Investigation into the Quantitative Assessment of Reserve Mobilization in Horizontal Well Groups Within the Southern Sichuan Shale Gas Reservoir

Abstract

The deep shale gas reservoirs of the southern Sichuan Basin exhibit high temperatures, high pressure, large stress differentials, and complex natural fracture systems. Since 2019, hydraulic fracturing technology in this region has evolved through four stages: exploratory fracturing, intensive limited-volume fracturing, tight spacing with controlled fluid and proppants, and balanced fracturing that combines long-section temporary plugging with short-section intensive cutting. Despite these advances, production remains suboptimal due to inefficient reserve utilization, a lack of quantitative methods for residual gas evaluation, and unclear identification of the remaining reserves. To address these challenges, we developed an integrated workflow combining dynamic production analysis, geomechanical modeling, and numerical simulation to evaluate representative fracturing techniques. Fracture propagation in the well group was modeled in the in-house hydraulic fracture simulator, ZFRAC, to assess fracture geometry, while production history and geological data were used to build calibrated reservoir simulation models. This enabled quantitative assessment of effective fracture parameters, reserve utilization, and residual gas distribution. The results show significant intra-stage heterogeneity driven by stress interference, effective fracture half-lengths of 60–105 m, and a cut-off ratio (proportion of effective fracture half-length to wetted fracture half-length) of 60–93%. Reserve utilization peaked at 60% for intensive limited-volume fracturing, while the efficacy of long-section temporary plugging was limited. These findings offer critical insights for optimizing infill strategies and enhancing sustainable shale gas development in southern Sichuan.

1. Introduction

China holds vast deep shale gas potential with 9.5 × 10¹² m³ in recoverable reserves, positioning it as a key area for future shale gas production growth. Sinopec’s W shale gas field, China’s first commercially developed deep shale gas field, reported 124.678 billion cubic meters of proven reserves in the Wufeng Formation–Longmaxi Member 1 (layers ①–⑥) period in 2018. By 2021, it achieved a first-phase production capacity of 1 billion cubic meters annually and is advancing toward a second-phase target of 2 billion cubic meters per year. Hydraulic fracturing, critical for transforming shale gas reservoirs, has evolved in the W field since 2019 into four main processes, which have been adopted by Sinopec. The four main processes are intensive limited-volume fracturing, tight spacing with controlled fluid and enhanced proppants, and balanced fracturing with long-section extreme temporary plugging and short-section intensive cutting [1] The W deep shale gas reservoir presents significant hydraulic fracturing challenges due to its complex geological and engineering properties, primarily in forming complex fracture networks, maintaining hydraulic fracture support, and managing frequent casing deformation. Development efforts have focused on iteratively addressing these issues through advancing fracturing technology and optimizing key parameters. Initially, before 2019, techniques inspired by foreign methods utilized fewer clusters per stage, medium displacement (13–15 m³/min), low sand intensity (0.78 m³/m), and gel fluid sanding, but these resulted in low horizontal section coverage and poor fracture conductivity [2]. To improve this, intensive limited-volume fracturing was introduced, optimizing fracture spacing based on regional fracability, bedding fractures, natural fracture development, and well patterns. Microseismic monitoring and simulations confirmed enhanced horizontal coverage, yet stable production remained elusive. The utilization of tight spacing with controlled fluid and enhanced proppants followed, increasing displacement, optimizing fluid, boosting sand intensity, and implementing dual temporary plugging at fracture mouths and interiors; this significantly improved parameters but increased casing deformation, hindering efficiency. To address this, a balanced fracturing approach was developed, employing long-section extreme temporary plugging and short-section intensive cutting. This method promotes lateral extension and vertical fracture height growth through denser segmentation, higher single-hole flow rates, and improved temporary plugging and diversion, enhancing overall fracturing effectiveness. Continuous exploration has refined these processes by optimizing cluster number, fracturing displacement, and sand intensity while maintaining or reducing fluid intensity, improving fracturing outcomes. However, the average well EUR remains below expectations due to uneven inter-well recovery, limited vertical fracture height, and insufficient fracture network complexity, leading to low reserve recovery. Due to unclear understanding of reserve development status in production areas, ambiguous distribution of remaining reserves, and the lack of quantitative methods to evaluate the remaining gas, the potential for development adjustments remains undefined. This poses significant challenges to rational production deployment and economically efficient recovery.

