A modeling method based on improved discrete fracture network

Abstract

The present invention provides a modeling method based on an improved discrete fracture network, comprising: S1, gridding the study area and obtaining data representing the intensity / density, morphology, and occurrence of fractures of various levels based on the geological data of the study area; S2, calculating the number of fracture slices within a single grid based on the intensity or density of the fractures within each grid, and generating the coordinates of the center points of each fracture slice; S3, grouping adjacent fracture slices within all grids into the same group based on the fracture shape, adjusting the scale and posture of each fracture slice based on the fracture shape, occurrence, and center point coordinates of each group of fracture slices to simulate complex fractures and generate a fracture network; S4, evaluating the rationality of the fracture network based on the geological data of the study area; if not, re-performing step S2. The present invention considers the spatial correlation of fracture slices, improves the accuracy of fracture simulation, and provides a reliable data volume for numerical simulation of fractured gas reservoirs.

Description

Technical Field

[0001] The present invention relates to the field of oil and gas exploration and development, and more particularly to a modeling method based on an improved discrete fracture network. Background Art

[0002] The most widely used model for modeling fractured reservoirs is the continuum model. The concept was first proposed by Barentblatt et al. in 1960. In 1963, Warrant and Root improved upon the continuum model and proposed the Warrant-Root model, introducing the concept of dual-porosity media. This model subsequently gained widespread application and became one of the most comprehensive fracture models available. In 1970, Louis proposed a completely new model, the discrete fracture network model, which marked a new era in fracture research. The discrete fracture network model differs significantly from the traditional Warrant-Root model. Each fracture slice defined in the discrete fracture network model has coordinates, occurrence, size, aperture, physical properties, and parameters corresponding to its host rock. The resulting fracture network model not only meets specific geostatistical requirements but also ensures that the fracture network is treated as a discrete object. It accurately reflects regional variations in fracture development and seepage characteristics, making it possible to develop precise fracture geometry and parameter models for reservoirs.

[0003] Discrete fracture network models use numerous discrete planar units (i.e., fracture slices) with defined orientations, lengths, and areas to characterize the distribution of fractures. There are two main approaches to modeling discrete fracture networks (DFNs): 1. Stochastic fracture modeling, which uses fractures as the "target prediction phase" and predicts their distribution using a variety of parameters constraining their development. 2. Fault-based fracture modeling, which uses self-similarity (local and global similarity) to predict the distribution of fractures near faults based on core fractures or faults.

[0004] Although three-dimensional fracture modeling technology has made certain progress in recent years, many problems still exist. Due to the unique complexity of the fractures themselves, the uncertainty of the factors controlling their development, the random nature of their formation, and the highly heterogeneous distribution, it is difficult to obtain accurate prediction results, making its research a difficult problem that needs to be solved urgently. Summary of the Invention

[0005] In view of the technical problems existing in the prior art, the present invention provides a modeling method based on an improved discrete fracture network, which takes into account the spatial correlation of fracture slices and improves the accuracy of fracture simulation.

[0006] According to a first aspect of the present invention, a modeling method based on an improved discrete fracture network is provided, comprising:

[0007] S1. Grid the study area and obtain data representing the intensity / density, morphology, and occurrence of fractures of various levels based on the geological data of the study area. Fracture morphologies include: isolated, curved, conjugate, or binary tree-like; fracture occurrence includes: extension, inclination, and orientation.

[0008] S2, calculating the number of crack slices in a single grid according to the intensity value or density value of the cracks in each grid, and generating the coordinates of the center point of each crack slice;

[0009] S3: All adjacent fracture slices in the grid are grouped together according to the fracture shape. The scale and posture of each fracture slice are adjusted based on the fracture shape, occurrence, and center point coordinates of each group of fracture slices to simulate complex fractures and generate a fracture network.

[0010] S4, evaluating whether the fracture network is reasonable based on the geological data of the study area. If it is not reasonable, re-execute step S2.

[0011] On the basis of the above technical solution, the present invention can also make the following improvements.

[0012] Optionally, step S1 is to grid the study area and obtain data bodies representing the intensity / density, morphology, and occurrence of fractures of various levels based on the geological data of the study area. The morphology of fractures includes: isolated, curved, conjugate, or binary tree. The occurrence of fractures includes: extension, inclination, and orientation. Specifically, the following are included:

[0013] Construct a model body corresponding to the scope of the study area, divide the study area into grids according to a preset grid size, and record the model body data through the grid index.

