daily fluid production.

An integrated analysis of the hydraulic fracture propagation patterns in areas with varying brittleness index magnitudes and the correlation between the brittleness index and average daily fluid production was conducted. It was determined that the rock’s brittleness has a significant impact on the fracture propagation patterns within it [17]. Specifically, in regions with higher overall brittleness, hydraulic fractures propagate more extensively, facilitating the formation of complex fracture networks through fracturing treatments, which is conducive to the production of oil and gas.

4.2. The Influence of Horizontal Stress Differences on Hydraulic Fracture Propagation

Hydraulic fracture propagation can be perceived as a process where stress competition with rock structure unfolds, with stress conditions playing a pivotal role in determining its propagation patterns. During stimulation, the direction of fracture development is governed by the in-situ stress state, with fractures typically extending along the maximum horizontal principal stress orientation. The horizontal stress difference coefficient is defined as the ratio of the difference between the maximum and the minimum horizontal principal stresses to the minimum horizontal principal stress. It effectively characterizes the difference of the two-way stress. The calculation formula is presented as Equation (12):

$$K_h={\displaystyle \frac{{\sigma }_H-{\sigma }_h}{{\sigma }_h}}$$

(12)

where $K_h$ is the dimensionless coefficient of horizontal stress contrast and ${\sigma }_H$ is the maximum horizontal principal stress, MPa. ${\sigma }_h$ is the minimum horizontal principal stress, MPa.

The propagation of hydraulic fractures at different perforation positions with different horizontal stress difference coefficients was observed, respectively. It can be seen from Figure 5 that the stress distribution varies significantly at the positions of the three boxes 4, 5, and 6. The stress difference at the position of the lower 6-1 sublayer is significantly lower, and the overall hydraulic fracture expands significantly downward. At the 4-2 sublayer, there are large stress differences both above and below, resulting in significant obstruction for the downward propagation of the hydraulic fracture below and limited downward expansion of the hydraulic fracture at the perforation position above [18,19]. Overall, the propagation of hydraulic fractures will be hindered in areas with large stress differences.

The study conducted a statistical analysis of the stress difference coefficient for trial oil wells within the operational area and correlated this with the corresponding average daily fluid production. The processed data were subjected to correlation analysis (Figure 6). The results revealed a notable negative correlation between the horizontal stress difference coefficient and the average daily fluid production.

Fracture propagation behavior under different stress difference coefficients and the correlation analysis between stress difference coefficient and average daily fluid production indicates that a smaller coefficient results in less resistance to hydraulic fracture extension, enabling more efficient creation of a volumetric fracture network via stimulation. Conversely, larger stress difference coefficients promote a more uniform fracture pattern with a marked directional bias, hindering the development of intricate fracture networks.

4.3. Impact of Natural Fractures on Hydraulic Fracture Propagation

The impact of natural fractures on hydraulic fracture propagation in the Yingxiongling mixed fine-grained sedimentary rock reservoir is complex, with quantitative analysis posing significant challenges. While the impact of natural fractures on hydraulic fracturing efficiency remains a topic of contention, studies have revealed that shear failure characterizes natural fracture opening, while hydraulic fractures bypass these fractures through tensile failure [20,21,22]. By employing the UFM model, we investigate the effect of distinct natural fractures on fracture development by altering the distance between them and the fracture length.

Varying the natural fracture spacing from 10 m to 60 m allows us to manipulate fracture density within the field, thereby observing the distinct patterns of hydraulic fracture growth under varying natural fracture conditions (Figure 7). The result highlights the significant influence of natural fracture density on hydraulic fracture expansion. As the density decreases, with an increase in fracture spacing, hydraulic fractures predominantly form long, linear fractures. Conversely, when the density is high, numerous but shorter branch fractures are observed. Notably, at a spacing of 20 m, hydraulic fractures exhibit efficient communication with the natural fractures, facilitating the formation of intricate fracture networks. These findings demonstrate that natural fracture density has a substantial impact on hydraulic fracture propagation.

