Numerical simulation method for shale condensate gas reservoir development based on universal pEDFM

Abstract

The present invention belongs to the technical field of gas field exploitation, and discloses a numerical simulation method for shale condensate gas reservoir development based on a universal pEDFM, comprising the following steps: S1, using the universal pEDFM to calculate the conductivity of the connection between various grids; S2, constructing a commercial numerical simulation software CMG equivalent model based on the universal pEDFM; S3, calling the nonlinear solver in CMG to solve the shale condensate gas reservoir component model. The present invention adopts the above-mentioned numerical simulation method for shale condensate gas reservoir development based on the universal pEDFM, which not only ensures the advantages of the universal pEDFM in calculation accuracy and processing of complex fracture networks, but also avoids the difficulty of developing additional high-performance nonlinear solvers, and provides a numerical simulation tool with the best theoretical comprehensive calculation performance for the development of fractured shale condensate gas reservoirs.

Description

Technical Field

[0001] The invention relates to the technical field of gas field exploitation, and in particular to a numerical simulation method for shale condensate gas reservoir development based on a universal pEDFM. Background Art

[0002] As an unconventional oil and gas resource, shale condensate gas reservoirs are of great significance to increasing my country's oil and gas resource output and achieving diversified energy supply. Compared with conventional gas reservoirs, their development process faces the following two situations: First, in the early stage of depletion development of shale condensate gas reservoirs, because the bottom hole pressure is lower than the dew point pressure, condensate will precipitate in the formation near the well, resulting in a sharp drop in production capacity. In the later stage of development, due to the continuous decrease in reservoir pressure, retrograde condensation damage is more serious, and gas injection and huff-and-puff are generally required to improve recovery. Second, the reservoir porosity and permeability are extremely low, and horizontal well fracturing development is required to increase production.

[0003] The dynamic simulation of gas injection energy replenishment in shale condensate gas reservoirs is of great significance for predicting the energy replenishment effect and designing energy replenishment schemes. The reservoir seepage model is a component model, and its seepage medium includes dense matrix and various fractures. Therefore, the key to the dynamic simulation of shale oil gas injection energy replenishment lies in how to carry out seepage simulation in fractured reservoirs. The commonly used numerical simulation frameworks for fractured reservoirs are: discrete fracture model (DFM / DFN), embedded discrete fracture model (EDFM) and pEDFM developed in recent years. DFM reduces the dimension of fractures and generates matching unstructured grids to make them located at the intersection between matrix grids. Compared with the dual medium model, DFM can more accurately describe the geometric characteristics of fractures and has higher calculation accuracy. However, due to the complex geometry of the complex fracture network formed by fractures, it is very difficult to generate corresponding matching grids, and it is difficult to avoid the generation of a large number of small-size grids near the intersection of fractures and the narrow area between fractures, which significantly increases the calculation cost and convergence difficulty.

[0004] EDFM can avoid the division of matching unstructured grids. It only needs to divide the matrix part into structured grids, embed discrete fractures into the matrix grid, and treat fractures as source and sink items in the matrix grid, which reduces the difficulty of matrix grid generation. At present, EDFM has been widely used in numerical simulation of fractured reservoirs.

[0005] However, EDFM cannot effectively handle the situation where the fracture grid permeability is lower than the matrix permeability. Therefore, pEDFM is proposed to solve this problem. This model projects the fracture grid embedded in the matrix grid onto the matrix grid intersection surface, adds additional matrix-fracture (mf) connection relationships, and weakens the connection conductivity between the original adjacent matrix grids. Moreover, pEDFM can solve the problem that EDFM produces significant errors when simulating two-phase flow in natural fracture reservoirs and cannot correctly describe the flux distribution. Through continuous improvement of pEDFM technology, a general pEDFM that can maintain the superiority of EDFM calculation in any seepage scenario is proposed, laying the theoretical foundation for the commercial application of pEDFM. However, when the general pEDFM is applied to complex component models (for example, shale condensate gas reservoir development containing more than ten components or pseudo-components and flash calculations of each component), there is a problem of high cost of developing high-performance solvers.

[0006] Therefore, the present invention proposes a numerical simulation method for shale condensate gas reservoir development based on a universal pEDFM. Summary of the invention

[0007] The purpose of the present invention is to provide a numerical simulation method for shale condensate gas reservoir development based on a universal pEDFM, which not only ensures the advantages of the universal pEDFM in calculation accuracy and processing of complex fracture networks, but also avoids the difficulty of developing additional high-performance nonlinear solvers, thereby providing a numerical simulation tool with the best theoretical comprehensive calculation performance for the development of fractured shale condensate gas reservoirs.