[0014] The sedimentary facies map, well-connected profile map and manually interpreted well sedimentary facies data in the geological data are digitized into the gridded model volume to establish a sedimentary facies model corresponding to the study area;

[0015] Core sampling, logging, seismic and production dynamic data from the study area are used to identify and describe core fractures and logging fractures, so as to obtain data bodies representing the intensity / density, morphology, occurrence and distribution of fractures of various orders in the study area. The morphology includes: isolated, curved, conjugate or binary tree, and the occurrence includes: extension, inclination and orientation.

[0016] Optionally, step S1 further includes constructing a fracture-related attribute model based on geological data of the study area, wherein the attribute model includes a fracture range model, a well data file, a fracture intensity model, a fracture orientation model, and a fracture inclination model; wherein,

[0017] Constructing a fracture range model, including: constructing a sedimentary facies model using a geological modeling algorithm based on lithofacies, bedding planes, and fault data. The sedimentary facies model is used to determine whether a grid belongs to the area to be simulated, thereby constraining the fracture range;

[0018] Establishing a well data file, including: identifying fractures on the imaging log based on the interpretation information of the imaging log in the study area, the identification content includes at least the fracture depth, dip, azimuth, strike, aperture, extension length, filling degree, and fracture number data, so as to deterministically establish a fracture model on the well;

[0019] Establishing a fracture strength model includes: constructing fracture data of different scales through core, well logging, seismic, dynamic data and literature research data, extracting strength curve data from the fracture data, coarsening the strength curve data into the established sedimentary facies model, using the coarsened strength curve data as conditional point data, using fault distance and structural curvature as constraints, and using Gaussian random simulation to establish a fracture strength model for the study area to characterize fracture density attributes;

[0020] Establish a fracture orientation model and a fracture inclination model, including: for the well location area of ​​the study area, quantitatively describe and calculate the fracture parameters of the single well target layer imaging logging to obtain the fracture dip, azimuth, opening, and extension length; for the non-well location area of ​​the study area, extract the dip and azimuth attributes through the ant volume data in the seismic data to form a fracture inclination model and a fracture orientation model to constrain the generated fracture dip and azimuth.

[0021] Optionally, step S2, calculating the number of crack slices in a single grid according to the intensity value or density value of the cracks in each grid, and generating the coordinates of the center points of each crack slice, includes:

[0022] Select a grid according to the grid index, obtain the coordinate range of the current grid, and determine whether the current grid has crack strength constraints:

[0023] If there is a crack strength constraint, the strength value of the crack in the current grid is obtained, and the number of crack slices in the current grid is calculated based on the crack strength value;

[0024] If there is no crack strength constraint, then set the crack density value and the density multiplier value in the three-dimensional direction (X, Y, Z) of the model body to determine the number of crack slices in the current grid;

[0025] Using the coordinate range of the current grid as a constraint, randomly generate the center point coordinates of each crack slice;

[0026] Traverse all the grids in the area to be simulated and obtain the coordinates of the center points of the crack slices in each grid.

[0027] Optionally, if there is no crack intensity constraint, and the crack density of some areas to be simulated is different from the overall crack density, then a separate crack density value is set for the area where the crack density is different from the overall crack density; or,

[0028] The phase control method is used to set different phase types for different density areas, and a mapping relationship between phase types and fracture density values ​​is established, thereby setting different fracture density values ​​for different areas.

[0029] Optionally, the number of crack slices in the current grid is calculated based on the intensity value of the cracks, and is calculated using the following formula (1):

[0030]

[0031] In formula (1), Sum_lf is the total number of cracks, n is the total number of grids in the study area, V i is the volume of the i-th grid, and Qi is the intensity attribute value of the i-th grid.