By altering the length of natural fractures in the study area from 20–80 m while maintaining a constant natural fracture density, the propagation behavior of hydraulic fractures under varying lengths of natural fractures was investigated [3]. As depicted in Figure 8, with the increase in the length of natural fractures, the number of branching fractures in the hydraulic fracture propagation tends to decrease, and the stimulated reservoir volume increases, facilitating the formation of a single long, straight fracture. However, when the length of natural fractures is too short, the length of branching fractures is also small. It can be observed that the length of natural fractures has a relatively minor impact on the propagation of hydraulic fractures.

Microfracture closure within rocks is macroscopically observed as an increase in shear and bulk moduli. To systematically assess the impact of natural fractures on hydraulic fracture propagation, this study utilizes the effective medium approach (NIA). An equation for the microfracture development index is derived, serving as a quantitative descriptor of natural fracture development. The explicit formula for this calculation are as follows:

$${\displaystyle \frac{K_O}{K}}=1+{\rho }_c{\displaystyle \frac{h}{1-2{\nu }_0}}\left(1-{\displaystyle \frac{{\nu }_0}{2}}\right)$$

(13)

$${\displaystyle \frac{G_O}{G}}=1+{\rho }_c{\displaystyle \frac{h}{1+{\nu }_0}}\left(1-{\displaystyle \frac{{\nu }_0}{5}}\right)$$

(14)

$$h={\displaystyle \frac{16(1-{{\nu }_0}^2)}{9(1-{\nu }_0/2)}}$$

(15)

where $K_O$ is the maximum bulk modulus of the reservoir, GPa. $K$ is the bulk modulus of the reservoir, GPa. ${\rho }_c$ is the microfracture development index. $G_O$ is the maximum shear modulus of the reservoir, GPa. $G$ is the shear modulus of the reservoir, GPa. ${\nu }_0$ is the Poisson’s ratio.

The microfracture development index of typical wells in the study area was calculated and a correlation analysis was conducted with average daily fluid production. The result is shown in Figure 9, and the analysis indicates a relatively good positive correlation between the microfracture development index and average daily fluid production.

An integrated analysis of the impact of natural fracture density and natural fracture length on the propagation of hydraulic fractures, as well as the correlation between different microfracture development indices and average daily fluid production rates, has yielded the following conclusion: Areas with higher microfracture indices exhibit more extensive hydraulic fracture propagation, which facilitates the formation of complex fracture networks through fracturing stimulation.

4.4. Experimental Evaluation of the Compressibility of Rock Mechanics in Different Lithologies

There is a significant difference in the mechanical properties among different lithologies in the Yingxiongling mixed fine-grained sedimentary rock reservoir development block [23]. By comparing the triaxial compression and microfracture experiment results of homogeneous sandstone, laminated dolomitic shale, and layered argillaceous shale, the ability of rocks with different structures to form fracture networks was evaluated.

As shown in Figure 10, by observing the stress-strain curves and fracture characteristics of sandstone, the results indicate that sandstone has a high uniaxial compressive strength, with the maximum pressure-bearing capacity approaching 200 MPa, and is relatively brittle. A comparative analysis of the fracture characteristics before and after the sandstone experiment reveals that a single fracture is formed after the rock breaks, with a relatively simple fracture morphology. By analyzing the stress-strain curves and fracture characteristics of layered argillaceous shale, the results show that the uniaxial compressive strength of layered argillaceous shale is slightly lower than that of sandstone, and it is also brittle. After fracturing, two intersecting fractures are formed, with a more complex fracture morphology than sandstone. The analysis of the stress-strain curves and fracture characteristics of laminated dolomitic shale reveals that it has the lowest uniaxial compressive strength, only half that of sandstone. From the stress-strain curve, it can be observed that laminated dolomitic shale has a longer fracture closure stage before reaching the elastic stage and still maintains load-bearing capacity after reaching the compressive strength, with the stress fluctuating downward, reflecting obvious plastic characteristics. The fractures are mainly open along the bedding planes, with different fractures intersecting each other, resulting in the most complex fracture morphology and the strongest fracture network formation ability. Each lithology was subjected to three rounds of triaxial compression and micro-fracture experiments, culminating in a total of nine experiments.

The experimental analysis concludes that laminated dolomitic shale has the strongest ability to form complex fracture networks, followed by layered argillaceous shale, while sandstone tends to form a single fracture after breaking.

