A numerical simulation method for water injection-induced dynamic fracture seepage considering the seepage mechanism

Abstract

This invention relates to a numerical simulation method for water injection-induced dynamic fracture seepage considering the percolation mechanism. The steps are as follows: Step 1: Model establishment; Based on the basic equations of oil-water two-phase flow, considering the dual effects of percolation and displacement on oil-water flow and water injection-induced dynamic fractures, the capillary force model, relative permeability model, and matrix fracture flow rate of the displacement and percolation processes are respectively incorporated into the flow control equations to establish a mathematical model of water injection-induced dynamic fracture oil-water two-phase seepage under the dual effects of percolation and displacement; Step 2: Model solution; The fully implicit finite difference method is used to obtain the dynamic fracture oil-water two-phase seepage field data considering the dual effects of percolation and displacement. This invention can more accurately characterize the actual seepage environment of the reservoir and has higher accuracy in predicting development indicators.

Description

Technical Field

[0001] This invention relates to a numerical simulation method for dynamic crack seepage induced by water injection, taking into account the seepage mechanism. Background Technology

[0002] Current research primarily focuses on the adsorption effect between static fractures and the matrix, with limited research on the seepage mechanism under the dual effects of pressurized dynamic fractures and matrix adsorption (capillary force and relative permeability adsorption displacement). Furthermore, the characterization of dynamic fractures currently only equates to fractures by altering the near-wellbore conductivity, without considering the geometrical opening and extension characteristics of dynamic fractures during water injection. At the seepage scale, based on effective stress and formation pressure, the change in fracture permeability with pressure is simply considered. Currently, there are few seepage models that consider the coupling effect of dynamic fractures and adsorption, resulting in unclear seepage laws governing adsorption displacement during water injection-induced fracture propagation, which cannot further guide field development practices. To address the problems of current numerical models, it is necessary to study a mathematical model that simultaneously considers dynamically propagating fractures and their adsorption effect with the matrix, embedding dual capillary force curves to accurately characterize the fluid seepage laws during induced fracture occurrence and their impact on development effectiveness. Summary of the Invention

[0003] The present invention aims to address the problems existing in the prior art by proposing a seepage mathematical model for oil-water two-phase flow in water-injection induced dynamic fractures that simultaneously considers the dual effects of permeation, adsorption, and displacement.

[0004] The technical solution of this invention is as follows:

[0005] A numerical simulation method for water injection-induced dynamic fracture seepage considering the seepage mechanism, comprising the following steps:

[0006] Step 1: Build the model;

[0007] Based on the basic equations of oil-water two-phase flow, considering the dual effects of adsorption and displacement on oil-water flow and water injection-induced dynamic fractures, the capillary force model, relative permeability model, and matrix fracture flow rate of the displacement and adsorption processes are respectively incorporated into the flow control equations to establish a mathematical model of oil-water two-phase flow under the dual effects of adsorption and displacement and water injection-induced dynamic fractures.

[0008] Step 2: Model solution. The fully implicit finite difference method is used to obtain dynamic fracture oil-water two-phase flow field data considering the dual effects of seepage and displacement.

[0009] The specific process of building the model is as follows:

[0010] (1) Establish the basic differential equation for oil-water two-phase flow;

[0011] Taking the production term into account, the basic differential equations for oil-water two-phase flow are as follows:

[0012]

[0013]

[0014] (2) Establish a double capillary force model for pressurized osmosis

[0015] The capillary force calculation model considering the displacement process is as follows:

[0016]

[0017] In the formula: P cq To account for the capillary pressure during the displacement process, MPa;

[0018] P d To displace the inlet pressure, MPa;

[0019] S w S represents water saturation, a dimensionless quantity. wr This represents the water saturation level when the relative permeability of water approaches 0.

[0020] Dimensionless; λ is the pore size distribution index, which is dimensionless;

[0021] The capillary force calculation model considering the percolation process is as follows:

[0022]

[0023] In the formula: P cs To account for the capillary pressure during the absorption process, MPa;

[0024] P neg The maximum capillary pressure, in MPa, is the maximum pressure at which water saturation reaches its maximum value on the osmotic curve.

