Core type adaptive SC-CO2 phase state stress collaborative fracturing method

Abstract

The invention relates to the technical field of oil and gas field exploitation, in particular to a core type adaptive SC-CO2 phase state stress collaborative fracturing method. The method specifically comprises the steps that fracturing experiments under various variable conditions are carried out, and fracturing pump pressure is obtained; the variable conditions comprise a core type, an injection fluid type, an injection flow rate and axial pressure and confining pressure conditions; converting the core type and the injected fluid type into virtual variables; establishing a multiple linear regression model by taking the fracture pump pressure as a dependent variable and taking the core type, the injection fluid type, the injection flow rate and the difference between the axial pressure and the confining pressure as independent variables; fitting by adopting a least square method to obtain a regression coefficient of each independent variable; performing t inspection to obtain a p value of each independent variable, and screening out the independent variables with the p values smaller than or equal to 0.2; and supercritical CO2 phase state stress synergistic fracturing is carried out. The method solves the problems that an existing method is unstable in effect in practical application, and the diversified requirements of different unconventional reservoirs are difficult to meet.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas field development, and particularly relates to a core type adaptive SC-CO2 phase state stress synergistic fracturing method. BACKGROUND

[0002] The development of unconventional reservoirs, especially shale oil and tight oil and gas, has long been faced with challenges such as low porosity, poor permeability and imperfect fracture network. Although the existing fracturing technology, especially the traditional water-based fracturing fluid and acid fracturing technology, has achieved certain results in the development of some conventional oil and gas fields, in the application of unconventional reservoirs, it often faces the problems of limited fracturing effect and insufficient complexity of fractures. For example, the main problem of water-based fracturing fluid is its poor permeability and weak fracture propagation ability, especially in low-permeability shale oil and tight sandstone, it is difficult for water-based fluid to form a complex fracture network, resulting in insufficient fracture ductility and affecting oil and gas recovery.

[0003] SC-CO2, the full name of which is Supercritical Carbon Dioxide, is also known as supercritical carbon dioxide. As a research hotspot in recent years, SC-CO2 fracturing technology can form more complex and deep fracture networks in low-permeability reservoirs due to its diffusion property as a gas and solubility as a liquid in the supercritical state.

[0004] Although SC-CO2 technology has made certain application progress in the development of shale oil and gas and tight oil and gas, the existing method still has the problems of insufficient core type adaptation and insufficient coordination control of formation stress conditions and SC-CO2 phase state. The current research generally ignores the synergistic effect between formation stress conditions and SC-CO2 fluid phase state, resulting in significant differences in fracture propagation behavior under different stress states. The above technical defects make the existing SC-CO2 fracturing method unstable in actual application, and it is difficult to meet the diversified needs of different unconventional reservoirs. SUMMARY

[0005] In order to solve the problems of insufficient core type adaptation and insufficient coordination control of stress conditions and SC-CO2 phase state in the existing SC-CO2 fracturing technology, which leads to unstable effect in actual application of the existing method and difficulty in meeting the diversified needs of different unconventional reservoirs, the present application provides a core type adaptive SC-CO2 phase state stress synergistic fracturing method.

[0006] To achieve the above purpose, the technical scheme of the present application is as follows.

[0007] The application provides a core type adaptive SC-CO2 phase state stress synergistic fracturing method, including the following steps: carrying out fracturing experiments under multiple variable conditions to obtain fracturing pump pressures under the multiple variable conditions; the variable conditions are core types and injection fluid types, injection flow rates, and axial pressure and confining pressure conditions; converting the core types and the injection fluid types into virtual variables; according to the virtual variables of the core types and the injection fluid types, establishing a multiple linear regression model with the fracturing pump pressures as dependent variables and differences between the core types and the injection fluid types, the injection flow rates, and the axial pressure and the confining pressure as independent variables; according to the established multiple linear regression model, performing fitting by using the least square method to minimize the sum of error squares between all observation values and prediction values, obtaining regression coefficients of each independent variable; performing t-test on the regression coefficients of each independent variable to evaluate the significance of each independent variable on the fracturing pump pressure, obtaining p values of each independent variable, and screening independent variables with p values less than or equal to 0.2; performing SC-CO2 phase state stress synergistic fracturing with the independent variables with p values less than or equal to 0.2 as experimental conditions.

[0008] Preferably, the core types are Q9 layer shale, Q1 layer shale, Q2 layer shale and sandstone; and the injection fluid types are clear water, liquid CO2 and supercritical CO2. Preferably, the Q9 layer shale, the Q1 layer shale and the Q2 layer shale are different layer shales in a geological profile, and the geological profile is mainly composed of laminated shale. In the application, the Q1 layer shale is Q1 layer laminated shale, the Q2 layer shale is Q2 layer laminated shale, the Q9 layer shale is Q9 layer laminated shale, and the sandstone is tight sandstone. Preferably, the method for converting the core types and the injection fluid types into virtual variables is to convert the core types and the injection fluid types into qualitative variables and then convert the qualitative variables into virtual variables.

[0009] Preferably, the axial pressure is 10 MPa to 15 MPa; and the confining pressure is 10 MPa to 15 MPa.

[0010] Preferably, the injection flow rate is 0.4 mL / min to 6 mL / min.

[0011] The application has the following beneficial effects: 1. The application proposes a more accurate and efficient fracturing method by comprehensively considering the synergistic effect of core types, stress conditions and supercritical CO2 fluid phase states, has significant advantages in improving fracturing effect, reducing energy consumption, increasing fracture complexity and improving oil and gas recovery rate compared with the prior art, and solves the problems of unstable effect in actual application and difficulty in meeting the diversified needs of different unconventional reservoirs.

[0012] 2. This invention adjusts the phase state of supercritical CO2 fluid (such as liquid, supercritical, or gaseous state) based on the physical properties of different core types, such as porosity, permeability, and mineral composition. This optimizes fluid permeability and fracture propagation effects to suit the specific reservoir conditions, significantly improving oil and gas recovery. By adapting to different core types, this invention ensures optimal performance of supercritical CO2 fluid in various reservoirs.

[0013] 3. This invention uses the relationship between axial pressure and confining pressure as the stress condition. By optimizing the synergistic effect between the stress condition and the phase state of the supercritical CO2 fluid, it is found that different formation stress states, such as axial pressure greater than confining pressure or axial pressure less than confining pressure, will affect the fracturing effect. Therefore, this invention proposes to adjust the phase state and injection conditions of the supercritical CO2 fluid under different stress conditions, thereby maximizing the complexity and depth of the fracture and improving the oil and gas recovery rate.

[0014] 4. This invention proposes a precise method for controlling fluid injection flow rate and pressure. Based on the physical properties of different reservoirs, such as permeability and porosity, the injection flow rate and pressure of supercritical CO2 are adjusted to achieve optimal fracture propagation. This method effectively avoids uneven permeation during fluid injection, ensuring uniform fluid penetration within the reservoir and optimizing fracture complexity, thereby improving fracturing efficiency.