[0032] Optionally, if there is no crack strength constraint, the crack density value and the density multiplier value of the model body in the three-dimensional direction (X, Y, Z) are set to determine the number of crack slices in the current grid, including:

[0033] For a single grid, obtain its grid size based on the three-dimensional directions X, Y, and Z of the model body, which are L_x, L_y, and L_z respectively. Assuming the overall initial density is Den_lf, the initial number of crack points Num_x, Num_y, and Num_z in the three directions of X, Y, and Z can be calculated by the following formula (2):

[0034]

[0035] Set the density ratios T_x, T_y, and T_z of the three-dimensional directions X, Y, and Z of the model body respectively. According to the density ratios T_x, T_y, and T_z in the X, Y, and Z directions, calculate the final number of crack points Fin_x, Fin_y, and Fin_z in the X, Y, and Z directions by the following formula (3):

[0036] Fin_x=Num_x*T_x

[0037] Fin_y=Num_y*T_y

[0038] Fin_z=Num_z*T_z (3);

[0039] The final total number of grid crack slices Sum_lf is calculated by the following formula (4):

[0040] Sum_lf=Fin_x*Fin_y*Fin_z (4).

[0041] Optionally, in step S1, based on the geological data of the study area, data bodies representing the occurrence and distribution of fractures of various levels are obtained, including:

[0042] In the observation of fracture parameters of core and imaging logging in the study area, the occurrence data volume of each fracture is obtained. The occurrence data volume includes:

[0043] Horizontal extension range, which represents the maximum and minimum horizontal extension ranges of the cracks in the study area and the corresponding horizontal horizontal extension range;

[0044] Vertical extension range, which represents the maximum and minimum vertical extension ranges of the cracks in the study area and the corresponding average vertical extension range;

[0045] Azimuth angle range, which represents the strike range of the fractures rotating around the vertical direction in the study area and the corresponding average azimuth angle range;

[0046] The tilt angle range characterizes the strike range of the fractures rotating around the horizontal plane and the corresponding average tilt angle range in the study area;

[0047] The distribution mode of the horizontal extension range, vertical extension range, azimuth angle range and inclination angle range of the cracks in the study area is obtained, and the distribution mode is random distribution, uniform distribution or normal distribution.

[0048] Optionally, step S3 groups all adjacent fracture slices in the grid into the same group based on the fracture shape, and adjusts the scale and posture of each fracture slice group based on the fracture shape, occurrence, and center point coordinates of each fracture slice group to simulate complex fractures and generate a fracture network, including:

[0049] According to the shape of the crack to be simulated and the coordinates of the center points of each crack slice, all adjacent crack slices in the grid are divided into the same group to obtain several groups of crack slices;

[0050] Randomly extract any group of fracture slices according to the distribution of the fractures to be simulated, and obtain the horizontal extension value, vertical extension value, azimuth angle value and inclination angle value of the current group of fracture slices extracted according to the fracture occurrence data or set data;

[0051] Based on the center point coordinates of the crack slice, the coordinates of the four vertices of each crack slice in the current group of crack slices extracted are calculated according to the horizontal extension value and the vertical extension value of the crack slice to establish an initial crack slice; the azimuth angle and inclination angle of the initial crack slice are adjusted according to the obtained azimuth angle value and inclination angle value to obtain a final crack slice;

[0052] Generate corresponding types of cracks according to the shape of the crack to be simulated;

[0053] Traverse all groups of crack slices in all grids to generate a crack network.

[0054] Optionally, generating a corresponding type of crack according to the shape of the crack to be simulated includes:

[0055] If the shape of the crack to be simulated is an isolated crack, the final crack slice is used as the crack simulation result;

[0056] If the shape of the crack to be simulated is a curved crack, the adjacent crack slices are spliced ​​end to end in sequence according to the positional relationship of the crack slices to simulate the curved crack;

[0057] If the shape of the crack to be simulated is a conjugate crack, then based on the positional relationship of the crack slices, two adjacent crack slices are divided into a group and share a common center coordinate point. The two crack slices use different azimuth angles and / or tilt angles to simulate the conjugate crack.

[0058] If the shape of the crack to be simulated is a binary tree crack, the parameters of the binary tree are defined: number of branches, extension length, branch length, and branch angle, and the crack slices in the same group are spliced ​​according to the defined binary tree parameters to simulate the binary tree crack.

