The Discovery of Fracture Tip-Driven Stress Concentration: A Key Contributor to Casing Deformation in Horizontal Wells

Abstract

Casing deformation (CD) is generally believed to be caused by the slip of fractures in the strata and its shear effects on horizontal wells. However, the casing deformation mode based on this theory cannot fully match the measurement data, and the differential deformation characteristics and the mechanism behind this phenomenon are not completely clear. To elucidate the mechanisms of CD and enhance prevention and control measures, the CD modes in Shunan Block were identified and deformation mechanisms of these modes were comprehensively investigated. Our research shows the following: (1) Under the mechanism of penetrating fracture shear deformation, CD exhibit obvious shear deformation, and the natural fractures near the intersection point with the wellbore are prone to form a higher risk of deformation. (2) Natural fractures with tips approaching the wellbore experience intense stress concentration (1.6 times higher than shear stress) during activation, resulting in compression and asymmetrical CD. (3) The shear deformation induced by penetrating fractures is 15.52 mm, while the fracture tip-induced compression deformation demonstrates a substantially greater magnitude at 44.17 mm. This compressive deformation exceeds the shear deformation by a factor of approximately 2.85. (4) The stress concentration at the fracture tip is highly sensitive to the injection rate. Hence, adherence to the “avoiding stress concentration” principle is crucial in hydraulic fracturing operations. The conclusion indicates that in addition to penetrating fracture shear deformation, fracture tip compression deformation is another significant mechanism that causes CD. This research finding can offer theoretical guidance for developing effective measures to prevent and control CD in the exploitation of deep shale gas.

1. Introduction

The development of deep shale gas is vital for the future of natural gas exploration and development [1,2]. The staged fracturing technology of a horizontal well is crucial for its effective extraction. With advancements in horizontal well staged transformation technology, there has been a significant increase in deep shale gas production in the Shunan block [3,4]. However, the reduction in cluster spacing and the enlargement of the fracturing scale have resulted in frequent casing deformation (CD) issues during fracturing operations, which is hindering the development of large-scale deep shale gas production [5]. The CD incidents that have occurred in the deep shale gas wells in the Shunan Block have reached 51.0%, with an overall casing deformation rate of 77.3% in well SN-H areas, which undoubtedly severely affects the efficient development of horizontal wells in deep shale gas reservoirs [6,7,8].

Research has been conducted on the CD of deep shale gas horizontal wells in the Sichuan Basin [9,10,11]. Research has found that the continuous injection of high displacement fracturing/drilling fluid leads to the activation of natural fracture systems, which are key factors leading to the deformation of casing and hindering normal fracturing operations [12,13,14,15]. Therefore, addressing CD induced by fracture activation has become a focus of studies. Xi et al. [16] believe that the asymmetric fractures and uneven stress generated during the fracturing process lead to the sliding of natural cracks, potentially causing shear deformation in casing. Yin et al. [17] and Zhang et al. [18] used underground lead models and combined them with three-dimensional seismic, microseismic, and logging data to determine that the deformation of casing in the Sichuan Basin is mainly caused by shear deformation caused by natural fractures sliding [19,20,21]. Xi et al. [22] and Chen et al. [23] discovered that the Changning-Weiyuan block develops high-angle natural fractures, with the majority of them being in a critical stress state. Elevated operating pressures may result in fracture activation, which could potentially lead to casing shear deformation. Lu et al. [24] constructed a numerical model to study the deformation of casing resulting from natural fracture slip shear failure. The model investigated the correlation between fracture slip and casing shear failure and explained the direct relationship between fracture slip and underground lead models [25,26]. The combination of the above research contributes to a more comprehensive understanding of the complex dynamics of fracture slip and its impact on casing mechanics.

Current studies on the deformation of casing in horizontal wells mainly focus on the shear deformation of casing caused by natural fractures and fault systems slipping [27,28,29]. Nevertheless, field data show a new form of compression deformation in deep shale gas horizontal well casings [30,31,32]. The mechanism of “natural fracture slip”—“shear casing deformation”—currently under extensive scrutiny, fails to account for the frequent incidence of compression casing deformation at horizontal well fracturing sites [9]. The insufficient differentiation of CD in deep shale gas horizontal wells, coupled with delayed formation mechanism research, has resulted in ineffective measures being implemented at current horizontal well fracturing sites to tackle CD issues. The issue of CD persists as a significant concern in the development of deep shale gas.