4.5. Evaluation Criteria for Compressibility Index of Yingxiongling Shale Oil

To quantify the compressibility of the Yingxiongling mixed fine-grained sedimentary rock reservoir, this paper considered three influencing factors: brittleness, horizontal stress difference, and natural fracture density. The compressibility index was established using the brittleness index, stress difference index, and microfracture development index. The weights of each influencing factor were calculated using the grey relational analysis method. The application of this method first involves normalizing and dimensionless processing of each parameter, determining the evaluation criteria for each influencing factor as the reference sequence, and treating each influencing factor as a comparative sequence [24]. The principle is primarily to calculate the distance between the trend of change in each comparative sequence and the trend of change in the reference sequence, thus determining the grey relational degree of each influencing factor. A smaller distance indicates a stronger correlation. The formula for calculating the grey relational degree is:

$${\xi }_i={\displaystyle \frac{min\left|x-x_0\right|+\rho max\left|x-x_0\right|}{\left|x_i-x_0\right|+\rho max\left|x-x_0\right|}}$$

(16)

where ${\xi }_i$ represents the grey relational degree of a specific comparison sequence at a given point about the reference sequence, expressed as a dimensionless quantity. $x$ denotes the value of the comparison sequence at a given point, also a dimensionless quantity. $x_i$ indicates the value of a particular comparison sequence at a given point. $x_0$ stands for the value of the reference sequence at a given point. $\rho$ is the resolution coefficient.

Utilizing the average daily fluid production rate as a reference for the complexity of hydraulic fracturing transformation, the brittleness index, stress difference index, and microfracture development index of each test oil and production well in the study area were statistically analyzed as comparative sequences. Multi-attribute correlation analysis was conducted using the grey relational method, yielding attribute weights of 0.3, 0.23, and 0.47, respectively. Consequently, the formula for calculating the compressibility index is as follows:

$$B_{I1}={\displaystyle \frac{B_I-B_{Imin}}{B_{Imax}-B_{Imin}}}$$

(17)

$$K_{h1}={\displaystyle \frac{K_{hmax}-K_h}{K_{hmax}-K_{hmin}}}$$

(18)

$$K_{a1}={\displaystyle \frac{K_a-K_{amin}}{K_{amax}-K_{amin}}}$$

(19)

$$F_I=0.3B_{I1}+0.23K_{h1}+0.47K_{a1}$$

(20)

where the $F_I$ is compressibility index, nondimensional. $B_{I1}$ is the normalized brittleness index. $K_{h1}$ is the nondimensional normalized horizontal stress difference coefficient. $K_{a1}$ is the nondimensional normalized micro-fracture development index.

The correlation between the compressibility index and the fluid production rate was analyzed (Figure 11). The analysis revealed that the calculated compressibility index exhibits a strong correlation with the fluid production rate, indicating that the compressibility index calculation method is reliable. In addition, there is a significant correlation between the overall density and compressibility index of rocks. A higher overall density of rocks may indicate a lower porosity, which, in turn, can affect the volume change of the rocks under compression. Both factors collectively influence the mechanical behavior of rocks, including aspects such as the stress-strain relationship, elastic modulus, and strength.

Based on the results of the rock compressive strength index and on-site understanding, the classification evaluation standard of the compressibility of the Yingxiongling shale reservoir is established (

Table 1. Evaluation criteria for the compressibility of Yingxiongling shale oil.

Classification of Compressibility	Class I	Class II	Class III
Base Material Brittleness ($B_{I1}$)	≥0.6	0.4~0.6	<0.4
Coefficient of Horizontal Stress Difference ($K_{h1}$)	≥0.7	0.3~0.7	<0.3
Natural Micro-Fracture Development Index ($K_{a1}$)	≥0.25	0.1~0.25	<0.1
Compressibility Index ($F_I$)	≥0.5	0.45~0.5	<0.45

). The evaluation standard mainly includes the brittleness of the shale rock, the coefficient of horizontal stress difference, the natural micro-fracture development index, and the rock compressive strength index. When the compressive strength index is greater than 0.5, it is classified as a Class I compressible region; when the compressive strength index is greater than 0.45 and less than 0.5, it is classified as a Class II compressible region; when the compressive strength index is less than 0.45, it is classified as a Class III compressible region [25,26,27,28,29].