[0015] 5. This invention proposes a supercritical CO2 fracturing fluid selection method based on core type optimization. By customizing different supercritical CO2 fracturing fluid compositions according to reservoir core types, such as shale and sandstone, this invention can improve the compatibility of the fracturing fluid with the reservoir rock, enhance fluid permeability and fracture propagation effect. By optimizing the fracturing fluid selection method, the efficiency and effectiveness of the fracturing process are improved. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of the multifluid pseudo-triaxial fracturing experimental device according to an embodiment of the present invention. Wherein, 1 is a triaxial fracturing gripper; 2 is a manual axial pressure pump; 3 is a manual confining pressure pump; 4 is a data acquisition and processing unit; 5 is a vacuum pump; 6 is a first piston container; 7 is a second piston container; 8 is a fracturing pump; 9 is a temperature measurement and control unit; 10 is a gas booster pump; 11 is a gas cylinder interface; V1 is a vent valve; V2 is a first pressure regulating valve; V3 is a second pressure regulating valve; V4 is a third pressure regulating valve; V5 is a fourth pressure regulating valve; V6 is a fifth pressure regulating valve; V7 is a sixth pressure regulating valve; V8 is a seventh pressure regulating valve; P1 is an axial pressure sensor; P2 is a confining pressure sensor; P3 is a fracturing pressure sensor; and P4 is a vacuum pressure sensor.

[0017] Figure 2 These are physical images of the core samples in the embodiments of the present invention. Among them, (a) is a physical image of a shale core sample; and (b) is a physical image of a sandstone core sample.

[0018] Figure 3 These are schematic diagrams showing the dimensions of the core samples and actual images of the processed core samples according to embodiments of the present invention. (a) is a schematic diagram showing the dimensions of the core samples; (b) is an actual image of the processed core samples.

[0019] Figure 4 This is a phase diagram of carbon dioxide.

[0020] Figure 5 These are pump pressure-time curves under different fracturing media. Among them, (a) is the pump pressure-time curve of shale layer Q9; (b) is the pump pressure-time curve of shale layer Q1 under different fracturing media; (c) is the pump pressure-time curve of shale layer Q2 under different fracturing media; and (d) is the pump pressure-time curve of tight sandstone under different fracturing media.

[0021] Figure 6 These are pump pressure-time curves for different shale layers, as well as for the Q1 layer shale and tight sandstone. Among them, (a) is the pump pressure-time curve for different shale layers; (b) is the pump pressure-time curve for the Q1 layer shale and tight sandstone.

[0022] Figure 7 These are pump pressure-time curves for the Q2 shale under different stress conditions.

[0023] Figure 8 The values ​​represent the fracturing pump pressures under different core types and under different stress conditions when water and liquid CO2 are injected. Among them, (a) represents the fracturing pump pressures under different stress conditions for different core types; and (b) represents the fracturing pump pressures under different stress conditions when water and liquid CO2 are injected.

[0024] Figure 9 It refers to the fracturing pump pressure for different core types at different injection flow rates.

[0025] Figure 10 These are the pump pressure-time curves of the Q9 shale and tight sandstone under different injection flow rates. Among them, (a) is the pump pressure-time curve of the Q9 shale under different injection flow rates; (b) is the pump pressure-time curve of the tight sandstone under different injection flow rates.

[0026] Figure 11 This is the result of regression analysis. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0028] In this invention, supercritical CO2 refers to carbon dioxide in a supercritical state, abbreviated as SC-CO2. The temperature and pressure of SC-CO2 both exceed the critical point of carbon dioxide, which is 31.1°C and 7.38 MPa. In this supercritical state, supercritical CO2 has both the density of a liquid and the diffusivity of a gas, and can effectively penetrate reservoir rocks at lower pressures, promote fracture propagation, and improve oil and gas release.

[0029] Phase regulation refers to the process of controlling temperature and pressure to switch fluids between different phases. In the supercritical state, supercritical CO2 possesses optimal permeability and solubility, making it suitable as a fracturing medium. Phase regulation allows for optimization of supercritical CO2 flowability based on reservoir core type, permeability, and stress conditions, thereby improving fracture propagation. Fracture propagation refers to the process by which injected fluid forms and extends a fracture network within the rock during fracturing, enhancing reservoir permeability and oil and gas flow channels; this is the core aspect of fracturing operations. Initiation pressure is the minimum pressure required for the fluid to overcome rock strength and initiate fractures. Due to its low viscosity and high diffusivity, supercritical CO2 has a lower initiation pressure than traditional water-based fracturing fluids under the same formation conditions, contributing to lower energy consumption in rock fracturing. Axial pressure is the stress acting vertically to the core, while confining pressure is the stress acting laterally to the rock formation; both influence the rock fracture mode and fracturing effect. Fluid injection velocity and injection pressure are the speed and pressure at which the fracturing fluid enters the reservoir, respectively, and have a crucial impact on fracture formation, propagation uniformity, and fracturing effectiveness. Fracture morphology refers to the structure and distribution of fractures formed by hydraulic fracturing, and its complexity directly affects oil and gas flow and recovery rate. Shale oil and tight sandstone, as typical unconventional reservoirs, are characterized by low porosity and low permeability, making extraction difficult and costly. Enhanced Oil Recovery (EOR) is a technology that improves recovery rates through methods such as gas injection. Supercritical CO2 fracturing, as an EOR method, can significantly improve extraction efficiency by optimizing fracture complexity and permeability.

[0030] Current supercritical CO2 fracturing technology still suffers from the following problems: Insufficient core type adaptability: Existing schemes often use uniform fracturing parameters, failing to fully consider the differences in mineral composition and porosity / permeability between different cores, such as shale and tight sandstone, leading to unstable fracturing effects. Insufficient stress control: Existing methods fail to systematically optimize fluid phase and injection strategies under different axial / confining pressure relationships, neglecting the crucial influence of actual formation stress on fracture formation and stability. Insufficient injection condition optimization: Injection velocity and pressure settings are relatively uniform, without differentiated design based on reservoir characteristics, easily leading to unsatisfactory fracture propagation or insufficient energy utilization. Unclear interaction mechanism between supercritical CO2 phase and formation: Insufficient research on the interaction between supercritical CO2 phase regulation and mineral and pore structure limits its potential for fracture network optimization in different core reservoirs. Lack of energy efficiency optimization: Existing research focuses mainly on increasing fracture complexity, without systematically considering the minimization of energy consumption, affecting the overall benefits of supercritical CO2 fracturing technology. To address this, the present invention proposes a supercritical CO2 phase-stress synergistic fracturing method adapted to different core types. By adjusting the fracturing medium and construction parameters according to the core type and stress conditions, the overall fracturing effect and energy efficiency are improved.

[0031] like Figure 1 A multi-fluid pseudo-triaxial fracturing experimental apparatus includes a triaxial fracturing gripper 1, a data acquisition and processing unit 4, a gas / liquid pressurization control system, a heating and insulation system, a pseudo-triaxial servo control system, and a gas cylinder interface 11. The multi-fluid pseudo-triaxial fracturing experimental apparatus is an existing triaxial fracturing experimental apparatus. The gas / liquid pressurization control system includes a manual axial pressure pump 2, a manual confining pressure pump 3, a vacuum pump 5, a first piston container 6, a second piston container 7, a fracturing pump 8, and a gas pressurization pump 10. The heating and insulation system includes a temperature measurement and control unit 9. The gas / liquid pressurization control system maintains the pressure of the triaxial fracturing gripper 1 at ≤50MPa. The heating and insulation system regulates the temperature range from room temperature to 100℃. The room temperature is 20℃ to 30℃. Vent valve V1, first pressure regulating valve V2, second pressure regulating valve V3, third pressure regulating valve V4, fourth pressure regulating valve V5, fifth pressure regulating valve V6, sixth pressure regulating valve V7, and seventh pressure regulating valve V8 are used to regulate the pressure. Axial pressure sensor P1, confining pressure sensor P2, fracturing pressure sensor P3, and vacuum pressure sensor P4 are used to display pressure.

