A three-dimensional physical simulation experiment method for carbon dioxide geological storage reservoir-cap system

Abstract

The present application provides a kind of three-dimensional physical simulation experiment method for carbon dioxide geological storage reservoir-cap system, belong to physical simulation technical field, the present application is by setting hierarchical measuring point and interface control layer in three-dimensional reservoir-cap physical model, using pressure rise rate contrast determination interface breakthrough, utilize pressure field space gradient matrix to identify preferential percolation channel, by multi-index fusion algorithm comprehensively capillary breakthrough, seepage breakthrough and macro breakthrough three levels criterion, in combination with history matching inversion method dynamic adjustment model parameter, establish breakthrough characteristic parameter set and realize closed capacity quantitative grading, solved the technical problem that cannot accurately identify the multilevel process of carbon dioxide breakthrough cap in three-dimensional physical simulation experiment and quantitatively evaluate the sealing performance.

Description

Technical Field

[0001] This invention belongs to the field of physical simulation technology, and more specifically, relates to a three-dimensional physical simulation experimental method for a carbon dioxide geological burial layer-caprock system. Background Technology

[0002] Geological storage of carbon dioxide is a crucial technological means to achieve carbon emission reduction. Its safety depends on the effective sealing capacity of the caprock. Traditional physical simulation experiments mainly employ one-dimensional or two-dimensional core displacement devices, determining caprock breach by measuring pressure changes and the time of gas emergence at the produced end. Current technologies lack precise monitoring methods for fluid intrusion at the reservoir-caprock interface, making it impossible to accurately capture the gradual evolution of carbon dioxide from capillary to macroscopic breach. Existing methods typically rely solely on the detection of gas at the produced end as a breach criterion, neglecting the intermediate processes of pressure field evolution within the caprock and the formation of preferential seepage channels, making it difficult to establish a quantitative relationship between breach characteristic parameters and sealing capacity. In other words, existing technologies suffer from the technical problem of failing to accurately identify the multi-level processes of carbon dioxide breaching the caprock and quantitatively evaluate sealing performance in three-dimensional physical simulation experiments. Summary of the Invention

[0003] In view of this, the present invention provides a three-dimensional physical simulation experimental method for a carbon dioxide geological buried storage layer-caprock system, which can solve the technical problem in the prior art that it is impossible to accurately identify the multi-level process of carbon dioxide breaking through the caprock and quantitatively evaluate the sealing performance in a three-dimensional physical simulation experiment.

[0004] This invention is implemented as follows: It provides a three-dimensional physical simulation experimental method for a carbon dioxide geological buried reservoir-caprock system, comprising: preparing a three-dimensional reservoir-caprock physical model; filling and compacting the experimental chamber in layers according to the order of lower reservoir zone, interface control layer, and upper caprock zone; setting reservoir zone measuring points, interface zone measuring points, and caprock zone measuring points on the sealed cap; after assembly, placing the three-dimensional reservoir-caprock physical model in a constant temperature chamber and applying confining pressure; after vacuuming the three-dimensional reservoir-caprock physical model, injecting simulated formation brine to saturation; injecting simulated formation brine at a constant flow rate using an injection metering pump and recording the pressure at each measuring point; calculating the initial permeability of the reservoir zone and the initial permeability of the caprock zone; establishing baseline seepage parameters; injecting carbon dioxide into the reservoir zone in constant flow or constant pressure mode; and collecting real-time data on the pressure at the reservoir zone measuring points and the interface zone. The pressure at the measuring points in the interface zone and the caprock zone is used to determine when carbon dioxide begins to invade the caprock. Carbon dioxide is continuously injected, and the pressure time series at each measuring point in the caprock zone is monitored. The spatial gradient matrix of the pressure field in the caprock zone is calculated to identify preferred seepage channels. When a continuous high-gradient channel forms and the gas flow rate at the production end exceeds 0.01 mL / min and remains stable, a macroscopic breakthrough is determined. The pressure at each measuring point and the cumulative injection volume at the moment of macroscopic breakthrough are recorded. The breakthrough pressure differential and invasion height are calculated, and a set of breakthrough characteristic parameters is established. Based on the pressure time series at each measuring point, the cumulative injection volume, and the cumulative production volume, the relative permeability model parameters and capillary pressure model parameters are adjusted using a historical fitting inversion method. The breakthrough confidence distribution is calculated using a multi-index fusion breakthrough judgment algorithm, and the dynamic sealing performance evaluation results are output.

[0005] The interface control layer is used to stabilize the geometry of the contact interface between the reservoir region and the caprock region. The material includes a microporous filter membrane, a sheet material, or a roughened interface layer, with a thickness of 0.5 μm to 2 mm and a pore size of 10 μm to 100 μm. Its function is to reduce the fluid bypass effect at the contact interface and improve the repeatability of the carbon dioxide intrusion process.

[0006] The baseline seepage parameters include the initial permeability of the reservoir zone and the initial permeability of the caprock zone, which are calculated using Darcy's law. The inputs to the calculation equation for the initial permeability of the reservoir zone include the injection flow rate, the simulated formation brine viscosity, the reservoir zone length, and the pressure difference at the measuring points in the reservoir zone. The inputs to the calculation equation for the initial permeability of the caprock zone include the injection flow rate, the simulated formation brine viscosity, the caprock zone length, and the pressure difference at the measuring points in the caprock zone.

[0007] The pressure rise rate at the reservoir measurement point is obtained by performing time difference calculation on the pressure time series of the reservoir measurement point, and the pressure rise rate at the interface measurement point is obtained by performing time difference calculation on the pressure time series of the interface measurement point. When the pressure rise rate at the interface measurement point exceeds twice the pressure rise rate at the reservoir measurement point, it indicates that carbon dioxide has broken through the top of the reservoir and begun to enter the caprock.

[0008] The intrusion height refers to the maximum vertical expansion distance of carbon dioxide after entering the caprock. The calculation method is to determine the highest position reached by the pressure response front by using linear interpolation based on the correspondence between the pressure response time and the vertical position of each measuring point in the caprock.

[0009] The spatial gradient matrix of the pressure field in the caprock is used to identify the preferred seepage channels within the caprock. The calculation method is to perform spatial difference calculation on the pressure values ​​of each measuring point in the caprock at the same time to obtain the horizontal gradient and the vertical gradient. When the gradient value of a certain area is continuously greater than 3 times the average gradient value, it is determined that there is a continuous high gradient channel in the area.

[0010] The macroscopic breakthrough moment refers to the moment when the gas flow rate at the extraction end exceeds 0.01 mL / min and remains stable. The macroscopic breakthrough moment is a key time node for evaluating the failure of the caprock sealing performance and is used to determine the breakthrough pressure difference and the breakthrough characteristic parameter set.

[0011] The breakthrough pressure difference refers to the difference between the pressure at the interface zone measuring point and the initial caprock top pressure at the moment of macroscopic breakthrough. When the breakthrough pressure difference is greater than 5 MPa, the caprock zone is judged to have high sealing performance. When the breakthrough pressure difference is between 2 MPa and 5 MPa, the caprock zone is judged to have medium sealing performance. When the breakthrough pressure difference is less than 2 MPa, the caprock zone is judged to have weak sealing performance.

[0012] The breakthrough characteristic parameter set includes macroscopic breakthrough time, breakthrough pressure difference, invasion height, cumulative injection volume, produced gas-liquid ratio, and cumulative produced volume, which are used to quantitatively characterize the migration and breakthrough behavior of carbon dioxide in the reservoir-caprock system.

[0013] The historical fitting and inversion method employs a parameterized relative permeability model and a capillary pressure model. The objective functions are the pressure time series of reservoir area measuring points, the pressure time series of interface area measuring points, the pressure time series of caprock area measuring points, the cumulative injection volume, and the produced gas-liquid ratio. An ensemble Kalman filter algorithm is used to adjust the model parameters online.

[0014] The relative permeability model parameters include residual water saturation, residual gas saturation, and pore connectivity index, while the capillary pressure model parameters include inlet pressure, pore size distribution index, and maximum capillary pressure.

[0015] The multi-index fusion breakthrough judgment algorithm integrates three levels of judgment criteria: capillary breakthrough, seepage breakthrough, and macroscopic breakthrough. The capillary breakthrough criterion is that the pressure at the top of the caprock reaches the capillary entry threshold. The seepage breakthrough criterion is that the pressure gradient inside the caprock is established and reaches the Darcy flow threshold. The macroscopic breakthrough criterion is that the gas phase flow rate at the extraction end exceeds 0.01 mL / min and remains stable.

[0016] The multi-index fusion breakthrough judgment algorithm uses a Bayesian change point detection algorithm to perform time series analysis on the produced gas-liquid ratio signal, identify the moment of sudden change, calculate the confidence level of breakthrough at each level, and obtain the breakthrough confidence distribution.

[0017] The dynamic sealing performance evaluation results include the sealing capacity level of the caprock, breakthrough risk assessment, and upper limit of safe injection pressure. The sealing capacity level of the caprock is divided into three levels: high sealing, medium sealing, and low sealing, based on the breakthrough pressure difference and intrusion height.