5. Analysis of Fracturing Factors and Parameter Optimization for the Law of Hydraulic Fracture Extension

The law of hydraulic fracture propagation is controlled not only by reservoir factors such as rock brittleness, interlayer stress differences, the degree of development of natural fractures, and different lithological types, but it is also influenced by various parameters of the fracturing operation. The effective formation of volumetric fractures and ensuring the stimulated reservoir volume (SRV) are fundamental to the stable production of shale oil. Given the characteristics of the Yingxiongling hybrid sedimentary shale reservoir, which includes high brittleness, significant interlayer stress differences, and well-developed microfractures, this study simulates the law of hydraulic fracture extension under different conditions of pump rate, sand injection intensity, fluid injection intensity, and cluster spacing [30]. The objective is to identify the optimal construction parameters and to optimize the related fracturing operation plans.

5.1. Pumping Rate

The pumping rate is a crucial parameter in the hydraulic fracturing process. Increasing the pumping rate effectively could reduce the filtration loss of the hydraulic fracturing fluid, enhance the net pressure of the fractures, promote the simultaneous extension of hydraulic fractures in different directions, and increase the stimulated reservoir volume. High pumping rates are beneficial for fracturing the formation, improving the formation permeability, and enhancing the flow capacity of the fractures.

To investigate the influence of the pumping rate on the hydraulic fracture propagation law, this study fixed the proppant loading at 3 m³/min, the clustering space at 10 m, and the fluid pumping rate at 30 m³/min. changing the pumping rate to 6 m³/min, 8 m³/min, 10 m³/min, 12 m³/min, 14 m³/min, and 16 m³/min, respectively. The horizontal well hydraulic fracturing was simulated using the UFM model (Figure 12). The results indicate that the propagation of the hydraulic fractures becomes more complete as the pumping rate increases, with the fracture length, width, and height increasing to varying degrees (

Table 2. Comparison of hydraulic fracture parameters under different pumping rates.

Pumping Rate	Hydraulic Fracture Length	Hydraulic Fracture Width	Hydraulic Fracture Height	Flow Conductivity	SRV
(m3/min)	(m)	(mm)	(m)	(mD·m)	(104 m3)
8	190	11.4	30.2	371	1710
10	206.6	12.6	31.1	596	1928
12	219.1	11.7	31.3	483	2024
14	229.2	12.0	31.3	529.5	2136
16	238.9	12.6	31.1	584.7	2250
18	243.6	12.6	31.5	573.5	2298

). When the pumping rate is less than 16 m³/min, the propagation of the hydraulic fractures changes significantly as the pumping rate increases. However, when the pumping rate exceeds 18 m³/min, the propagation does not change noticeably. The optimal pumping rate is determined to be 16 m³/min, as it provides a more uniform treatment effect and is the most cost-effective and efficient option based on the overall economic and treatment performance considerations.

5.2. Cluster Spacing

Cluster spacing is also an important factor affecting the propagation of hydraulic fractures. Current research suggests that during the fracturing process of shale oil reservoirs, it is appropriate to reduce the cluster spacing, and employing horizontal well multi-stage fracturing technology can effectively enhance the complexity of hydraulic fractures. However, the reduction in cluster spacing will result in an overall increase in the number of clusters, which in turn will increase the volume of fracturing fluid and sand used, leading to higher fracturing costs.

The cluster spacings were set to 2.5 m, 5 m, 10 m, 15 m, and 20 m, respectively, to analyze the propagation patterns of hydraulic fractures under different cluster spacings, using a single zone as an example (Figure 13) [31]. Simulation outcomes revealed that a 5 m cluster spacing limited the effectiveness of individual clusters, with stress shadow effects causing varied fracture lengths, some reaching greater lengths than others, leading to incomplete overall fracture development (

Table 3. Comparison of hydraulic fracturing parameters with different cluster spacing.