[0032] A core type-adaptive SC-CO2 phase stress-assisted fracturing method is described below: Step 1: Conduct fracturing experiments under multiple variable conditions to obtain the fracturing pump pressure under multiple variable conditions; the variable conditions are core type and injected fluid type, injection flow rate, and axial pressure and confining pressure conditions.

[0033] Specifically, the core samples consisted of Q9 shale, Q1 shale, Q2 shale, and sandstone; the injected fluids were clear water, liquid CO2, and supercritical CO2. The Q9, Q1, and Q2 shale layers represent different strata within the geological profile, which is predominantly composed of laminated shale.

[0034] Core sample preparation: Shale and sandstone cores were used in the experiment to ensure similar physical properties and consistent drilling direction. After cutting and grinding, standard cylindrical core samples with a diameter of Φ1=50mm and a height of h1=100mm were prepared, such as... Figure 2 Following this, drilling and fracturing pipe bonding were performed to simulate the actual wellbore fracturing process: an injection hole with a diameter of Φ2=10mm and a depth of h2=45mm was drilled at the center of the core sample end face, using PEEK tubing as the fracturing pipe. Before bonding, O-rings were used to fix the pipe head, leaving a 3mm long, 8mm diameter open hole section. After injecting a special adhesive to fix the fracturing pipe, it was allowed to cure for 48 hours to ensure bonding strength and sealing. Finally, excess tubing and adhesive were trimmed to ensure a smooth core sample end face for easy injection of fracturing media. See the attached image for the finished core sample. Figure 3 .

[0035] Fracturing Experiments: A multi-fluid pseudo-triaxial fracturing experimental setup was used to systematically conduct fracturing experiments on shale core samples from Q9, Q1, and Q2 shale layers, as well as tight sandstone samples, under different axial and confining pressures, injection velocities, and fracturing media conditions. The fracturing media were water, liquid CO2, or supercritical CO2. The specific procedure is as follows: Instrument Installation and Debugging: Place the prepared core sample in the center of the triaxial fracturing gripper and tightly wrap it with a high-temperature resistant rubber sleeve. Adjust the use of a conventional rubber sleeve and a carbon dioxide-resistant rubber sleeve according to the experimental protocol. Then, tighten and seal the core gripper with bolts and rubber sealing rings, and connect the fluid inlet and outlet pipelines, temperature sensor, and pressure sensor. Turn on the power to the experimental setup and start the control panel, connecting it to the computer data acquisition system. Perform a pipeline airtightness test: Introduce nitrogen or clean water at a certain pressure into the system, and check each section to confirm that the entire high-pressure pipeline system is leak-free.

[0036] Axial and confining pressure loading: Simultaneously activate the confining and axial pressure loading systems to apply both confining and axial pressures to the core sample. Once the axial pressure reaches the preset value, set the axial pressure to 10 MPa or 15 MPa and the confining pressure to 10 MPa or 15 MPa. Then, lock the axial pressure control valve to maintain a constant axial pressure and continue applying the confining pressure to the set value. This set confining pressure must be greater than the estimated maximum fluid injection pressure to ensure effective sealing of the core sample during subsequent fracturing media injection.

[0037] Pressure adjustment, system preheating, and temperature control: Based on the injected fracturing medium, the pressurization value of the gas pressurization system was adjusted. In the experiment, liquid CO2 was set to 5 MPa; supercritical CO2 to 10 MPa, and the gas was transferred to the fracturing pump through pipelines. The preheating switch was turned on, and the system temperature was set to the experimental design values: 20℃ for liquid CO2 and 60℃ for supercritical CO2. Heating was continued for more than 1 hour to ensure that the overall temperature inside the core sample, the injection pipeline, and the fracturing medium all reached and stabilized at the temperature corresponding to the target phase. The target phase was liquid or supercritical.

[0038] After the temperature stabilized, fracturing medium injection and fracturing process monitoring began. All relevant valves were ensured to be open, and the high-pressure pump was started to inject the target fracturing medium into the core sample at the set injection flow rate. The injection pressure was monitored and recorded in real time as a function of time; the injection pressure is the pump pressure. Once the core sample fractured, as indicated by a peak in the pressure curve followed by a significant decrease and subsequent stabilization, fracturing medium injection was terminated, the experiment ended, and the fracturing pump pressure was obtained.

[0039] System depressurization, cooling, and sample recovery: After the experiment, the system was depressurized, cooled, and the samples were recovered. The system was shut down safely in sequence: the heating device was stopped, the fracturing medium cylinder valve was closed, the high-pressure pump was stopped, and finally, the main power supply was disconnected. Subsequently, the confining pressure and axial pressure were slowly and sequentially removed. The fractured core sample was carefully removed and recovered for subsequent analysis.

[0040] To investigate the effects of different injected fluid types, axial and confining pressures, and injection velocities on the fracturing characteristics of different core types, fracturing experiments with different injected fluid types were designed. CT scans were performed on some of the fracturing core samples, and quantitative fracture properties were calculated using quantitative fracture analysis methods. Specific experimental designs are shown in Tables 1 and 2.

[0041] Table 1. Experimental schemes for laminated shale with different core types.

[0042] Note: Axial pressure / containing pressure of 10 / 10 indicates that axial pressure = containing pressure; axial pressure / containing pressure of 15 / 10 indicates that axial pressure > containing pressure; axial pressure / containing pressure of 10 / 15 indicates that axial pressure < containing pressure.

[0043] Table 2 Experimental schemes for core types of tight sandstone

[0044] Note: Axial pressure / containing pressure of 10 / 10 indicates that axial pressure = containing pressure; axial pressure / containing pressure of 15 / 10 indicates that axial pressure > containing pressure; axial pressure / containing pressure of 10 / 15 indicates that axial pressure < containing pressure.

[0045] Analysis of Results from Different Fracturing Media: To systematically investigate the influence of different fracturing media on core fracture characteristics, comparative experiments were conducted on Q9 shale, Q1 shale, Q2 shale, and tight sandstone under unified basic conditions to reveal the mechanism of action of different media. The unified basic conditions were axial pressure / confining pressure = 10.00 MPa / 10.00 MPa, and the injection flow rate was matched according to the characteristics of the fracturing media: supercritical CO2 was 4 mL / min, and liquid CO2 and water were 0.4 mL / min. The Q9 shale was classified as P1, P6, and P7; the Q1 shale as P8, P11, and P12; the Q2 shale as P17, P20, and P21; and the tight sandstone as S1, S6, and S7.

[0046] To ensure that the CO2 fluid used in the experiment maintains its intended phase state, the carbon dioxide phase diagram is consulted before fluid injection, such as... Figure 4 Constant temperature heating equipment and gas pressurization equipment are required to adjust the pressure and temperature. The initial pressure of supercritical CO2 is 10 MPa, and the temperature is 60℃; the initial pressure of liquid carbon dioxide is 5 MPa, and the temperature is 20℃. From Figure 5 As shown in the pump pressure-time curve, the pump pressure gradually increases with increasing injection time, reaches a peak, and then rapidly decreases due to fracture initiation and propagation to the sample surface. The fracturing pump pressure exhibits a consistent gradient with the type of fracturing medium: clear water > liquid CO2 > supercritical CO2, and this trend remains stable across different core types. The specific data are as follows: Q9 shale: supercritical CO2 / P1 group is 32.91 MPa, liquid CO2 / P6 group is 35.65 MPa, and clear water / P7 group is 36.83 MPa; Q1 shale: supercritical CO2 / P8 group is 34.96 MPa, liquid CO2 / P11 group is 38.76 MPa, and clear water / P12 group is 39.18 MPa; Q2 shale: supercritical CO2 / P17 group is 34.73 MPa, liquid CO2 / P20 group is 37.30 MPa, and clear water / P21 group is 38.31 MPa; tight sandstone: supercritical CO2 / S1 group is 34.12 MPa, liquid CO2 / S7 group is 43.59 MPa, and clear water / S6 group is 44.50 MPa.