A comprehensive survey elucidates that North American shale gas reservoirs predominantly utilize high-density well configurations with minimal inter-well spacing, predisposing them to inter-well interference phenomena [3]. In the Permian Basin, infill well strategies demonstrate that production from parent wells modifies in situ stress orientation and intensity within the pressure propagation envelope, thereby influencing optimal infill well placement [4,5]. In the Eagle Ford Shale, four-dimensional stress field simulations evaluate the pressure dissipation effects of parent wells on child well fracturing, informing strategic decisions on infill timing and spacing based on stress dynamics [6,7]. Geomechanical modeling further scrutinizes inter-well stress interactions in the Marcellus Shale, establishing optimal well spacing through correlations between spacing and fracture overlap ratios [8,9]. Cube well architectures enhance reserve mobilization across and between strata, mitigating stress shadowing and pressure interference during both fracturing and production phases. Nonetheless, investigations into reserve mobilization, residual gas distribution, and quantitative assessments under diverse fracturing conditions remain sparse. In China, the Fuling shale gas field—Sinopec’s pioneering commercially developed marine shale gas reservoir—has delineated residual gas distribution through integrated geological-engineering analyses and dynamic monitoring, addressing heterogeneous gas distribution and complex stress regimes in cube development optimizations [10,11,12]. Further research is imperative to quantify residual gas distribution and recovery in deep shale gas reservoirs characterized by elevated temperatures, pressures, stress differentials, and localized natural fracture networks.

In summation, this investigation synergizes numerical fracturing and production simulations. Leveraging sophisticated geological modeling, a robust numerical simulation framework is constructed. Anchored by an advanced geomechanical model, hydraulic fracture propagation is meticulously simulated for archetypal well group across four distinct fracturing scenarios, elucidating three-dimensional fracture extension in terms of length and height. The dynamic production profiles of gas wells under diverse fracturing methodologies are rigorously assessed to delineate optimal fracture network parameters post-fracturing. Furthermore, the reserve mobilization efficacy of representative well group under these four differentiated processes is quantitatively evaluated.

2. Methodology

This study employs hydraulic fracturing simulations capable of modeling rock deformation, intrafracture fluid flow, and hydraulic fracture propagation. By accounting for the influence of natural fractures on hydraulic fracture growth, activated natural fractures interconnect with hydraulic fractures to form complex fracture networks [13,14]. The flow dynamics within the complex fracture network are simulated using Embedded Discrete Fracture Modeling [15]. The integrated workflow of hydraulic fracturing simulation and production numerical simulation is as shown in Figure 1.

2.1. Hydraulic Fracture Propagation Simulation

The in-house hydraulic fracturing simulator used in the study couples complex rock deformation and fluid flow behaviors, solving a system of nonlinear equations at each time step to determine the length and extension direction of new elements. The Displacement Discontinuity Method (DDM) is adopted for boundary element discretization of fractures. For fracture problems, displacement or stress boundary conditions are typically defined on crack surfaces, along with any remote boundary conditions.

Moreover, the simultaneous solution of the equations considers stresses induced by mechanical interactions with nearby elements. Therefore, due to the normal and shear stress boundary conditions on each element $i$ determined by the fluid flow solution, the normal and shear displacement discontinuities $D_n^j$ and $D_s^j$ at each element $j$ can be determined, respectively, by solving a system of linear equations, as follows:

$$\left\{\begin{array}{c}{\sigma }_n^i=\sum _{j=1}^{N_f}{G}^{ij}{C}_{ns}^{ij}{D}_s^j+\sum _{j=1}^{N_f}{G}^{ij}{C}_{nn}^{ij}{D}_n^j\\ {\sigma }_s^i=\sum _{j=1}^{N_f}{G}^{ij}{C}_{ss}^{ij}{D}_s^j+\sum _{j=1}^{N_f}{G}^{ij}{C}_{sn}^{ij}{D}_n^j\end{array}\right.$$