[0025] m, n, and α are the model fitting parameters;

[0026] S iw The bound water saturation is dimensionless; S or Residual oil saturation, dimensionless;

[0027] The phase permeation model considering the displacement process is as follows:

[0028]

[0029]

[0030] Where: K rwq The relative permeability of the aqueous phase during the displacement process is dimensionless.

[0031] roqTo account for the relative permeability of the oil phase during the displacement process, dimensionless;

[0032] The relative permeation model considering the osmosis process is as follows:

[0033]

[0034]

[0035] In the formula: ros The relative permeability of the aqueous phase is dimensionless to account for the osmosis process;

[0036] K rws The relative permeability of the oil phase during the absorption process is dimensionless.

[0037] Based on the basic seepage model, the dual effects of adsorption and displacement on oil-water seepage can be considered by substituting the capillary force model and relative permeability model of the displacement and adsorption processes into the seepage control equation.

[0038] (3) Establishing a numerical characterization model for artificial fracturing fractures in oil and water wells (EDFM)

[0039] The governing equations for the embedded discrete crack numerical model are:

[0040]

[0041]

[0042] In the formula: The matrix porosity is dimensionless. The crack porosity is dimensionless.

[0043] S l,m The saturation of phase l in the matrix is ​​dimensionless. l,f The saturation of phase l in the crack is dimensionless.

[0044] m Let mD be the matrix permeability. f Let mD be the crack permeability.

[0045] k rl, k is the relative permeability of phase l in the matrix, dimensionless; rl,f Let be the relative permeability of phase l in the fracture, dimensionless;

[0046] l represents the oil phase or the aqueous phase; o, w; μ l The viscosity of phase l is given in mPa·s; ρ l g represents the specific gravity of phase l, in MPa / m; represents the depth, in meters.

[0047] p l,m p represents the current formation pressure of phase l in the matrix, in MPa;l,f Let be the current formation pressure of phase l in the fracture, in MPa;

[0048] q m The flow rate (m) is the flow rate between the reservoir matrix and the embedded fractures. 3 / d;

[0049] q f The flow rate m is the flow rate between the fracture and the reservoir matrix. 3 / d;

[0050] q m =-q f ;

[0051] Where, q f The calculation process includes four cases;

[0052] The four cases represent three types of non-adjacent connection NNCs and the conductivity coefficient between the fracture and the wellbore; the definitions of the three types of non-adjacent connection NNCs are as follows:

[0053] ①NNC Type I: Connection between the crack segment and the matrix mesh it passes through;

[0054] ②NNC Type II: Connection between adjacent crack segments within a crack;

[0055] ③NNC Type III: Connection between intersecting different crack segments;

[0056] The crack flow rate between two meshes connected by a certain type of NNC is expressed as:

[0057] q f =λ l N NNC ΔP (15)

[0058] In the formula:

[0059] λ l For phase l mobility, mD·mPa -1 ·s -1 ;

[0060] T NNC Here, NNC represents the conductivity coefficient, mD·m;

[0061] ΔP is the pressure difference between adjacent grid cells, in MPa;

[0062]

[0063] In the formula: k NNC NNC permeability, mD;

[0064] A NNCFor NNC contact area, m 2 ;

[0065] d NNC For the feature distance between the connection and NNC, m;

[0066] ①NNC Type I

[0067] The matrix and embedded mesh crack conductivity coefficients for NNC type I are:

[0068]

[0069] In the formula: T mf The matrix and embedded mesh crack conductivity coefficient is given by mD·m.

[0070] A mf m is the contact area between the matrix and the embedded mesh crack. 2 ;

[0071] k mf The harmonic average of matrix permeability and crack permeability, mD;

[0072] d mf The mean normal distance from the matrix to the crack plane is in meters (m).

[0073]

[0074] The average normal distance d from the matrix to the crack plane mf for:

[0075]

[0076] In the formula: dv is the infinitesimal volume element within the grid block, m 3 ;

[0077] x n Let m be the distance from the infinitesimal volume element to the normal direction of the crack.