[0059] The present invention provides an improved discrete fracture network modeling method, which belongs to the field of oil and gas exploration and development technology. The model body is refined into a grid body, and a corresponding number of independent fracture slices are set according to the specific conditions of each grid. By grouping the independent fracture slices and adding shape category attributes to each individual fracture slice, various types of fractures can be established, such as: an isolated fracture with only one fracture slice, a fracture with a curved shape connected by multiple fracture slices, a conjugate fracture where two fracture slices intersect, and a lightning-type fracture where multiple fracture slices are distributed in a binary tree shape, thereby establishing a fracture model that is more consistent with the actual geological conditions. Because the discrete fracture network modeling method is used, not only does the established fracture model have different shapes, coordinates, sizes, orientations, and openings, but it also takes into account the reliability of sampling parameters such as the shape, size, inclination, and dip angle of the fracture, as well as the spatial correlation of the fracture slices, to establish a fracture model that is closer to the actual fracture style, improve the modeling accuracy of fracture gas reservoirs, more accurately predict high-quality seepage channels, and provide a reliable data body for numerical simulation of fracture gas reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of a process flow based on an improved discrete fracture network modeling method provided by the present invention;

[0061] Figure 2 Schematic diagram of a deterministic wellbore fracture model established in a certain embodiment;

[0062] Figure 3 A schematic diagram of a study area model established in a certain embodiment;

[0063] Figure 4 A schematic diagram of a crack strength model for a research area established in a certain embodiment;

[0064] Figure 5 A schematic diagram of a fracture azimuth attribute model for a study area established in a certain embodiment;

[0065] Figure 6 A schematic diagram of a fracture inclination angle attribute model for a study area established in a certain embodiment;

[0066] Figure 7 To show the intention of horizontal extension of cracks;

[0067] Figure 8 To show the intention of vertical extension of cracks;

[0068] Figure 9 Schematic diagram of the crack direction;

[0069] Figure 10 Schematic diagram of crack inclination;

[0070] Figure 11 Schematic diagram of randomly generated crack slice center points within the grid;

[0071] Figure 12 Schematic diagram of curved crack;

[0072] Figure 13 Schematic diagram of conjugate morphological cracks;

[0073] Figure 14 This is a schematic diagram of a lightning-shaped crack. DETAILED DESCRIPTION

[0074] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0075] Figure 1 The main flow chart of the modeling method based on the improved discrete fracture network provided by the present invention is as follows: Figure 1 As shown, the method includes S1 to S4:

[0076] S1. Grid the study area and obtain data representing the intensity / density, morphology, and occurrence of fractures of various levels based on the geological data of the study area. Fracture morphologies include: isolated, curved, conjugate, or binary tree-like; fracture occurrence includes: extension, inclination, and orientation.

[0077] S2, calculating the number of crack slices in a single grid according to the intensity value or density value of the cracks in each grid, and generating the coordinates of the center point of each crack slice;

[0078] S3: All adjacent fracture slices in the grid are grouped together according to the fracture shape. The scale and posture of each fracture slice are adjusted based on the fracture shape, occurrence, and center point coordinates of each group of fracture slices to simulate complex fractures and generate a fracture network.

[0079] S4, evaluating whether the fracture network is reasonable based on the geological data of the study area. If it is not reasonable, re-execute step S2.

[0080] It is understandable that although three-dimensional fracture modeling technology has made certain progress in recent years, it still has many problems. Due to the unique complexity of the fracture itself, the uncertainty of the factors that control its development, and the random nature of its formation, it is impossible to obtain ideal prediction results by only using a single subject knowledge or a single prediction method, making its research a difficult problem that needs to be solved urgently. The difficulties in fracture research are mainly reflected in: 1. The complexity of the causes of fractures; 2. The multi-factor nature of control and influence; 3. The randomness of formation and development; 4. The high degree of heterogeneity of distribution. These fracture modeling problems restrict the accuracy of fracture simulation and require further in-depth research. Based on the defects in the background technology, taking into account the spatial correlation of fracture slices, the embodiment of the present invention proposes a modeling method based on an improved discrete fracture network.

[0081] This embodiment provides an improved discrete fracture network modeling method, which belongs to the field of oil and gas exploration and development technology. The model body is refined into a three-dimensional grid body, and a corresponding number of independent fracture slices are set according to the specific conditions of each grid. By grouping the independent fracture slices and adding shape category attributes to each individual fracture slice, various types of fractures can be established, such as: isolated fractures with only one fracture slice, fractures with a curved shape connected by multiple fracture slices, conjugate fractures with two fracture slices intersecting, and lightning-shaped fractures with multiple fracture slices distributed in a binary tree-like pattern, thereby establishing a fracture model that is more consistent with the actual geological conditions. Because it uses a discrete fracture network modeling method, not only does the established fracture model have different shapes, coordinates, sizes, orientations, and apertures, but it also takes into account the reliability of sampling parameters such as fracture shape, size, dip, and inclination, as well as the spatial correlation of fracture slices, to establish a fracture model that is closer to the actual fracture pattern, improve the modeling accuracy of fractured gas reservoirs, more accurately predict high-quality seepage channels, and provide a reliable data body for numerical simulation of fractured gas reservoirs.