Hence, in this study, a response map depicting the natural fractures–microseismic events–casing deformation was constructed by integrating fracture development and distribution characteristics with microseismic, logging, and in situ stress data, and the differential characteristics of CD in different regions and wellbore positions were determined. Additionally, focusing on the induced stress field and the slip of fracture systems, the interaction mechanism between hydraulic fractures and natural fractures was investigated. The formation mechanism of differential CD in deep shale gas horizontal wells was determined, and the risk of CD induced by the activation of natural fractures was reassessed. The research finding can offer theoretical guidance for developing effective measures to prevent and control casing deformation in the exploitation of deep shale gas.

2. Analytical and Numerical Model

2.1. Characteristics of Casing Deformation

2.1.1. Casing Deformation in Deep Shale Gas Horizontal Wells

The Shunan block is located on the southern wing of the Weiyuan slope, which is a transitional zone from the Weiyuan single slope to the Luzhou low steep anticline. The Shunan block belongs to the deep-water shelf sedimentary facies, with a continuous thickness of 7–8.5 m for Class I reservoirs. In terms of regional structure, the study area is located in the southwest Sichuan low-fold tectonic belt, which is subjected to the multi-directional compression and superposition of the Longmen Mountain, Jiangnan Xuefeng Mountain, Dalou Mountain, and Daba Mountain orogenies. The area exhibits a complex fault system with a network of developed fractures (Figure 1) [5].

The interpretation results of multi-arm caliper imaging logging show that more than 50 wellbore CD events have occurred since the fracturing development process of the SN-H21 well platform. According to the severity of the deformation of the inner diameter of the casing, the CD in the study area is categorized into three levels: Class A (inner diameter > 85 mm), Class B (54 mm < inner diameter < 85 mm), and Class C (inner diameter < 54 mm). The statistical analysis reveals that over 50 instances of CD were observed at the SN-H21 well platform, predominantly classified as Class A deformations, with a few cases categorized as Class B deformations. Furthermore, severe Class C deformation events were identified at SN-H21-1, SN-H21-3, and SN-H21-4 (Figure 2).

2.1.2. Differential Characteristics of Casing Deformation

The superposition map of microseismic events–natural fracture–CD events show that the three typical CDs occurred in SN-H21-1. Among these, a significant Class C CD was identified at the 17th fracturing interval (4754 m). The superposition map revealed the presence of a natural fracture close to the Class C CD event, with the fracture tip approaching the CD point without directly intersecting or crossing the wellbore (Figure 3). Furthermore, at positions 1–2 and 1–3 of the CD, a natural fracture was observed intersecting and crossing through the wellbore. The microseismic data indicate that these three natural fractures are all activated by hydraulic fractures. The deformation of the casing in SN-H21-1 is primarily attributed to the activation of natural fractures due to hydraulic fracturing. However, the mechanism of action of fractures differs. The deformation of the Class C at position 1–1 is primarily induced by natural fractures with tips approaching the wellbore, while the CDs at positions 1–2 and 1–3 are associated with a natural fracture intersecting the wellbore at an angle and traversing through it.

The CD of SN-H21-2 to SN-H21-2-4 is similar to that of SN-H21-1. The series of CDs of Class A observed are primarily induced the activation of natural fractures that intersect the wellbore obliquely and cause shearing, while the significant CDs of Class B and Class C are mainly induced by the natural fractures with tips approaching the wellbore, which initiate and propagate, resulting in stress concentration (Figure 3).

Furthermore, statistical results show that most of the CD events in SN-H21 well platform are located at locations where natural fractures are developed, with the incidents related to natural fractures accounting for as high as 94% (Figure 4a). This indicates that the CDs in SN-H21 are mainly caused by the activation of natural fractures, which is consistent with conclusions drawn from the fracturing site. However, prior research often fails to consider the differences in the role played by different natural fractures in inducing CD, particularly those near the wellbore but without direct intersection. These natural fractures significantly contribute to CD during fracturing procedures [33].

Strong energy event points and significant CDs predominantly occur at the fracture tip. Specifically, all three Class C and 50% of Class B CDs observed at the SN-H21 well platform that occurred at the tip of the fracture approached the wellbore, but did not intersect obliquely with the wellbore (Figure 4b). Preliminary speculation suggests that the initiation and propagation of natural fractures during hydraulic fracturing result in strong stress concentration at the fracture tip, leading to the compression of the wellbore and significant CDs.

