A supercritical carbon dioxide plume migration simulation method

Abstract

The application discloses a supercritical carbon dioxide plume migration simulation method, comprising the following steps of constructing a well area adaptive geological model, optimizing a Peng-Robinson state equation and a Brokaw viscosity model, coupling a Darcy-Brinkman multi-physical field control equation, setting a refined grid and a boundary condition, carrying out four-parameter sensitivity analysis of porosity, permeability, injection rate and capillary pressure, and verifying the model precision through a Buckley-Leverett analytical solution. The application realizes high-fidelity simulation of supercritical carbon dioxide plume dynamic migration in a specific well area, the maximum error is less than or equal to 0.229%, and provides quantitative support for CCUS engineering injection optimization and sealing safety evaluation.

Description

Technical Field

[0001] This invention relates to the field of carbon capture, utilization and storage technology, and in particular to a method for simulating supercritical carbon dioxide plume transport. Background Technology

[0002] Carbon capture, utilization, and storage (CCUS) technology, as a key pathway to achieving carbon neutrality, has been implemented in engineering practices and pilot tests in multiple mining fields. For example, Hungary's gas field initially injected carbon dioxide into depleted dry gas reservoirs to improve recovery rates, the K12-B offshore gas field in the Netherlands conducted a field test of reinjecting carbon dioxide into the original gas reservoir, and Australia has advanced the CO2CRC geological storage project. In terms of numerical simulation research, the Niger Delta reservoir supercritical carbon dioxide geological storage project proposed a simplified modeling method, Leonenko and Keith explored the impact of saline water injection on the supercritical carbon dioxide dissolution process using numerical methods, and Juanes et al. focused on the mechanism of the relative permeability hysteresis effect on plume transport behavior. Although the above studies have promoted the development of supercritical carbon dioxide geological storage simulation technology to some extent, existing methods still have significant limitations. First, most models fail to fully consider the complex phase behavior of supercritical carbon dioxide in deep reservoirs under high temperature and high pressure conditions, especially lacking precise adaptation to specific well area geological conditions (such as heterogeneous porosity and permeability distribution), resulting in simulation results that are difficult to reflect the real reservoir dynamics. Secondly, current simulation methods have shortcomings in the dynamic coupling between geological parameters and fluid properties. Some simplified models neglect key physical processes and cannot accurately capture the spatiotemporal evolution characteristics of plume transport, while some highly complex models lack systematic sensitivity analysis of the interactions of multiple factors. Furthermore, due to the disconnect between model construction and actual engineering data, there are often significant discrepancies between simulation predictions and field observations. This makes it difficult to directly use simulation results to guide injection parameter optimization or sequestration safety assessment, increasing engineering trial-and-error costs and implementation risks. Fundamentally, existing technologies have not yet established an integrated simulation system that combines "well-area measured data—multi-physics equation coupling—key parameter sensitivity quantification," making it difficult to balance simulation relevance, computational accuracy, and engineering practicality. There is an urgent need for a new simulation method that can accurately analyze the dynamic transport laws of supercritical carbon dioxide plumes and support engineering decision-making. Summary of the Invention

[0003] This invention addresses the shortcomings of existing carbon capture, utilization, and storage (CCUS) numerical simulation techniques, such as insufficient adaptability to well areas, coarse characterization of multiphase flow transport, and weak engineering guidance capabilities. It provides a simulation method based on supercritical carbon dioxide plume transport. Specifically, these shortcomings include: general models are difficult to embed measured geological parameters for specific well areas, leading to significant deviations between simulation results and actual reservoir responses; physical property parameters often use constant values ​​or simplified relationships, failing to reflect the nonlinear PVT characteristics of supercritical carbon dioxide in high-temperature, high-pressure deep environments; the flow control equations are singular and do not couple the transition mechanism between wellbore free flow and matrix seepage; and there is a lack of systematic sensitivity analysis of key geological and engineering parameters, making it difficult to quantify their impact on plume sweep efficiency. Therefore, this invention constructs an integrated simulation method of "data-driven—equation coupling—parameter sensitivity—verification closed loop," achieving high-fidelity analysis of the dynamic transport process of supercritical carbon dioxide in target well reservoirs.