[0018] The upper limit of the safe injection pressure is the maximum injection pressure that ensures carbon dioxide does not exceed the caprock zone, and 80% of the pressure difference exceeding the caprock is taken as the upper limit of the safe injection pressure.

[0019] The three-dimensional reservoir-caprock physical model consists of a pressure-bearing shell, a sealing cap, an experimental chamber, fastening screws, sealing screws, inlet and outlet ports, a confining pressure isolation membrane, and a rotating support base. The experimental chamber has dimensions of 40cm in length, 40cm in width, and 20cm in height. Inside the experimental chamber, a partitioned limiting structure forms a superimposed structure of the lower reservoir region and the upper caprock region.

[0020] This invention achieves layered monitoring of the carbon dioxide intrusion process by setting monitoring points in the reservoir, interface, and caprock zones within a three-dimensional reservoir-caprock physical model, combined with a stable contact interface controlled by an interface control layer. This solves the problem of existing technologies being unable to accurately capture the interface breakthrough process. This invention uses pressure rise rate comparison to determine interface breakthrough, utilizes the spatial gradient matrix of the pressure field to identify preferential seepage channels, and establishes a breakthrough confidence distribution by integrating three levels of judgment criteria: capillary breakthrough, seepage breakthrough, and macroscopic breakthrough, overcoming the limitations of a single criterion. This invention dynamically adjusts the relative permeability and capillary pressure model parameters through a history fitting inversion method, and achieves quantitative and graded evaluation of sealing capacity by combining a breakthrough feature parameter set. In summary, this invention solves the technical problem mentioned in the background art of being unable to accurately identify the multi-level process of carbon dioxide breakthrough in the caprock and quantitatively evaluate sealing performance in three-dimensional physical simulation experiments. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method of the present invention.

[0022] Figure 2 This is a schematic diagram of the overall structure of a three-dimensional physical simulation experimental device.

[0023] Figure 3 This is a front view of the structure of a three-dimensional reservoir-caprock physical model.

[0024] Figure 4 This is a top view of the structure of a three-dimensional reservoir-caprock physical model.

[0025] Figure 5 This is a cross-sectional view of a three-dimensional reservoir-caprock physical model.

[0026] The reference numerals in the attached figures are explained as follows: 1. Injection metering pump; 2. Confining pressure loading pump; 3. Carbon dioxide gas cylinder; 4. Vacuum pump; 5. Valve; 6. First intermediate container; 7. Second intermediate container; 8. Constant temperature chamber; 9. Three-dimensional reservoir-caprock physical model; 10. Temperature sensor; 11. Gas-liquid separation and production collector; 12. Computer; 13. Pressure sensor; 14. Sealing cap; 15. Rotary support base; 16. Pressure-bearing outer shell; 17. Three-dimensional experimental chamber; 18. Fastening screw; 19. Sealing screw; 20. Interface control layer; 21. Pressure measuring point; 22. Inlet port; 23. Outlet port; 24. Caprock medium; 25. Reservoir medium. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below.

[0028] like Figure 1 The diagram shows a flowchart of a three-dimensional physical simulation experimental method for a carbon dioxide geological burial reservoir-caprock system provided by the present invention. This method includes the following steps:

[0029] S1. Prepare a three-dimensional reservoir-caprock physical model. Fill and compact the experimental chamber in layers according to the order of lower reservoir zone, interface control layer and upper caprock zone. Set the reservoir zone measuring points, interface zone measuring points and caprock zone measuring points on the sealing cap. After the assembly is completed, place the three-dimensional reservoir-caprock physical model in a constant temperature chamber and apply confining pressure.

[0030] S2. Vacuum treatment is performed on the three-dimensional reservoir-caprock physical model, followed by injection of simulated formation brine to saturation. Simulated formation brine is injected at a constant flow rate using an injection metering pump, and the pressure at each measuring point is recorded. The initial permeability of the reservoir area and the initial permeability of the caprock area are calculated, and baseline seepage parameters are established.

[0031] S3. Inject carbon dioxide into the reservoir area in constant flow or constant pressure mode, and collect the pressure at the reservoir area measuring point, the interface area measuring point, and the caprock area measuring point in real time. When the pressure rise rate at the interface area measuring point exceeds twice the pressure rise rate at the reservoir area measuring point, it is determined that carbon dioxide has begun to invade the caprock.

[0032] S4. Continuously inject carbon dioxide and monitor the pressure time series at each measuring point in the caprock area. Calculate the spatial gradient matrix of the pressure field in the caprock area. When a continuous high gradient channel is formed and the gas flow rate at the extraction end exceeds 0.01 mL / min and remains stable, it is judged as a macroscopic breakthrough.

[0033] S5. Record the reservoir zone measuring point pressure, interface zone measuring point pressure, caprock zone measuring point pressure and cumulative injection volume at the moment of macroscopic breakthrough, calculate the breakthrough pressure difference and invasion height, and use gas-liquid separator and production collector to measure the produced gas-liquid ratio and cumulative production volume in real time, and establish a set of breakthrough characteristic parameters.

[0034] S6. Based on the time series of pressure at measuring points in the reservoir area, the interface area, and the caprock area, as well as the cumulative injection and production, the relative permeability model parameters and capillary pressure model parameters are adjusted using the historical fitting inversion method. The breakthrough confidence distribution is calculated through a multi-index fusion breakthrough judgment algorithm, and the dynamic sealing performance evaluation results are output.

[0035] The interface control layer is used to stabilize the geometry of the contact interface between the reservoir and caprock regions. The material includes microporous filter membranes, sheet materials, or roughened interface layers with a thickness of 0.5 mm to 2 mm and a pore size of 10 μm to 100 μm. Its function is to reduce the fluid bypass effect at the contact interface and improve the repeatability of the carbon dioxide intrusion process.

[0036] The baseline permeability parameters include the initial permeability of the reservoir zone and the initial permeability of the caprock zone, calculated using Darcy's law. The input to the initial permeability calculation equation for the reservoir zone includes the injection flow rate, simulated formation brine viscosity, reservoir zone length, and pressure difference at the reservoir zone measuring points; the output is the initial permeability of the reservoir zone. The input to the initial permeability calculation equation for the caprock zone includes the injection flow rate, simulated formation brine viscosity, caprock zone length, and pressure difference at the caprock zone measuring points; the output is the initial permeability of the caprock zone. The baseline permeability parameters are used for comparative analysis of changes in permeability capacity during subsequent carbon dioxide injection.

[0037] The pressure rise rate at the reservoir measurement points is obtained by performing time difference calculation on the time series of pressure at the reservoir measurement points, and the pressure rise rate at the interface measurement points is obtained by performing time difference calculation on the time series of pressure at the interface measurement points. When the pressure rise rate at the interface measurement points exceeds twice the pressure rise rate at the reservoir measurement points, it indicates that carbon dioxide has broken through the top of the reservoir and begun to enter the caprock.

[0038] The intrusion height refers to the maximum vertical expansion distance of carbon dioxide after entering the caprock. It is calculated by using linear interpolation to determine the highest position reached by the pressure response front based on the correspondence between the pressure response time and the vertical position of each measuring point in the caprock. The intrusion height reflects the ability of carbon dioxide to overcome capillary resistance and move upward.

[0039] The spatial gradient matrix of the pressure field in the caprock is used to identify the preferred seepage channels within the caprock. The calculation method is to perform spatial difference calculation on the pressure values ​​of each measuring point in the caprock at the same time to obtain the horizontal gradient and the vertical gradient. When the gradient value of a certain area is continuously greater than 3 times the average gradient value, it is determined that there is a continuous high gradient channel in the area.

[0040] The macroscopic breakthrough moment refers to the moment when the gas flow rate at the extraction end exceeds 0.01 mL / min and remains stable. The macroscopic breakthrough moment is a key time node for evaluating the failure of the caprock sealing performance and is used to determine the breakthrough pressure difference and the breakthrough characteristic parameter set.

[0041] The breakthrough pressure difference refers to the difference between the pressure at the interface zone measuring point and the initial top pressure of the caprock at the moment of macroscopic breakthrough. The breakthrough pressure difference reflects the minimum driving pressure required for carbon dioxide to break through the caprock zone and is a key indicator for evaluating the sealing capacity of the caprock zone. When the breakthrough pressure difference is greater than 5 MPa, the caprock zone is judged to have high sealing capacity; when the breakthrough pressure difference is between 2 MPa and 5 MPa, the caprock zone is judged to have medium sealing capacity; when the breakthrough pressure difference is less than 2 MPa, the caprock zone is judged to have weak sealing capacity.

[0042] The breakthrough characteristic parameter set includes macroscopic breakthrough time, breakthrough pressure difference, invasion height, cumulative injection volume, produced gas-liquid ratio, and cumulative produced volume. The breakthrough characteristic parameter set is used to quantitatively characterize the migration and breakthrough behavior of carbon dioxide in the reservoir-caprock system, providing experimental data support for closure assessment and safety boundary determination.

[0043] The historical fitting and inversion method employs a parameterized relative permeability model and a capillary pressure model. The objective functions are the pressure time series at reservoir measurement points, interface measurement points, caprock measurement points, cumulative injection volume, and produced gas-liquid ratio. An ensemble Kalman filter algorithm is used to adjust the parameters of the relative permeability model and the capillary pressure model online, resulting in dynamic relative permeability and capillary pressure curves that are consistent with experimental data. The relative permeability model parameters include residual water saturation, residual gas saturation, and pore connectivity index, while the capillary pressure model parameters include inlet pressure, pore size distribution index, and maximum capillary pressure.