The deformation of the casing of the SN-H21 well platform is primarily attributed to the activation of natural fractures due to hydraulic fracturing. Class A casing deformations, which are not severe, are mainly induced by natural fractures intersecting obliquely with the wellbore and causing shearing. Conversely, severe deformations in Class B and particularly in Class C, which halt the further construction of the wellbore, are predominantly induced by stress concentration resulting from the natural fracture tip approaching the wellbore.

2.2. Finite Element Analysis Model

2.2.1. Stress Calculation Model

(1)

Fluid dynamics equations

During the hydraulic fracturing process in reservoirs, the expansion of hydraulic fractures interacts with natural fractures, activating them and causing both slip and further propagation [34,35] (Figure 5). During the activation process of natural fractures, it is assumed that the tangential fluid parallel to the fracture surface is an incompressible Newtonian fluid, and its flow law follows the Newtonian fluid pressure transfer formula, which satisfies the lubrication theory and the continuity equation [36]:

$$q={\displaystyle \frac{d^3}{12\mu }}\nabla p_i$$

(1)

$${\displaystyle \frac{\partial d}{\partial t}}+\nabla \cdot (-{\displaystyle \frac{d^3}{12\mu }}\nabla p_i)+(q_t+q_b)=Q_{inj}$$

(2)

where μ is the dynamic viscosity of fracturing water, Pa·s; d is the crack opening, m; q is the tangential volume flow of fracturing water, m³/s; $\nabla p_f$ is the fluid pressure gradient along the cohesive elements, Pa/m; and q_t and q_d are the flow rates of fracturing water flowing into and out of the upper and lower surface of the cohesive elements, m³/s, which, respectively, reflect the leakoff from the fracture surface to the adjacent layer. For non-permeable rock formations, q_t = q_d = 0.

The normal flow of cohesive elements can be calculated as follows:

$$\left\{\begin{array}{l}{q}_t=c_t(p_i-p_t)\\ q_b=c_b(p_i-p_b)\end{array}\right.$$

(3)

where P_t and P_b are the pore pressure in the adjacent porous elastic material at the top and bottom of the cohesive elements, respectively, Pa; c_t and c_b are the corresponding leakoff coefficients, m³/(Pa·s); and P_i is the fluid pressure acting on the fracture surface, Pa.

Substituting Equations (1) and (3) into Equation (2) results in the Reynolds lubrication equation

$${\displaystyle \frac{\partial d}{\partial t}}+c_t(p_i-p_t)+c_b(p_i-p_b)={\displaystyle \frac{1}{12\mu }}\nabla \cdot (d^3\nabla p_f)+Q_{inj}$$

(4)

(2)

Stress field of natural fracture

According to the theory of fracture mechanics, after the activation of natural fractures by hydraulic fracturing, the change in the stress field of the natural fractures can be expressed as [37]

$${\sigma }_x={\sigma }_{\min }-{\displaystyle \frac{K_I}{\sqrt{2\pi r}}}\cos {\displaystyle \frac{\theta }{2}}(1+\sin {\displaystyle \frac{\theta }{2}}-\sin {\displaystyle \frac{3\theta }{2}})$$

(5)

$${\sigma }_y={\sigma }_y^{’}-{\displaystyle \frac{K_I}{\sqrt{2\pi r}}}\cos {\displaystyle \frac{\theta }{2}}(1-\sin {\displaystyle \frac{\theta }{2}}-\sin {\displaystyle \frac{3\theta }{2}})$$

(6)

$${\tau }_{xy}=-{\displaystyle \frac{K_I}{\sqrt{2\pi r}}}\sin {\displaystyle \frac{\theta }{2}}\cos {\displaystyle \frac{3\theta }{2}}\cos {\displaystyle \frac{\theta }{2}}$$

(7)

$${\sigma }_y^{’}={\sigma }_{\max }{\cos }^2\alpha +{\sigma }_v{\sin }^2\alpha$$

(8)

$$K_I=△p\sqrt{\pi a}$$

(9)

By combining the above equations, the normal stress and shear stress that hydraulic fractures experience when encountering natural fractures can be obtained [38,39]:

$$\tau =\left({\sigma }_x-{\sigma }_y\right)\cos \left({\displaystyle \frac{\pi }{2}}-\beta \right)\sin \left({\displaystyle \frac{\pi }{2}}-\beta \right)+{\tau }_{xy}({\cos }^2({\displaystyle \frac{\pi }{2}}-\beta )-{\sin }^2({\displaystyle \frac{\pi }{2}}-\beta ))$$