[0008] To achieve the above object, the present invention provides a numerical simulation method for shale condensate gas reservoir development based on a universal pEDFM, comprising the following steps:

[0009] S1. Use the general pEDFM to calculate the conductivity of connections between various grids;

[0010] S2, construct a CMG equivalent model based on the commercial digital simulation software of the general pEDFM;

[0011] S3. Call the nonlinear solver in CMG to solve the shale condensate gas reservoir component model.

[0012] Preferably, based on EDFM, pEDFM establishes additional connections and weakens original connections by projecting cracks in the matrix grid onto the matrix grid surface. In any seepage scenario, the general pEDFM algorithm uses a micro-shift method to obtain the projected configuration of high-conductivity cracks, and uses a non-projected conductivity multiplier method to process low-conductivity cracks.

[0013] Preferably, when the matrix permeability is isotropic, the conductivity of the increased mf connection in pEDFM is as follows:

[0014]

[0015] in, and They are f i and f j exist projection on is the crack grid f j The permeability, is the crack grid f j The opening degree, is the matrix grid m i With matrix grid m j The intersection of is m j The penetration rate, is m j arrive distance;

[0016] The conductivity of the original mm connection is weakened to:

[0017]

[0018] in, is m i arrive The distance is the matrix grid m i With matrix grid m k The intersection of is the matrix grid m k arrive distance;

[0019] By halving the conductivity of the matrix grid, the conductivity of the original mf connection is weakened to (m i ,f i ) as an example, the conductivity is as follows:

[0020]

[0021] Based on the improved pEDFM, a connection is established between two crack meshes projected on the same matrix mesh intersection but in different matrix meshes. The connection is assumed to be (f i ,f j ), the conductivity of the connection is calculated as:

[0022]

[0023] in, and The crack grids f i and f jThe projected area on the intersection of the same matrix grid;

[0024] The general pEDFM uses a non-projection method, namely the conductivity multiplier method, to deal with low-conductivity fractures. For the connection intersected by low-conductivity fractures, the updated conductivity of the connection is:

[0025]

[0026] Among them, T ij is the original conductivity based on EDFM, T′ ij is the updated conductivity of the general pEDFM, T f is the conductivity of the low conductivity fracture, k f is the permeability of low conductivity fractures, w f is the opening of the low conductivity fracture, A ij is the seepage area of ​​the corresponding connection.

[0027] Preferably, in step S2, based on the inter-grid connections and corresponding conductivity values ​​obtained by the general pEDFM, a CMG model equivalent to the real reservoir pEDFM model is constructed, which needs to meet the following conditions:

[0028] (1) The volume of each effective grid in the CMG model is the same as the pore volume of the corresponding grid in the real reservoir model;

[0029] (2) The connection table between the effective grids in the CMG model is the same as the connection table between the corresponding grids obtained based on the general pEDFM of the real reservoir model;

[0030] (3) The well index of the grid where the injection and production wells are located in the CMG model is the same as the well index calculated for the corresponding grid in the real model.

[0031] Preferably, an equivalent CMG model is constructed. For the two-dimensional pEDFM model, a corresponding one-dimensional CMG model is first constructed so that the constructed one-dimensional CMG model meets the above three requirements and is equivalent to the real model. The specific method is as follows:

[0032] S21, let the grid sizes of the i-th matrix grid in the x, y, and z directions be Δx i , Δy i and Δz i Since CMG requires the mesh size of the one-dimensional model in the y direction and the z direction to be the same value, the mesh size of the one-dimensional model in the y direction and the z direction is set to the true Δy of all meshes. i and Δz i The average or the most digit of , where i = 1, 2, ..., n m ; may be recorded as In order to keep the grid volume unchanged, the Δx of the i-th matrix grid in the CMG one-dimensional model is i Calculated as:

[0033]

[0034] Among them, V m,i is the volume of the ith matrix grid;

[0035] S22. The crack grid is placed behind the matrix grid in the one-dimensional model, and its size in the y and z directions still needs to be and The corresponding Δx i Calculated as:

[0036]

[0037] Among them, V f,i is the volume of the i-th crack grid;

[0038] S23. The size of the invalid grid in the x direction takes an arbitrary value.

[0039] Preferably, in step S3, the calculation process of the general pEDFM and CMG is integrated as follows:

[0040] S31. By judging whether the permeability is higher than the average permeability of the matrix system, the fracture is classified into high conductivity type F p and low conductivity type F np ; Wherein, the subscripts p and np respectively indicate that the corresponding types of cracks will be processed by the projection method and the non-projection method respectively;

[0041] S32, ignore F np , use the nudge method to get F np The projection configuration of each fracture grid is based on the original pEDFM to obtain various connections between grids and calculate the corresponding conductivity, which is recorded as cont1. According to the improved pEDFM, when there are multiple high-conductivity fracture grids projected onto the same matrix grid surface, the connection between any two fracture grids in different matrix grids is constructed, which is recorded as cont2, so that the grid connection cont = cont1∪cont2 of the reservoir model is obtained.