[0078] V is the volume of the mesh block, m 3 ;

[0079] ②NNC Type II

[0080] The conductivity coefficient between adjacent fracture segments in NNC type II is:

[0081]

[0082]

[0083]

[0084] In the formula: The conductivity coefficient between adjacent crack segments is mD·m;

[0085] k f1 The permeability of adjacent fracture segment 1 is given by mD; w f1 Let m be the aperture of the adjacent crack segment 1;

[0086] k f2 The permeability of adjacent fracture segment 2 is given by mD; w f2 Let m be the opening of the adjacent crack segment 2;

[0087] L int The length of the intersection line of the crack segments, in meters;

[0088] d f1 The distance from the center of adjacent crack segment 1 to the intersection line is m;

[0089] d f2 The distance from the center of adjacent crack segment 2 to the intersection line is m;

[0090] The formula for calculating intersecting cracks is:

[0091]

[0092]

[0093] Where: dS i It is a micro-element of area;

[0094] S i Let m be the area of ​​crack segment i. 2 ;x n The distance from the area element to the intersection line is in meters (m).

[0095] ③NNC Type III

[0096] The inter-segment conductivity coefficient of intersecting fracture segments in NNC type III is:

[0097]

[0098]

[0099]

[0100] In the formula: The conductivity coefficient between intersecting crack segments is mD·m;

[0101] A c Let m be the common surface of two intersecting crack segments. 2 ;

[0102] k f1 ′ represents the permeability of intersecting fracture segment 1, mD; k f2′ represents the permeability of intersecting fracture segment 1, in mD;

[0103] d seg1 Let be the distance from the centroid of intersecting crack segment 1 to the common surface, in meters (m).

[0104] d seg2 Let m be the distance from the centroid of the intersecting crack segment 2 to the common surface;

[0105] ④ Conductivity coefficient between fracture and wellbore

[0106] Based on the Peaceman mathematical model, the conductivity coefficient between fractures and the wellbore, and the equivalent well coefficient of the fracture mesh are derived as follows:

[0107]

[0108]

[0109] In the formula: WI f The conductivity coefficient between the fracture and the wellbore is mD·m;

[0110] r e m; r represents the equivalent well coefficients for the fractured mesh. w Let be the radius of the wellbore, in meters (m).

[0111] Δθ is the central angle of the radial well contained within the fracture, i.e., Δθ = 2π;

[0112] L f h f Here, represents the length and height of the crack, respectively, in meters.

[0113] To address the issue of dynamic cracking induced by water injection, dynamic crack parameters are introduced. and k f ′:

[0114] Among them, dynamic fracture permeability k f 'for

[0115]

[0116] In the formula: fi Let mD be the initial fracture permeability.

[0117] γ f Crack permeability stress-sensitive modulus, MPa -1 ;

[0118] p fi Initial fracture pressure, MPa; p f Pressure within the crack, MPa;

[0119] k miLet mD be the matrix permeability.

[0120] Among them, dynamic crack porosity for

[0121]

[0122] In the formula: The crack porosity is dimensionless. The initial crack porosity is dimensionless.

[0123] C f The crack compressibility coefficient is given in MPa. -1 ;

[0124] p f The pressure inside the crack is expressed in MPa.

[0125] In addition, the boundary conditions and initial conditions of the model are as follows:

[0126] 1) Boundary conditions

[0127] ①External boundary conditions

[0128] Constant pressure gradient external boundary condition:

[0129]

[0130] In the formula: The normal pressure gradient along the boundary direction is expressed in MPa / m.

[0131] const is a constant;

[0132] ②Inner boundary conditions

[0133] Bottom hole pressure:

[0134] P| rw =P wf (x,y,z)=const(35)

[0135] In the formula: P| rw To determine the bottom hole pressure, in MPa;

[0136] P wf Here, is the bottom hole flowing pressure, in MPa; const is a constant.

[0137] 2) Initial conditions

[0138] Assuming the reservoir pressure at the initial moment of the numerical simulation is a known function P0(x,y,z), the initial conditions for the reservoir pressure in the numerical simulation are as follows:

[0139] P(x,y,z,0)=P0(x,y,z)(36) During the development process, when the oil and water phases flow in the reservoir, it is necessary to determine the saturation field of a certain phase at the initial moment of development. The initial condition for water saturation is:

[0140] S w =S w0 (x,y,z)(37)

[0141] In the formula: S w This is a dimensionless parameter representing the water saturation at any point in the reservoir's various media systems.

[0142] P is a parameter representing the pressure at any point in the reservoir's various media systems, expressed in MPa.