[0044] The multi-index fusion breakthrough judgment algorithm integrates three levels of judgment criteria: capillary breakthrough, seepage breakthrough, and macroscopic breakthrough. The capillary breakthrough criterion is that the pressure at the top of the caprock reaches the capillary entry threshold; the seepage breakthrough criterion is that the pressure gradient inside the caprock is established and reaches the Darcy flow threshold; and the macroscopic breakthrough criterion is that the gas phase flow rate at the production end exceeds 0.01 mL / min and remains stable. The produced gas-liquid ratio signal is analyzed by time series analysis using a Bayesian change point detection algorithm to identify abrupt change moments, calculate the confidence level of breakthrough at each level, and obtain the breakthrough confidence distribution.

[0045] The dynamic sealing performance evaluation results include the caprock sealing capability level, breakthrough risk assessment, and safe injection pressure limit. The caprock sealing capability level is divided into three levels: high sealing, medium sealing, and low sealing, based on the breakthrough pressure differential and intrusion height. The breakthrough risk assessment is a comprehensive judgment made by the breakthrough confidence distribution and the breakthrough characteristic parameter set. The safe injection pressure limit is the maximum injection pressure that ensures carbon dioxide does not exceed the caprock zone, and 80% of the breakthrough pressure differential is taken as the safe injection pressure limit.

[0046] The three-dimensional physical simulation experimental device consists of an injection metering pump, a confining pressure loading pump, a vacuum pump, a carbon dioxide gas cylinder, valves and manifolds, a first intermediate container, a second intermediate container, a constant temperature chamber, a three-dimensional reservoir-caprock physical model, a gas-liquid separator and production collector, a pressure sensor, a temperature sensor, and a computer. The first intermediate container is used to store simulated formation brine, and the second intermediate container is used to store buffer solution. The injection metering pump is connected to the first intermediate container, the second intermediate container, and the three-dimensional reservoir-caprock physical model via pipelines, and is used to perform water saturation after vacuuming, establish the initial water content, perform baseline seepage testing of simulated formation brine, and perform displacement operations before carbon dioxide injection on the three-dimensional reservoir-caprock physical model. The vacuum pump is connected to the three-dimensional reservoir-caprock physical model via pipelines, and is used to evacuate the three-dimensional reservoir-caprock physical model to remove residual gas and increase saturation. The confining pressure loading pump is connected to the external confining pressure chamber of the three-dimensional reservoir-caprock physical model via pipelines, and is used to apply constant confining pressure to the three-dimensional reservoir-caprock physical model, thereby simulating the in-situ stress environment. The carbon dioxide gas cylinder is connected to the... The pressure boosting unit is connected to the three-dimensional reservoir-caprock physical model to achieve constant flow or constant pressure carbon dioxide injection into the model. The outlet of the three-dimensional reservoir-caprock physical model maintains the target pressure through a backpressure valve and is connected to a gas-liquid separator and production collector to achieve produced gas-liquid separation. The three-dimensional reservoir-caprock physical model is the core component of the device. During the experiment, the three-dimensional reservoir-caprock physical model is placed in a constant temperature chamber to control the temperature of the three-dimensional physical simulation experiment. The three-dimensional reservoir-caprock physical model extends along the reservoir region, interface region, and caprock region. Measurement points are set up in the reservoir zone, interface zone, and caprock zone. Each measurement point is sealed with plugging screws when not in use. During the experiment, it is connected to a pressure sensor or sampling branch as needed. The gas-liquid separator and production collector is equipped with a metering module, which is connected to a computer via pipeline to realize real-time metering and data recording of produced gas and produced liquid. The signal terminals of the pressure sensor and temperature sensor are both connected to the computer to realize real-time monitoring and synchronous acquisition of pressure and temperature at the reservoir zone, interface zone, and caprock zone measurement points.

[0047] The three-dimensional reservoir-caprock physical model consists of a pressure-bearing shell, a sealing cap, an experimental chamber, fastening screws, sealing screws, inlet and outlet ports, a confining pressure isolation membrane, and a rotating support base. The sealing cap has reservoir zone, interface zone, and caprock zone measuring points distributed according to a certain rule. The upper part of the measuring points is sealed by sealing screws when no experiment is being conducted, and the lower end of the measuring points is connected to the experimental chamber. The experimental chamber serves as the space for partitioned filling of the reservoir and caprock media, as well as for simulating carbon dioxide migration, invasion, and breakthrough experiments. The interior of the experimental chamber forms a superimposed structure of the lower reservoir zone and the upper caprock zone through a partitioned limiting structure. The dimensions of the experimental chamber are length × width × height: 40cm × 40cm × 20cm. During filling, the three-dimensional reservoir-caprock physical model is first flipped so that the bottom surface faces upwards, the bottom fastening screws are unscrewed, and the bottom plate along with the lower wall of the experimental chamber is removed. The filling and construction were carried out in the order of reservoir zone – interface control layer – caprock zone. The filling steps included: designing the thickness ratio and target porosity-permeability grade of the reservoir zone and caprock zone, and preparing the reservoir medium and caprock medium; making measurement marks in the experimental chamber, filling the reservoir medium in layers according to the marks and compacting it into shape; laying the interface control layer on the top surface of the reservoir zone to stabilize the interface geometry and improve experimental repeatability; filling and compacting the caprock medium in layers above the interface control layer to form a continuous and dense structure in the caprock zone; after filling, laying a sealing film on the bottom surface to ensure that the filler is in close contact with the lower wall of the experimental chamber, and then reinstalling the lower wall of the experimental chamber and tightening the fastening screws to complete the filling and assembly; after assembly, rotating and resetting the three-dimensional reservoir-caprock physical model by rotating the support base so that the caprock zone is vertically above the reservoir zone, forming a three-dimensional structure of lower reservoir zone – upper caprock zone.

[0048] The specific implementation methods of the above steps are described in detail below. The specific implementation method of step S1 is as follows: First, determine the thickness ratio of the reservoir zone to the caprock zone according to the experimental design requirements. Typically, the reservoir zone thickness is set to 20cm–25cm, and the caprock zone thickness is set to 15cm–20cm, with a thickness ratio between 1.5:1 and 3:1. Flip the three-dimensional reservoir-caprock physical model so that the bottom surface faces upwards, unscrew the bottom fastening screws, and remove the bottom plate along with the lower wall of the experimental chamber. Prepare the reservoir medium. A high-porosity, high-permeability medium with a porosity of 25%–35% and a permeability of 500mD–2000mD is selected. Use a ruler to make measurement marks inside the experimental chamber. Fill the reservoir medium in layers according to the marked positions, with each layer controlled to a thickness of 2cm–3cm. After filling, compact the medium using a compaction tool at a pressure of 0.5MPa–1.0MPa to ensure that the reservoir zone forms a uniform and continuous pore structure. An interface control layer with a thickness of 0.5 mm to 2 mm and a pore size of 10 μm to 100 μm is laid on the top surface of the reservoir zone. During laying, it is ensured that the interface control layer adheres tightly to the top surface of the reservoir zone without air bubbles. A caprock medium is then filled above the interface control layer. This caprock medium is a low-porosity, low-permeability medium with a porosity of 5% to 15% and a permeability of 0.01 mD to 1 mD. It is also filled and compacted in layers, with each layer controlled to a thickness of 1 cm to 2 cm. After filling, a sealing membrane is laid flat on the bottom surface. The lower wall of the experimental chamber is then reinstalled and the fastening screws are tightened, with a tightening torque controlled at 20 N·m to 30 N·m. The three-dimensional reservoir-caprock physical model is rotated and reset using a rotating support base, so that the caprock zone is vertically above the reservoir zone. The three-dimensional reservoir-caprock physical model is placed in a constant temperature chamber, with the target temperature set to 50℃ to 80℃. Hydraulic oil is injected into the confining pressure chamber using a confining pressure loading pump to apply confining pressure to 10MPa–30MPa. The confining pressure value is determined based on the target simulation depth; for every 100m increase in simulation depth, the confining pressure increases by approximately 2.5MPa. The purpose of this step is to establish a three-dimensional reservoir-caprock physical model that conforms to the stratigraphic stacking relationship, and to achieve temperature and pressure stress conditions similar to the underground burial environment through temperature and confining pressure control.