[0047] Based on supercritical CO2, the pump pressure increase of liquid CO2 ranged from 7.40% to 27.75%, while that of clear water ranged from 10.31% to 30.42%. The increase in pressure was generally higher in tight sandstone than in shale, reflecting the greater sensitivity of low-permeability cores to the type of fracturing medium. The low viscosity of supercritical CO2 significantly reduces its flow resistance within the core, enabling rapid diffusion along bedding planes or microfractures to form a uniform pore pressure field. Taking the Q9 shale layer as an example, the pore pressure transmission efficiency under supercritical CO2 was higher than that under clear water, resulting in a 11.91% reduction in fracturing pump pressure. Liquid CO2 has a slightly higher viscosity and lower permeability, with pump pressure falling between the two. Supercritical CO2 at 60℃ and 10MPa possesses both gas diffusivity and liquid solubility, with a diffusion coefficient three times that of liquid CO2, allowing it to efficiently penetrate micropores, expand the effective radius of action, and thus reduce fracturing pressure. Liquid CO2 is phase-stable at 20℃ and 5MPa, but its diffusion capacity is limited, resulting in a weaker pore pressure increase effect.

[0048] Water readily reacts with clay minerals in shale, temporarily enhancing the cementation strength of bedding planes and further increasing fracturing resistance. The pump pressure during fracturing with water in the Q1 shale layer is 1.08% higher than that during fracturing with liquid CO2; this effect is almost negligible in tight sandstone with a clay content of <5%.

[0049] Water readily undergoes hydration reactions with the clay minerals in shale, leading to a temporary increase in the cementation strength of the bedding planes. This further increases fracturing resistance, resulting in a generally higher pump pressure for water than for CO2. During fracturing of the Q1 shale layer with water, hydration caused the pump pressure to be 1.08% higher than that of liquid CO2. This effect is almost negligible in tight sandstone due to its clay content of <5%.

[0050] Analysis of Core Result Results of Different Types: To investigate the fracturing initiation characteristics of different types of cores during hydraulic fracturing, layered shale and tight sandstone of layers Q1, Q2, and Q9 were selected as research objects, and comparative experiments were conducted under the same experimental conditions. During the experiment, the axial pressure / confining pressure was maintained at 10 MPa / 10 MPa, the injected fluid was supercritical CO2, and the injection flow rate was 4 mL / min to ensure the singularity of variables. The influence on the fracturing pump pressure was analyzed only by changing the core type and the shale layer selected.

[0051] Analysis of core results from shale with different laminar textures: From Figure 6 From the fracture pressure and pump pressure-time curves in (a), the shale from different layers shows certain differences. The specific data are as follows: In the layered shale, the fracture pump pressure of the P1 group in layer Q9 is 32.91 MPa, that of the P8 group in layer Q1 is 34.96 MPa, and that of the P17 group in layer Q2 is 34.73 MPa.

[0052] Comparing different shale layers, the Q9 shale layer exhibited the lowest fracturing pump pressure, followed by the Q1 layer, while the Q2 layer had a slightly lower fracturing pump pressure than the Q1 layer. This result indicates that even within the same type of lamellar shale, different layers exhibit varying mechanical properties. This difference is primarily due to the sedimentary environment. Shales at different strata experienced different pressures, temperatures, and mineral composition depositional conditions during their formation, leading to variations in their density, pore structure, and bedding characteristics. The Q9 shale layer, mainly formed in a shallow marine sedimentary environment, has a higher clay mineral content and weaker interlayer cementation. In contrast, the Q1 and Q2 shale layers, formed in deeper sedimentary environments, have increased siliceous or calcareous cement content, resulting in a denser bedding structure. Therefore, higher pump pressures are required to fracture the samples during fracturing.

[0053] Looking at the specific data, the fracturing pump pressure of each shale layer shows a gradient distribution of "Q9 shale < Q2 shale < Q1 shale": the fracturing pump pressure of group P1 in Q9 is 32.91 MPa, group P17 in Q2 is 34.73 MPa, and group P8 in Q1 is 34.96 MPa. The difference between Q2 and Q9 is 1.82 MPa, and the difference between Q1 and Q2 is 0.23 MPa. The relative difference between Q2 and Q9 is 5.53%, and the relative difference between Q1 and Q2 is 0.66%, indicating that the difference between Q9 and the other two layers is more significant, while the difference between Q1 and Q2 is relatively minor.

[0054] From the perspective of stratigraphic formation environment, the core differences in stratigraphy stem from variations in pressure, temperature, and mineral composition during sedimentation. The Q9 shale primarily formed in a shallow marine sedimentary environment, where the sedimentation rate is slow, allowing clay minerals such as montmorillonite and illite to accumulate significantly, reaching contents exceeding 40%. The layered structure of these clay minerals results in weak cementation between shale bedding planes, small interlayer spacing, and good connectivity, creating preferential channels for fluid infiltration. When supercritical CO2 is injected, the fluid readily expands rapidly along the bedding planes, reducing the overall rock's resistance to fracturing, thus resulting in the lowest fracturing pump pressure.

[0055] In contrast, the Q1 and Q2 shale layers formed in a deeper sedimentary environment with greater water depth and higher pressure. During sedimentation, the content of rigid minerals such as siliceous minerals (e.g., quartz) and calcareous minerals (e.g., calcite) increased significantly. These rigid minerals filled the gaps in the bedding planes through cementation, making the bedding structure more compact and reducing connectivity. At this point, the permeation pathway of supercritical CO2 was restricted, requiring greater matrix strength to fracture the rock. Therefore, the fracturing pump pressure of the Q1 and Q2 shale layers was significantly higher than that of the Q9 shale layer. Specifically, the Q1 shale layer, due to its higher silica content and stronger cementation, had a slightly higher fracturing pump pressure than the Q2 shale layer. Although the difference was only 0.23 MPa, it reflects the influence of subtle differences in mineral composition on mechanical properties.

[0056] Further verification using rock physical properties revealed that the porosity and permeability of the Q9 shale layer were higher than those of the Q1 and Q2 shale layers. This higher porosity and permeability facilitated the penetration of supercritical CO2 into the rock, reducing effective stress through pore elasticity and further explaining the lower fracturing pump pressure. Conversely, the low porosity and permeability of the Q1 and Q2 shale layers exacerbated the difficulty of fluid penetration, requiring higher pump pressures to reach the fracturing critical value.

[0057] Analysis of core samples from laminated shale and tight sandstone: From Figure 6 From the fracture pressure and pump pressure-time curves of the Q1 shale and tight sandstone in (b), the fracture pump pressures of the laminated shale and tight sandstone exhibit a "similar but differentiated" characteristic: the fracture pump pressure of the tight sandstone S1 group is 34.12 MPa; among the laminated shale, the fracture pump pressure of the P8 group of the Q1 shale is 34.96 MPa, the fracture pump pressure of the P17 group of the Q2 shale is 34.73 MPa, and the fracture pump pressure of the P1 group of the Q9 shale is 32.91 MPa. Quantitative analysis shows that the fracture pump pressure of the Q1 shale is 0.84 MPa higher than that of the tight sandstone, an increase of 2.46%; the difference in fracture pump pressure of the Q2 shale narrows to 0.61 MPa, an increase of 1.79%; and the fracture pump pressure of the Q9 shale is 1.21 MPa lower, an increase of 3.55%.