[0143] The specific process of solving the model is as follows:

[0144] After performing finite difference on the seepage control equation, the discrete model is numerically solved. The calculation process is as follows: (1) Mesh generation and initialization: The matrix is ​​meshed and the properties of the matrix and cracks are initialized, and the conductivity is calculated.

[0145] (2) Crack mesh generation and embedding: Based on the crack mesh coordinates, the cracks are embedded into the matrix mesh;

[0146] (3) Calculate the macroscopic seepage parameters of the cracks: Calculate the permeability, porosity, relative permeability and other parameters of the cracks according to different crack states;

[0147] (4) Calculate the cross-flow rate: Calculate the conductivity and cross-flow rate between the crack and the matrix, and between the cracks, and construct the coefficient matrix of the linear equation system;

[0148] (5) Solve the seepage differential equation: Substitute the calculated crossflow rate into the seepage differential equation and solve it implicitly.

[0149] (6) Output results: Calculate up to the final time and output the pressure and saturation fields.

[0150] The technical advantages of this invention are as follows:

[0151] This invention establishes a mathematical model for dynamic fracture seepage in low-permeability reservoirs by considering the propagation of dynamically induced fractures and their interaction with the matrix, while utilizing two capillary force curves embedded with suction and displacement. Geometrically, the model equates the dynamic opening and closing of fractures to the activation and annihilation of the fracture grid over time. Furthermore, at the seepage scale, by introducing dynamic porosity and permeability of fractures and discretizing and numerically solving the seepage model, it reveals the laws governing the absorption and seepage during fracture formation and their impact on development performance. This allows for a numerical characterization of the coupling mechanism between water-injection-induced fractures and absorption and seepage at the reservoir scale, providing a more accurate representation of the actual reservoir seepage environment and resulting in higher accuracy in predicting development indicators. Attached Figure Description

[0152] Figure 1 This is a schematic diagram for calculating the flow rate between the matrix and the crack.

[0153] Figure 2 This is a dynamic characterization process for embedded discrete cracks.

[0154] Figure 3 This is a flowchart of dynamic crack calculation.

[0155] Figure 4 This diagram illustrates the process of dynamic crack formation induced by water injection.

[0156] Figure 5 This is a schematic diagram showing the relationship between effective pressure and permeability in the core.

[0157] Figure 6 This is a schematic diagram showing the relationship between effective pressure and permeability in the core.

[0158] Figure 7 The curves show the relationship between the degree of core extraction with the injection volume for different fracture sizes.

[0159] Figure 8 To simulate and calculate the production system considering the seepage behavior under the influence of induced cracks and dual capillary forces.

[0160] Figure 9 This is a contour map of the matrix pressure field.

[0161] Figure 10 This is a field contour map of the water saturation of the matrix.

[0162] Figure 11 This is a graph showing the change in bottom hole pressure over time, with and without considering the effect of seepage.

[0163] Figure 12 The graph shows the changes in percolation rate and cumulative percolation over time, with and without considering percolation.

[0164] Figure 13The graph shows the changes in oil production rate and cumulative oil production over time, regardless of whether the dual effects of pressurized permeation and adsorption are considered. Detailed Implementation

[0165] Example 1—Model Validation (Comparison of Experimental Results of Water Injection-Induced Crack Initiation)

[0166] (1) Numerical Model Construction

[0167] To verify the accuracy of the water injection-induced fracture model constructed above, a core numerical model at the experimental scale was constructed based on the water injection-induced fracture initiation experiment in a low-permeability reservoir, and the water injection-induced fracture experimental process of the core was simulated.

[0168] The constructed numerical model has a mesh size of 22×22×33. The model porosity is 7.9%, permeability is 0.845 mD, crude oil viscosity is 1 mPa·s, and pressure is set to 4 MPa based on experimentally obtained breakthrough pressure. The total salinity of water is 30000 mg / L, and the model temperature is set to 25℃. Other parameter settings refer to the default values ​​of the black oil model; therefore, this factor is considered when setting other properties, and corresponding minor adjustments are made to the model parameter settings. When setting up the rock sample, a square core with a size of 2.5 cm × 2.5 cm × 5 cm is used to multiply the injection and production by appropriate coefficients to match the laboratory data conditions. A water injection well is set at the front end of the rock sample for fracture simulation.