(10)

$$\sigma ={\sigma }_x{\sin }^2\beta +{\sigma }_y{\cos }^2\beta +2{\tau }_{xy}{\sin }^2\beta -\cos \beta$$

(11)

where σ_x and σ_y represent the normal stresses along the x-axis and y-axis, MPa; τ_xy is the shear stress, MPa; σ_max and σ_min are the maximum and minimum horizontal principal stresses, respectively, MPa; K_I is the stress intensity factor; a is the half-length of the natural fracture, m; Δp is the net pressure within the fracture, MPa; r represents the distance from the midpoint of the fracture to its tip, m; θ indicates the polar coordinate of a specific location relative to the fracture tip, °; and β is the angle formed by the crack and the wellbore, °.

(3)

Casing stress

As the cement slurry hardens, the casing and the formation create a unified elastic entity. The stress distribution within the casing is influenced by various factors, including in situ stress, internal casing pressure, and stress induced by hydraulic fracturing, and can be expressed as follows [40,41]:

$$\left\{\begin{array}{l}{\sigma }_r={\displaystyle \frac{r_1^2{r}_2^2(q_1-p_i)}{r_2^2-r_1^2}}{\displaystyle \frac{1}{r^2}}+{\displaystyle \frac{r_1^2{p}_i-r_2^2{q}_1}{r_2^2-r_1^2}}\\ {\sigma }_0={\displaystyle \frac{r_1^2{r}_2^2(q_1-p_i)}{r_2^2-r_1^2}}{\displaystyle \frac{1}{r^2}}+{\displaystyle \frac{r_1^2{p}_i-r_2^2{q}_1}{r_2^2-r_1^2}}\\ {\sigma }_i={\displaystyle \frac{r_1^2{r}_2^2(q_1-p_i)}{r_2^2-r_1^2}}{\displaystyle \frac{1}{r^2}}+{\displaystyle \frac{r_1^2{p}_i-r_2^2{q}_1}{r_2^2-r_1^2}}\end{array}\right.$$

(12)

$$q_1={\displaystyle \frac{f_4{f}_8{p}_0+(f_6+f_7)f_1{p}_i}{(f_2+f_3)(f_6+f_7)-f_4{f}_5}}$$

(13)

where p_o is the synthetic vector sum of natural cracks and stress fields, MPa; P_i is the internal pressure of the casing wall, MPa; r₁ and r₂ are the inner and outer radii of the casing, mm; r₃ is the inner radius of CS, mm; r₄ is the inner radius of the surrounding rock in millimeters; f₁~f₈ are coefficients; σ_r is radial stress; σ_θ is tangential stress; and σ_t is axial stress, MPa.

2.2.2. The CD Model of Strata-Fracture-Casing

(1)

Model structure and boundary conditions

We assume that the casing is axially oriented perpendicular to the slip surface and that both the casing and the cement ring are concentric cylindrical structures. For the convenience of grid division, the formation is considered a hollow cylindrical structure, thus forming the casing–cement ring formation system. To eliminate the boundary effect, the diameter and height of the cylinder was set much larger than the diameter of the casing. Without considering the initial process of slip surfaces caused by the failure of the rock layer and cement sheath, a contact relationship is established between the slip surfaces, which are supported and lubricated by the fracturing fluid, where the mechanical interaction is relatively weak.

To simulate the adhesion and friction between the casing and the cement sheath, as well as between the cement sheath and the formation, a contact relationship is established. The normal contact is rigid, while the tangential contact follows Coulomb friction, with the friction coefficient set at 0.7. To ensure both calculation accuracy and efficiency, a grid with a dense-to-sparse transition along the direction from the shear center to the outer boundary is set up, utilizing the three-dimensional eight-node reduced integration element C3D8R, with a total of approximately 160,000 elements (Figure 6).

(2)

Material constitutive

Due to the significant deformation of the casing during the shearing process, the wellbore adopts an elastic–plastic constitutive model, which follows the linear isotropic hardening rule (a plastic modulus of 5 GPa) after plastic deformation. Some areas of the casing cement and shale (stratum) also undergo significant deformation, and the Drucker–Prager plastic constitutive model (D-P plastic model) is used. The D-P constitutive theory is commonly employed to describe the pressure-dependent yield behavior of geotechnical materials. To simulate the mechanical properties of the cement sheath and stratum after yielding failure, an ideal plastic model (where the internal friction angle is equal to the dilation angle) is utilized.