Cluster Spacing	Hydraulic Fracture Length	Hydraulic Fracture Width	Hydraulic Fracture Height	Flow Conductivity
(m)	(m)	(mm)	(m)	(mD·m)
5	162.3	9.1	35.6	323.6
10	207.8	12.6	39.1	389.2
15	185.5	11.7	38.6	364.8
20	183.7	10.8	37.9	347.5

). At 10 m, fractures extended adequately laterally but featured fewer branch fractures. A 15 m spacing facilitated more extensive fracture development and branch formation, whereas at 20 m, the close-cutting effect was inadequate, and near-well optimization was suboptimal. Balancing treatment effectiveness and economic considerations, a cluster spacing range of 10–15 m was identified as optimal for efficient shale oil stimulation.

5.3. Proppant Loading

Proppant loading is a crucial factor affecting hydraulic fracture propagation during hydraulic fracturing, and it significantly influences the fracture length, width, and conductivity. Generally, higher proppant loading implies that the proppant in the fracturing fluid has a higher density, which is more likely to effectively support the fracture during the closure process, thereby enhancing the fracture’s conductivity [25,27].

To investigate the impact of proppant loading on hydraulic fracturing propagation, we kept the pumping rate and fluid set and designed proppant loading levels of 2 m³/m, 2.5 m³/m, 3 m³/m, and 3.5 m³/m. We conducted hydraulic fracturing extension simulations using a horizontal well fracture propagation model and discovered that a proppant loading of 3 m³/m resulted in the most complete fracture extension (Figure 14). Simulation results are shown in

Table 4. Comparative analysis of hydraulic fracture parameters at different proppant loading.

Proppant Loading	Hydraulic Fracture Length	Hydraulic Fracture Width	Hydraulic Fracture Height	Flow Conductivity
(m3/m)	(m)	(mm)	(m)	(mD·m)
2	173.3	10.8	31.2	409.7
2.5	176.0	11.0	32.8	395.4
3	226.0	12.1	36.9	355.9
3.5	167.4	10.8	32.5	453.4

. When the proppant loading was less than 3 m³/m, increasing the proppant loading effectively improved the hydraulic fracturing reconstruction efficiency. However, when the proppant loading increased to 3.5 m³/m, the fracture extension was hindered, and the possible reason for this phenomenon is that the proppant loading and its corresponding fluid intensity and pumping rate were not matched, leading to a decline in the reconstruction efficiency. Therefore, we determined that 3 m³/m is the optimal proppant loading.

5.4. Fluid Pumping Rate

The fluid pumping rate, defined as the total volume of liquid injected per unit length, is a critical factor that directly affects the quality of the reformation during hydraulic fracturing operations. An increased fluid pumping rate can effectively promote hydraulic fracture expansion, increase the pressure applied to the rock, and create more complex fractures, thereby enhancing the oil and gas production capacity. The fluid pumping rate also influences the branching and direction of hydraulic fractures. Properly increasing the fluid pumping rate may lead to the extension of fractures in multiple directions, forming more branching fractures, increasing the flow pathways for oil and gas, and ultimately improving the recovery rate [28,29]. However, excessively increasing the fluid pumping rate can significantly increase the cost of hydraulic fracturing, and may not achieve the desired results. Therefore, it is necessary to study the influence of fluid pumping rate on the expansion rules of hydraulic fractures to optimize the fluid pumping rate and achieve the best remodeling effect.

Under a fixed proppant concentration and pumping rate, the fluid pumping rate was designed at 20 m³/m, 25 m³/m, 30 m³/m, 35 m³/m, and 40 m³/m, and a hydraulic fracturing propagation simulation was conducted (Figure 15). Simulation results are shown in

Table 5. Comparison of hydraulic fracture parameters across different fluid pumping rates.

Fluid Pumping Rate	Hydraulic Fracture Length	Hydraulic Fracture Width	Hydraulic Fracture Height	Flow Conductivity
(m3/m)	(m)	(mm)	(m)	(mD·m)
20	203.4	10.5	34.4	474.9
25	214.7	12.5	32.6	569.1
30	229.2	12.0	31.3	529.5
35	244.5	13.2	30.9	537.3
40	248.1	14.0	30.4	643.5

. The analysis of the simulation results showed that increasing the fluid pumping rate could effectively enhance the propagation effect of hydraulic fractures. When the fluid pumping rate increased to 35 m³/m, the hydraulic fracture propagation was optimal, and the overall fracture propagation was complete. Although further increasing the fluid pumping rate could increase the length of some fractures, considering the economic cost, the optimal fluid pumping rate was determined to be 35 m³/m.