[0042] S33, use the non-projected conductivity multiplier method to update the F in cont np The conductivity of each connection that intersects;

[0043] S34. Based on cont and the corresponding conductivity, the corresponding CMG equivalent model is constructed, so as to call the CMG component model nonlinear solver to calculate the global nonlinear equation group for the development of shale condensate gas reservoirs, and obtain the pressure, saturation, concentration distribution of each component and well dynamic data.

[0044] Therefore, the present invention adopts the above-mentioned general pEDFM-based shale condensate gas reservoir development numerical simulation method, and the beneficial effects are as follows:

[0045] (1) The numerical simulation method of shale oil gas injection based on the integration of general pEDFM and CMG does not need to generate a grid that matches the fracture network morphology, reducing the computational cost and convergence difficulty. Compared with EDFM, it can simulate the impact of high and low conductivity fractures in the formation on the seepage field with high precision. Compared with LGR, it can simulate more complex fracture morphologies.

[0046] (2) The present invention uses the commercial software CMG for calculation, avoiding the difficulty of developing an additional high-performance nonlinear solver. In theory, it is a numerical simulation tool with the best comprehensive calculation performance for the study of gas injection and energy replenishment in fractured condensate gas reservoirs.

[0047] (3) The present invention verifies that the computational accuracy of the general-purpose pEDFM in shale condensate gas reservoir depletion development and multi-round throughput development is significantly higher than that of EDFM, and that it can effectively handle complex reservoir models with complex fracture networks and multi-well production. This reveals the feasibility, necessity and great potential of pEDFM, which has significant computational superiority over the most commonly used EDFM, in actual mine applications.

[0048] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a model overview of the commonly used numerical simulation framework for fractured reservoirs; among them, (a) is the EDFM model; (b) is the pEDFM model;

[0050] Figure 2 It is a schematic diagram of the one-dimensional CMG model built based on pEDFM;

[0051] Figure 3 is the reservoir model diagram;

[0052] Figure 4 is the phase permeability curve data used in Example 1; where (a) is the oil-water phase permeability curve; (b) is the gas-liquid phase permeability curve;

[0053] Figure 5is a comparison of well dynamic data calculated by EDFM, pEDFM and reference model; where (a) is the oil production rate; (b) is the gas production rate; (c) is the water production rate;

[0054] Figure 6 is the comparison of well performance calculated by EDFM, pEDFM and reference model; where (a) is the oil production rate; (b) is the gas production rate; (c) is the water production rate;

[0055] Figure 7 is a comparison of the pressure distributions calculated by different methods in Example 2; where (a) is the pressure of the reference model; (b) is the pressure of the pEDFM; (c) is the pressure of the EDFM;

[0056] Figure 8 Comparison of oil saturation distributions calculated by different methods in Example 2; (a) is the oil saturation of the reference model; (b) is the oil saturation of pEDFM; (c) is the oil saturation of EDFM;

[0057] Fig. 9 Comparison of gas saturation distributions calculated by different methods in Example 2; (a) is the gas saturation of the reference model; (b) is the gas saturation of pEDFM; (c) is the gas saturation of EDFM;

[0058] Fig.10 is a schematic diagram of the reservoir model in Example 3;

[0059] Fig.11 The pressure and saturation distributions after 3 years and 10 years calculated by the general pEDFM in Example 3 are compared; (a) is the pressure distribution after 3 years; (b) is the pressure distribution after 10 years; (c) is the oil saturation distribution after 3 years; (d) is the oil saturation distribution after 10 years; (e) is the gas saturation distribution after 3 years; (f) is the gas saturation distribution after 10 years; (g) is the water saturation distribution after 3 years; (h) is the water saturation distribution after 10 years;

[0060] Fig.12 This is the comparison of oil production, gas production and water production rates calculated by the general pEDFM in Example 3; where (a) is the oil production rate; (b) is the gas production rate; (c) is the water production rate;

[0061] Fig.13 It is a flow chart of the numerical simulation method for shale condensate gas reservoir development based on the universal pEDFM of the present invention. DETAILED DESCRIPTION

[0062] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0063] 1. Basic seepage model and control equations.

[0064] 1. Flow equation: The material balance discrete equation of each component in the oil phase, gas phase and water component is as follows:

[0065]

[0066] Among them, N i is the mole fraction of component i per unit grid volume, is the mole fraction of water per unit grid volume, D is the depth, n c is the number of components, p is the pressure, P cog is the capillary force between oil and gas, P cwo is the capillary force between oil and water, q is the injection or production rate, T j is the conductivity of phase j, V is the grid volume, y ij is the mole fraction of component i in phase j, γ is the specific gravity, Δt is the time step, is a function, the superscript n and n+1 are the previous time step and the next time step respectively. For explicit grids, the superscript m represents the nth time step, and for fully implicit grids, the superscript m represents the n+1th time step. It is assumed that there is no mass transfer between hydrocarbons and water phases.