[0049] The specific implementation of step S2 is as follows: First, a vacuum pump is used to evacuate the three-dimensional reservoir-caprock physical model for 2 to 4 hours, achieving a vacuum level of -0.09 MPa to -0.095 MPa. After evacuation, the injection metering pump is turned on, and simulated formation brine is injected into the three-dimensional reservoir-caprock physical model at a constant flow rate of 0.5 mL / min to 2 mL / min. The simulated formation brine has a salinity of 50 g / L to 200 g / L. During the injection process, the outlet end is monitored in real time for liquid outflow. When simulated formation brine continuously and stably flows out of the outlet end at a flow rate of more than 95% of the injected flow rate, the three-dimensional reservoir-caprock physical model is considered to have reached saturation. After saturation, the outlet valve is closed, and simulated formation brine continues to be injected at a constant flow rate. The pressure at the reservoir and caprock measuring points is recorded using pressure sensors. Once the pressure stabilizes, the initial permeability of the reservoir and caprock is calculated according to Darcy's law. When calculating the initial permeability of the reservoir zone, the input parameters include the injection flow rate, the simulated formation brine viscosity, the reservoir zone length, and the pressure difference at the measuring points within the reservoir zone. The output parameter is the initial permeability of the reservoir zone. Similarly, when calculating the initial permeability of the caprock zone, the input parameters include the injection flow rate, the simulated formation brine viscosity, the caprock zone length, and the pressure difference at the measuring points within the caprock zone. The output parameter is the initial permeability of the caprock zone. Darcy's law describes the seepage behavior of fluids in porous media, establishing a quantitative relationship between permeability and driving force by measuring pressure difference and flow rate. The purpose of this step is to establish the initial saturation state and baseline seepage parameters of the three-dimensional reservoir-caprock physical model, providing a benchmark for comparison of seepage capacity changes during subsequent carbon dioxide injection.

[0050] The specific implementation of step S3 involves injecting carbon dioxide into the reservoir area using a carbon dioxide gas cylinder and a pressure stabilization unit. In constant flow mode, the injection flow rate is set to 0.1 mL / min to 1 mL / min; in constant pressure mode, the injection pressure is set to the initial reservoir area measuring point pressure plus 3 MPa to 8 MPa. During injection, pressure sensors synchronously collect the pressure at measuring points in the reservoir area, interface area, and caprock area at time intervals of 1 to 5 seconds, forming three pressure time series. Time difference calculations are performed on the reservoir area measuring point pressure time series to calculate the ratio of the pressure difference between adjacent moments to the time interval, yielding the reservoir area measuring point pressure rise rate. The same method is used to calculate the interface area measuring point pressure rise rate for the interface area measuring point time series. When the interface area measuring point pressure rise rate exceeds twice the reservoir area measuring point pressure rise rate, it indicates that the lateral diffusion of carbon dioxide within the reservoir area is essentially complete and it begins to accumulate at the interface. Buoyancy causes carbon dioxide to overcome capillary resistance and enter the caprock area; at this point, it is determined that carbon dioxide has begun to invade the caprock. The threshold for the rate of pressure rise is set at 2 times, based on extensive experimental statistical results. When the multiple is less than 2, the pressure response in the interface region may originate from pressure propagation from the reservoir region; when the multiple is greater than 2, the pressure response in the interface region mainly originates from direct contact and entry of carbon dioxide. The purpose of this step is to identify the initiation time of carbon dioxide intrusion from the reservoir region to the caprock region through pressure response characteristics, providing a time reference for calculating the intrusion height and breakthrough pressure differential.

[0051] The specific implementation of step S4 involves continuously injecting carbon dioxide while simultaneously monitoring the pressure time series at each measuring point in the caprock. Spatial difference calculations are performed on the pressure values ​​at each measuring point in the caprock at the same time. The ratio of the pressure difference between adjacent measuring points in the horizontal direction to the distance between measuring points is calculated to obtain the horizontal gradient. Similarly, the ratio of the pressure difference between adjacent measuring points in the vertical direction to the distance between measuring points is calculated to obtain the vertical gradient. The horizontal and vertical gradients of all measuring points are combined to form the spatial gradient matrix of the caprock pressure field. For each element in the spatial gradient matrix of the caprock pressure field, its ratio to the average gradient value is calculated. When the gradient values ​​at three or more consecutive measuring points in a certain area are consistently greater than three times the average gradient value for a duration exceeding 10 minutes, it is determined that a continuous high gradient channel exists in that area. The formation of a continuous high gradient channel indicates that a stable seepage path for carbon dioxide has been established within the caprock, and the sealing capacity of the caprock begins to fail. Simultaneously, the gas phase flow rate of the gas-liquid separator and the gaseous collector is monitored. When the gas phase flow rate at the extraction end exceeds 0.01 mL / min and remains stable for more than 5 minutes, it is determined to be a macroscopic breakthrough. The gas phase flow rate threshold of 0.01 mL / min is determined based on the detection accuracy of the gas phase flow meter and the experimental noise level; gas phase signals below this threshold are difficult to distinguish from system noise. The purpose of this step is to comprehensively determine the timing and path of carbon dioxide breakthrough in the caprock region by combining the spatial distribution characteristics of the pressure field in the caprock region and the gas phase response at the extraction end, providing a data foundation for establishing a set of breakthrough characteristic parameters.

[0052] The specific implementation of step S5 involves recording the pressure at the reservoir measuring points, the interface measuring points, the caprock measuring points, and the cumulative injection volume at the macroscopic breakthrough moment. When calculating the breakthrough pressure differential, the input parameters are the interface measuring point pressure and the initial caprock top pressure at the macroscopic breakthrough moment, and the output parameter is the breakthrough pressure differential, which equals the interface measuring point pressure at the macroscopic breakthrough moment minus the initial caprock top pressure. When calculating the invasion height, the pressure response time of each measuring point in the caprock region is extracted; the pressure response time is defined as the moment when the pressure rise at the measuring point exceeds 5% of the initial pressure. Based on the vertical position coordinates of each measuring point in the caprock region and the pressure response time, a linear interpolation method is used to determine the highest position reached by the pressure response front; this highest position is the invasion height. The gas-liquid separator and production collector measure the produced gas phase volume and produced liquid phase volume in real time, and calculate the produced gas-liquid ratio, which is equal to the produced gas phase volume divided by the produced liquid phase volume. The cumulative production volume is the sum of the produced gas phase volume and the produced liquid phase volume. The breakthrough time, breakthrough pressure differential, invasion height, cumulative injection volume, produced gas-liquid ratio, and cumulative produced volume are combined to form a breakthrough characteristic parameter set. The purpose of this step is to quantitatively characterize the key characteristic parameters of carbon dioxide breakthrough in the caprock region, providing experimental data support for evaluating the caprock's sealing performance and determining its safety boundaries.

[0053] The specific implementation of step S6 involves adjusting the parameters of the relative permeability model and the capillary pressure model using a history fitting inversion method. This method is based on the comparison and optimization of numerical simulation and experimental data. First, a numerical model of the three-dimensional reservoir-caprock physical model is established, embedding parameterized relative permeability and capillary pressure models within this model. The relative permeability model parameters include residual water saturation, residual gas saturation, and pore connectivity index, with initial values ​​set based on core test results. The capillary pressure model parameters include inlet pressure, pore size distribution index, and maximum capillary pressure, with initial values ​​set based on mercury intrusion porosimetry (MIM) results. An ensemble Kalman filter algorithm is used to adjust the relative permeability and capillary pressure model parameters online. This algorithm estimates the parameter uncertainty through the statistical characteristics of ensemble members and updates the parameter distribution based on experimental observation data. The objective function is the sum of squared errors between the numerical simulation results and the experimental data, using the pressure time series at reservoir, interface, and caprock measurement points, the cumulative injection volume, and the produced gas-liquid ratio. The objective function was minimized through iterative optimization, resulting in dynamic relative permeability and capillary pressure curves consistent with experimental data. A multi-index fusion breakthrough judgment algorithm was used to calculate the breakthrough confidence distribution. This algorithm integrates three levels of judgment criteria: capillary breakthrough, seepage breakthrough, and macroscopic breakthrough. The capillary breakthrough criterion is that the pressure at the top of the caprock reaches the capillary entry threshold, which is calculated based on the capillary pressure curve and interfacial tension. The seepage breakthrough criterion is that a pressure gradient is established within the caprock and reaches the Darcy flow threshold, which is calculated based on the initial permeability and carbon dioxide viscosity of the caprock. The macroscopic breakthrough criterion is that the gas phase flow rate at the produced end exceeds 0.01 mL / min and remains stable. A Bayesian change point detection algorithm was used to perform time series analysis on the produced gas-liquid ratio signal. This algorithm identifies abrupt changes in the time series based on Bayesian inference and calculates the posterior probability distribution of these abrupt changes. The confidence levels for each level of capillary breakthrough, seepage breakthrough, and macroscopic breakthrough were calculated, and a comprehensive breakthrough confidence distribution was obtained. The caprock sealing capacity is classified into levels based on breakthrough pressure differential and intrusion height. A breakthrough pressure differential greater than 5 MPa is considered high sealing capacity, between 2 MPa and 5 MPa is considered medium sealing capacity, and less than 2 MPa is considered low sealing capacity. Breakthrough risk is assessed using breakthrough confidence distribution and a set of breakthrough characteristic parameters. The upper limit of safe injection pressure is calculated, equal to the breakthrough pressure differential multiplied by 0.8. The dynamic sealing performance evaluation results are output, including the caprock sealing capacity level, breakthrough risk assessment, and upper limit of safe injection pressure. This step aims to obtain dynamic relative permeability and capillary pressure curves through historical fitting and inversion methods, and to quantify the uncertainty of the breakthrough process using a multi-index fusion breakthrough judgment algorithm. This provides theoretical support for the assessment of the sealing properties of geological burial cover zones and the determination of safety boundaries.