(1)

In the formula, ${\sigma }_n$ represents the normal stress, in MPa; ${\sigma }_s$ represents the shear stress, in MPa; $N_f$ represents the number of elements, dimensionless; $D_s$ represents the shear displacement discontinuity, in m; Dₙ represents the normal displacement discontinuity, in m; $C_{ns}$, $C_{nn}$, $C_{ss}$, $C_{sn}$ represent the elastic influence coefficient matrix; $G^{ij}$ represents the correction factor considering the finite fracture height [16], and its calculation formula is as follows:

$$G^{ij}=1-{\displaystyle \frac{d_{ij}^{\beta }}{{\left[d_{ij}^2+{\left(\frac{H}{\alpha }\right)}^2\right]}^{\frac{\beta }{2}}}}$$

(2)

In the formula, $d_{ij}$ represents the distance between the $i$th and $j$th micro-elements, in m; H represents the fracture height, in m; $\alpha$ and $\beta$ represent empirical constants.

As the fracturing fluid is injected, the fluid pressure within the fracture, the fracture width, and the stress intensity factor at the fracture tip continuously increase. When the rock deformation at the fracture tip reaches the critical point, the fracture opens and continues to propagate along a specific direction. According to linear elastic fracture mechanics, when the stress intensity factor at the tip of an opening-mode fracture is equal to the rock fracture toughness, the fracture initiates. The fracture propagation direction follows the maximum circumferential stress criterion, satisfying the following equation:

$$\theta =2a\tan \left({\displaystyle \frac{1}{4}}{\displaystyle \frac{K_I}{K_{II}}}\pm {\displaystyle \frac{1}{4}}\sqrt{{\left({\displaystyle \frac{K_I}{K_{II}}}\right)}^2+8}\right)$$

(3)

In the formula, $a$ is the half-length of the micro-element, in m; $K_I$ and $K_{II}$ are functions of the shear and normal displacement discontinuities at the fracture tip, Young’s modulus, and Poisson’s ratio, which can be obtained through Olson’s formula:

$$K_I={\displaystyle \frac{0.806E\sqrt{\pi }}{4(1-v^2)\sqrt{2a}}}{D}_n$$

(4)

$$K_{II}={\displaystyle \frac{0.806E\sqrt{\pi }}{4(1-v^2)\sqrt{2a}}}{D}_s$$

(5)

In the formula, $E$ is the Young’s modulus, in MPa; $v$ is the Poisson’s ratio, dimensionless.

2.2. Embedded Discrete Fracture Model

The Embedded Discrete Fracture Model (EDFM) meticulously delineates hydraulic and natural fractures by representing them as quadrilateral planes integrated within structured grid systems, while preserving structured matrix grids, encompassing both orthogonal and corner-point configurations [17]. This methodology excels in modeling intricate fracture networks, maintaining precise geometric fidelity without necessitating the transformation of matrix grids into unstructured formats [15,18]. The EDFM framework employs a six-step process for grid delineation and conductivity computation, as illustrated in Figure 2.

• Screen out all matrix grids in contact with fracture pieces.
• Generate the intersection surfaces between the fracture plane and matrix grids, which may be triangles, quadrilaterals, pentagons, or hexagons.
• Cut the intersection surfaces according to the fracture shape.
• Calculate the transmissibility of the matrix grid–polygon intersection surface.
• Search for the connections of polygons inside the fracture and calculate the conductivity between polygons.
• Search for connections between fractures: first, find the intersection line between fractures, then find the polygons through which the intersection line passes in the two fractures, respectively, and calculate the transmissibility between polygons sharing the intersection line.