[0004] This invention provides a method for simulating supercritical carbon dioxide plume transport, comprising the following steps:

[0005] S1: Constructing a well-adaptive geometric and geological model. In the numerical simulation platform COMSOL Multiphysics, a cuboid reservoir computational domain of 500 m × 500 m × 100 m was established to represent the effective sealing layer of the target well area. At the geometric center of this computational domain, a cylindrical gas injection well channel with a diameter of 0.1 m was vertically positioned. Furthermore, a core computational region was defined within 50 m above and below the injection wellbore to precisely capture the non-Darcy flow and pressure disturbance characteristics in the near-wellbore zone caused by high-speed injection. The discrete data points of porosity and permeability variation with depth obtained from well logging interpretation of the target well area are imported. In the global definition module, a continuous function f(x, y, z) with spatial coordinates (x, y, z) as independent variables is constructed through cubic spline interpolation or radial basis function interpolation. This function represents the heterogeneous porosity field and permeability field in the entire computational domain, respectively. The input data of the interpolation function comes from the core analysis or well logging inversion results of at least one monitoring well in the target well area to ensure that the geological properties of the model are consistent with the actual reservoir structure.

[0006] S2: Optimize the phase state and property model of supercritical carbon dioxide. In the material definition module, carbon dioxide is set as a compressible fluid, and the thermodynamic property calculation interface is enabled. The Peng-Robinson equation of state (PR-EOS) is used to calculate the pressure-volume-temperature (PVT) relationship of supercritical carbon dioxide in the supercritical state, ensuring that the nonlinear change of density under different formation pressures and temperatures is accurately calculated. The expression is as follows:

[0007] Where R is the universal gas constant, T is the absolute temperature, v is the molar volume, and a(T) and b are physical property parameters related to the critical temperature and critical pressure.

[0008] Furthermore, the Brokaw viscosity model is introduced to calculate the dynamic viscosity μ, which is in the form of:

[0009]

[0010]

[0011] in For low-pressure ideal gas viscosity, This is the critical viscosity reference value. For density comparison, , For the purpose of temperature comparison, A, B, C, and D are all empirical coefficients;

[0012] Specifically, by combining the measured geothermal gradient and static rock pressure gradient in the target well area, temperature and pressure are defined in the variable nodes, and the background components of formation water are used as the input of the mixed fluid to correct the collision integral parameters in the Brokaw model. Finally, the physical property parameter expression is constructed so that the viscosity is updated in real time with spatial location.

[0013] S3: Construct a multiphysics coupled governing equation system. In the porous media flow module, apply Darcy's law to the rock matrix region: Describe low-velocity seepage behavior; apply the Brinkman equation to the gas injection wellbore and the near-wellbore high-permeability transition zone (i.e., the core computational domain): By simultaneously considering viscous diffusion and Darcy resistance terms, the coupling mechanism between free flow and porous media flow is accurately characterized. Through a multiphysics interface, parameters such as porosity, permeability, density, and viscosity are set as functions of the solution variables, achieving bidirectional coupling of the mass conservation equation and momentum conservation equation during transient solution. The mass conservation equation includes mass source terms for the supercritical carbon dioxide and water phases, and a relative permeability model is introduced.

[0014]

[0015] To characterize the mutual interference effect of two-phase flow.

[0016] S4: Set boundaries and initial conditions and perform mesh generation. The inner wall of the injection well is set as the inlet boundary, the supercritical carbon dioxide phase volume fraction is specified as 1.0, and the mass flow rate or volumetric rate is set according to the injection regime; the six outer faces of the model are set as constant pressure boundaries, with pressure values ​​taken from the original formation pressure of the target well area. In the multiphase flow interface, the inlet capillary pressure function is defined:

[0017]

[0018] in For the inlet capillary pressure, denoted as the porosity distribution index, and S as the residual saturation. During the mesh generation stage, a user-controlled free tetrahedral mesh strategy was adopted: an extremely fine mesh (cell size ≤ 0.5 m) was deployed within a 5 m radius around the injection well; a medium-density mesh was used in the 5–50 m transition zone; and a sparse mesh (cell size ≥ 5 m) was used beyond 50 m. Simultaneously, a three-layer boundary layer mesh was added at the interface between the Darcy and Brinkman regions, with the first layer having a height of 0.1 m and a growth factor of 1.2, to accurately resolve abrupt changes in the pressure gradient.