[0054] It should be noted that the first key technical idea of ​​this invention is to use a three-dimensional reservoir-caprock physical model structure with upper and lower layers stacked to simulate the real process of carbon dioxide accumulating at the top of the reservoir area and invading into the caprock area. Traditional single-core holder devices can only realize one-dimensional or quasi-one-dimensional seepage, which cannot reflect the lateral migration and top accumulation process of carbon dioxide in the reservoir area, nor can it reflect the influence of the geometry of the interface between the reservoir area and the caprock area and the cross-boundary seepage boundary on the breakthrough behavior. This invention constructs a stacked structure of the lower reservoir area and the upper caprock area by partitioning and filling, and sets an interface control layer at the interface, so that carbon dioxide migrates upward under the drive of buoyancy and accumulates at the interface, which realistically reproduces the three-dimensional spatial distribution and migration path of carbon dioxide during underground burial, overcoming the limitation of traditional devices in being unable to reflect three-dimensional spatial effects. The second key technical idea is to obtain the spatiotemporal distribution characteristics of the pressure field through real-time monitoring of pressure at multiple measuring points in the reservoir area, interface area, and caprock area. Traditional pressure monitoring devices primarily focus on the inlet and outlet ends, obtaining pressure data from only a limited number of locations. This makes it difficult to reflect the pressure propagation patterns within the caprock and the formation process of preferential seepage channels. This invention addresses this by setting multiple measuring points in the reservoir, interface, and caprock regions, acquiring real-time pressure time series data at each point, and constructing a spatial gradient matrix of the caprock pressure field through spatial difference operations. This achieves high-resolution characterization of the pressure front propagation process and breakthrough path, providing a data foundation for the quantitative identification of the breakthrough mechanism. The third key technical approach is to employ a history fitting inversion method and a multi-index fusion breakthrough judgment algorithm to achieve dynamic relative permeability curve inversion and quantitative evaluation of the breakthrough process. Traditional methods rely on steady-state or non-steady-state methods to measure relative permeability, resulting in long measurement cycles and reliance on one-dimensional assumptions, making them unsuitable for complex three-dimensional flows and transient breakthrough processes. This invention achieves a closed-loop analysis from experimental data to capillary performance evaluation by historically fitting numerical simulations with experimental data, using an ensemble Kalman filter algorithm to adjust model parameters online, and obtaining dynamic relative permeability and capillary pressure curves consistent with experimental data. Furthermore, it identifies the breakthrough moment using a Bayesian variable point detection algorithm, quantifying the uncertainty of the breakthrough process. The synergistic effect of these three key technical approaches lies in organically combining three-dimensional physical simulation, multi-point pressure monitoring, and data inversion analysis, forming a complete technical chain from experimental design to data acquisition and mechanism identification. The three-dimensional overlay structure provides a realistic physical carrier for the spatiotemporal evolution of the pressure field; multi-point pressure monitoring provides high-resolution constraint data for historical fitting and inversion; and historical fitting and inversion, along with multi-index fusion judgment, transform experimental observations into quantitative capillary performance evaluation results. The synergy of these three aspects enables this invention to realistically reproduce the entire process of carbon dioxide migration, intrusion, and breakthrough under near-underground temperature and pressure stress conditions, and to identify key control factors affecting the caprock's sealing performance at the mechanistic level. This provides experimental basis and theoretical guidance for the safety design of geological burial sites and the optimization of injection conditions.

[0055] It should be noted that this invention also solves the following technical problem: the poor repeatability of experiments caused by the instability of the reservoir-caprock contact interface in existing physical simulation experiments. This invention addresses this by laying an interface control layer with a thickness of 0.5 mm to 2 mm and a pore size of 10 μm to 100 μm on the top surface of the reservoir region. This interface control layer, made of microporous membranes, thin sheet materials, or roughened interface layers, can stabilize the geometry of the contact interface, reduce fluid bypass effects at the interface, and ensure that the carbon dioxide intrusion process strictly follows the vertical transport law. This avoids pressure response distortion and breakthrough time determination deviations caused by interface fluctuations, thereby significantly improving the data consistency and comparability of results between different batches of experiments, and providing an experimental basis for establishing a reliable benchmark for evaluating sealing performance.

[0056] Furthermore, this invention addresses the problem of the difficulty in quantitatively determining the upper limit of safe injection pressure in existing technologies. By establishing a set of breakthrough characteristic parameters, including macroscopic breakthrough time, breakthrough pressure differential, intrusion height, cumulative injection volume, produced gas-liquid ratio, and cumulative produced volume, and combining this with relative permeability curves and capillary pressure curves obtained through historical fitting and inversion, this invention classifies the breakthrough pressure differential into three levels: high sealing (above 5 MPa), medium sealing (2 MPa to 5 MPa), and low sealing (below 2 MPa). 80% of the breakthrough pressure differential is taken as the upper limit of safe injection pressure. This method combines experimental observations of macroscopic breakthroughs with microscopic two-phase flow parameters, establishing a quantitative conversion relationship from experimental data to engineering application parameters, and providing clear operational boundaries for pressure management and risk control in carbon dioxide geological storage projects.

[0057] Specifically, the principle of this invention is as follows: The invention solves the aforementioned technical problems by establishing a comprehensive monitoring system covering the entire process from interface intrusion to macroscopic breakthrough. The interface control layer eliminates the fluid bypass effect at the contact interface, allowing the interface measurement points to accurately reflect the moment when carbon dioxide breaks through the top of the reservoir. When the pressure rise rate at the interface measurement point exceeds twice the pressure rise rate at the reservoir measurement point, it indicates that carbon dioxide has overcome the capillary resistance at the interface and begun to intrude into the caprock. By calculating the spatial difference in pressure at different measurement points, the spatial gradient matrix of the pressure field in the caprock region can identify continuous high-gradient channels formed within the caprock. These channels are the preferential migration paths for carbon dioxide, reflecting the controlling effect of caprock heterogeneity on the breakthrough process. The multi-index fusion breakthrough judgment algorithm integrates three levels of physical criteria: capillary entry threshold, Darcy flow establishment, and gas phase flow rate at the extraction end. It identifies the confidence level at the breakthrough moment through Bayesian variable point detection, avoiding accidental misjudgment due to a single index. The historical fitting inversion method uses pressure and flow data from the entire process to constrain the relative permeability and capillary pressure models, making the model parameters consistent with the experimental data, thereby establishing a quantitative relationship between breakthrough pressure difference and sealing capacity.

[0058] The following provides a specific embodiment 1 of the present invention. The specific implementation of step S1 in this embodiment 1 is the same as that described above, and will not be repeated in detail here. The specific implementation of other steps is described in detail below.

[0059] The specific implementation of step S2 includes the calculation process of initial permeability and the establishment of baseline seepage parameters based on Darcy's law. Initial permeability of the reservoir area. The calculation formula is expressed as follows:

[0060] ;

[0061] In the formula, Initial permeability of the reservoir region, in units of ; The simulated formation brine injection flow rate is given by the metering pump, in units of... The flow control module of the metering pump is used to set and record the data. To simulate the viscosity of formation brine, the unit is... The viscosity was obtained by measuring it using a viscometer under experimental temperature conditions. The length of the reservoir region is expressed in units of 1000 m / s. The dimensions are determined based on the experimental chamber size and filling design. The cross-sectional area of ​​the reservoir region is expressed in units of... The dimensions are calculated from the cross-sectional dimensions of the experimental chamber. The pressure difference at the measuring points in the reservoir area is expressed in units of... The initial permeability of the caprock is obtained by measuring the pressure difference between the inlet and outlet points of the reservoir. The calculation formula is expressed as follows:

[0062] ;

[0063] In the formula, The initial permeability of the caprock is given in units of... ; The length of the cap layer is given in units of 1000 m. ; The pressure difference at the measuring points in the caprock zone is expressed in units of... Baseline seepage parameters are obtained by measuring the pressure difference between the bottom and top measuring points in the caprock region. and For subsequent Comparative analysis of changes in seepage capacity during injection.

[0064] The specific implementation of step S3 includes the calculation of the pressure rise rate and Intrusion detection. Rate of pressure rise at reservoir measuring points. It is obtained through time difference calculation, and the formula is expressed as follows:

[0065] ;

[0066] In the formula, for The rate of pressure rise at the reservoir measuring point at any given time, in units of ; for Pressure at reservoir measuring points at any given time, in units of The data is collected in real time via a pressure sensor. For time steps, the unit is The empirical value is 10. Rate of pressure rise at the interface measuring point. The calculation formula is expressed as follows:

[0067] ;

[0068] In the formula, for The rate of pressure rise at the measuring point in the interface area at any given time, in units of ; for Pressure at measurement points in the interface area at any given time, in units of .when At that time, the judgment It begins to invade the caprock.