3. Field Application

3.1. Overview

The W gas field, situated in the Western Sichuan Low-Steep Structural Belt, features a gentle structure (amplitude < 300 m) with a “two depressions and one uplift” configuration and minimal fault development (see Figure 3). The production area has a burial depth of 3580–3880 m, with the main body exceeding 3700 m. During the Wufeng Formation–Longmaxi Member 1 sedimentary period, the area was a deep-water shelf environment, fostering organic-rich shale development. High-quality shales (TOC > 2%) are consistently distributed. Layers ① to ④ in the production area average 36.9 m thick, with Class I reservoirs (layers ② to ③¹⁾ averaging 5.76 m, exhibiting stable lateral and planar distribution, though slightly thinner on the eastern and western margins. These layers are classified as Class I and II reservoirs.

3.2. Geological and Engineering Parameters

A 3D geological model of the W gas field extracts a section covering three well pads (A, B, C) with nine wells, including those affected by fracture hits (initiating and receiving wells). The model uses a 20 m × 10 m planar grid and spans layers ① to ⑥, subdivided into 28 sub-layers (①, ②, ③¹, ③², ③³, ④, ⑤, ⑥). Grid counts are 431,200 for pad A, 404,712 for pad B, and 492,156 for pad C. For pad B, mechanical properties (see Figure 4) include an average Young’s modulus of 22.67 GPa, Poisson’s ratio of 0.22, maximum horizontal principal stress of 98.63 MPa, and minimum horizontal principal stress of 85.22 MPa. The formation pressure coefficient ranges between 1.94 and 2.06, with a reservoir temperature of 135 °C. Core data from the Weirong field indicate a Langmuir adsorption pressure of 7.25 MPa and an average Langmuir volume of 2.07 m³/t. Calculations show total reserves for layers ① to ④ in pad A at 2.50 billion m³, with 73% free gas and a reserve abundance of 6.49 × 10⁸ m³/km². Reserve abundances for layers ① to ④ are 6.30 × 10⁸ m³/km² for pad B and 6.08 × 10⁸ m³/km² for pad C.

3.3. Post-Fracture Network Model

Initially, data on the initial sand volume and temporal displacement variations for each fracturing stage are assimilated within the horizontal well fracturing report, seamlessly integrating it into the fracturing simulator’s input file. Leveraging the XGBoost machine learning algorithm, the uncertain parameter permutations generated by the input simulator alongside simulation outcomes are utilized as training samples, coupled with sophisticated sampling techniques for iterative optimization and learning. This streamlined process autonomously refines the fitting error of the pumping pressure, ultimately yielding the uncertainty combination most congruent with the actual pumping pressure and the corresponding intricate hydraulic fracture morphology across the entire well section. The uncertain parameters encompass perforation aperture, the number of effective perforations, wellbore friction coefficient, and casing pressure gradient. Evaluation criteria comprise precise alignment with pumping pressure during stable construction displacement, accurate capture of instantaneous shut-in pressure, and an overall fitting error below 10%.

For the intensified limited-volume fracturing phase, Wells A-1 and A-2 were designated. For the fortification phase, employing tight spacing with meticulously controlled fluid dynamics and augmented proppant application, Wells B-1, B-2, and B-3 were selected. In the advanced fortification phase, a short-section intensive cutting methodology was implemented for Wells C-3 and C-4, while a long-section extreme temporary plugging technique was applied to Wells C-1 and C-2 (refer to

Table 1. Operation parameters for different fracturing techniques.

Different Phase	Frac Technique	Well ID	Lateral Length (m);	Stage Count	Cluster Count	Pumping Rate (m3/min)	Fluid Intensity Per Stage (m3/m)	Proppant Concentration (m3/m)
Improvement	Intensive limited-volume fracturing	A-1	1439.40	20	6–7	12–15	32.33	1.27
		A-2	1469.20	20	5–6	11–15	32.01	1.17
Strengthening	Tight spacing with controlled fluid and enhanced proppants	B-1	1429.63	19	4/8	16.5–18	23.94	2.02
		B-2	1441.44	19	8	16.7–17.5	22.98	2.09
		B-3	1426.78	20	4/8	15–17.5	23.57	2.01
Strengthening Plus	Stage-balanced	C-1	1041	11	11–21	13.5–16.5	26.51	1.36
		C-2	1334.50	14	9–20	13.5–16.0	28.16	1.42
		C-3	1372.40	28	4–7	16.5–20	28.49	1.80
		C-4	1493	35	3–6	14.5–19.6	30.06	1.98