[0169] (2) Numerical simulation verification of water injection-induced crack initiation experiment

[0170] The process of dynamic crack formation in numerical models is as follows: Figure 4 As shown, the cracks continued to expand as water was injected, eventually forming a through crack;

[0171] Based on the simulation results of the numerical model, the relationship between injection pressure and permeability can be obtained as follows: Figure 5 As shown;

[0172] By fitting and comparing the numerical model with the water injection-induced fracture experiment of the core, it can be found that the pressure curve of the above numerical model during water injection shows frequent fluctuations, just like the experimental pressure curve, and multiple breakthrough characteristics appear. The breakthrough pressure of the former is between 2.8 MPa and 5 MPa, and the core is the first breakthrough.

[0173] For the model, after water injection and pressure buildup, the injection pressure rises sharply, approaching its peak at an effective pressure of 3.0 MPa–3.3 MPa. The pressure reaches the fracture initiation pressure but does not yet open the fracture. As water injection continues, at 2.3 minutes, the injection pressure increases to 4.5 MPa, reaching the model's breakthrough pressure and initiating microcracks. After reaching the breakthrough pressure, the pressure decreases and then, with increasing water injection volume and duration, pressure fluctuations rise, the fracture continues to extend, and permeability increases accordingly. Finally, at 10 minutes, the fracture extends to the core surface, at which point the injection pressure remains stable. Therefore, the numerical simulation results of the water injection-induced fracture initiation experiment accurately match the experimental conclusions, further demonstrating the relationship between pressure and induced fracture initiation.

[0174] (3) Verification of the accuracy of commercial software and numerical models

[0175] To verify the superiority of the water injection-induced fracture model established in this paper over existing commercial numerical simulation software in dynamic fracture simulation, this section utilizes commercial software (Tnavigator) to establish a numerical model of fractured rock cores. Water injection-induced fracture initiation experimental simulations were conducted, yielding the following results: Figure 6 The pressure chart is shown. Through fitting and comparison, the model built by Tnavigator can only generate static cracks when simulating water injection-induced fractures. Therefore, it cannot accurately describe the crack initiation and the timing of microcrack formation. Compared with the dynamic cracks built by the aforementioned model, the accuracy of static cracks is not high enough; it can only simply simulate the process of water injection producing cracks and cannot characterize the dynamic propagation of cracks. Therefore, the numerical model built above can be used as a model to verify the accuracy of the experiment.

[0176] Example 2—Model Validation (Comparison of Results from Dynamic Immersion Experiments under Pressure)

[0177] (1) Numerical Model Construction

[0178] To verify the accuracy of the water injection-induced fracture model considering the seepage mechanism, an experimental study on the influencing factors of the coupling effect between pressurized dynamic seepage and fracture in low-permeability reservoirs was conducted. Based on the established numerical model, a core numerical simulation was constructed, and the experimental results were compared with the numerical simulation results to verify the accuracy of the numerical model in pressurized dynamic seepage. The numerical model retains the original mesh properties, and Model 1 is set with a porosity of 11.9% and a permeability of 0.421 × 10⁻⁶. -3 μm 2 The formation porosity of Model 2 was set to 10.9%, and the permeability to be 0.381 × 10⁻⁶. -3 μm 2 Model 3 was set with a porosity of 11.3% and a permeability of 0.357 × 10⁻⁶. -3 μm 2Rock sample 4 was set with a porosity of 11.6% and a permeability of 0.244 × 10⁻⁶. -3 μm 2 The formation temperature was uniformly set to 50℃, and brine was selected as the injection medium at the front end of the rock samples. The total mineralization of the formation water was 78000 mg / L. Model 1 was used as the matrix, and fracture simulations were performed on models 2, 3, and 4, with fracture sizes set as 1 / 3 fracture, 2 / 3 fracture, and through fracture, respectively, while other parameters remained unchanged. Four numerical models were created. After the models were built, production simulations were conducted, with the injection well at the front end and the production well at the back end. Water injection was stopped when the water cut of the production well reached 98%. The well was then left to stagnate for 12 hours before resuming water injection, and the relationship between the recovery rate and the injection volume was recorded at different times.