[0019] S5: Design a sensitivity analysis experiment for key parameters. Parameterized scanning was configured in the research nodes to examine the effects of four types of parameters on plume transport in sequence: (1) Porosity base, with three levels: 20%, 35%, and 50%; (2) Permeability gradient, defined as the ratio of horizontal permeability to vertical permeability, with values ​​of 1:1, 5:1, and 10:1; (3) Injection rate, set to 3 kg / s, 5 kg / s, and 8 kg / s; (4) Inlet capillary pressure threshold P_{c0}, with values ​​of 0.1 MPa, 0.5 MPa, and 1.0 MPa. The BDF (Backward Differentiation Formula) adaptive time-step transient solver was used, with the total simulation duration set to 4 years (i.e., 1.26144 × 10⁻⁶). 8 The time step is initially set to 1 day, with a maximum step size of no more than 30 days. In the post-processing module, the spatiotemporal evolution data of CO2 saturation are dynamically extracted, and the plume leading edge position, three-dimensional swept volume, and diffusion morphology indices at t=1 year, 2 years, and 4 years are calculated.

[0020] S6: Model Validation and Accuracy Control. The CO2 saturation distribution curve was extracted along the vertical axis (x=250 m, y=250 m) at the center of the computational domain and compared with the analytical solution of the Buckley-Leverett one-dimensional displacement theory. The analytical solution was calculated based on the same initial saturation, relative permeability index, and injection rate. The maximum relative error between the numerical and analytical solutions at the saturation front was calculated. The grid density and time step tolerance were iteratively adjusted until the maximum relative error was ≤ 0.229%, which was used as the criterion for model validity. The 0.229% threshold was determined based on multiple grid convergence tests to ensure a balance between numerical dissipation and physical diffusion.

[0021] The core mathematical model used in this invention also includes a CO2 solubility model in formation water.

[0022]

[0023]

[0024] In the formula, The standard Henry's constant for water, Let be the fugacity coefficient of water, calculated from the equation of state for an ideal gas. =1, The total pressure of the system is expressed in Pa. Standard partial molar volume, in units of .

[0025] The beneficial effects of this invention are:

[0026] The well-area adaptability modeling step S1 directly maps measured porosity and permeability data into a spatially continuous field, ensuring that the computational domain geometry, wellbore structure, and reservoir heterogeneity are consistent with the target well area, thus resolving the representativeness issue caused by the "one-size-fits-all" approach of general models. The property model optimization step S2 combines the PR-EOS model with an environmentally corrected Brokaw model, ensuring that CO2 density and viscosity are within the range of 25–100. The dynamic updates within the MPa and 40–120°C range avoid systematic errors introduced by constant physical property assumptions. The multiphysics coupling step S3 applies the Darcy-Brinkman equation in sections, accurately reproducing the full-scale flow transition from wellbore turbulence to matrix laminar flow, overcoming the limitation of the single Darcy equation failing in the near-wellbore region. The sensitivity analysis step S5 quantifies the laws that the swept volume increases by more than 2 times when the porosity increases from 20% to 50%, and the injection rate increases by 60%, leading to an increase of about 45% in the leading edge propulsion speed, providing direct quantitative basis for injection parameter optimization. The verification step S6, through rigorous comparison with the Buckley-Leverett analytical solution, controls the maximum error of the model to within 0.229%, ensuring the reliability of subsequent prediction results.