). Leveraging comprehensive structural, natural fracture, and in situ stress models, and accounting for the influence of natural fracture distribution and stress shadowing on fracture morphology, the pumping pressure was meticulously calibrated to align with the actual pumping schedule. Subsequently, post-fracturing fracture network propagation simulations were conducted for nine wells (see Figure 5, Figure 6, Figure 7 and Figure 8). The average fracture height for Well A-1 is 16.02 m, with an average fracture half-length of 111.48 m; for Well A-2, the average fracture height is 16.74 m, with an average fracture half-length of 113.35 m.

Fracture propagation simulations elucidate that stress interference precipitates suboptimal stimulation across all stages, manifesting distinct fracture distribution patterns contingent on the fracturing methodologies employed (refer to Figure 9). Beyond short-cluster configurations, augmenting stage length and cluster density amplifies the near-well fracture network’s area and intricacy. Conversely, the fracture propagation distance (fracture length) orthogonal to the wellbore diminishes on a per-cluster basis, signaling a reduction in stimulated reservoir volume. This attenuation may curtail fracture-controlled reserves, ultimately attenuating the estimated ultimate recovery (EUR). Consequently, long-stage intensive fracturing enhances near-well complexity but compromises far-field extension. In contrast, short-cluster intensive fracturing outperforms long-stage extreme diversion in achieving superior far-field stimulation. Nonetheless, the constrained fluid volume per cluster in the two wells of pad C limited the enhancement of the stimulated volume (see Figure 10).

3.4. Production Performance Analysis

3.4.1. Analysis of Gas Flow in Shale and Complex Fracture Network Characteristics

Figure 11 illustrates the log-log pseudo-pressure buildup curves for Well F-X, an archetypal deep shale gas well in the adjacent Block F. The 1/4 slope segment signifies bilinear flow within secondary fractures and the matrix, succeeded by a 1/2 slope segment indicative of linear flow in the matrix. Notably, the bilinear flow manifests early, underscoring the elevated conductivity of secondary fractures. This initial flow regime is ephemeral and obscured by wellbore storage effects. Figure 12 depicts the log-log pressure buildup curve for Well YC-X, a representative deep shale gas well in the proximal YC Block. The flow dynamics reveal a linear flow segment with a 1/2 slope and a bilinear flow segment with a 1/4 slope. The diminished permeability of secondary fractures markedly extends the duration of linear flow within them. During the shut-in phase, the pressure response predominantly reflects the early flow pattern, culminating in bilinear flow within secondary fractures and the matrix, without transitioning to late linear flow in the matrix. Consequently, fracturing in this region tends to generate simplistic fracture networks, resulting in constrained fracture complexity.

Evaluation of well testing and production efficacy in Block W elucidates that deep shale gas wells subjected to nascent fracturing methodologies were hampered by diminished fracture network intricacy and constrained stimulated reservoir volume (SRV). Subsequent adoption of an innovative balanced fracturing paradigm markedly enhanced near-wellbore fracture complexity. The linear flow regime manifested approximately 10 days earlier (relative to 30–60 days with controlled near-wellbore expansion fracturing), accompanied by elevated initial normalized production rates. The normalized production curve exhibited characteristic slope transitions from −1/2 to −1 (Figure 13). Nevertheless, under the new technique, the normalized production curve shifted downward, with the −1 slope segment, indicating SRV extent, showing a reduction.

3.4.2. Inversion of Effective Fracture Network Parameters Based on Production Dynamic Analysis After Fracturing

Production analysis indicates fracture half-lengths ranging from 60 to 105 m, with conductivities between 4.2 and 11.5 mD-m. Compared to fracture propagation simulations, fracture cut off ratios reach 60–93% (refer to

Table 2. Parameters of effective fracture networks after frac for typical well groups under different fracturing processes.