[0082] The present invention is now further described based on the improved discrete fracture network modeling method for an ultra-deep fractured gas reservoir in a certain work area in a specific implementation scenario.

[0083] The data of this embodiment are rich coring, logging, seismic, production dynamic data and other data of the study area (corresponding to the work area) and the gridded work area.

[0084] Work Area Scale: The specifications are columns (number of columns) * rows (number of rows) * layers (vertical grid) (columns = 314, rows = 41, layers = 151), with an original grid number of 1,943,974 points. In the three-dimensional coordinates of the address model, the X coordinate spacing is cellsize1X (cellsize1X = 100m), the Y coordinate spacing is cellsize1Y (cellsize1Y = 99m), and the Z coordinate spacing is cellsize1Z (cellsize1Z = 2m).

[0085] The grid is designed as a standard rectangular grid. The index of each grid is valid and unique. The model data is recorded through the grid index, that is, the model data is associated with the index / coordinate value of the grid.

[0086] 1. Establishment of geological database of the work area

[0087] According to the well data and its scope of the study area and the requirements of geological modeling, the study area is divided into grids of appropriate size (314*41*151), and the index of each grid is valid and unique.

[0088] The sedimentary facies maps, well-connected profiles and manually interpreted well sedimentary facies data in the geological data are digitized into the grid model body to obtain the position of each well in the grid and to establish a sedimentary facies model corresponding to the study area.

[0089] Make full use of the abundant coring, logging, seismic and production dynamic data in the study area to identify and describe core fractures and logging fractures, and divide the fracture levels according to the size and identifiability of the fractures to obtain the intensity / density, morphology, and occurrence data of the fractures of different levels in the study area, as well as the distribution of the fractures. For example, the morphology of the fractures includes: isolated fractures, curved fractures, conjugate fractures or binary tree fractures (i.e., lightning-type fractures), the occurrence of the fractures includes: extension range, inclination angle and azimuth angle, and the overall distribution of the fracture properties includes random distribution, uniform distribution and / or normal distribution. It is understandable that fractures may have various forms, such as isolated fractures with only one fracture sheet, fractures with a curved shape connected by multiple fracture sheets, conjugate fractures with two fracture sheets intersecting, lightning-type fractures with multiple fracture sheets distributed in a binary tree shape, etc.

[0090] Fractures of different scales are characterized and the model volume data is recorded through grid indexing to complete the establishment of the geological database.

[0091] Using data from coring, well logging, seismic data, and production dynamics in the study area, we identify and describe fractures in core and well logging data. We then categorize fractures according to their size and identifiability, obtaining data on the intensity / density, morphology, and occurrence of fractures of different levels, as well as their distribution. We characterize fractures of varying scales and record model data using grid indexing to complete the construction of a geological database.

[0092] 2. Establishment of relevant attribute body model

[0093] The attribute models that need to be prepared include: ① fracture range model, ② well data file, ③ fracture intensity model, ④ fracture orientation model and ⑤ fracture inclination model.

[0094] ① Crack range model, such as Figure 3 As shown in the figure, it is a file that represents the scope of the study area, which represents the grid index value aggregation of all non-invalid value areas in the work area, and is used to control the development area during fracture simulation.

[0095] When establishing a fracture range model, the phase matrix model is selected as the fracture range model, that is, a sedimentary facies model constructed using a geological modeling algorithm based on lithofacies, bedding planes, and fault data. Selecting the phase matrix model of the study area as the range constraint model allows for determining whether a grid is a valid value to determine whether it is the area to be simulated. On the other hand, it also allows for phase-controlled fracture modeling based on the specific values ​​of the sedimentary facies model. Multi-scale fracture modeling can be implemented, for example, by setting different phase type values ​​for different regions and determining the fracture scale range corresponding to the specific phase type value, thereby achieving the goal of modeling fractures of different scales in different regions. Multi-density fracture modeling can also be implemented, by setting different phase type values ​​for different density regions and establishing a corresponding relationship between different phase type values ​​and fracture density development ranges, thereby achieving the goal of establishing fracture models of different densities in different regions.