[0027] In summary, this invention constructs a reproducible, verifiable, and engineering-applicable supercritical carbon dioxide plume transport simulation system through the complete technology chain from S1 to S6. This not only improves the ability of numerical simulation to reproduce geological conditions in specific well areas, but also realizes the quantitative analysis of the coupling of multiple physical mechanisms and the influence of key parameters, providing a high-precision technical tool for the design of CCUS project injection schemes, assessment of storage capacity, and prediction of long-term safety. Attached Figure Description

[0028] Figure 1 This is a flowchart of the method of the present invention;

[0029] Figure 2 This is a map showing the distribution of carbon dioxide saturation in the formation;

[0030] Figure 3A cloud map showing carbon dioxide transport beneath a 20% porosity formation;

[0031] Figure 4 A cloud map showing carbon dioxide transport beneath a 35% porosity formation;

[0032] Figure 5 A cloud map showing carbon dioxide transport beneath a 50% porosity formation;

[0033] Figure 6 Comparative slices of carbon dioxide transport in formations with different porosities;

[0034] Figure 7 This is a diagram showing the evolution of carbon dioxide sweep range under formations with different porosities. Detailed Implementation

[0035] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0036] This invention provides a simulation method for supercritical carbon dioxide plume transport. The specific implementation process relies on the COMSOL Multiphysics simulation platform. Based on actual geological and production data of a target well area, a high-fidelity three-dimensional numerical model is constructed. Through multi-parameter sensitivity analysis and rigorous verification mechanisms, accurate analysis of the dynamic transport behavior of supercritical carbon dioxide in deep reservoirs is achieved. The following is in conjunction with the appendix... Figures 3 to 7 The specific embodiments of the present invention will be described in detail below, along with the accompanying reference numerals.

[0037] First, in step S1, a well-adaptive geometric and geological model is constructed. In the geometry nodes of the COMSOL software, the matrix rock computational domain is constructed first. A cuboid solid with dimensions of 500m × 500m × 100m is created to represent the target reservoir area. At the geometric center, a gas injection well channel is constructed using a cylinder tool, setting the bottom radius of the cylinder to 0.05m (i.e., a diameter of 0.1m), with a height extending throughout the entire model. To capture the dramatic flow changes in the near-wellbore zone, a core computational region is extracted within 50m above and below the injection port using Boolean operations, forming an independent fluid injection domain. In the globally defined interpolation function, actual geological production data of the target well area (including porosity and permeability distribution data with depth interpreted from well logging) are imported. The interpolation function is mapped onto the spatial coordinates (x, y, z) of the geometric model, and a continuous spatial function f(x, y, z) is established with these coordinates as independent variables. This defines the heterogeneous porosity and permeability fields throughout the entire computational domain, ensuring that the model reproduces the actual geological conditions of the well area to the greatest extent possible. This achieves a high-precision reconstruction of the heterogeneous reservoir structure.

[0038] Proceed to step S2, Material Library Retrieval and Definition: In the material node, add "Carbon Dioxide" as a fluid material. Thermodynamic Equation of State Settings: In the fluid property settings, enable the thermodynamic property calculation function. Select the Peng-Robinson Equation of State (EOS) to characterize the pressure-volume-temperature (PVT) relationship of carbon dioxide in the supercritical state, ensuring that the nonlinear change in density under different formation pressures and temperatures is accurately calculated. The Peng-Robinson Equation of State takes the following form:

[0039]

[0040] In the formula, This is the universal gas constant. This is a temperature-dependent attraction parameter, reflecting intermolecular van der Waals forces. It is related to the critical temperature and critical pressure, and decreases as temperature increases. .

[0041] Viscosity model correction: The Brokaw mixing model is introduced to calculate the dynamic viscosity of carbon dioxide. Its expression is:

[0042]

[0043]

[0044] In the formula, The viscosity is that of an ideal gas at low pressure, neglecting intermolecular interactions. This is a critical viscosity reference value, a characteristic viscosity defined based on critical state parameters. For density comparison, .

[0045] By using COMSOL's variable nodes, the measured temperature and pressure gradient formulas of the target well reservoir, as well as the background values ​​of reservoir fluid components, are input. The standard viscosity model is then specifically corrected and parameterized, and expressions for physical property parameters that dynamically change with temperature and pressure environments are constructed to improve simulation accuracy.

[0046] In step S3, the physics interface is selected: the porous media flow module is chosen in the model wizard. For the rock matrix region, a Darcy's law interface is added to describe low-velocity seepage; for the wellbore and near-wellbore high-permeability region, a Brinkman equation interface is added to describe the transition characteristics between free fluid and porous media flow. Multidimensional numerical model construction and coupling: Through multiphysics nodes, the coupling relationship between the mass source and momentum source is established to achieve real-time bidirectional transmission and feedback of formation parameters (permeability, porosity) and fluid properties (density, viscosity).