Well ID	Inner Zone Permeability mD	Periphery Permeability mD	Stimulated Area ha	Frac Half Length m	FC mD-m	Cut-off Ratio %
A-1	3.63 × 10−4	1 × 10−5	23.40	100.73	5.95	90.36
A-2	2.50 × 10−4	1 × 10−5	23.70	101.80	4.15	89.81
B-1	6.50 × 10−4	1 × 10−5	16.59	93.25	9.69	92.14
B-2	4.00 × 10−4	1 × 10−5	18.16	79.36	4.87	89.46
B-3	6.70 × 10−4	1 × 10−5	23.40	105.00	11.54	91.66
C-1	8.41 × 10−4	1 × 10−5	12.49	81.50	10.60	93.27
C-2	6.49 × 10−4	1 × 10−5	10.68	60.14	10.89	93.65
C-3	7.41 × 10−4	1 × 10−5	12.62	66.55	9.55	62.89
C-4	6.11 × 10−4	1 × 10−5	12.54	62.27	5.39	60.46

). The extreme diversion technique in long stages achieved the highest cut off ratio (93%), followed by fluid-controlled high-sand-concentration intensive fracturing (91%) and early-stage intensive limited-volume fracturing (90%), while short-cluster intensive fracturing yielded the lowest ratio (62%).

3.5. Study on Reserve Utilization of Typical Well Pads

Employing the post-fracturing fracture simulation framework, and accounting for high-pressure pressure–volume–temperature (PVT) behavior, layer-specific multiphase permeability, and stress-sensitive properties across strata—in conjunction with actual wellbore trajectories, perforation intervals, and historical production chronicles—a rigorous history matching process is undertaken to calibrate model parameters. This facilitates the projection of well-specific production indices and the evaluation of reserve recovery efficiencies across wells and geological layers.

3.5.1. Production History Matching

The bottom-hole flowing pressure is calibrated under a consistent production rate. Utilizing predefined fracture parameter configurations, production history matching for gas wells is executed. For instance, in pad B, the bottom-hole flowing pressure is precisely aligned with a constant production rate (refer to Figure 14).

3.5.2. Production Forecasting

Utilizing the history-matched model, with sustained production and a bottom-hole flowing pressure threshold of 2 MPa, the recoverable reserves for individual wells are forecast over a 20-year horizon. The prognostic outcomes indicate that the cumulative production of single wells by the end of the 20-year period ranges between 0.50 and 0.85 billion cubic meters(see Figure 15). With the exception of Well B-3, the progression from the initial intensive cutting process through the fluid-controlled sand enhancement technique to the extended-length extreme temporary plugging method reveals a declining trend in estimated ultimate recovery (EUR). However, following the implementation of short-section precision-intensive cutting, the EUR exhibits a notable resurgence.

3.5.3. Analysis of Reserve Utilization Status

Drawing on statistical analysis with a 10% pressure decline, the well group of pad A has sustained production for 56 months. Planar evaluation of reserve exploitation reveals that layers ①–③¹ exhibit the highest degree of reserve utilization, approximating 45%, while layer ③² demonstrates a marginally reduced utilization rate of 37.83%. The mean utilization degree across layers ①–③² is calculated at 39.61% (as depicted in Figure 16). Vertically, the predominant reserve exploitation is concentrated within layers ①–③², whereas layer ④ remains virtually untapped.

Leveraging statistical data reflecting a 10% pressure decline, following two decades of production from the well group of pad A, planar assessment of reserve exploitation reveals that layers ①–③² exhibit the most pronounced utilization degree, surpassing 60%. In contrast, layer ③³ demonstrates a comparatively deficient utilization rate of 25.58%. The mean utilization degree across layers ①–③³ is calculated at 50.73% (as illustrated in Figure 17). Vertically, the primary reserve exploitation is concentrated within layers ①–③², with suboptimal utilization of layer ③³ and negligible exploitation of layer ④.