[0058] From the perspective of the core structure's core type influence, laminated shale exhibits significant anisotropy. The presence of bedding planes gives the rock strength a directional characteristic, with lower strength parallel to the bedding planes and higher strength perpendicular to them. When supercritical CO2 is injected, if the fluid permeates along the bedding planes, such as in the Q9 shale layer, the low strength of the bedding planes can reduce fracturing resistance, resulting in a lower pump pressure than sandstone. However, if the bedding is dense, such as in the Q1 and Q2 shale layers, the fluid struggles to extend along the bedding planes and must overcome the high strength perpendicular to the bedding planes, leading to a slightly higher pump pressure than sandstone. Dense sandstone, on the other hand, is uniformly cemented by minerals such as quartz and feldspar, lacking obvious bedding or directionality, and exhibits a stable strength distribution. Supercritical CO2 penetration has no preferred path and must uniformly overcome the matrix strength.

[0059] Meanwhile, the shale from different layers exhibits significant differences in mineral composition, further influencing the fracturing pressure. The clay minerals in the Q9 shale may experience slight swelling upon exposure to supercritical CO2, weakening the bedding plane cementation and indirectly reducing fracturing resistance. The Q1 and Q2 shale layers have a high silica content of approximately 30%–45%, exhibiting strong chemical stability and resistance to CO2 reaction, resulting in stable bedding cementation and strong fracturing resistance. The dense sandstone has a high quartz content of 70%–80%, exhibiting strong chemical inertness and stable matrix strength, with pump pressure less affected by fluid-mineral interactions. This mineral-fluid interaction makes the pump pressure difference between the Q9 shale and sandstone more significant, while the differences between the Q1 and Q2 shale and sandstone are more dependent on structural characteristics.

[0060] Analysis of Supercritical CO2-Induced Fracturing Results under Different Stress Conditions: To investigate the influence of different stress conditions on the fracture characteristics of cores, three stress states were set up in the experiment: axial pressure = confining pressure, axial pressure > confining pressure, and axial pressure < confining pressure. When axial pressure = confining pressure, axial pressure / confining pressure = 10 MPa / 10 MPa; when axial pressure > confining pressure, axial pressure / confining pressure = 15 MPa / 10 MPa; and when axial pressure < confining pressure, axial pressure / confining pressure = 10 MPa / 15 MPa. Under these three stress states, fracturing experiments were conducted on lamellar shale and tight sandstone of the Q9, Q1, and Q2 shale layers, respectively. The injected fluid types included supercritical CO2, liquid CO2, and clear water to comprehensively explore the relationship between stress conditions and fracturing pump pressure.

[0061] From the overall experimental results of laminated shale, the influence of stress conditions on the core fracture pump pressure shows a significant regularity. For example... Figure 7 and Figure 8For lamellar shale, when supercritical CO2 was injected, the fracturing pump pressures of the three experimental groups for the Q9 shale (P1, P2, and P3) were 32.91 MPa, 35.29 MPa, and 37.75 MPa, respectively; and the fracturing pump pressures of the three experimental groups for the Q2 shale (P17, P18, and P19) were 34.73 MPa, 40.2 MPa, and 45.09 MPa, respectively. The pattern showed that the fracturing pump pressure was highest when the axial pressure was lower than the confining pressure, followed by the lower pressure when the axial pressure was higher than the confining pressure, and lowest when the axial pressure equaled the confining pressure.

[0062] For tight sandstone, in the three experimental groups during supercritical CO2 injection, the fracturing pump pressures for groups S1, S2, and S3 were 34.12 MPa, 39.94 MPa, and 36.9 MPa, respectively. The pump pressures exhibited a pattern of highest pressure when axial pressure > confining pressure, followed by pressure when axial pressure < confining pressure, and lowest pressure when axial pressure = confining pressure, which differs from the pattern observed in shale. Due to the strong homogeneity of the tight sandstone structure, it does not exhibit the significant anisotropy similar to shale and lacks the abundant vertical bedding found in shale, making its mechanical properties more sensitive to axial stress. When axial pressure > confining pressure, the high axial stress significantly enhances the rock's compressive strength in the axial direction, thus requiring a higher pump pressure to fracture. Conversely, when axial pressure < confining pressure, the increased circumferential stress has a relatively weaker restraining effect on the sandstone, resulting in a lower fracturing pump pressure than when axial pressure > confining pressure.

[0063] This pattern aligns with the minimum principal stress theory in rock mechanics, according to the HF criterion, i.e. P f =3 σ 3- σ 1+ T 0- α p p ,in, P f Indicates the rupture pressure; σ 1 indicates axial compression; σ 3 indicates confining pressure; T 0 represents the tensile strength of the rock; α Indicates the porosity elastic coefficient; p p This represents pore pressure. The fracturing pressure of rock is mainly controlled by the minimum principal stress. When the axial pressure < confining pressure, the confining pressure becomes the maximum principal stress, and the axial pressure becomes the minimum principal stress. At this time, the circumferential constraint on the rock is enhanced, and a higher pumping pressure is required to overcome the constraint and cause the rock to fracture. When the axial pressure > confining pressure, the axial pressure becomes the maximum principal stress, and the confining pressure becomes the minimum principal stress. The axial constraint is enhanced, but since the direction of the axial pressure is consistent with the direction of fluid injection, it offsets part of the constraint effect to a certain extent, making the fracturing pumping pressure lower than when the axial pressure < confining pressure. When the axial pressure = confining pressure, the rock is in an isotropic stress state, and the constraint effect is relatively weak, so the fracturing pumping pressure is the lowest.

[0064] from Figure 8 Comparing the different fluid types in (b), in the Q1 shale layer, when liquid CO2 was injected, the fracturing pump pressure was 38.76 MPa for group P11, 39.5 MPa for group P13, and 39.25 MPa for group P15; when clear water was injected, the fracturing pump pressure was 39.18 MPa for group P12, 40.43 MPa for group P14, and 39.53 MPa for group P16. The fracturing pump pressure variation trends of the two fluids under different stress conditions are similar to those of supercritical CO2, both being slightly higher when axial pressure > confining pressure and lower when axial pressure < confining pressure, and the fracturing pump pressure of clear water is generally higher than that of liquid CO2.

[0065] Analysis of Experimental Results of Supercritical CO2 Fracture Induction at Different Injection Rates: To investigate the effect of injection rate on core fracture characteristics, experiments were conducted under the conditions of axial pressure = confining pressure and axial pressure / confining pressure = 10 MPa / 10 MPa. Three injection rates of 2 mL / min, 4 mL / min, and 6 mL / min were set, and supercritical CO2 was injected into Q9 shale, Q1 shale, and tight sandstone, respectively. The fracture pump pressure was measured to clarify the relationship between injection rate and fracture pressure. Figure 9 Overall, the experimental results show that the injection rate has a significant impact on the fracture pump pressure of different core samples. The low-permeability Q1 shale and tight sandstone exhibit a "single-tone increase in pump pressure as injection rate increases" pattern; while the high-permeability Q9 shale shows a "V-shaped trend of first decreasing and then increasing" pattern.