[0067] N i The relationship with porosity, phase molar density, saturation and composition is as follows:

[0068] N i =φ(ρ o S o y io +ρ g S g y ig )i=1,…,n c (3)

[0069]

[0070] Where φ is the porosity, S is the fluid saturation, and ρ is the density.

[0071] 2. Saturation equation: Saturation and N i and ρ m (m=o,g,w) relationship is as follows:

[0072]

[0073] 3. Phase equilibrium equation: Flash evaporation is used in the component model to calculate the molar fraction of component i in the gas and liquid phases. When the system is in equilibrium, the chemical potential of each component in each phase is equal. The thermodynamic equilibrium equation expressed in fugacity is as follows:

[0074] f i V -f i L =φ i V y i p-φ i L x i p=0 (8)

[0075] Introducing the equilibrium constant K, mole fraction x i and i The following conditions are met:

[0076]

[0077] Among them, f i L is the fugacity of component i in the liquid phase, f i L is the fugacity of component i in the gas phase, φ i V is the fugacity coefficient of component i in the gas phase, φ i L is the fugacity coefficient of component i in the liquid phase.

[0078] Substituting the PR state equation into equation (8) yields the calculated fugacity coefficient as follows:

[0079]

[0080] The present invention considers 11 components in shale condensate gas reservoirs, as shown in Table 1.

[0081] Table 1 Relevant parameters of each component of reservoir fluid

[0082]

[0083]

[0084] From the above equations and the component types and parameters in Table 1, it can be seen that the component model for shale condensate gas reservoir development is relatively complex, and the work cost of developing corresponding high-performance nonlinear solvers is relatively high. Current commercial reservoir numerical simulation software (such as CMG, ECLIPSE, tnavigator, etc.) already includes such nonlinear solvers. Therefore, the present invention uses the general pEDFM as a pre-processing operation, that is, using the various grid connections and conductivity obtained by the general pEDFM, and importing these connections into the commercial numerical simulation software by constructing non-adjacent connections, thereby calling the nonlinear solver in the commercial numerical simulation software to solve the component flow model efficiently and robustly, so as to apply the general pEDFM to various reservoir seepage models including shale condensate gas reservoir development at the lowest possible cost.

[0085] like Fig.13 As shown, the numerical simulation method for shale condensate gas reservoir development based on the universal pEDFM of the present invention specifically includes the following steps:

[0086] S1. Use the general pEDFM to calculate the conductivity of connections between various grids.

[0087] Before the introduction of pEDFM, EDFM was a commonly used numerical simulation framework for fractured reservoirs in mining fields. EDFM mainly includes three types of connections between grids, namely, the connection between two adjacent matrix grids, the connection between the matrix grid and the fracture grids it contains, and the connection between adjacent fracture grids; among them, the connection between a specific two grids is represented by (·,·).

[0088] like Figure 1 As shown in (a), the first type of connection includes (m i ,m j ), (m i ,m k ); The second type of connection includes (m i ,f i ), (m j ,f j ), (m k ,f k ); The third type of connection includes (f i ,f k ).

[0089] For the first type of connection, (m i ,m j ) as an example, when using two-point flux estimation (TPFA), calculate (m i ,m j ) is as follows:

[0090]

[0091] Where A(·) represents the area operator, is m i and m j The intersection of and The grid m i and grid m j The permeability, and The grid m i Center and Grid m j Center to distance; it can be seen that (m i ,m j ) has a conductivity equal to m i The conductivity and m j Half the harmonic mean of the conductivity.

[0092] For the second type of connection, (m i ,f i ) as an example, calculate (m i ,f i ) is as follows:

[0093]

[0094] in, and The crack grids f i The permeability, area and opening of is the crack grid f i Center to matrix grid m i The distance between the centers, x and n, are respectively the distance from a point in the grid to the crack grid f i The vector of the center and the normal vector of the crack surface, is the matrix grid m i volume.

[0095] For the third type of connection, it is generally necessary to subdivide it into two cases. The first case is the connection between two adjacent crack grids on the same crack surface, with (f i ,f k ) as an example, the conductivity is as follows:

[0096]

[0097] in, and The crack grids f i The permeability, opening and height of The crack grids f k The permeability, opening and height of and The crack grids f i and the crack grid f k to f i and f k The distance on the intersection surface.

[0098] The second case is the connection between multiple fracture grids that may exist in the same matrix grid. In this case, star transformation is generally used for calculation.

[0099] EDFM has been widely used in the industry due to its simplicity and practicality. However, it has been found to produce significant errors in a large number of seepage scenarios. pEDFM was proposed to address many limitations of EDFM.