[0047] Brinkman equation:

[0048]

[0049] In the formula, This is the viscous diffusion term, describing the velocity gradient caused by fluid viscosity. This is the Darcy resistance term, representing the permeability resistance of the porous medium to the flow. This is the pressure gradient driving term.

[0050] Penetration model:

[0051]

[0052] in the formula Residual gas saturation This refers to penetration rate.

[0053] In multiphysics nodal computation models, the coupling between Darcy's law and Brinkman's equations is implemented as follows:

[0054]

[0055] In the formula, It is a non-stationary term. This is the convective acceleration term, used to account for inertial effects.

[0056] Darcy's Law Describing the seepage velocity vector With pressure gradient The relationship, in the formula, This is the seepage velocity vector, with units of m / s. This represents the pressure gradient, expressed in Pa / m.

[0057] Boundary conditions and initial conditions settings:

[0058] Inlet boundary: Select the inner wall of the carbon dioxide injection well section as the inlet boundary, set the mass flow rate or injection rate in combination with the injection regime, and specify the phase volume fraction at the inlet.

[0059] Pressure boundary: Set the outer boundary of the model to the original pressure boundary or constant pressure boundary of the formation.

[0060] Capillary pressure: In a multiphase flow setup, the inlet capillary pressure function is input to characterize the displacement resistance during the initial injection phase. Capillary pressure model:

[0061]

[0062] in the formula For the inlet capillary pressure, denoted as the porosity distribution index, and S as the residual saturation.

[0063] Mesh generation: User-controlled meshes are used in the mesh nodes. Extremely fine free tetrahedral meshes are used around the injection well, while sparser meshes are used in the far-well region. Boundary layer meshes are set at the interfaces between different physics fields to capture gradient changes, balancing computational accuracy and efficiency.

[0064] Step S4: Parametric Scan Setup: Under the study node, add a parametric scan step. Set four sets of comparative experimental variables: porosity baseline, permeability gradient, injection rate, and inlet capillary pressure threshold.

[0065] Transient solver configuration: Select the transient solver. Set the time step algorithm to adaptive step size (BDF or Generalized-alpha), and set the total simulation duration to 4 years (enter the time range in seconds or days).

[0066] Calculation and Tracking: Run calculations to dynamically track the diffusion trajectory of carbon dioxide plumes in three-dimensional space under different parameter combinations.

[0067] Post-processing visualization: In the results node, create 3D surface maps and slice maps, visualize the saturation distribution cloud map, and extract isosurface data to quantify the evolution of the affected area.

[0068] Step S5: Analytical Solution Comparison: Based on the Buckley-Leverett one-dimensional displacement theory, extract the saturation distribution data along the central axis of the numerical model. Error Analysis: Compare the saturation leading edge position obtained from the numerical simulation with the theoretical position calculated by the Buckley-Leverett analytical solution. Calculate the root mean square error or maximum relative error of both.

[0069] Accuracy assessment: Adjust the grid density and time step tolerance until the verification results show that the maximum error between the model calculation results and the analytical solution is strictly controlled within 0.229%. This is used as the basis for judging the effectiveness of the model and ensuring the reliability of subsequent prediction data.

[0070] In the geological model building phase, the simulation data was first systematically processed to obtain porosity and permeability data for the vertical strata. Based on the Comsol software platform, the model mesh accuracy was set to 500m×500m×100m, generating a total of 769,999 domain elements, 75,970 boundary elements, and 2,435 edge elements.

[0071] Based on this, a set of comparative experiments were set up: porosity sensitivity analysis (20%, 35%, 50%).