[0066] like Figure 10 (a) For the Q9 shale, the fracture pump pressure was 33.91 MPa when the injection flow rate of P4 was 2 mL / min, the fracture pump pressure of P1 dropped to 32.91 MPa when the injection flow rate was 4 mL / min, and the fracture pump pressure of P5 rose to 34.84 MPa when the injection flow rate was 6 mL / min, showing a V-shaped trend of first decreasing and then increasing. This may be related to the pore structure of the Q9 shale. The Q9 shale may have more natural fractures and high-permeability bedding. When the injection flow rate is low, as low as 2 mL / min, supercritical CO2 has enough time to penetrate along the natural fractures and bedding, resulting in a relatively uniform pore pressure distribution and a high fracturing pump pressure. When the injection flow rate increases to 4 mL / min, the fluid begins to accumulate inside the core while penetrating, the pore pressure gradient is moderate, and the expansion of natural fractures and the generation of new fractures reach equilibrium, causing the fracturing pump pressure to drop to the minimum. When the injection flow rate further increases to 6 mL / min, the fluid accumulation rate exceeds the penetration rate, the pore pressure rises rapidly, and the fracturing pump pressure increases.

[0067] For the Q1 shale, the fracturing pump pressure was 31.09 MPa at an injection flow rate of 2 mL / min for group P9, 34.96 MPa for group P8 at 4 mL / min, and increased to 38.72 MPa for group P10 at 6 mL / min, showing a significant increase with increasing injection flow rate. This is because at higher injection flow rates, the accumulation rate of supercritical CO2 inside the core is faster than its permeation and diffusion rate, leading to a rapid increase in pore pressure and a rapid decrease in the effective stress of the rock, thus reaching the fracturing condition in a shorter time, manifested as an increase in fracturing pump pressure. Furthermore, at high injection flow rates, the impact of the fluid on the core is enhanced, which may exacerbate the propagation of micro-fractures inside the rock, further promoting fracturing. However, due to the excessively rapid pressure accumulation, the overall fracturing pump pressure is actually higher.

[0068] like Figure 10 In (b), the experimental results for the tight sandstone in groups S1, S4, and S5 show a similar trend to those for the Q1 shale. The fracturing pump pressure in group S4 was 33.27 MPa at an injection flow rate of 2 mL / min, in group S1 it was 34.12 MPa at an injection flow rate of 4 mL / min, and in group S5 it increased to 38.45 MPa at an injection flow rate of 6 mL / min, monotonically increasing with increasing injection flow rate. This is because tight sandstone has low porosity and poor permeability, resulting in weak permeation and diffusion of supercritical CO2 within it. Regardless of the injection flow rate, the injected fluid mainly accumulates; the higher the injection flow rate, the faster the pressure accumulates, and the higher the fracturing pump pressure. Compared to shale, the fracture pump pressure of tight sandstone varies more significantly with the injection rate. The fracture pump pressure at 6 mL / min is 5.18 MPa higher than that at 2 mL / min. The difference is 7.63 MPa for the Q1 shale and 0.93 MPa for the Q9 shale. This indicates that the better the permeability of the core type, the more complex the effect of the injection rate on the fracture pump pressure.

[0069] Comparing different core types, at the same injection flow rate, the fracturing pump pressure of tight sandstone is similar to or slightly higher than that of shale. For example, at 4 mL / min, the fracturing pump pressure of group S1 (34.12 MPa) is close to that of group P1 (32.91 MPa) and group P8 (34.96 MPa). At 6 mL / min, the fracturing pump pressure of group S5 (38.45 MPa) is lower than that of shale P10 in layer Q1 (38.72 MPa) but higher than that of group P5 in layer Q9 (34.84 MPa). This indicates that the influence of core type on fracturing pump pressure is interactive with the injection rate. At low injection flow rates, the fracturing pump pressure of shale may be higher or lower than that of sandstone, while at high injection flow rates, the fracturing pump pressure of shale in layer Q1 may be higher than that of sandstone.

[0070] Step 2: Convert the core type and injected fluid type into dummy variables; based on the dummy variables of core type and injected fluid type, establish a multiple linear regression model with fracturing pump pressure as the dependent variable and the differences between core type, injected fluid type, injection velocity, and axial pressure and confining pressure as independent variables; fit the established multiple linear regression model using the least squares method to minimize the sum of squared errors between all observed and predicted values, and obtain the regression coefficient of each independent variable; perform a t-test on the regression coefficient of each independent variable to evaluate the significance of each independent variable on fracturing pump pressure, obtain the p-value of each independent variable, and screen out independent variables with p-value ≤ 0.2; use independent variables with p-value ≤ 0.2 as experimental conditions to conduct supercritical CO2 phase stress-coordinated fracturing.

[0071] Multivariate Analysis of Fracturing Pump Pressure: The experiment collected data on various variables, including injection velocity, core type, injected fluid type, and axial / confining pressure conditions. By establishing a multiple linear regression model for regression analysis, the influence of various variables in the experiment can be quantified, which has certain application value for optimizing fracturing effects and improving fracturing efficiency. Before conducting the regression analysis, the data needs to be processed to ensure it meets the requirements. The experimental data includes both qualitative variables, such as core type and injected fluid type, and quantitative variables, such as injection velocity and axial / confining pressure. Qualitative variables need to be converted into dummy variables so that the regression analysis can identify their influence on the fracturing pump pressure. In the dummy variable encoding, to avoid "perfect collinearity" that would prevent the regression model from being solved, a baseline group needs to be set. In this experiment, there are two qualitative variables: core type and injected fluid type. The Q9 shale core type and the clear water injected fluid type are set as the baseline groups. No dummy variables are assigned, and the correlation coefficients obtained can be used to represent the increase or decrease effect relative to the Q9 shale or clear water.

[0072] The method for converting core type and injected fluid type into dummy variables is to use core type and injected fluid type as qualitative variables and convert them into dummy variables. Specifically, the method is as follows: a baseline group is set, and no dummy variables are assigned to the baseline group; dummy variables are set for the qualitative variables that exclude the baseline group; the method for setting dummy variables for the qualitative variables that exclude the baseline group is as follows: when the corresponding core type is selected, the dummy variable is set to 1, otherwise it is 0; when the corresponding injected fluid type is selected, the dummy variable is set to 1, otherwise it is 0.

[0073] Qualitative variables primarily include core type and injected fluid type. In the regression processing of the multivariate analysis, core types include Q9 shale, Q1 shale, Q2 shale, and sandstone, with Q9 shale set as the baseline group. Injected fluid types include clean water, liquid CO2, and supercritical CO2, with clean water set as the baseline group. A dummy variable is assigned to each core type; for example, for Q1 shale, the dummy variable representing Q1 shale is set to 1, otherwise 0. Injected fluid types include water, liquid CO2, and supercritical CO2. A dummy variable is assigned to each fluid type. For example, for supercritical CO2, the dummy variable is 1 if this fluid is used, otherwise 0; water and liquid CO2 are coded in the same way.

[0074] The quantitative variables mainly include axial pressure / confining pressure and injected fluid velocity. These two quantitative variables are specific values, making direct regression fitting possible. With the fracture pump pressure as the dependent variable and the injection velocity, core type, injected fluid type, and the difference between axial pressure and confining pressure as independent variables, the purpose of establishing the regression model is to explore the relationship between the dependent variable and multiple independent variables. A multiple linear regression model can be used to establish the regression equation, as follows: .

[0075] in, P 裂 It is the pressure of the rupture pump; X 流速 It is the injection flow rate; X Q1 It is a dummy variable for the Q1 shale layer; X Q2 It is a dummy variable for the Q2 shale layer; X 砂岩 It is a dummy variable for sandstone; X 轴压 / 围压 It is the difference between axial pressure and confining pressure; X SC-CO2 It is a dummy variable for supercritical CO2; X 液态CO2 It is a dummy variable for liquid CO2; This is the error term, representing the random error in the model. β 0、 β 1. β 2. β 3. β 4. β 5. β 6 and β 7 represent the regression coefficients of the corresponding independent variables. X Q1 , X Q2 , X砂岩 These are dummy variables corresponding to the core type; X SC-CO2 , X 液态CO2 These are dummy variables corresponding to the fluid type. In this invention, the error involves, on the one hand, the error caused by variables not considered in the experimental design, and on the other hand, the error caused by instruments and measurements during the experiment.