The mechanical properties of the reservoir are also a key factor affecting the deformation of horizontal well casing. Therefore, to authentically capture formation characteristics, the numerical model integrates petrophysical parameters derived from field-acquired wireline log data within the target interval. These lithology-specific mechanical properties are subsequently incorporated into the network model, ensuring an accurate representation of subsurface rock behavior (Figure 7). The initial maximum and minimum horizontal principal stresses, based on logging data, were set at 85 MPa and 79 MPa, while the vertical principal stress was set at 86 MPa. The inner and outer diameters were specified as 121.4 mm and 139.7 mm, respectively (

Table 1. Basic parameters of finite element numerical simulation.

Category	Parameters	Values
Mechanical parameters	Rock density, D	2400 kg/m3
	Tensile strength, TO	2.3 MPa
	Cohesion, c	0.06 MPa
	Friction angle, φ	40°
	Young’s modulus, E	30.0 GPa
	Friction coefficient, μ	0.5
	Poisson’s ratio	0.21
In situ stress	Max horizontal principal stress, SH	85 MPa
	Min horizontal principal stress, Sh	79 MPa
	Vertical principal stress	86 MPa

).

(3)

Model construction

An interpretation of multi-arm wellbore imaging logging reveals that the CD of Class C at the 17th section resulted from stress concentration at the fracture tip causing compression deformation. The CD of Class B at the 22nd section is due to fracture shear causing shearing. The variations in load on the strata–fracture–casing system before and after the CDs at the 17th and 22nd sections of the well are analyzed. The CD models were then created to evaluate the casing deformation associated with the strata–fracture–casing connection (Figure 8).

Near Class B at the 22nd section, there is a group of natural fractures that intersect the wellbore obliquely. To simulate the effect of this natural fracture on the wellbore, a corresponding natural fracture was constructed in a three-dimensional model of CD (Figure 9).

Near the 17th section (4395 m) of well SN-H21-1, a set of natural fractures developed with their tips close to the wellbore but not intersecting with the wellbore, and a Class C casing deformation was formed at the tip of this natural fracture (Figure 8).

To simulate the effect of this natural fracture on the wellbore, a corresponding natural fracture was constructed in a three-dimensional model of CD. The pink part represents the natural fractures with tips approaching the wellbore, while the blue part represents reservoir rocks. In the simulation, the wellbore and formation were fixed, and the fractures sheared and slipped horizontally (Figure 10).

2.2.3. Model Validation

The microseismic data from the research area indicates that there is an asymmetrical distribution of casing deformation in the 17th fracturing section of this well. Therefore, using microseismic data and logging data, a casing deformation model for SN-H21-1 was established, and the casing deformation characteristics of the WY46-1 well were obtained through finite element numerical simulation (Figure 11).

The results of the multi-arm logging indicate that after the natural fractures are activated, the casing will undergo severe deformation, demonstrated by a distinct “S” shape bending. A comparison between the numerical simulation of casing deformation and the field measurements reveals a strong correlation in morphological features. Although there is a certain error (7%) in the amount of casing deformation, this error is within an acceptable range. The consistency between the simulation results and the experimental results highlights the reliability and effectiveness of the modeling method employed.

3. Simulation Results and Analysis

3.1. Penetrating Fracture Shear Deformation

The simulation results indicate that along the direction of shear slip within the fracture, the reduction in the inner diameter of the casing demonstrates distinct shear deformation characteristics. Far from the fracture plane, the Von-Mises stress level in the casing is low, and the casing inner diameter almost does not change. In contrast, in proximity to the intersection of the fracture plane and the wellbore, external loads become concentrated, resulting in the Von-Mises stress within the casing surpassing the yield strength. Consequently, the casing inner diameter decreases sharply, leading to the shear deformation of the casing (Figure 12).

The simulation results indicate a significant increase in CD with fracture slip. However, strengthening the casing by increasing its strength and thickness results in minimal impact on casing deformation, with variations of only up to 5% (Figure 13). The mechanism of penetrating fracture shear deformation mainly relies on the shear slip distance, rather than being correlated with casing strength and thickness.