[0179] (2) Numerical simulation verification of dynamic permeation experiment of fractured core

[0180] Numerical model results and experimental results on the relationship between the degree of core recovery with injection volume for different fracture sizes are as follows: Figure 7 As shown in the figure, after fitting and comparing the results with the dynamic permeation experiments of fractured cores, it can be seen that the numerical model results of dynamic permeation have a very high degree of fit with the experimental results. From the numerical model results, matrix model 1 had a recovery rate of 48.68% before well shut-in and a final recovery rate of 50.65% after well shut-in, representing an increase of 1.97% in permeation recovery. Model 2 with 1 / 3 fractures had a recovery rate of 56.4% before well shut-in and a final recovery rate of 59.93% after well shut-in, representing an increase of 3.53% in permeation recovery. Model 3 with 2 / 3 fractures had a recovery rate of 61.58% before well shut-in and a final recovery rate of 67.15% after well shut-in, representing an increase of 5.57% in permeation recovery. Model 4 with through fractures had a recovery rate of 40% before well shut-in and a final recovery rate of 45.89% after well shut-in, representing an increase of 5.89% in permeation recovery.

[0181] The model verified the experimental conclusion that fractures can improve the degree of displacement and production, and the degree of production increases with the increase of fractures. However, after the fractures penetrate through, they form high-permeability channels, which reduces the degree of displacement and production. The contribution of well-drainage to the degree of production increases with the length of the fractures. This is because the longer the model is, the larger its contact area with the injected water, which increases the permeation effect.

[0182] Specific examples

[0183] Based on the capillary force curves considering both adsorption and displacement, this study explores the impact of adsorption on the seepage field and production dynamics. Numerical model examples considering and not considering adsorption are established, with fractures set as dynamic fractures. The production regime and reservoir numerical simulation parameters remain consistent with the previous ones. The operating regime is water injection-pump-pump, i.e., a continuous cycle of three operating regimes: constant pressure production, water injection, and well shut-in. The production regime is as follows: Figure 8 As shown.

[0184] By comparing the matrix pressure field ( Figure 9 It can be observed that the impact of percolation on pressure is significant depending on whether percolation is considered. When percolation is considered, more water enters the matrix, maintaining a higher pressure in the matrix. Therefore, water injection that considers percolation has the effect of replenishing the energy of the reservoir.

[0185] By comparing the contour maps of the matrix saturation field, it can be found that ( Figure 10 The effect of osmosis on water saturation is considered, with and without consideration. This is because, considering osmosis, due to capillary forces, the injected water can penetrate deeper into the matrix, resulting in a higher water saturation.

[0186] Figure 11 The graph shows the change in bottom hole pressure over time, considering both pressurized seepage and suction effects. Analysis of the bottom hole pressure curves reveals that the pressure gradually increases during water injection, and decreases after well shut-in as the injected water gradually enters the matrix. The difference in bottom hole pressure is not significant when considering seepage because the injection pressure is much greater than the capillary force of seepage; therefore, the effect of seepage on bottom hole pressure is not significant.

[0187] Figure 12 To illustrate the changes in water production rate and cumulative water production over time, considering whether seepage is taken into account, during the first constant-pressure production, the water production rate without considering seepage is higher than that with seepage. This is because, during the first extraction, the capillary force considering seepage is equal to the resistance during extraction, resulting in a lower water production. However, for the cumulative water production throughout the entire production process, the cumulative water production without considering seepage is higher than that with seepage.

[0188] Figure 13 To investigate the changes in oil production rate and cumulative oil production over time with and without considering percolation, a comparison of the oil production rate and cumulative oil production with and without considering percolation reveals that as production continues, the oil production rate decreases regardless of whether percolation is considered. However, the rate of decrease in the oil production rate is reduced through water injection energy storage.