[0076] The multiple linear regression model was fitted using the least squares method, minimizing the sum of squared errors between all observed and predicted values. This method yields the regression coefficients for each independent variable, reflecting their specific impact on the rupture pump pressure. Regression analysis was performed using the statsmodels library in Python based on experimental data. The results of the regression analysis are shown in Table 3.

[0077] Table 3 Regression Analysis Results

[0078] The magnitude and sign of the regression coefficients indicate the specific impact of each independent variable on the rupture pump pressure. The regression coefficient of the intercept is... β 0 = 18.3971, which is the baseline value for the multiple linear regression model, indicating that when all independent variables are 0, the baseline value for the rupture pump pressure is 18.3971 MPa. The regression coefficient for the injection flow rate is... β The p-value of 1 = 0.3016 indicates that for every 1 mL / min increase in injection flow rate, the fracturing pump pressure increases by 0.3016 MPa. However, the p-value for injection flow rate is 0.759, much greater than the significance level of 0.05, indicating that the effect of injection flow rate on fracturing pump pressure is not significant in this experiment. Compared with shale layer Q9, the p-value of layer Q1 is 0.705, much greater than 0.05, indicating that layer Q1 shale has no significant effect on fracturing pump pressure. Compared with shale layer Q9, the regression coefficient of layer Q2 shale is... β The regression coefficient 3 = -4.4718, but its p-value is 0.181, indicating that the Q2 shale layer has some influence on the fracturing pump pressure, but this influence is not significant under the current experimental conditions. The regression coefficient of the tight sandstone is... β The value of 4 = 0.4520 indicates that compared to the Q9 shale layer, the tight sandstone slightly increases the fracturing pump pressure, but the p-value of 0.891 shows that the effect of the tight sandstone on the fracturing pump pressure is relatively weak and not significant in this experiment. The difference between axial pressure and confining pressure has a more significant effect on the fracturing pump pressure, with a regression coefficient of [missing value]. β The regression coefficient for supercritical CO2 was 0.9412, with a p-value of 0.067, close to the significance level of 0.05. This indicates that a large difference between axial pressure and confining pressure may lead to a higher rupture pump pressure, and changes in stress conditions can have a significant impact on rupture pressure.β The coefficient 6 = 6.0581 indicates that when the fluid is supercritical CO2, the fracturing pump pressure changes by approximately 6.06 MPa compared to other fluid types, resulting in better fracturing initiation and propagation. The regression coefficient for liquid CO2 is... β The p-value of 7 = 12.3390 and the p-value of 0.023 indicate that liquid CO2 has the most significant impact on the fracturing pump pressure. This is the factor with the greatest influence on the fracturing pump pressure in this experiment, indicating that liquid CO2 has high permeability and fracture propagation capacity during fracturing. Although supercritical CO2 showed some influence in this experiment, with a p-value of 0.113, its influence was slightly smaller than that of liquid CO2.

[0079] like Figure 11 The regression analysis results provide several statistics to evaluate the performance of the multiple linear regression model. 2 A p-value of 0.953 indicates that the multiple linear regression model can explain 95.3% of the rupture pump pressure variation. This demonstrates that the multiple linear regression model fits the data very well and can accurately reflect the impact of various independent variables on the rupture pump pressure. A t-test of the regression coefficients can assess the significance of each independent variable on the rupture pump pressure. Independent variables with p-values ​​≤ 0.2, such as liquid CO2 fluid and supercritical CO2 fluid, indicate that these factors have a significant impact on the rupture pump pressure. Independent variables with p-values ​​> 0.5, such as injection velocity, Q1 shale layer, and sandstone, indicate that these factors have no significant impact on the rupture pump pressure under the experimental conditions.

[0080] In summary, the SC-CO2 phase stress-coordinated fracturing method adapted to core type provided by the embodiments of the present invention couples core type identification, phase locking of fracturing medium, stress control and injection flow rate optimization. The phase calibration of the injected fluid is completed on the ground, and the geostress path is matched according to the characteristics of the core sample. The controlled initiation and propagation of fractures are achieved by step injection.

[0081] To quantify the impact of injected fluid type on fracturing pump pressure, experiments were conducted under the same geostress condition of axial pressure / confining pressure = 10 MPa / 10 MPa. The phase control parameters for supercritical CO2 were 60℃ / 10 MPa, and for liquid CO2 and clear water, they were 20℃ / 5 MPa, ensuring the fracturing medium was in the target phase before injection to avoid runaway fluctuations caused by phase transitions. The fracturing pump pressure of liquid CO2 increased by approximately 7.40%–27.75% compared to supercritical CO2, while the fracturing pump pressure of clear water increased by approximately 10.31%–30.42% compared to supercritical CO2. The increase in pressure was generally higher in tight sandstone than in shale. The results indicate that low-permeability core types are more sensitive to the type of injected fluid.

[0082] To quantify the impact of core type on fracturing pump pressure, lamellar shale was subdivided into Q1, Q2, and Q9 layers. Under the same operating conditions and fracturing medium, the control parameters were set as follows: supercritical CO2, injection flow rate of 4 mL / min, and axial pressure / confining pressure = 10 MPa / 10 MPa. The results showed that the Q9 layer had the lowest fracturing pump pressure at 32.91 MPa; the Q2 layer had the second lowest at 34.73 MPa; and the Q1 layer had the highest at 34.96 MPa, forming a low, medium, and high fracturing pump pressure gradient. This invention transforms the high porosity and permeability advantages of the Q9 layer into a lower baseline fracturing pressure, and based on this, provides differentiated fracturing operation recommendations for lamellar shale with different core layers, significantly improving the adaptability and stability of the scheme in reservoirs with multiple core types.

[0083] Stress and phase state work together to achieve controllable fracture initiation and fracture guidance. Under supercritical CO2 injection conditions, the Q9 and Q2 shale layers exhibit consistent stress sensitivity patterns: the fracture pump pressure is highest when axial pressure < confining pressure, followed by the fracture pump pressure when axial pressure > confining pressure, and lowest when axial pressure = confining pressure. Taking the Q2 shale layer as an example, the fracture pump pressures for axial pressure = confining pressure, axial pressure > confining pressure, and axial pressure < confining pressure are 34.73 MPa, 40.20 MPa, and 45.09 MPa, respectively. This invention, by predicting the stress field during the design phase and linking the supercritical CO2 phase state with the injection strategy, achieves a target combination of minimum fracture initiation and stable fracture propagation. In dense sandstone, the fracture pump pressure exhibits a different pattern: the highest when axial pressure > confining pressure, the second highest when axial pressure < confining pressure, and the lowest when axial pressure = confining pressure. Specifically, the fracture pump pressure is 34.12 MPa when axial pressure = confining pressure, 39.94 MPa when axial pressure < confining pressure, and 36.90 MPa when axial pressure > confining pressure. Based on this, the embodiments of the present invention employ differentiated loading paths and phase windows in sandstone to avoid relying solely on shale experience and improve fracture controllability and guidance.