[0100] pEDFM projects the cracks in the matrix grid onto the matrix grid surface to establish additional mf connections and ff connections and weaken the original connections. It projects the original cracks embedded in the matrix grid onto the intersection surface of the matrix grid, thereby converting the original EDFM into an approximate DFM for processing. In essence, pEDFM is a model between EDFM and DFM. It has the advantages of EDFM in terms of computational domain mesh discretization and DFM in terms of its applicability to a wider range of seepage scenarios. Figure 1 As shown in (b), f i For example, pEDFM will f i Projected along the x and y directions to At this point, pEDFM will actually Figure 1 The original model in (a) is transformed into Figure 1 Compared with EDFM, f i Need to be with m j and m k There is a connection between them, and the corresponding seepage area is f i The projected area of ​​​​(m j ,f i )、(m k ,f i ) connection occupies and Part of the area, weakening the original (m i ,m j )、(m i ,m k ) seepage area. At the same time, due to f i and f j All Upward projection, f i and f j There is also a connection between.

[0101] Based on the general pEDFM algorithm that can maintain its superiority over EDFM calculation in any seepage scenario, the nudge method is used to obtain the projection configuration of high-conductivity fractures, and the non-projection "conductivity multiplier method" is used to deal with low-conductivity fractures.

[0102] When the matrix permeability is isotropic, such as Figure 1 As shown in (b), the conductivity of the increased mf connection in pEDFM is as follows:

[0103]

[0104] in, and They are f i and f j exist Projection on. and The crack grids f j The permeability and opening of is the matrix grid m i With matrix grid m j The intersection of is m j The penetration rate, is m j arrive distance.

[0105] The conductivity of the original mm connection is weakened to:

[0106]

[0107] in, is m i arrive The distance is the matrix grid m i With matrix grid m k The intersection of is the matrix grid m k arrive distance.

[0108] By halving the conductivity of the matrix grid, the conductivity of the original MF connection is weakened, with (m i ,f i ) as an example, the conductivity is as follows:

[0109]

[0110] The improved pEDFM points out that it is necessary to establish a connection between two crack meshes projected on the same matrix mesh intersection but in different matrix meshes. It can be assumed that the connection is (f i ,fj ), the conductivity of the connection is calculated as:

[0111]

[0112] in, and The crack grids f i and f j The projected area on the intersection surface of the same matrix grid.

[0113] In addition, the general pEDFM uses a non-projection method, which can be called the “conductivity multiplier method”, to deal with low-conductivity fractures. For the connection intersected by the low-conductivity fracture, the updated conductivity of the connection is:

[0114]

[0115] Among them, T ij is the original conductivity based on EDFM, T′ ij is the updated conductivity of the general pEDFM, T f is the conductivity of the low conductivity fracture, k f is the permeability of low conductivity fractures, w f is the opening of the low conductivity fracture, A ij is the seepage area of ​​the corresponding connection.

[0116] The universal pEDFM can effectively solve the situation where EDFM and original pEDFM will produce significant calculation errors, and achieve higher simulation accuracy, significantly reduce the complexity of additional ff connections between fracture grids projected onto the intersection of the same matrix grid, avoid the difficulty of developing a universal algorithm for strictly constructing ff connections, and lay the algorithm foundation for the commercialization of pEDFM. Therefore, the universal pEDFM is the current numerical simulation framework for fractured reservoirs with the best comprehensive performance.

[0117] S2. Construct an equivalent model based on the universal pEDFM to call the commercial digital simulation software CMG

[0118] Based on the inter-grid connections and corresponding conductivity values ​​obtained by the general pEDFM, a CMG model equivalent to the real reservoir pEDFM model needs to meet the following requirements:

[0119] (1) The volume of each effective grid in the CMG model is the same as the pore volume of the corresponding grid in the real reservoir model;

[0120] (2) The connection table between effective grids in the CMG model (i.e., the connection relationship between grids and the corresponding conductivity) is the same as the connection table between corresponding grids in the real reservoir model;

[0121] (3) The well index of the grid where the injection and production wells are located in the CMG model is the same as the well index calculated for the corresponding grid in the real model.

[0122] In order to facilitate the construction of an equivalent CMG model, a corresponding one-dimensional CMG model can be constructed for the two-dimensional pEDFM model. CMG generally does not support the assignment of grid sizes in the x, y, and z directions to each grid on the one-dimensional model. For a one-dimensional model extending along the x direction, the grid sizes of the model in the y and z directions need to be the same value. Therefore, in order to make the constructed CMG one-dimensional model meet the above three requirements to make it equivalent to the real model, the specific approach is as follows:

[0123] S21, let the grid sizes of the i-th matrix grid in the x, y, and z directions be Δx i , Δy i and Δz i Since CMG requires the mesh size of the one-dimensional model in the y direction and the z direction to be the same value, the mesh size of the one-dimensional model in the y direction and the z direction is set to the true Δy of all meshes. i and Δz i (i=1,2,…,n m ) can be expressed as In order to keep the grid volume unchanged, the Δx of the i-th matrix grid in the CMG one-dimensional model is i Calculated as:

[0124]

[0125] Among them, V m,i is the volume of the ith matrix grid.