The relationship between the slip distance and the approach angle of fractures indicates that the maximum slip distance occurs when the approach angle of the fracture (angle with the maximum horizontal principal stress) is 30°, and the risk of CD is highest; when the approaching angle of natural fractures is greater than 70°, the fractures are difficult to be activated, and there is no risk of CD in the wellbore (Figure 14a). The slip distance of the fracture varies in a parabolic shape along the fracture propagation path, with the maximum slip distance occurring at the midpoint of the fracture (Figure 14b). At the intersection of the wellbore and the fracture, the closer it is to the midpoint of the fracture, the greater the slip distance of the fracture, and the higher the risk of CD.

Therefore, it is evident that under the mechanism of penetrating fracture shear deformation, the natural fractures characterized by medium-to-low approach angles and intersection points with the wellbore near the center of the fracture are prone to form a higher risk of CD. Conversely, fractures with higher approach angles or intersections near both ends of the fracture pose a lower risk of CD.

3.2. Fracture Tip Compression Deformation

In contrast to shear deformation in the casing caused by fracture sliding, the deformation of the casing caused by the propagation of the fracture tip is manifested as compression deformation under compressive stress. As the fracture tip propagates forward, the stress concentration is greater, causing a more obvious deformation of the casing on one side of the fracture tip. Consequently, compared with the symmetrical deformation formed by casing shearing induced by fracture sliding, the deformation of the casing subjected to compressive stress exhibits an asymmetrical distribution pattern (Figure 15).

Furthermore, by comparing the deformation behaviors of casings across varying injection rates, it was observed that as the injection rate increased, the compression deformation of the casing along the direction of stress concentration at the fracture tip gradually increased. At an injection rate of 8 m³/min, the casing exhibited a minimal deformation of 9.78 mm. Upon increasing the injection rate to 16 m³/min, the deformation of the casing significantly increased to 44.17 mm. The injection rate doubled, leading to a more than 4-fold increase in CD. This is consistent with the changes in the stress field at the fracture tip; that is, with the increase in the injection rate, both the stress concentration degree and range at the fracture tip show an increasing trend, resulting in a gradual rise in the degree of CD. Therefore, for formations with natural fractures, the first principle to strictly follow during hydraulic fracturing operations is to “avoid stress concentration”.

3.3. Comparison of Shear Deformation and Compression Deformation

When comparing the casing deformations caused by shear and compression deformations in the same fracturing construction method and geological conditions, it was found that the maximum shear deformation of the casing induced by penetrating fracture is approximately 15.52 mm, whereas the maximum deformation of the casing induced by the propagation of fracture tip can reach 44.17 mm, significantly exceeding the CD from penetrating fracture shearing (Figure 16).

The multi-arm caliper imaging logs from SN-H21-1, sections 17 (4754 m) and 22 (4395 m), reveal that at the point of maximum deformation in section 17 (with a standard inner diameter of 114.3 mm), the inner diameter measures a maximum of 129.29 mm, an average of 115.19 mm, and a minimum of 102.49 mm. Correspondingly, for section 17, the respective inner diameter values are 156.9 mm (maximum), 132.5 mm (average), and 116.67 mm (minimum). The maximum deformations in sections 17 and 22, measuring 42.6 mm and 14.99 mm, respectively, closely align with the outcomes of numerical simulations, demonstrating high comparability.

Additionally, the imaging logging results indicate that the primary deformation pattern of the casing in section 22 of SN-H21-1 is shear deformation, while the CD in section 17 exhibits a pronounced compression shape (Figure 17). A comparative analysis of numerical simulations and multi-arm caliper imaging log interpretations confirms that penetrating fractures are prone to causing shear deformation, while the propagation of the fracture tip compresses the casing, resulting in an enhanced compression deformation of the casing.

Thus, it can be seen that the compressive stress formed at the fracture tip is significantly higher than the shear stress generated by the shear sliding of penetrating fracture, leading to the propagation of fracture tips near the wellbore, which is more likely to cause higher-grade CD. Consequently, during fracturing operations, priority should be given to monitoring natural fractures whose fracture tips are approaching to the wellbore. Targeted preventative and mitigation strategies should be implemented to address CD mechanisms associated with these fractures.

4. Discussion

4.1. The Key Contributor to Casing Deformation in Horizontal Wells

The simulation results indicate that after a natural fracture was activated, stress exhibits a bimodal distribution along the fracture path, with stress concentration at both ends of the fracture and low horizontal stress along the horizontal section. Furthermore, the stress variations at the fracture tip under different injection rates reveals that the stress concentration at the fracture tip is highly sensitive to the injection rate. As the injection rate increases, both the degree and extent of stress concentration at the fracture tip exhibit a gradual upward trend (Figure 18).