[0084] The injection flow rate showed a monotonically positive correlation with the fracturing pump pressure. At an axial pressure / confining pressure of 10 MPa / 10 MPa, the fracturing pump pressure of the Q1 shale layer increased monotonically with increasing injection flow rate, as did that of the tight sandstone. At an injection flow rate of 2 mL / min... -1 4 mL·min -1 6 mL·min -1Below, the fracturing pump pressures of the Q1 shale were 31.09 MPa, 34.96 MPa, and 38.72 MPa, respectively, while those of the tight sandstone were 33.27 MPa, 34.12 MPa, and 38.45 MPa, respectively. Based on this, the present invention sets an upper limit for the injection flow rate and a stepped acceleration strategy. This utilizes the higher injection flow rate pressure difference to drive accelerated fracture penetration while avoiding excessive pressure accumulation that could lead to increased initiation pressure and instability. This achieves dynamic optimization of low-threshold fracturing initiation, sustainable propagation, and controllable complexity across different core types. Cross-core type comparisons show that the fracturing pump pressures of sandstone and shale are similar but differentiated at the same injection flow rate, suggesting an interaction between the injection flow rate effect and core type. Based on this, the present invention establishes a two-dimensional parameter table of core type × flow rate, providing a theoretical basis for parameter adjustment.

[0085] Mechanistic studies show that supercritical CO2 has low viscosity and high diffusion. Its diffusion coefficient is approximately three times that of liquid CO2, resulting in faster pore pressure transmission and a wider effective radius of action. Clear water, due to the hydration of clay minerals, temporarily increases the cementation strength of bedding planes, raising the initiation pressure. This invention incorporates four factors—core type, injected fluid type, difference between axial and confining pressure, and injection velocity—into a multiple linear regression model. This model quantifies the contribution weight of each independent variable to the fracturing pump pressure, supporting rapid cross-block migration and prior parameter setting, such as the positive contribution of the axial / confining pressure difference to the fracturing pump pressure and statistical test results of the overall influence of injection velocity.

[0086] Experimental data show that when using supercritical CO2 as the fracturing medium, the initiation pressure of supercritical CO2 is significantly lower than that of traditional water-based fracturing fluids under different core types, and it is adjustable under different reservoir core types, thereby optimizing the fracturing process and improving oil and gas recovery. Under the same stress conditions, the initiation pressure of supercritical CO2 is 15%–30% lower than that of water-based fluids, and the difference is more significant as reservoir permeability decreases. The initiation pressure of shale oil reservoirs is lower than that of tight sandstone, indicating that supercritical CO2 has better permeability and fracture propagation in low-permeability strata. The initiation pressure of supercritical CO2 in shale oil reservoirs is 32.91 MPa, while that of water-based fluids is 40.20 MPa, a reduction of approximately 18.3%; the initiation pressure of supercritical CO2 in tight sandstone is 34.12 MPa, while that of water-based fluids is 39.94 MPa, a reduction of approximately 14.7%.

[0087] Under different stress conditions, the fracturing initiation pressure of supercritical CO2 is generally lower than that of water-based fluids. When the axial pressure is greater than the confining pressure, the fracturing initiation pressure of supercritical CO2 is lower, indicating that it is more likely to promote fracture formation; when the axial pressure is less than the confining pressure, the fracturing initiation pressure of supercritical CO2 decreases further, and its advantage becomes more prominent. Supercritical CO2 fluid can adjust its fracturing initiation pressure according to changes in formation stress to achieve optimal fracturing effect.

[0088] Supercritical CO2 can significantly reduce fracturing pressure, mainly due to its fluidity and solubility in rocks. The low viscosity and high diffusivity of supercritical CO2 make it easier to penetrate microfractures and induce propagation. Fluid phase regulation optimizes fracturing pressure under different reservoir conditions. Supercritical CO2 can react with clay minerals in the rock formation, altering bedding structure or cementation strength and reducing fracturing strength. Compared to traditional water-based fracturing fluids, the use of supercritical CO2 fluid significantly reduces fracturing pressure under different experimental conditions and can adapt to different reservoir core types and stress conditions, demonstrating good adaptability and stability. Therefore, the supercritical CO2 fluid phase regulation and stress condition optimization method proposed in this invention can significantly improve the efficiency of unconventional reservoir development, reduce energy consumption, and has good environmental friendliness and application prospects.

[0089] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A core type-adaptive SC-CO2 phase stress-coordinated fracturing method, characterized in that, Includes the following steps: Conduct fracturing experiments under multiple variable conditions to obtain the fracturing pump pressure under multiple variable conditions; The variable conditions are core type and injected fluid type, injection velocity, and axial pressure and confining pressure conditions; Convert the core type and injected fluid type into dummy variables; Based on the dummy variables of core type and injected fluid type, with fracture pump pressure as the dependent variable and the differences between core type, injected fluid type, injection velocity, axial pressure and confining pressure as independent variables, a multiple linear regression model is established. Based on the established multiple linear regression model, the least squares method is used for fitting, minimizing the sum of squared errors between all observed and predicted values, and obtaining the regression coefficient of each independent variable; A t-test was performed on the regression coefficient of each independent variable to assess the significance of each independent variable on the rupture pump pressure, obtain the p-value of each independent variable, and screen out independent variables with p-values ​​≤ 0.2; Supercritical CO2 phase stress-coordinated fracturing was carried out under experimental conditions with an independent variable having a p-value ≤ 0.

2.

2. The SC-CO2 phase stress-coordinated fracturing method adapted to core type as described in claim 1, characterized in that, The core samples are composed of Q9 shale, Q1 shale, Q2 shale, and sandstone; the injected fluids are water, liquid CO2, and supercritical CO2.

3. The SC-CO2 phase stress-coordinated fracturing method adapted to core type as described in claim 2, characterized in that, Shale layers Q9, Q1, and Q2 are shale layers at different strata in the geological profile, and the geological profile is dominated by lamellar shale.

4. The SC-CO2 phase stress-coordinated fracturing method adapted to core type as described in claim 2, characterized in that, The method for converting core type and injected fluid type into dummy variables is to use core type and injected fluid type as qualitative variables and convert the qualitative variables into dummy variables.

5. The SC-CO2 phase stress-coordinated fracturing method adapted to core type as described in claim 4, characterized in that, The method for converting core type and injected fluid type into dummy variables is as follows: Set up a baseline group, which is not assigned any dummy variables; set dummy variables for qualitative variables that exclude the baseline group. The method for setting dummy variables for qualitative variables excluding the baseline group is as follows: When the corresponding core type is selected, the dummy variable is set to 1; otherwise, it is set to 0. When the corresponding injection fluid type is selected, the dummy variable is set to 1; otherwise, it is set to 0.

6. The SC-CO2 phase stress-coordinated fracturing method adapted to core type as described in claim 1, characterized in that, The multiple linear regression model is as follows: ； in, P 裂 It is the pressure of the rupture pump; X 流速 It is the injection flow rate; X Q1 It is a dummy variable for the Q1 shale layer; X Q2 It is a dummy variable for the Q2 shale layer; X 砂岩 It is a dummy variable for sandstone; X 轴压 / 围压 It is the difference between axial pressure and confining pressure; X SC-CO2 It is a dummy variable for supercritical CO2; X 液态CO2 It is a dummy variable for liquid CO2; It is an error; β 0、 β 1. β 2. β 3. β 4. β 5. β 6 and β 7 represent the regression coefficients of the corresponding independent variables.

7. The SC-CO2 phase stress-coordinated fracturing method adapted to core type as described in claim 1, characterized in that, Axial pressure is 10MPa~15MPa; confining pressure is 10MPa~15MPa.

8. The SC-CO2 phase stress-coordinated fracturing method adapted to core type as described in claim 1, characterized in that, The injection flow rate is 0.4 mL / min to 6 mL / min.

Images