[0069] The specific implementation of step S4 includes the calculation of the spatial gradient matrix of the pressure field in the caprock region and the determination of macroscopic breakthrough. Spatial gradient matrix of the pressure field in the caprock region. To identify preferred seepage channels, the calculation employs a spatial difference method with a horizontal gradient. and vertical gradient The formula is expressed as follows:

[0070] ;

[0071] ;

[0072] In the formula, for Time Covering Zone Line number Horizontal pressure gradient at the measuring points, in units of ; This represents the vertical pressure gradient, in units of... ; for Time Covering Zone Line number Pressure at measuring points, unit: ; This is a dimensionless index for the measurement points in the cover layer region. This is a column index for measurement points in the cover layer area, dimensionless. The distance between measuring points in the horizontal direction is expressed in units of 1. The arrangement of measuring points in the cover layer area is determined accordingly. The vertical distance between measuring points is expressed in units of 1. The combined gradient magnitude of the gradient matrix The statement is as follows:

[0073] ;

[0074] In the formula, for Time of the first Line number The combined pressure gradient modulus of the measuring points, in units of The average value of the combined gradient modulus of all measuring points in the caprock region. The calculation formula is expressed as follows:

[0075] ;

[0076] In the formula, for The average value of the combined gradient modulus of all measuring points in the cover layer region at any given time, in units of ; The number of measuring points in the cover layer is dimensionless and determined according to the measuring point layout plan. This represents the number of measurement points in the cover layer region, dimensionless. When a certain measurement point... Furthermore, if the duration exceeds 300 seconds, a continuous high-gradient channel is considered to exist in the region. (Gas flow rate at the extraction end) The data is obtained in real time through the metering module of the gas-liquid separator and collector, and the unit is... The criteria for judging a macro breakout are: The moment when the flow rate exceeds 0.01 mL / min and remains stable is recorded as the macroscopic breakthrough moment. The unit is .

[0077] The specific implementation of step S5 includes the calculation of breakthrough characteristic parameters. Injection cumulative amount. The calculation formula is expressed as follows:

[0078] ;

[0079] In the formula, for Moment Cumulative injection amount, in units of ; for Moment Injected flow rate, in units of The flow control and metering devices of the injection system record data in real time. For integration variables, the unit is... Breakthrough pressure differential The calculation formula is expressed as follows:

[0080] ;

[0081] In the formula, To overcome the pressure difference, the unit is... ; A moment for macro breakthrough Pressure at the interface area measuring point, unit: ; The initial caprock pressure, in units of In the injection Intrusion height obtained from prior measurements. The calculation is performed using linear interpolation, and the formula is expressed as follows:

[0082] ;

[0083] In the formula, for The intrusion height in the cap layer region, in units of ; For the first The coordinates of the vertical measuring points, in units of ; This is a vertical measuring point index, dimensionless; For the first Pressure response time at each vertical measuring point, in units of The time when the pressure at the measuring point exceeds the initial pressure by 10% is defined as the moment when the pressure at that measuring point exceeds the initial pressure. and The first The location coordinates and pressure response time of each vertical measuring point. Extracted gas-liquid ratio. The calculation formula is expressed as follows:

[0084] ;

[0085] In the formula, for The gas-liquid ratio produced at any given time is dimensionless. The total extracted gas volume is expressed in units of... Real-time measurement is achieved through the metering module of the gas-liquid separator and the collector. The total volume of liquid extracted is expressed in units of... Cumulative extraction volume The calculation formula is expressed as follows:

[0086] ;

[0087] In the formula, for Cumulative output at any given time, in units of .

[0088] The specific implementation of step S6 includes history fitting inversion and a multi-index fusion breakthrough judgment algorithm. History fitting inversion uses an ensemble Kalman filter algorithm to adjust the parameters of the relative permeability model and the capillary pressure model. State vector The statement is as follows:

[0089] ;

[0090] In the formula, For the first The state vector at any given time; For time indexing, dimensionless; The residual water saturation is dimensionless and ranges from 0.1 to 0.4. The residual gas saturation is dimensionless and ranges from 0.05 to 0.3. The pore connectivity index is dimensionless and ranges from 1.5 to 4.0. To enter pressure, the unit is ; It is a dimensionless pore size distribution index, ranging from 0.3 to 2.0. Maximum capillary pressure, unit: Observation vector The statement is as follows:

[0091] ;

[0092] In the formula, For the first The observation vector at time; For the first Time, in units The multi-indicator fusion breakthrough judgment algorithm integrates three levels of judgment criteria, uses Bayesian change point detection to identify abrupt changes, and achieves breakthrough confidence. The calculation formula is expressed as follows:

[0093] ;

[0094] In the formula, To achieve a comprehensive breakthrough confidence level, the value is dimensionless and ranges from 0 to 1; For capillary breakthrough confidence level, dimensionless; The breakthrough confidence level for seepage is dimensionless; The confidence level for macroscopic breakthroughs is dimensionless. , , The weighting coefficients are dimensionless and satisfy the following conditions: The empirical values ​​are 0.2, 0.3, and 0.5 respectively. Maximum safe injection pressure. The calculation formula is expressed as follows:

[0095] ;

[0096] In the formula, The upper limit of the injection pressure for safety is specified in units of... Take 80% of the breakthrough pressure difference as the safety threshold.

[0097] To better understand and implement this invention, the following is a specific application scenario of this invention, Example 2:

[0098] A technical team is responsible for deep saline aquifers. The geological site selection evaluation task requires a quantitative evaluation of the caprock sealing performance of the sandstone-mudstone interbedded system in the target area. This team used the experimental method described in this invention to conduct research on a reservoir-caprock combination with a reservoir depth of 2800m, an average reservoir permeability of 350mD, and an average caprock permeability of 0.08mD. Intrusion and breakthrough simulation experiment.

[0099] like Figure 2 As shown, the experimental setup consists of an injection metering pump 1, a confining pressure loading pump 2, and... The system consists of: gas cylinder 3, vacuum pump 4, valve 5, first intermediate container 6, second intermediate container 7, constant temperature chamber 8, three-dimensional reservoir-caprock physical model 9, gas-liquid separation and production collector 11, computer 12, pressure sensor 13, and temperature sensor 10. The first intermediate container 6 stores simulated formation brine with a salinity of 85 g / L and a volume of 5 L, equipped with a liquid level monitoring device to display the remaining liquid volume in real time. The second intermediate container 7 stores buffer solution with a volume of 3 L, primarily used for... Prior to injection, the reservoir area is flushed to reduce residual brine saturation and ensure subsequent... Stability of the injection process. The injection metering pump 1 is a high-precision plunger pump with a flow control accuracy of 0.01 mL / min and a pressure output range of 0–50 MPa. It is connected to the first intermediate container 6 and the second intermediate container 7 via a three-way valve 5, allowing for switching of the injection medium as needed during the experimental stage. The pump's outlet is connected to the inlet port 22 of the three-dimensional reservoir-caprock physical model 9 via a high-pressure pipeline, used for water saturation, baseline seepage testing, and pre-injection replacement operations on the three-dimensional reservoir-caprock physical model 9. The vacuum pump 4 has a pumping capacity of 2 L / min and an ultimate vacuum of -0.098 MPa. It is connected to the outlet port 23 of the three-dimensional reservoir-caprock physical model 9 via a vacuum pipeline. Before water saturation, it performs a vacuuming operation on the model to remove residual gas and increase saturation. A vacuum gauge is installed on the vacuuming pipeline for real-time monitoring of the vacuum level. The confining pressure loading pump 2 is a constant pressure plunger pump with an output pressure range of 0 to 60 MPa and a pressure fluctuation of less than 0.1 MPa. It is connected to the confining pressure cavity inside the pressure-bearing shell 16 outside the three-dimensional reservoir-caprock physical model 9 through a confining pressure pipeline. The confining pressure cavity and the three-dimensional experimental chamber 17 are pressure-transmitting but fluid-isolated through a flexible confining pressure isolation membrane, so as to apply constant confining pressure to the physical model 9, thereby simulating the in-situ stress environment and ensuring that the medium inside the model is in a stress state similar to that of the actual reservoir. Gas cylinder 3 is a high-pressure steel cylinder containing gas with a purity of 99.9%. Gas is connected to inlet port 22 of the three-dimensional reservoir-caprock physical model 9 via a pressure-stabilizing and boosting unit consisting of a pressure reducing valve and a booster pump. The booster pump can increase the gas pressure to 35 MPa, and constant flow injection is achieved through a mass flow controller or constant pressure injection through a pressure controller, thereby achieving constant flow or constant pressure injection of the physical model 9. injection.