Statistically analyzing the data reveals that stress concentration at the tip of the natural fracture reaches 138 MPa, indicating a 56 MPa increase compared to the 86 MPa observed in the horizontal fracture section. This high stress concentration results in significantly higher compressive stress (138 MPa) on the wellbore than the shear stress (86 MPa) from the shear-activated natural fracture intersecting the wellbore through the horizontal fracture section. Consequently, the stress concentration at the fracture tip presents a higher risk of CD and more severe deformation activation (Figure 19).

The shear and slippage of natural fractures intersecting the wellbore, along with the initiation and propagation of a fracture tip approaching the wellbore, are two crucial mechanisms leading to CD in deep shale gas wells in the research region. When natural fractures with tips approaching the wellbore are activated by hydraulic fracturing, the fracture tip starts to propagate and propagate, leading to a buildup of substantial stress at the fracture tip. This stress compresses the wellbore, creating a squeezing effect that causes significant compression deformation (Figure 20b). In the case of natural fractures that intersect the wellbore obliquely and penetrate it, the activation of these fractures results in shear stress from sliding in the horizontal section, causing the shear deformation of the casing (Figure 20b).

Experts and scholars widely acknowledge that natural fractures/faults constitute a critical factor in the casing deformation of shale gas wells. Chen et al. [42] investigated casing deformation in the Weiyuan area, revealing that fracturing fluid preferentially infiltrates natural fractures in shale formations through seepage channels, inducing shear slip and subsequent casing shear deformation. The spatial alignment of deformed casing positions with slip fractures/faults detected via microseismic monitoring (in both strike and azimuth) confirms that the deformation is caused by natural cracks/fault shear slip. Through 3D imaging analysis using multi-arm caliper logging, Li et al. [43] identified S-shaped casing distortions characterized by vertical displacement, indicative of shear misalignment caused by natural fracture slippage. This supports the conclusion that natural fracture/fault slip during hydraulic fracturing is the primary driver of severe casing deformation. Dong et al.’s [44] analysis of 38 casing deformation cases in the Silurian Longmaxi shale formation (Sichuan Basin) revealed bidirectional displacement at deformation points via lead seal imprints, consistent with multi-arm caliper data, both exhibiting shear deformation signatures from natural fracture slippage. In summary, the prevailing academic consensus attributes casing deformation predominantly to shear-induced casing distortion triggered by hydraulic fracturing-activated slip along natural fractures/minor faults.

In a statistical analysis of casing deformation (CD) characteristics in deep shale gas horizontal wells in the Luzhou area, Jin et al. [5] compared 44 CDs across 18 sample wells with fracture prediction results. Of these, 21 CDs were located within fracture zones, while 23 occurred outside but proximal to predicted fractures. Accounting for fracture prediction accuracy, less than 50% of CDs could be attributed to the shear activation of natural fractures intersecting wellbores. This implies that some CDs correlate with natural fractures whose tips are near (but not intersecting) the wellbore. Meng et al. [9], investigating CDs in Sichuan Basin (Weiyuan, Changning and Zhaotong block), identified intense tectonic stresses with high horizontal stress anisotropy (~20 MPa), abundant natural fractures, and prominent shear deformation and compression deformation modes under aggressive hydraulic fracturing. However, existing studies have failed to adequately explain the increasing field prevalence of compression-induced deformation or provide a comparative analysis of its differences from shear-dominated casing deformation (CD). This paper’s analysis of the SN-H21 well platform in southern Sichuan reveals that 30% of the total CDs (including all Class C and 50% of Class B CDs) were induced by stress concentrations at fracture tips. Additionally, numerical simulations reveal that the hydraulic fracturing-induced propagation of natural fractures generates stress concentrations at fracture tips, which subject casings to significant non-uniform external collapse loads, resulting in compression deformation and amplifying casing deformation severity in horizontal wellbores under equivalent operational conditions.

It can be concluded that the proposal of the deformation mechanism involving fracture tip stress compression is somewhat of an improvement on the current ideas surrounding this mechanism, which previously stated that the CDs primarily cause casing shear deformation through natural fracture slip. Severe CDs in the research area are primarily attributed to the mechanism of fracture tip stress compression or the combination of fracture tip stress compression and penetrating fracture stress shear. Thus, proposing the deformation mechanism of fracture tip stress compression is of great significance for implementing targeted on-site prevention and control measures in the research area.