[0126] S22, such as Figure 2 As shown, the crack grid (green) is placed behind the matrix grid (red) in this one-dimensional model, and its size in the y and z directions still needs to be set to and The corresponding Δx i Calculated as:

[0127]

[0128] Among them, V f,i is the volume of the i-th crack grid.

[0129] S23. The size of the invalid grid in the x direction can be assigned an arbitrary value.

[0130] The connections and conductivity between each grid obtained by the general pEDFM are included in the data file of the model as non-neighboring connection (NNC) files. It should be pointed out that the NNC in CMG corresponds to the "SCONNECT" keyword (i.e. special connection), and its conductivity value is the conductivity value calculated with permeability units of mD and length units of m.

[0131] For permeability and porosity, the physical property parameter values ​​of each real grid can be directly assigned to the corresponding grid in the CMG one-dimensional model, and the parameter values ​​of the invalid grid can be assigned any value.

[0132] The above method of assigning the size of each grid in the x, y and z directions in the CMG one-dimensional model may cause the well index calculated by the grid where the well is located to be different from the well index of the original real model. However, when the grid size of each grid in the y and z directions is , this difference is quite limited in most cases.

[0133] S3. Call the nonlinear solver in CMG to solve the shale condensate gas reservoir component model. The specific process is as follows:

[0134] S31. By judging whether the permeability is higher than the average permeability of the matrix system, the fracture is classified into high conductivity type F p and low conductivity type F np Among them, the subscripts p and np indicate that the corresponding types of cracks will be processed by the projection method and the non-projection method respectively.

[0135] S32, ignore F np , use the nudge method to get F np The projection configuration of each fracture grid is based on the original pEDFM to obtain various connections between grids and calculate the corresponding conductivity, which is recorded as cont1. According to the improved pEDFM, when there are multiple high-conductivity fracture grids projected onto the same matrix grid surface, the connection between any two fracture grids in different matrix grids is constructed, which is recorded as cont2, so that the grid connection cont = cont1∪cont2 of the reservoir model is obtained.

[0136] S33, use the non-projected conductivity multiplier method to update the F in cont np The conductivity of each connection that intersects;

[0137] S34. Based on cont and the corresponding conductivity, the corresponding CMG equivalent model is constructed, so as to call the CMG component model nonlinear solver to calculate the global nonlinear equation group for the development of shale condensate gas reservoirs, and obtain the pressure, saturation, concentration distribution of each component and well dynamic data.

[0138] Among them, steps S31 to S34 are the processing flow of the general pEDFM on the real reservoir model; step S35 is the process of calling CMG for simulation calculation based on the general pEDFM pre-processing information.

[0139] Example

[0140] The present invention designs three examples to demonstrate that the universal pEDFM can significantly improve the simulation accuracy of shale condensate gas reservoir development compared with EDFM.

[0141] The second example is that these two examples are used to demonstrate that pEDFM can significantly improve the simulation accuracy compared with EDFM under the conditions of high conductivity, low conductivity fractures, depletion development, and multiple rounds of N2 injection, revealing the significance of this study. Finally, the third example shows the calculation effect of this method applied to complex fracture networks.

[0142] Example 1: Depletion development of a shale condensate gas reservoir with high conductivity natural fractures.

[0143] like Figure 3 As shown in Figure 2, the reservoir model size is 1000m×500m×10m, the matrix grid size is 10m×10m×10m, and there is a fractured horizontal well with 7 fractures (marked in blue) and 14 high conductivity natural fractures (marked in red) in the reservoir. The basic parameters of the reservoir and fractures used in the simulation are listed in Table 2. Figure 4 The figure shows the phase permeability curve used in this example. Table 3 lists the relevant parameters of each component of the reservoir fluid. Using the PR state equation and flash calculation without considering the capillary pressure, it can be obtained that when the reservoir temperature is 175℃, the GOR is 927m3 / m3 and the dew point pressure is 55.13×103kPa.