[0100] For example Figure 3As shown, the three-dimensional reservoir-caprock physical model 9 is the core component of this device. During the experiment, it is placed in a constant temperature chamber 8. The constant temperature chamber 8 adopts a PID temperature control system with a temperature control accuracy of ±0.5℃ and a temperature range of room temperature to 150℃, which can simulate the formation temperature conditions at different burial depths. The three-dimensional reservoir-caprock physical model 9 has multiple pressure measurement points 21 distributed along the reservoir zone, interface zone, and caprock zone. The pressure measurement points 21 are stainless steel pipes with an inner diameter of 2mm, which pass through the sealing cap 14 and extend to different positions inside the cavity. Each pressure measurement point 21 is sealed with a sealing screw 19 to prevent leakage when not being tested. During the experiment, the sealing screw 19 is unscrewed, and the pressure measurement point 21 is connected to the pressure sensor 13 through a quick connector to realize synchronous monitoring of the pressure at different positions inside the model. The gas-liquid separator and collector 11 is connected to the outlet port 23 of the physical model 9 via a high-pressure hose. A backpressure valve is installed at the outlet port 23 to maintain the outlet pressure of the model. The pressure setting range of the backpressure valve is 0-30 MPa, and the pressure control accuracy is 0.05 MPa. It is used to maintain the outlet pressure of the model and realize the gas-liquid separation of the produced fluid. The gas-liquid separator and collector 11 adopts a gravity separation structure. The gas phase is discharged from the top and measured by a gas flow meter, while the liquid phase is discharged from the bottom and collected and its volume is measured by a liquid collection bottle. It is equipped with a metering and recording module, which is connected to the computer 12 via a data acquisition card. The sampling frequency is 1 Hz, realizing real-time monitoring of the produced gas and produced liquid. Pressure sensor 13 is a strain gauge pressure sensor with a range of 0–40 MPa, an accuracy class of 0.1, and a response time of less than 10 ms. A total of 12 sensors are configured and connected to different pressure measurement points 21 in the reservoir, interface, and caprock regions. Temperature sensor 10 is a platinum resistance temperature sensor with a temperature measurement range of 0–200℃ and an accuracy of ±0.2℃. It is installed on the inner wall of the constant temperature chamber 8 and the surface of the three-dimensional reservoir-caprock physical model 9. The signal terminals of pressure sensor 13 and temperature sensor 10 are connected to the data acquisition system of computer 12 through shielded cables to realize real-time monitoring and synchronous acquisition of pressure and temperature at different measurement points.

[0101] like Figure 4As shown, the three-dimensional reservoir-caprock physical model 9 consists of a pressure-bearing shell 16, a sealing cap 14, a three-dimensional experimental chamber 17, fastening screws 18, sealing screws 19, a rotating support base 15, an inlet port 22, and an outlet port 23. The pressure-bearing shell 16 is made of high-strength stainless steel with a wall thickness of 15cm, capable of withstanding a maximum internal and external pressure difference of 50MPa. The sealing cap 14 has 16 measuring points 21 distributed in a 4×4 grid pattern, with a spacing of 10cm, covering the main monitoring area of ​​the model. The upper part of the measuring points 21 is sealed by the sealing screws 19 when no experiment is being conducted. The sealing screws 19 use polytetrafluoroethylene (PTFE) sealing rings to ensure reliable sealing. The lower end of the measuring points is connected to the three-dimensional experimental chamber 17. The three-dimensional experimental chamber 17 serves as the reservoir-caprock partition filling construction and for conducting experiments. The space for simulating migration, intrusion, and breach is designed with a partitioned limiting structure forming a stacked structure of a lower reservoir zone and an upper caprock zone. The limiting structure utilizes detachable stainless steel partitions with evenly distributed through-holes to allow fluid passage while restricting media movement. The dimensions of the three-dimensional experimental chamber 17 are set at 40cm x 40cm x 20cm (length x width x height), with an effective volume of 32L. The inlet port 22 is located at the center of the bottom of the pressure-bearing outer shell 16 and connects to the bottom of the reservoir zone via a 6mm inner diameter stainless steel pipe. The outlet port 23 is located above the top caprock zone and connects via an 8mm inner diameter stainless steel pipe. Both ports are equipped with quick-connect fittings for easy pipe connections. The rotating support base 15 uses a bearing-type rotating mechanism, allowing the model to rotate 360 ​​degrees. The base is equipped with angle markings and a locking device to fix the model's posture during filling and experimental phases.

[0102] Based on the lithological characteristics of the target reservoir-caprock assemblages, the technical team prepared reservoir medium 25 and caprock medium 24 using quartz sand and bentonite in different ratios. The reservoir layer was designed to be 12 cm thick, using quartz sand with a particle size of 0.3–0.5 mm mixed with 3% bentonite and compacted, with a target permeability of 350 mD. The caprock layer was designed to be 6 cm thick, using fine sand with a particle size less than 0.05 mm mixed with 15% bentonite and compacted, with a target permeability of 0.08 mD. Figure 5As shown, during filling, the physical model 9 is first flipped so that its bottom surface faces upwards. The bottom fastening screw 18 is unscrewed, and the lower wall of the cavity 17 is removed. Then, the filling is carried out in sections according to the sequence of reservoir – interface – cap layer. According to the design dimensions, a ruler is used to make measurement marks inside the experimental cavity 17. The reservoir medium 25 is filled and compacted in layers according to the marks to form a continuous reservoir structure. An interface control layer 20 is laid on the top surface of the reservoir. A polytetrafluoroethylene microporous filter membrane with a pore size of 50μm and a thickness of 1mm is used to ensure that it is tightly attached to the top surface of the reservoir 25 to stabilize the interface geometry and improve the repeatability of the experiment. The cap layer medium 24 is filled and compacted in layers above the interface control layer 20 to form a continuous and dense cap layer structure. After filling, a sealing film is laid flat on the model surface to ensure that the filler is tightly attached to the lower wall of the cavity. Then, the lower wall of the cavity and the sealing cap are reinstalled, and the fastening screw 18 is tightened to complete the filling and sealing assembly. After assembly, the physical model 9 is rotated and reset by rotating the support base 15, so that the filled cap layer area is located vertically above the reservoir area, forming a three-dimensional structure of lower reservoir-upper cap layer.

[0103] Before the experiment, the technical team placed the three-dimensional reservoir-caprock physical model 9 in a constant temperature chamber 8, setting the temperature to 85℃ to simulate the formation temperature of the target reservoir. A confining pressure of 28 MPa was applied to the model using a confining pressure loading pump 2 to simulate the in-situ stress environment. Subsequently, a vacuum pump 4 was used to evacuate the model, achieving a vacuum level of -0.095 MPa and maintaining it for 2 hours. After evacuation, simulated formation brine from the first intermediate container 6 was injected into the model at a flow rate of 0.5 mL / min using a metering pump 1, continuing for 48 hours until the model reached complete saturation. After saturation, the technical team injected simulated formation brine into the model at a constant flow rate of 1.0 mL / min, recording the pressure data at measuring points 21 in the reservoir, interface, and caprock regions in real time. After stable flow, the pressure differential at the reservoir measuring points stabilized at 0.42 MPa, and the pressure differential at the caprock measuring points stabilized at 8.75 MPa. Based on Darcy's law, the initial permeability of the reservoir zone is 348 mD, and the initial permeability of the caprock zone is 0.079 mD, which is basically consistent with the design target. Baseline seepage parameters were established.

[0104] After baseline testing was completed, the technical team used the buffer solution in the second intermediate container 7 to replace the model, draining some of the formation brine from the reservoir area. After the replacement was completed, Gas cylinder 3 is connected to model inlet port 22, and supercritical state gas is injected into the reservoir at a constant flow rate of 2.0 mL / min. During the injection process, computer 12 collected pressure data in real time from measuring points 21 in the reservoir, interface, and caprock regions. In the first 15 minutes after injection began, the pressure at the reservoir measuring points rapidly increased to 12.5 MPa, the pressure at the interface measuring points increased to 8.2 MPa, while the pressure at the caprock measuring points remained relatively unchanged. As injection continued, the rate of pressure increase at the interface measuring points accelerated significantly. At the 23rd minute of injection, the rate of pressure increase at the interface measuring points reached 0.35 MPa / min, more than twice the rate of increase of 0.16 MPa / min at the reservoir measuring points. The technical team determined... It begins to invade the caprock region.

[0105] Continue injecting The pressure changes at each measuring point 21 in the caprock region were continuously monitored. A total of 8 measuring points were set up in the caprock region, distributed at different vertical heights and horizontal positions. By calculating the spatial gradient matrix of the pressure field in the caprock region, the technical team discovered a continuous high-gradient channel in the left-central region of the model, where the pressure gradient value consistently exceeded the average gradient value by 3.8 times. At the 38-minute mark of injection, the gas-liquid separation and extraction collector 11 detected a gas flow rate exceeding 0.01 mL / min at the extraction end, which continued to rise steadily. The technical team determined this to be a macroscopic breakthrough. The breakthrough characteristic parameters recorded at this time are shown in Table 1.

[0106] Table 1 Breakthrough Feature Parameters

[0107]

[0108] Based on the breakthrough pressure differential values ​​in Table 1, the technical team determined that the caprock has moderate sealing properties. By analyzing the correspondence between the pressure response time and the vertical position of each measuring point in the caprock area, linear interpolation was used to determine that the highest position reached by the pressure response front was 4.8 cm, i.e., the intrusion height was 4.8 cm, indicating... It has a strong ability to overcome capillary resistance and move upwards.

[0109] Based on experimentally obtained pressure time series data from reservoir, interface, and caprock regions, as well as cumulative injection volume and produced gas-liquid ratio data, the technical team employed an ensemble Kalman filter algorithm for historical data fitting and inversion. By adjusting the relative permeability model and capillary pressure model parameters online, the inversion yielded residual water saturation of 0.28, residual gas saturation of 0.15, pore connectivity index of 2.3, inlet pressure of 0.35 MPa, pore size distribution index of 1.8, and maximum capillary pressure of 6.5 MPa. These parameters showed good self-consistency with the experimental data.