[0096] ② Well data file, which is the well fracture data interpreted in the Petrel software work area, mainly contains the x, y, z coordinate values ​​of the fracture slice, well name, dip, azimuth, opening and the extension length data of a single fracture slice. The fracture slice established based on this file is a deterministic fracture model. The well data file mainly comes from the interpretation of the imaging logging in the study area. The fractures on the imaging logging are identified. The content that needs to be identified mainly includes the fracture depth, dip, azimuth, strike, opening, extension length, filling degree, and fracture number data body. Figure 2 The figure shows a local enlarged schematic diagram of the deterministic wellbore fracture model. By identifying these wellbore fracture data, the following can be deterministically established: Figure 2 The fracture model above the well is shown.

[0097] ③ Crack strength model, such as Figure 4 The figure shown here shows a fracture strength attribute model constructed using Gaussian random simulation in Petrel software based on coarsening of intensity curve data as conditional points, with fault distance and structural curvature as constraints. This model primarily characterizes fracture density properties. The fracture strength model reveals the mechanical properties and stability of fractures by modeling and analyzing fracture parameters. This model considers factors such as fracture density, range, and location, and employs mathematical and physical methods for description and calculation. To establish the fracture strength model, fracture data at various scales were constructed using dynamic data, core data, well logging, seismic data, and literature research. Intensity curve data was then extracted from this fracture data and coarsened into the established sedimentary facies model. The coarsened intensity curve data served as conditional points, with fault distance and structural curvature as constraints. A Gaussian random simulation method was used to establish the fracture strength model for the study area.

[0098] ④ Crack orientation model, such as Figure 5The figure below shows a fracture orientation attribute model created in Petrel software using ant volume data extracted from seismic data. This model represents the fracture azimuth attribute. The fracture orientation model attribute values ​​represent the strike (azimuth) of the fracture slice at each grid point. For example, starting from the core point of a fracture slice (e.g., the coordinates of the center point of the fracture slice), the angle of rotation of the fracture slice counterclockwise relative to true north ranges from 0° to 360°. Well location and horizon data are the foundation of a three-dimensional geological model. By quantitatively describing and calculating fracture parameters from imaging logging of target layers in a single well in the study area, information such as fracture dip, azimuth, aperture, and extension can be obtained. For fracture orientation information in non-well locations within the study area, seismic ant volume data was selected. Dip and azimuth attributes were extracted from the ant volume data to form fracture dip attribute volumes and fracture azimuth attribute volumes. These constraints constrain the generated fracture dip and azimuth, ensuring that the simulated fracture occurrence more closely matches the actual geological conditions.

[0099] ⑤Crack tilt model, such as Figure 6 Figure 2 shows a fracture dip attribute model extracted from ant volume data in the seismic data using the Petrel software. This model represents the fracture dip angle attribute. The fracture dip model attribute value represents the tilt angle of the fracture slice at each grid point. This represents the degree of inclination of the fracture slice relative to the horizontal plane, ranging from -90° to 90°. The method for obtaining the fracture dip attribute model for the study area is similar to that for the azimuth attribute model and will not be further elaborated here.

[0100] 3. Obtaining fracture occurrence data (extension, strike, dip) and distribution parameters