6. Volumetric Fracturing Stimulation Techniques

Utilizing a geological-engineering integration platform, a method for selecting zone clusters based on reservoir quality (porosity and oil saturation) and completion quality (minimum stress and compressibility index) along the wellbore is adopted. Segments with similar quality are grouped into the same fracturing zone to reduce property variations within the zone, aiming to achieve uniform stimulation (Figure 16). Furthermore, based on the comprehensive quality of the fracturing zones, differentiated stimulation contents are formulated.

The brittle nature of the Yingxiongling mixed fine-grained sedimentary rock reservoir is relatively high, with the potential to form complex damage patterns. Locally developed fractures, significant natural fractures near the wellbore, and zones with natural fractures can exhibit stress responses during fracturing, leading to fluid loss upon connection with natural fractures. This results in proppant deposition, increasing the risk of sand blocking and casing deformation [26]. In the far field, the extension of artificial fractures encounters small natural fractures, effectively increasing the complexity of the fracture network. The weak plane structure of the lamellar shale is complex, which is conducive to the formation of a three-dimensional fracture network. However, the high-frequency sedimentary cycle changes limit the vertical effective extension of the fractures. The significant horizontal stress difference is not favorable for the formation of volume fractures, and the high stress gradient may restrict fracture width. The large variations in interlayer stress differences constrain the vertical extension of fractures.

Currently, a volume fracturing technology with a core of “close-cut multi-cluster multi-zone + limited flow perforation + high displacement + reverse composite + variable viscosity slick water with high-intensity continuous sand injection” is adopted.

In response to the vertical stacking of the Yingxiongling mixed fine-grained sedimentary rock reservoir, with complex superposition of “layered shale oil” and “lamellar shale oil”, the fracturing scale (artificial fractures within a range of 20–40 m) shows the development of bedding. A comprehensive approach using “gel + slick water” reverse composite modification is adopted, with gel creating long fractures to “suppress near and expand far”, and slick water increasing fracture complexity. Different transformation strategies are adopted for different types of shale reservoirs. For lamellar shale oil, the reverse composite fracturing concept of “gel + slick water” is used, with gel breaking through bedding and slick water increasing complexity. For layered shale oil, which has high brittle minerals and less influence from bedding, the entire process uses slick water to maximize the complexity of modification.

Given the characteristics of the Yingxiongling mixed fine-grained sedimentary rock reservoir, such as “super-thick, high stress, large two-way stress difference, and strong heterogeneity”, the volume modification strategy of “close-cut, high-displacement construction, high-intensity sand injection, ultimately limited flow, and reverse-composite of gel-slick water” is adhered to. The “three-stage” artificial fracture active control approach is maintained. Figure 17 illustrates the fracturing construction curve of the seventh stage of the YY2H15-2 well (Figure 17). The first stage shows active control of near-well zone fracture morphology: The combination of “limited flow perforation and rapid increase to the maximum allowable displacement” strengthens the hydraulic fracturing initiation capability in high-stress and high-carbonate shale oil reservoirs. The combination of “directional perforation, high displacement, and high-viscosity preflush with large zone” enhances the vertical layer-penetrating ability. The second stage is active control of the far-field fracture morphology: After a high-viscosity large zone, the combination of “tail-chasing low-viscosity slick water preflush with large zone and pulsating stepped small particle size proppant with low sand ratio sand plug” polishes the near-well zone hydraulic fractures and expands the fracture volume in the far-well zone [32,33]. The third stage is the control of fracture morphology and connectivity: The combination of “mixed particle size, high-intensity ceramic sand tail-chasing, and fiber suspension sand” improves the vertical uniform support of proppants at the fracture mouth and near-well zone.