[0110] The technical team further employed a multi-index fusion breakthrough determination algorithm to comprehensively analyze the experimental results. Capillary breakthrough occurred at the 23rd minute of injection, when the pressure at the top of the caprock reached the capillary entry threshold. Seepage breakthrough occurred at the 32nd minute, when the pressure gradient within the caprock was established and reached the Darcy flow threshold. Macroscopic breakthrough occurred at the 38th minute, when the gas flow rate at the produced end exceeded the threshold and remained stable. Time-series analysis of the produced gas-liquid ratio signal using a Bayesian change point detection algorithm identified the abrupt change moments of the three breakthrough levels. The calculated confidence levels were 0.92 for capillary breakthrough, 0.88 for seepage breakthrough, and 0.95 for macroscopic breakthrough. The overall breakthrough confidence distribution indicates that the breakthrough determination results are reliable.

[0111] Based on the above analysis, the technical team output the dynamic sealing performance evaluation results. The cap layer sealing capability of this reservoir-cap assembly is classified as medium sealing. The breakthrough risk assessment shows a significant breakthrough risk when the injection pressure exceeds 4.2 MPa. The upper limit of the safe injection pressure is taken as 80% of the breakthrough pressure differential, i.e., 3.4 MPa. This value is used to ensure... Without exceeding the maximum injection pressure of the capping layer.

[0112] This invention represents a significant technological advancement over traditional one-dimensional or two-dimensional physical simulation methods. Traditional methods, limited by geometric dimensions, cannot accurately reflect... Preferred migration paths and heterogeneous breakthrough characteristics in three-dimensional space lead to systematic biases in the evaluation of caprock sealing performance. This invention addresses this by constructing a three-dimensional reservoir-caprock physical model. Three-dimensional spatial monitoring of the intrusion process can accurately identify the formation location and evolution process of continuous high-gradient channels. The introduction of the interface control layer effectively solves the problem of poor experimental repeatability caused by the fluid bypass effect at the reservoir-cap interface in traditional methods, and significantly improves the reliability of experimental results by stabilizing the interface geometry. The multi-index fusion breakthrough judgment algorithm establishes a complete breakthrough identification system from three levels: capillary breakthrough, seepage breakthrough, and macroscopic breakthrough, overcoming the problem of judgment lag or misjudgment caused by traditional methods that rely on only a single index to determine breakthroughs. The history fitting inversion method realizes the dynamic identification of key parameters such as relative permeability and capillary pressure, providing reliable input parameters for numerical simulation and solving the problem of traditional static parameter testing and dynamic... The problem of a disconnect in the displacement process.

[0113] It should be noted that the variables involved in this invention are explained in detail in Table 2.

[0114] Table 2 Variable Explanation Table

[0115]

[0116] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A three-dimensional physical simulation experimental method for a carbon dioxide geological burial reservoir-caprock system, characterized in that, The process includes preparing a three-dimensional reservoir-caprock physical model. Within the experimental chamber, the model is layered and compacted according to the sequence of lower reservoir zone, interface control layer, and upper caprock zone. Measurement points for the reservoir zone, interface zone, and caprock zone are set on the sealed cap. After assembly, the three-dimensional reservoir-caprock physical model is placed in a constant-temperature chamber and confining pressure is applied. After vacuuming the model, simulated formation brine is injected until saturated. Simulated formation brine is injected at a constant flow rate using a metering pump, and the pressure at each measurement point is recorded. The initial permeability of the reservoir zone and caprock zone is calculated to establish baseline seepage parameters. Carbon dioxide is injected into the reservoir zone in constant-flow or constant-pressure mode. The pressure at the measurement points in the reservoir zone, interface zone, and caprock zone is collected in real time. When the pressure at the interface zone measurement point rises... When the rate of carbon dioxide intrusion into the caprock exceeds twice the rate of pressure rise at the reservoir monitoring points, it is determined that carbon dioxide has begun to invade the caprock. Carbon dioxide is continuously injected while the pressure time series at each monitoring point in the caprock area is monitored. The spatial gradient matrix of the pressure field in the caprock area is calculated to identify the preferred seepage channels. When a continuous high gradient channel is formed and the gas phase flow rate at the production end exceeds 0.01 mL / min and remains stable, it is determined to be a macroscopic breakthrough. The pressure at each monitoring point and the cumulative injection volume at the moment of macroscopic breakthrough are recorded. The breakthrough pressure difference and invasion height are calculated, and a set of breakthrough characteristic parameters is established. Based on the pressure time series, cumulative injection volume, and cumulative production volume at each monitoring point, the parameters of the relative permeability model and the capillary pressure model are adjusted using the historical fitting inversion method. The breakthrough confidence distribution is calculated through a multi-index fusion breakthrough judgment algorithm, and the dynamic sealing performance evaluation results are output.

2. The three-dimensional physical simulation experimental method for a carbon dioxide geological burial layer-caprock system according to claim 1, characterized in that, The interface control layer is used to stabilize the geometry of the contact interface between the reservoir region and the caprock region. The material includes a microporous filter membrane, sheet material or roughened interface layer, with a thickness of 0.5 mm to 2 mm and a pore size of 10 μm to 100 μm. Its function is to reduce the fluid bypass effect at the contact interface and improve the repeatability of the carbon dioxide intrusion process.

3. The three-dimensional physical simulation experimental method for a carbon dioxide geological burial reservoir-caprock system according to claim 2, characterized in that, The baseline seepage parameters include the initial permeability of the reservoir zone and the initial permeability of the caprock zone, which are calculated using Darcy's law. The inputs to the calculation equation for the initial permeability of the reservoir zone include the injection flow rate, the simulated formation brine viscosity, the reservoir zone length, and the pressure difference at the measuring points in the reservoir zone. The inputs to the calculation equation for the initial permeability of the caprock zone include the injection flow rate, the simulated formation brine viscosity, the caprock zone length, and the pressure difference at the measuring points in the caprock zone.

4. The three-dimensional physical simulation experimental method for a carbon dioxide geological burial layer-caprock system according to claim 3, characterized in that, The pressure rise rate at the reservoir measurement points is obtained by performing time difference calculation on the time series of pressure at the reservoir measurement points, and the pressure rise rate at the interface measurement points is obtained by performing time difference calculation on the time series of pressure at the interface measurement points. When the pressure rise rate at the interface measurement points exceeds twice the pressure rise rate at the reservoir measurement points, it indicates that carbon dioxide has broken through the top of the reservoir and begun to enter the caprock.

5. The three-dimensional physical simulation experimental method for a carbon dioxide geological burial reservoir-caprock system according to claim 4, characterized in that, The intrusion height refers to the maximum vertical expansion distance of carbon dioxide after it enters the caprock. The calculation method is to determine the highest position reached by the pressure response front by using linear interpolation based on the correspondence between the pressure response time and the vertical position of each measuring point in the caprock.

6. The three-dimensional physical simulation experimental method for a carbon dioxide geological burial reservoir-caprock system according to claim 5, characterized in that, The spatial gradient matrix of the pressure field in the caprock is used to identify the preferred seepage channels within the caprock. The calculation method is to perform spatial difference calculation on the pressure values ​​of each measuring point in the caprock at the same time to obtain the horizontal gradient and the vertical gradient. When the gradient value of a certain area is continuously greater than 3 times the average gradient value, it is determined that there is a continuous high gradient channel in the area.

7. The three-dimensional physical simulation experimental method for a carbon dioxide geological burial reservoir-caprock system according to claim 6, characterized in that, The macroscopic breakthrough moment refers to the moment when the gas flow rate at the extraction end exceeds 0.01 mL / min and remains stable. The macroscopic breakthrough moment is a key time node for evaluating the failure of the caprock sealing performance and is used to determine the breakthrough pressure difference and the breakthrough characteristic parameter set.

8. The three-dimensional physical simulation experimental method for a carbon dioxide geological burial reservoir-caprock system according to claim 7, characterized in that, The breakthrough pressure difference refers to the difference between the pressure at the interface zone measuring point and the initial caprock top pressure at the moment of macroscopic breakthrough. When the breakthrough pressure difference is greater than 5 MPa, the caprock zone is judged to have high sealing performance. When the breakthrough pressure difference is between 2 and 5 MPa, the caprock zone is judged to have medium sealing performance. When the breakthrough pressure difference is less than 2 MPa, the caprock zone is judged to have weak sealing performance.

9. A three-dimensional physical simulation experimental method for a carbon dioxide geological burial layer-caprock system according to claim 8, characterized in that, The breakthrough characteristic parameter set includes macroscopic breakthrough time, breakthrough pressure difference, invasion height, cumulative injection volume, produced gas-liquid ratio, and cumulative produced volume, which are used to quantitatively characterize the migration and breakthrough behavior of carbon dioxide in the reservoir-caprock system.

10. A three-dimensional physical simulation experimental method for a carbon dioxide geological burial layer-caprock system according to claim 9, characterized in that, The historical fitting and inversion method adopts a parameterized relative permeability model and capillary pressure model, with the pressure time series of reservoir area measuring points, interface area measuring points, caprock area measuring points, cumulative injection volume, and produced gas-liquid ratio as objective functions, and uses an ensemble Kalman filter algorithm to adjust the model parameters online.

Images