[0072] Physical parameters and boundary conditions used in the simulation

[0073] Parameter name Simulated numerical Penetration 1*10^-13[m^2] Porosity 0.2 Initial reservoir pressure 9 [MPa] Capillary inlet pressure 1*10^4[Pa] Aperture distribution index 2 Bound water saturation 0.3 Bound carbon dioxide saturation 0.01 Hysteresis curvature parameter 1.5 Land coefficient 1 Salt water viscosity 6.81 * 10^-4 [Pa*s] Salt water density 1083.67 [kg / m^3] carbon dioxide viscosity 33.25 * 10^-6 [Pa*s] carbon dioxide density 412.39 [kg / m^3]

[0074] exist Figure 3 The carbon dioxide migration cloud map of the 20% porosity formation shows that the plume is concentrated near the injection point. Due to the low porosity, the migration and diffusion range is small. The high concentration areas (brightly colored areas) on slices 1 and 2 are compact and confined to the periphery of the injection point, reflecting the strong constraint of the low porosity formation on fluid seepage. Figure 4 For formations with 35% porosity, the diffusion range of the carbon dioxide plume is significantly increased compared to the 20% porosity condition. The high-concentration zone at the injection point extends outwards, and the color gradient band on the slice is wider, indicating that under medium porosity, the formation permeability is improved, providing a smoother channel for carbon dioxide transport, and the plume shows a trend of diffusion towards the far field. Figure 5 The study demonstrates the migration characteristics of formations with 50% high porosity, where the carbon dioxide plume diffusion is most complete. The high-concentration zone around the injection point expands significantly and forms a wider permeability zone on the slice, further reducing fluid flow resistance and allowing the plume to migrate rapidly and over a large area in the high-porosity medium.

[0075] Comparison of simulation results Figure 6 , 7 As can be seen, when the porosity is low, the absolute space that the formation can hold carbon dioxide is small. Even if the injected fluid can flow, the plume diffusion is limited due to the "insufficient storage space," resulting in a narrow range of impact. The injected carbon dioxide is easily saturated locally and cannot continue to be effectively displaced and diffused. With the increase of porosity, the storage space increases, and carbon dioxide has more "storage sites." The plume can fill and displace in a wider pore network, and the range of impact expands. When the porosity reaches 0.5, the storage space is sufficient, carbon dioxide can diffuse fully, contact more formation areas, and the sweep rate reaches a higher level.

[0076] In summary, this invention, through the aforementioned complete technology chain and driven by geological data from a specific well area, achieves a closed-loop process from geometric modeling, property correction, equation coupling, boundary setting, mesh optimization, parameter scanning to model verification. The generated simulation results not only highly reproduce the actual reservoir response, but also... Figures 3 to 7 The visualization output intuitively reveals the quantitative influence of key parameters such as porosity on plume transport, providing directly applicable technical support for the optimization of injection parameters and the assessment of storage safety in CCUS engineering.

[0077] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A method for simulating supercritical carbon dioxide plume transport, characterized in that, Includes the following steps: S1: Construct a heterogeneous geological model based on the actual geological production data of the target well area, including establishing the geometric entity of the reservoir calculation domain, constructing the gas injection well channel, and using spatial interpolation functions to map the porosity and permeability data interpreted by well logging to the calculation domain, forming a heterogeneous porosity field and permeability field that vary with spatial coordinates; S2: Define the phase state and property model of supercritical carbon dioxide in the simulation platform, including using the Peng-Robinson equation of state to describe the pressure-volume-temperature relationship of carbon dioxide under reservoir temperature and pressure conditions, and introducing the Brokaw mixing model to correct the dynamic viscosity of carbon dioxide for well environment. S3: Establish a multi-physics coupled numerical calculation model. Darcy's law is used to describe low-velocity seepage in the rock matrix region, and Brinkman's equation is used to describe the transition characteristics of free fluid and porous media flow in the wellbore and near-well high-permeability region. Real-time bidirectional transmission of formation parameters and fluid properties is achieved through the coupling of mass source and momentum source. S4: Conduct multi-parameter sensitivity comparative analysis, set porosity base, permeability difference, injection rate and inlet capillary pressure threshold as comparative experimental variables, use transient solver to simulate the diffusion trajectory of supercritical carbon dioxide plume in three-dimensional space, and extract saturation distribution and sweep range evolution data under different parameter combinations. S5: The numerical simulation results were verified based on the Buckley-Leverett one-dimensional displacement theory. The calculated saturation leading edge position was compared with the analytical solution. By adjusting the grid density and time step tolerance, the maximum relative error was controlled within 0.229%, thus confirming the effectiveness of the model.