4.2. CD Risk Assessment and Prevention Measures

The simulation results indicate that the slip distance of natural fractures is related to the approach angle and propagation path. Under the mechanism of penetrating fracture shear deformation, natural fractures (fracture 1–1) with medium-to-low approach angles and intersection points with the wellbore near the center of the fracture are prone to form a higher risk of CD, classified as high-risk fractures for CD. Conversely, fractures (fracture 1–2) with higher approach angles or intersections near both ends of the fracture pose a lower risk of CD (Figure 21).

Currently, research on the deformation mechanism of wellbores induced by the activation of natural fractures largely focuses on the penetrating fracture shear deformation mechanism, overlooking the deformation mechanism caused by fracture tip stress compression. Consequently, in existing studies, a prevalent approach in assessing risks associated with CD due to natural fracture activation is to categorize the natural fractures (fracture 2–1) with tips approaching but not intersecting the wellbore as posing low or no risk (Figure 21). Based on the numerical simulations, the superposition of microseismic events–natural fractures–casing deformation, and the imaging logging data, it can be understood that the natural fractures approaching but not intersecting the wellbore pose a high risk of inducing CD. Consequently, these fractures should be reclassified as high risk for CD, rather than being designated as low risk or risk-free (Figure 21).

Natural fractures (e.g., fracture 2–1) with tips approaching but not intersecting the wellbore can induce significant casing deformation (CD). Thus, a CD detection methodology and prevention measures targeting fracture tip-induced compression deformation mechanism was developed. First, the pre-fracturing risk assessment should classify intervals with such natural fractures as high-risk zones. Risks can be mitigated by reducing injection rates or applying temporary plugging to lower fluid pressure within fractures, thereby avoiding stress concentrations at fracture tips and minimizing CD risks.

During fracturing in these intervals, real-time monitoring and process optimization are critical. Stress release (instantaneous pressure drop upon fracture opening) and stress accumulation (pressure increase with continued injection) during fracture activation cause distinct pressure fluctuations in fracturing curves. Furthermore, the reactivation of these fractures triggers intense microseismic events. Consequently, if strong pressure fluctuations coincide with high-magnitude microseismic events near fracture tips, injection rates should be immediately reduced or operations paused to alleviate fluid pressure and prevent further stress concentrations, thereby averting CD.

5. Conclusions

(1)

Penetrating fracture shear deformation and fracture tip compression deformation are two key mechanisms leading to CD in deep shale gas wells of the study area. There are significant differences between the two in terms of their formation mechanisms and morphology. Research shows that the severe CDs in the research area are primarily attributed to the mechanism of fracture tip stress compression or the combination of fracture tip stress compression and penetrating fracture stress shear. Thus, proposing the deformation mechanism of fracture tip stress compression is of great significance for implementing targeted on-site prevention and control measures in the research area.

(2)

The slip distance of natural fractures is related to the approach angle and propagation path. Under the mechanism of penetrating fracture shear deformation, natural fractures with medium-to-low approach angles and intersection points with the wellbore near the center of the fracture are prone to form a higher risk of CD. These fractures can be categorized as high risk for casing deformation. Conversely, fractures with higher approach angles or intersections near both ends of the fracture pose a lower risk of CD and can be classified as low risk for CD.

(3)

Compared with the symmetrical deformation formed by casing shearing induced by fracture sliding, the CD subjected to compressive stress exhibits an asymmetrical distribution pattern, which increases the risk and level of induced casing deformation. The simulation results show that stress concentration formed at the fracture tip is highly sensitive to the injection rate. The greater the injection rate, the higher the deformation of the casing. Therefore, for formations with the fracture development, strict adherence to the first principle of “avoiding stress concentration” should be followed during hydraulic fracturing operations.

(4)

The proposed fracture propagation stress compression casing deformation (FP-SCCD) mechanism provides an improved understanding of casing deformation mechanisms. Based on this mechanism, a comprehensive casing deformation (CD) detection methodology and prevention strategy were developed. When strong pressure fluctuation coincides with high-magnitude microseismic events near fracture tips, an immediate reduction in injection rates or temporary operation suspension is recommended. This responsive measure effectively alleviates fluid pressure buildup and mitigates stress concentration, thereby preventing potential CD incidents.