[0144] Table 2 Basic parameters of reservoir and fracture in Example 1

[0145] parameter Value parameter Value Model size(m) 1000×600×10 Perforation cluster number 7 Original reservoir pressure (kPa) 75000 Cluster spacing (m) 100 Original reservoir temperature (℃) 175 Crack half length (m) 150 Matrix permeability (mD) 0.005 Crack height (m) 10 Matrix porosity 5％ Fracture permeability (mD) 100000 Initial water saturation 40％ Natural fracture permeability (mD) 10000 Matrix compressibility (1 / kPa) <![CDATA[1.8×10 -5 ]]> Fracture Porosity 40％ Reservoir thickness (m) 10 Crack opening (m) 0.005 Reservoir depth (m) 6000 Crack compression coefficient (1 / kPa) <![CDATA[1.8×10 -5 ]]>

[0146] Table 3 Relevant parameters of each component of reservoir fluid

[0147]

[0148] Since the distribution of fractures in this example is parallel to the coordinate line, the simulation results of the local grid refinement model (LGR) can be used as the reference solution. Set the bottom hole flow pressure of the horizontal well to 44000kPa for 2500 days of depletion development. According to the dew point pressure calculated above, this bottom hole flow pressure will cause condensate to precipitate. Figure 5 The figure shows the comparison of the oil, gas and water production rate curves calculated by the LGR reference model, pEDFM and EDFM.

[0149] From the comparison, we can see that in the process of depletion development of shale condensate gas reservoirs, the simulation results of pEDFM match the reference solution much better than those of EDFM, which significantly improves the calculation accuracy. In addition, based on the observation of the actual calculation process, it is found that the calculation speed of pEDFM is also significantly faster than that of EDFM during the period when the condensate begins to precipitate.

[0150] Example 2: Numerical example of multiple cycles of N2 injection into a shale condensate gas reservoir with low conductivity fractures.

[0151] Compared with Example 1, the high-conductivity natural fractures are changed into low-conductivity fractures with a permeability of 0 mD (i.e., flow barriers), and the production system is set as follows: N2 is injected at a rate of 20,000 cubic meters per day for one month, followed by 10 days of well soaking, and then the bottom hole pressure is set at 44,000 kPa for 325 days of production, a total of 365 days as a throughput cycle, a total of 6 cycles of throughput (a total of 6 years), and finally 4 years of exhaustion development. Figure 6 The following table compares the daily oil production, daily gas production and daily water production curves calculated by the LGR reference model, pEDFM and EDFM respectively. Figure 7 , Figure 8 and Fig. 9 The figure shows a comparison of the pressure, oil saturation and gas saturation distributions calculated by the three methods after 10 years.

[0152] It can be seen that: whether from the production rate or from the pressure and saturation distribution, the matching degree between pEDFM and the reference solution is far greater than that of EDFM, and the calculation accuracy of pEDFM is much higher than that of EDFM, while the calculation error of EDFM is very significant. This calculation result once again reveals the importance of studying the numerical model of gas injection and energy replenishment in shale condensate gas reservoirs based on pEDFM for improving the calculation accuracy.

[0153] Example 3 demonstrates the calculation effect of the method of the present invention when applied to a complex fracture network.

[0154] like Fig.10 As shown in Figure 2, the reservoir model includes two fractured horizontal wells to demonstrate the application effect of the universal pEDFM in complex fracture networks and multi-well production. The working system of the two fractured wells is the same as that of Example 2, first implementing 6 rounds of N2 injection and then continuing 4 years of depletion development. The permeabilities of the fractures and natural fractures are 40000mD and 10000mD respectively, and other reservoir, fracture and fluid parameters are the same as those of Example 1. Fig.11 The figure shows the pressure and oil, gas and water saturation distribution at the end of the first three throughput cycles (i.e., 3 years later) and 10 years later. Fig.12 The figure shows the oil, gas and water production rate curves obtained for the two production wells.

[0155] It can be seen that pEDFM can effectively handle complex fracture networks and multi-well production scenarios, revealing the feasibility, necessity and great potential of pEDFM, which has significant computational superiority over the most commonly used EDFM, in actual mine applications.