2. The method for simulating supercritical carbon dioxide plume transport according to claim 1, characterized in that: The construction of the heterogeneous geological model specifically includes: A cuboid reservoir computational domain with dimensions of 500m×500m×100m was constructed in COMSOL software; A cylindrical gas injection well channel with a radius of 0.05m and a height that runs through the entire model is constructed at the geometric center of the computational domain; The core calculation region is defined by using Boolean operations to extract a 50m area above and below the gas injection port; and Import the porosity and permeability distribution data of the target well area with depth into the global definition, and establish a continuous interpolation function f(x,y,z) with spatial coordinates (x,y,z) as independent variables.

3. The method for simulating supercritical carbon dioxide plume transport according to claim 1, characterized in that: The Peng-Robinson equation of state is in the form of: Where p is pressure, V is molar volume, T is temperature, R is the universal gas constant, and a(T) and b are substance-specific parameters.

4. The method for simulating supercritical carbon dioxide plume transport according to claim 3, characterized in that: The Brokaw mixing model is used to correct the dynamic viscosity of supercritical carbon dioxide, and its expression is as follows: in, For low-pressure ideal gas viscosity, This is the critical viscosity reference value. For density comparison, , For temperature comparison, A, B, C, and D are all empirical coefficients.

5. The method for simulating supercritical carbon dioxide plume transport according to claim 1, characterized in that: In the multiphysics coupled numerical calculation model, the coupling between Darcy's law and the Brinkman equation is implemented as follows: Darcy's law is applied in the rock matrix region. Describing the seepage velocity vector With pressure gradient The relationship is given by k, where k is the penetration rate. For viscosity; in the wellbore and near-wellbore region, the Brinkman equation is used. Describing fluid motion, where This is the viscous diffusion term. For Darcy's resistance term, This is the pressure gradient driving term.

6. The method for simulating supercritical carbon dioxide plume transport according to claim 1 or 5, characterized in that: The four sets of comparative experimental variables used in the multi-parameter sensitivity analysis are as follows: Porosity baseline: 20%, 35%, 50%; Permeability differentials: Different orders of magnitude differences are set based on measured data from the well area; Injection rate: Set according to the CCUS engineering injection protocol; as well as Inlet capillary pressure threshold: 1×10 4 Pa, and a capillary pressure model was adopted. The calculation is performed, where λ is the porosity distribution index and S is the residual saturation. The inlet capillary pressure.

7. The method for simulating supercritical carbon dioxide plume transport according to claim 1, characterized in that: The verification of the numerical simulation results based on the Buckley-Leverett one-dimensional displacement theory specifically includes: Extract the saturation distribution data of the numerical simulation results along the central axis of the model; The theoretical position of the saturation front is calculated based on the Buckley-Leverett one-dimensional displacement theory; Calculate the root mean square error or maximum relative error between the numerical simulation results and the analytical solution; and Iteratively adjust the grid density and time step tolerance until the maximum relative error is ≤0.229%, then determine that the model meets the accuracy requirements.

8. The method for simulating supercritical carbon dioxide plume transport according to claim 2, characterized in that: The mesh partitioning adopts a user-controlled mesh strategy: A free tetrahedral ultra-fine grid is used around the gas injection well; Sparse grids are used in the far-well region; as well as A boundary layer mesh is set at the interface between Darcy's law and the Brinkman equation to capture gradient changes.

9. The method for simulating supercritical carbon dioxide plume transport according to claim 1 or 5, characterized in that: The transient solver employs an adaptive step-size algorithm, including the BDF method or the Generalized-alpha method, with a total simulation duration set to 4 years to accurately track the dynamic transport process of the supercritical carbon dioxide plume.

10. The method for simulating supercritical carbon dioxide plume transport according to claim 1, characterized in that: The post-processing visualization includes generating a three-dimensional saturation distribution cloud map, a multi-section slice map, an isosurface extraction map, and a comparison map of carbon dioxide sweep range under different porosity conditions, which are used to quantitatively analyze the characteristics of plume diffusion and the sensitivity influence of various parameters.

Images