[0101] Before conducting fracture simulation, it is necessary to obtain the occurrence data and distribution patterns of fractures of various levels based on the geological data of the study area. In the observation of fracture parameters of core and imaging logging in the study area, the occurrence data of each fracture required for fracture simulation is obtained, for example: ① Horizontal extension range: such as Figure 7 As shown in Fig. 2, it characterizes the horizontal extension length of the fracture sheet. With the core point of the fracture sheet (e.g., the center coordinate point) as the origin, the initial position of the fracture sheet is located at a position perpendicular to the Y axis and passing through the origin. The extension length in the X-axis direction is the horizontal extension distance. In the observation of fracture parameters of core and imaging logging, the horizontal extension range of the fracture sheet in the study area and the corresponding average extension range can be obtained; ② Vertical extension range: as shown in Fig. 2. Figure 8 As shown in Figure 2, it represents the extension length of the fracture sheet in the vertical direction. The extension length of the fracture sheet in the Z-axis direction is the vertical extension distance. In the observation of fracture parameters of core and imaging logging, the vertical extension range of the fracture sheet in the study area and the corresponding average extension range can be obtained; ③ Strike (azimuth) angle: as shown in Figure 2 Figure 9As shown in the figure, when the fracture slice is in the initial position, the Z axis passing through the core point is used as the rotation axis, and the fracture slice direction is rotated clockwise by a certain angle. In the observation of fracture parameters of core and imaging logging, the range of the fracture slice direction (azimuth) and the corresponding average direction (azimuth) of the study area can be obtained. The setting value here is only valid when there is no fracture orientation model. When both data are available, the data of the fracture orientation model takes precedence; ④ Tilt angle: as Figure 10 As shown in the figure, after the strike (azimuth) angle of the fracture slice is adjusted, the straight line passing through the core point of the fracture slice, perpendicular to the Z axis and passing through the fracture slice is used as the rotation axis, and the fracture slice is rotated clockwise around the positive direction of the X axis by a certain angle, which is the inclination angle of the fracture slice. In the core and imaging logging fracture parameter observation, the range of the fracture slice inclination angle and the corresponding average inclination angle range in the study area can be obtained. The setting value here is only valid when there is no fracture inclination model. When both have data, the data of the fracture inclination model takes precedence. ⑤ Data distribution: In the core and imaging logging fracture parameter observation, all observed fracture-related data are counted to obtain the distribution mode of the data (random distribution, uniform distribution, normal distribution). In the subsequent fracture simulation process, the parameter values ​​of each group of fracture slices are randomly extracted according to this data distribution mode to complete the fracture simulation and ensure that the simulation results are consistent with the actual data.

[0102] 4. Crack density parameter setting:

[0103] The determination of the crack density can be determined by the crack strength model, or the number of crack slices can be determined by setting the crack density parameter value.

[0104] As mentioned above, the establishment of the fracture strength model can be constructed through dynamic data, core data, well logging, seismic data and literature research data, and the strength curve data can be extracted from the fracture data. The strength curve data is then coarsened into the established model body. The coarsened strength curve data is used as the conditional point data. With the fault distance and structural curvature as the constraints, the Gaussian random simulation method is used to establish the fracture strength model of the study area. For the calculation of the fracture strength value and the number of fractures, refer to the calculation formula (1):

[0105]

[0106] Where Sum_lf is the total number of cracks, n is the total number of grids in the study area, Vi is the volume of the i-th grid, and Qi is the strength attribute value of the i-th grid.

[0107] When the study area lacks a crack strength attribute model, the number of crack slices can be determined by setting the crack density parameter. For a single grid, the specific calculation method is the ratio of the scale value of the single grid in the three directions to the set crack density value, which is the number of crack slices in a certain direction in the current single grid. The cumulative multiplication of the number of cracks in the three directions of X, Y, and Z is the number of crack slices in the current grid; the number of crack slices in a certain direction can also be changed by setting the density multiplier in the X, Y, and Z directions. The default value is 1. The larger the value, the more crack points there are. The number of crack points in a single direction = (single grid volume / overall initial density value) * single direction density multiplier. The specific calculation process is as follows:

[0108] For a single grid, obtain its grid size based on the three-dimensional directions X, Y, and Z of the model body, which are L_x, L_y, and L_z respectively. Assuming the overall initial density is Den_lf, the initial number of crack points Num_x, Num_y, and Num_z in the three directions of X, Y, and Z can be calculated by the following formula (2):

[0109]

[0110] Set the density ratios T_x, T_y, and T_z of the three-dimensional directions X, Y, and Z of the model body respectively. According to the density ratios T_x, T_y, and T_z in the X, Y, and Z directions, calculate the final number of crack points Fin_x, Fin_y, and Fin_z in the X, Y, and Z directions by the following formula (3):

[0111] Fin_x=Num_x*T_x

[0112] Fin_y=Num_y*T_y

[0113] Fin_z=Num_z*T_z (3);

[0114] The final total number of grid crack slices Sum_lf is calculated by the following formula (4):

[0115] Sum_lf=Fin_x*Fin_y*Fin_z (4).