[0156] Therefore, the present invention adopts the above-mentioned numerical simulation method for shale condensate gas reservoir development based on pEDFM and CMG, which not only ensures the advantages of the general pEDFM in calculation accuracy and processing of complex fracture networks, but also avoids the difficulty of developing additional high-performance nonlinear solvers, and provides a numerical simulation tool with the best comprehensive calculation performance in theory for the development of fractured shale condensate gas reservoirs.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A numerical simulation method for shale condensate gas reservoir development based on a general pEDFM, characterized in that: The following steps are involved: S1. Use the general pEDFM to calculate the conductivity of connections between various grids; S2. Based on the inter-grid connections and corresponding conduction coefficient values ​​obtained by the universal pEDFM, the commercial digital model software CMG equivalent model based on the universal pEDFM needs to meet the following requirements: (1) The volume of each effective grid in the CMG model is the same as the pore volume of the corresponding grid in the real reservoir model; (2) The connection table between the effective grids in the CMG model is the same as the connection table between the corresponding grids obtained based on the general pEDFM of the real reservoir model; (3) The well index of the grid where the injection and production wells are located in the CMG model is the same as the well index calculated for the corresponding grid in the real model; Construct an equivalent CMG model. For the two-dimensional pEDFM model, first construct a corresponding one-dimensional CMG model so that the constructed one-dimensional CMG model meets the three requirements in (1)-(3) above and is equivalent to the real model. The specific steps are as follows: S21, let the grid sizes of the i-th matrix grid in the x, y, and z directions be Δx i , Δy i and Δz i Since CMG requires the mesh size of the one-dimensional model in the y direction and the z direction to be the same value, the mesh size of the one-dimensional model in the y direction and the z direction is set to the true Δy of all meshes. i and Δz i The mean or median of m ; may be recorded as In order to keep the grid volume unchanged, the Δx of the i-th matrix grid in the CMG one-dimensional model is i Calculated as: Among them, V m,i is the volume of the ith matrix grid; S22. The crack grid is placed behind the matrix grid in the one-dimensional model, and its size in the y and z directions still needs to be and The corresponding Δx i Calculated as: Among them, V f,i is the volume of the i-th crack grid; S23, the size of the invalid grid in the x direction takes an arbitrary value; S3. Call the nonlinear solver in CMG to solve the shale condensate gas reservoir component model.

2. The method for numerical simulation of shale condensate gas reservoir development based on universal pEDFM according to claim 1, characterized in that: Based on EDFM, pEDFM establishes additional connections and weakens the original connections by projecting the cracks in the matrix grid onto the matrix grid surface. In any seepage scenario, the general pEDFM algorithm uses the nudge method to obtain the projected configuration of high-conductivity cracks, and uses the non-projected conductivity multiplier method to deal with low-conductivity cracks.

3. The numerical simulation method for shale condensate gas reservoir development based on universal pEDFM according to claim 2, characterized in that: When the matrix permeability is isotropic, the conductivity of the increased mf junction in pEDFM is as follows: Where A(·) represents the area operator, and They are f i and f j In l ij projection on; is the crack grid f i The permeability, is the crack grid f j The permeability, is the crack grid f i The opening degree, is the crack grid f j The opening, l ij is the matrix grid m i With matrix grid m j The intersection of is the grid m i The permeability, is the grid m j The permeability, is m j To ij distance; The conductivity of the original mm connection is weakened to: Where A(·) represents the area operator, l ij is m i and m j The intersection of is m i To ij The distance, l ik is the matrix grid m i With matrix grid m k The intersection of is the matrix grid m k To ik distance; By halving the conductivity of the matrix grid, the conductivity of the original mf connection is weakened to (m i ,f i ) as an example, the conductivity is as follows: Among them, A fi is the crack grid f i area; is the crack grid f i Center to matrix grid m i the distance between centers; is the matrix grid m i Volume; Based on the improved pEDFM, a connection is established between two crack meshes projected on the same matrix mesh intersection but in different matrix meshes. The connection is assumed to be (f i ,f j ), the conductivity of the connection is calculated as: in, and The crack grids f i and f j The projected area on the intersection of the same matrix grid; The general pEDFM uses a non-projection method, namely the conductivity multiplier method, to deal with low-conductivity fractures. For the connection intersected by low-conductivity fractures, the updated conductivity of the connection is: Among them, T ij is the original conductivity based on EDFM, T ij ′ is the updated conductivity of the general pEDFM, T f is the conductivity of the low conductivity fracture, k f is the permeability of low conductivity fractures, w f is the opening of the low conductivity fracture, A ij is the seepage area of ​​the corresponding connection.

4. The numerical simulation method for shale condensate gas reservoir development based on universal pEDFM according to claim 1, characterized in that: In step S3, the calculation process of the general pEDFM and CMG is integrated as follows: S31. By judging whether the permeability is higher than the average permeability of the matrix system, the fracture is classified into high conductivity type F p and low conductance type F np ; Wherein, the subscripts p and np respectively indicate that the corresponding types of cracks will be processed by the projection method and the non-projection method respectively; S32, ignore F np , use the nudge method to get F np The projection configuration of each fracture grid is based on the original pEDFM to obtain various connections between grids and calculate the corresponding conductivity, which is recorded as cont1. According to the improved pEDFM, when there are multiple high-conductivity fracture grids projected onto the same matrix grid surface, the connection between any two fracture grids in different matrix grids is constructed, which is recorded as cont2, so that the grid connection cont = cont1∪cont2 of the reservoir model is obtained. S33, use the non-projected conductivity multiplier method to update the F in cont np The conductivity of each connection that intersects; S34. Based on cont and the corresponding conductivity, the corresponding CMG equivalent model is constructed, so as to call the CMG component model nonlinear solver to calculate the global nonlinear equation group for the development of shale condensate gas reservoirs, and obtain the pressure, saturation, concentration distribution of each component and well dynamic data.

Images