Claims

1. A numerical simulation method for water injection-induced dynamic fracture seepage considering the seepage mechanism, characterized in that: The steps are as follows: Step 1: Build the model; Based on the basic equations of oil-water two-phase flow, considering the dual effects of adsorption and displacement on oil-water flow and water injection-induced dynamic fractures, the capillary force model, relative permeability model, and matrix fracture flow rate of the displacement and adsorption processes are respectively incorporated into the flow control equations to establish a mathematical model of oil-water two-phase flow under the dual effects of adsorption and displacement and water injection-induced dynamic fractures. The specific process for establishing the model is as follows: (1) Establish the basic differential equation for oil-water two-phase flow; Taking the production term into account, the basic differential equations for oil-water two-phase flow are as follows: (3) (4) (2) Establish a double capillary force model for pressurized osmosis The capillary force calculation model considering the displacement process is as follows: (5) In the formula: To account for the capillary pressure during the displacement process, MPa; To displace the inlet pressure, MPa; The water saturation level is dimensionless. The water saturation is the degree of water saturation when the relative permeability of water approaches 0, and is dimensionless. is a dimensionless index representing the pore size distribution. The capillary force calculation model considering the percolation process is as follows: (6) In the formula: To account for the capillary pressure during the absorption process, MPa; The maximum capillary pressure, in MPa, is the maximum pressure at which water saturation reaches its maximum value on the osmotic curve. , Parameters for model fitting; To constrain water saturation, dimensionless; Residual oil saturation, dimensionless; The phase permeation model considering the displacement process is as follows: (7) (8) In the formula: The relative permeability of the aqueous phase during the displacement process is dimensionless. To account for the relative permeability of the oil phase during the displacement process, dimensionless; The relative permeability model considering the osmosis process is as follows: (9) (10) In the formula: The relative permeability of the aqueous phase is dimensionless to account for the osmosis process; The relative permeability of the oil phase during the absorption process is dimensionless. Based on the basic seepage model, the dual effects of adsorption and displacement on oil-water seepage can be considered by substituting the capillary force model and relative permeability model of the displacement and adsorption processes into the seepage control equation. (3) Establish a numerical characterization model for artificial fracturing fractures in oil and water wells (EDFM) The governing equations for the embedded discrete crack numerical model are: (11) (12) In the formula: The matrix porosity is dimensionless. The crack porosity is dimensionless. The saturation of phase l in the matrix is ​​dimensionless. The saturation of phase l in the crack is dimensionless. The matrix permeability is expressed in mD. Let mD be the crack permeability. The relative permeability of phase l in the matrix is ​​dimensionless. Let be the relative permeability of phase l in the fracture, dimensionless; l represents the oil phase or the aqueous phase, o and w; Let l be the viscosity of phase l, in mPa·s; The specific gravity of phase l is expressed in MPa / m. Let m be the depth. ρ represents the current formation pressure of phase l in the matrix, in MPa; Let be the current formation pressure of phase l in the fracture, in MPa; The flow rate (m) is the flow rate between the reservoir matrix and the embedded fractures. 3 / d; The flow rate m is the flow rate between the fracture and the reservoir matrix. 3 / d; ； in, The calculation process includes four cases; The four cases represent three types of non-adjacent connection NNCs and the conductivity coefficient between the fracture and the wellbore; The three types of non-adjacent connections (NNCs) are defined as follows: NNC Type I: Connection between the crack segment and the matrix mesh it passes through; NNC Type II: Connection between adjacent crack segments within a crack; NNC Type III: Connection between intersecting different crack segments; The crack flow rate between two meshes connected by a certain type of NNC is expressed as: (15) In the formula: For phase l mobility, mD·mPa -1 ·s -1 ; Here, NNC represents the conductivity coefficient, mD·m; The pressure difference between adjacent grid cells, in MPa; (16) In the formula: NNC permeability, mD; For NNC contact area, m 2 ; For the feature distance between the connection pair and the NNC, m; NNC Type I The matrix and embedded mesh crack conductivity coefficients for NNC type I are: (17) In the formula: The matrix and embedded mesh crack conductivity coefficient is given by mD·m. m is the contact area between the matrix and the embedded mesh crack. 2 ; The harmonic average of matrix permeability and crack permeability, mD; d mf The mean normal distance from the matrix to the crack plane is in meters (m). (18) The average normal distance d from the matrix to the crack plane mf for: (19) In the formula: Let m be a volume element within a mesh block. 