During the extraction of shale oil, high-pressure fluids are employed to fracture the rock, creating a complex fracture network that enhances the mobility of oil and gas. This process establishes effective pathways for oil and gas seepage. By controlling the Stimulated Reservoir Volume (SRV), the efficiency of oil and gas seepage can be improved. The actual fracturing pump injection displacement of the YY2H platform is 16 m³, the sand addition intensity is 2.9 m³/m, the liquid application intensity is 30 m³/m, and the cluster spacing is controlled at about 10 m. As shown in Figure 18a, To accurately evaluate the overall fracturing stimulation effect of the YY2H platform, the distribution of the fracture volume and network of the YY2H platform was characterized by using the model, which has a high matching degree with the actual microseismic monitoring data. It can be seen that the stimulated fracture network has good adaptability to the well spacing of 500 m, avoiding the problem of inter-well pressure interference and fully exploiting the target area. As shown in Figure 18b–d, we used the stimulated volume to characterize the seepage area of oil and gas [34,35]. Based on the simulation of the fracture network, the stimulated volumes of the four wells were finely delineated by using the high-precision three-dimensional geomechanics model, and the distribution of key attributes such as permeability and oil saturation within the stimulated area was determined at the same time. Through calculation, the overall SRV of the YY2H platform reached 100 million cubic meters, and the average SRV of a single well reached 25 million cubic meters. The permeability around the fractures will increase after volume stimulation, and the fluidity of oil and gas in the rock will be improved. At the same time, it can be seen that the oil saturation around the stimulated area is relatively high, achieving the effect of precise stimulation.

The understanding of fracture propagation control factors, the standards for compressibility evaluation, and the main technological processes developed in this study have effectively guided the optimization of stimulation schemes and the implementation of the pilot test platform for the shale oil exploration wells at Yinxiongling. Based on the optimization results of construction parameters and the volumetric modification strategy, the fracturing modification was conducted on the YY2H platform. The YY2H platform exhibited continuous oil flow upon initial mining, with an average of 3.5 days to see oil, and the daily oil production on the platform remained stable at over 30 tons. In the first two months of production, the cumulative oil production exceeded 2500 tons (Figure 19). This further confirms the viability and efficiency of employing volume stimulation techniques to unleash the production potential of the Yingxiongling mixed fine-grained sedimentary rock reservoir.

7. Conclusions

(1)

Higher brittleness, smaller horizontal stress difference, and higher natural fracture density lead to more complete hydraulic fracture expansion, facilitating the formation of a complex fracture network through modification. Establish a compressibility evaluation criterion that takes into account brittleness, horizontal stress difference, and natural fracture density. The weightings of these three factors are 0.23, 0.3, and 0.47, respectively, with the compressibility index showing a good correlation with the average daily fluid production.

(2)

By comparing the results of triaxial compression and micro-cracking experiments on homogeneous sandstone, laminated rocks, and thin-bedded rocks from the Yinxiongling shale oil reservoir, an evaluation was conducted on the fracturing network formation capabilities of rocks with varying structures. The findings indicated that the fracturing network formation capabilities for different lithologies were ranked as follows: laminated dolomitic shale > layered argillaceous shale > sandstone.

(3)

Considering the characteristics of the Yingxiongling mixed fine-grained sedimentary rock reservoir, which include high brittleness, large stress difference, and high fracture density, conduct research on the impact of different construction parameters on hydraulic fracture expansion. Comparative analysis reveals that the optimal injection rate in the Yingxiongling area is 16 m³/min, the optimal cluster spacing is 10–15 m, and the optimal sanding intensity is 3 m³/m, with the optimal fluid intensity being 35 m³/m.

(4)

Based on the study of hydraulic fracture expansion rules in the Yingxiongling mixed fine-grained sedimentary rock reservoir, coupled with its reservoir characteristics, a three-stage modification strategy is proposed: pre-inject high viscosity fluid to create the main fracture, inject low viscosity fluid segments to add sand to create volumetric fractures, and continuously carry sand with high viscosity fluid to maintain the conductivity of the fracture network. We conducted a simulation study of this volumetric fracture network modification on four wells at the YY2H platform. The resulting stimulated reservoir volume aligned with the oil-gas seepage area and corresponded to the actual 500 m well spacing, effectively achieving full reservoir exploitation. The YY2H platform’s daily oil production rate remained stable at over 30 tons, and the cumulative production exceeded 2500 tons within the first two months of operation.