Utilizing analytical data reflective of a 10% pressure decline, the well group of pad B has sustained production operations for 29 months. Areal assessment of reserve exploitation elucidates that layers ① through ③1 manifest the paramount degree of reserve utilization, exceeding 30%, whereas layer ③² registers a utilization rate of 31.73% (as illustrated in Figure 18). The mean utilization degree across layers ① through ③² is computed at 34.93%. Vertically, the principal reserve exploitation is localized within layers ① through ③², with layers ③³ and ④ exhibiting negligible exploitation.

After 20 years of production with a 10% pressure drop, planar analysis of pad B’s well group shows layers ①–③² with the highest reserve utilization (>60%), while layer ③³ is at 37.3%, averaging 55.08% across layers ①–③³ (Figure 19). Vertically, utilization concentrates in layers ①–③², with minimal use in layer ③³ and none in layer ④.

Using data from a 10% pressure drop after 19 months of production, planar analysis shows layers ①–③1 have the highest reserve utilization, over 30%, while layer ③² has slightly lower utilization at 28.8% (Figure 20). The average utilization for layers ①–③² is 32.8%. Vertically, utilization is concentrated in layers ①–③², with minimal utilization in layers ③³ and ④.

Based on a 10% pressure drop statistic, after 20 years of production from the well group of pad C, planar analysis of reserve utilization shows that the utilization degree of layers ①–③² is the highest, exceeding 50%, while that of layer ③³ is relatively poor at 29.19%. The average utilization degree of layers ①–③³ is 43.32% (as shown in Figure 21). Vertically, the main utilization focuses on layers ①–③², with poor utilization of layer ③³ and basically no utilization of layer ④.

3.5.4. Comparison of Reserve Utilization Degrees Under Different Fracturing Processes

Reserve utilization varies across fracturing techniques. Disregarding complexities such as well spacing and stage omissions, the early intensive cutting process achieves the highest utilization degree for layers ①–③³ at 60%, whereas the long-section extreme temporary plugging process yields suboptimal results (see Figure 22). Recovery degrees within the utilized reserves differ by process: the short-section fine intensive cutting technique exhibits a narrower utilization scope yet boasts a superior recovery rate within that scope.

4. Conclusions

(1)

We have developed an integrated workflow combining hydraulic fracturing simulation, production performance analysis, and reservoir numerical simulation. First, hydraulic fracture propagation was modeled for typical well groups under different fracturing techniques, with initial fracture geometry determined through AI-based matching of actual pumping pressure data. Second, production performance analysis combined with historical production matching comprehensively characterized post-fracturing effective fracture network parameters. Finally, this approach quantified reserve recovery status in typical pad wells across three fracturing techniques and revealed distribution patterns of remaining gas. This integrated workflow can be extended to the stack shale gas block in the southern Sichuan Basin for conducting cube development reserve evaluation and residual gas distribution research.

(2)

Fracture propagation modeling results for deep shale gas reservoirs in the southern Sichuan Basin indicate that, due to the influence of inter-fracture stress shadowing, the extent of intra-stage stimulation remains suboptimal. Additionally, the intra-stage fracture network morphologies exhibit significant variation across diverse hydraulic fracturing methodologies.

(3)

Based on the production dynamic analysis, the fracture half-length derived from post-fracturing parameter inversion spans from 60 to 105 m, with fracture conductivity ranging between 4.2 and 11.5 mD-m. In contrast to the outcomes of fracture propagation simulations, the fracture cut off ratio is 60% to 93.65%.

(4)

A comparative analysis of reserve utilization across various fracturing methodologies reveals distinct disparities in exploitation efficacy. Disregarding intricate variables such as inter-well distances and incomplete stages, the exploitation rate for strata ① through ③³ peaks at 60% under the initial intensive segmentation technique. Conversely, the reserve exploitation efficiency under the extended-interval extreme diversion method exhibits suboptimal performance.