3 ; Let m be the distance from the infinitesimal volume element to the normal direction of the crack. m is the volume of the mesh block. 3 ; NNC Type II The conductivity coefficient between adjacent fracture segments in NNC type II is: (20) (21) (22) In the formula: The conductivity coefficient between adjacent crack segments is mD·m; k f1 The permeability of adjacent fracture segment 1 is given by mD; w f1 Let m be the aperture of the adjacent crack segment 1; k f2 The permeability of adjacent fracture segment 2 is given by mD; w f2 Let be the aperture of the adjacent crack segment 2, in meters. The length of the intersection line of the crack segments, in meters; The distance from the center of adjacent crack segment 1 to the intersection line is m; The distance from the center of adjacent crack segment 2 to the intersection line is m; The formula for calculating intersecting cracks is: (23) (24) In the formula: S1~S4 are the areas of crack segments 1~4, in meters. 2 ; NNC Type III The inter-segment conductivity coefficient of intersecting fracture segments in NNC type III is: (25) (26) (27) In the formula: The conductivity coefficient between intersecting crack segments is mD·m; Let m be the common surface of two intersecting crack segments. 2 ; Let mD be the permeability of the intersecting fracture segment 1. Let mD be the permeability of the intersecting fracture segment 1. Let be the distance from the centroid of intersecting crack segment 1 to the common surface, in meters (m). Let m be the distance from the centroid of the intersecting crack segment 2 to the common surface; Conductivity coefficient between fracture and wellbore Based on the Peaceman mathematical model, the conductivity coefficient between fractures and the wellbore, and the equivalent well coefficient of the fracture mesh are derived as follows: (30) (31) In the formula: The conductivity coefficient between the fracture and the wellbore is mD·m; is the equivalent well coefficient for the fractured mesh, m; Let be the radius of the wellbore, in meters (m). The central angle of the radial well contained within the fracture is... ; , Here, the length and height of the crack are respectively, in meters; To address the issue of dynamic cracking induced by water injection, dynamic crack parameters are introduced. and : Among them, dynamic fracture permeability for (32) In the formula: Let mD be the initial fracture permeability. Crack permeability stress-sensitive modulus, MPa -1 ; Initial pressure within the crack, MPa; Pressure within the crack, MPa; The matrix permeability is expressed in mD. Among them, dynamic crack porosity for (33) In the formula: The crack porosity is dimensionless. The initial crack porosity is dimensionless. The crack compressibility coefficient is given in MPa. -1 ; The pressure inside the crack is MPa; In addition, the boundary conditions and initial conditions of the model are as follows: 1) Boundary conditions ① External boundary conditions Constant pressure gradient external boundary condition: (34) In the formula: The normal pressure gradient along the boundary direction is expressed in MPa / m. It is a constant; ② Internal boundary conditions Bottom hole pressure: (35) In the formula: To determine the bottom hole pressure, in MPa; The bottom hole flowing pressure is in MPa. It is a constant; 2) Initial conditions Assume the reservoir pressure at the initial moment of the numerical simulation is a known function. The initial conditions for pressure in the reservoir numerical simulation are: (36) During development, when oil and water flow in the reservoir, it is necessary to determine the saturation field of a certain phase at the initial moment of development. The initial conditions for water saturation are as follows: (37) In the formula: This is a dimensionless parameter representing the water saturation at any point in the reservoir's various media systems. P is a parameter representing the pressure at any point in the reservoir's various media systems, expressed in MPa. Step 2: Model solution. The fully implicit finite difference method is used to obtain dynamic fracture oil-water two-phase flow field data considering the dual effects of seepage and displacement.

2. The numerical simulation method for water injection-induced dynamic fracture seepage considering the seepage mechanism according to claim 1, characterized in that: The specific process of solving the model is as follows: After performing finite difference calculations on the seepage control equations, the discrete model is numerically solved. The calculation process is as follows: (1) Mesh generation and initialization: The matrix is ​​meshed and the physical properties of the matrix and cracks are initialized, and the conductivity is calculated; (2) Crack mesh generation and embedding: Based on the crack mesh coordinates, the cracks are embedded into the matrix mesh; (3) Calculate the macroscopic seepage parameters of the cracks: calculate the permeability, porosity and relative permeability of the cracks according to different crack states; (4) Calculate the cross-flow rate: Calculate the conductivity and cross-flow rate between the crack and the matrix, and between the cracks, and construct the coefficient matrix of the linear equation system; (5) Solve the seepage differential equation: Substitute the calculated crossflow rate into the seepage differential equation and solve it implicitly. (6) Output results: Calculate up to the final time and output the pressure and saturation fields.

Images