[0116] The improved new method of the present invention can also specify whether there is an area with a high number of cracks. When a certain area in the study area is a high-development or low-development area of ​​cracks, it indicates that the crack density in this area is different from the overall area, and the crack density value needs to be set separately, that is, the local area is encrypted or sparsely processed. At the same time, the range of the encrypted area also needs to be input. There are two ways to define the range of the encrypted area. One is based on the grid index range in the three directions of X, Y, and Z in the three-dimensional coordinate system; the other is to establish a range model body based on the set of non-null value areas made in Petrel, that is, using the phase control method to set different phase types for different density areas, establish a mapping relationship between phase type and crack density value, and then set different crack density values ​​for different areas.

[0117] 5. Crack simulation:

[0118] With the matrix phase model as the range constraint, the fracture simulation is carried out by adopting the grid-by-grid simulation method. The simulation of a single fracture can be divided into two steps: point process and indicative process.

[0119] Step 1: Click the process.

[0120] For the study area, obtain the coordinate range of each grid point.

[0121] Select a grid according to the grid index, obtain the coordinate range of the selected current grid, and determine whether there is a crack strength constraint at the current grid point (that is, whether there is crack strength value data):

[0122] If there is a crack strength constraint, the strength value of the crack in the current grid is obtained, and the number of crack slices in the current grid is calculated based on the strength value of the crack. The calculation process refers to the calculation formula (1) described above;

[0123] If there is no crack strength constraint, then set the crack density value and the density multiplier value of the model body in the three-dimensional direction (X, Y, Z) to determine the number of crack slices in the current grid. The calculation process refers to the calculation formulas (2) to (4) described above.

[0124] like Figure 11 The figure shows a schematic diagram of randomly generating the center points of crack slices within a grid. The coordinates of the center points of each crack slice are randomly generated using the X, Y, and Z coordinate ranges of the current grid as constraints.

[0125] Traverse all the grids in the area to be simulated and obtain the coordinates of the center points of the crack slices in each grid.

[0126] Step 2: Demonstration process.

[0127] After determining the center point of a single fracture slice, each fracture slice is grouped. The grouping principle is as follows: all adjacent fracture slices in the grid are grouped together according to the fracture shape. Then, the scale and posture of each fracture slice group are adjusted based on the fracture shape, occurrence data volume, and the coordinates of the center point of each fracture slice group to simulate complex fractures and generate a fracture network. The specific process is as follows:

[0128] According to the shape of the crack to be simulated and the coordinates of the center points of each crack slice, all adjacent crack slices in the grid are divided into the same group to obtain several groups of crack slices, each group of crack slices represents one crack;

[0129] Randomly extract the attribute values ​​of any group of crack slices according to the distribution of the cracks to be simulated, and obtain the horizontal extension value, vertical extension value, azimuth angle value, and inclination angle value of the current group of crack slices extracted based on the crack occurrence data (such as extension range / crack size, azimuth angle, inclination angle) or set attribute data. The horizontal extension value and vertical extension value can represent the size of the crack;

[0130] Based on the center point coordinates of the crack slice, the coordinates of the four vertices of each crack slice in the current set of crack slices extracted are calculated according to the horizontal extension value and vertical extension value (crack size) of the crack slice to establish an initial crack slice; the azimuth and inclination angles of the initial crack slice are adjusted according to the obtained azimuth and inclination angle values ​​to obtain a final crack slice;

[0131] Generate corresponding types of cracks according to the shape of the crack to be simulated; where:

[0132] If the shape of the crack to be simulated is an isolated crack, the final crack slice is used as the crack simulation result;

[0133] like Figure 12 The figure shows a schematic diagram of a curved crack. If the shape of the crack to be simulated is a curved crack, the adjacent crack slices are spliced ​​end to end in sequence according to the positional relationship of the crack slices to simulate the curved crack.

[0134] like Figure 13 The figure shows a schematic diagram of a conjugate morphology crack. If the shape of the crack to be simulated is a conjugate crack, then based on the positional relationship of the crack slices, two adjacent crack slices are divided into a group and share a common center coordinate point. The two crack slices use different azimuth angles and / or tilt angles to simulate the conjugate crack.

[0135] For any combination of crack patterns, such as Figure 14The figure shows a schematic diagram of a binary tree-like (lightning-shaped) crack. For example, if the crack to be simulated is a binary tree-like crack (lightning crack), the crack can be considered to be a tree-like structure formed by continuous growth and splitting. By defining the parameters of the binary tree: the number of branches, extension length, branch length, and branch angle, and splicing the crack slices in the same group according to the defined binary tree parameters, complex cracks can be reproduced.

[0136] Traverse all groups of crack slices in all grids to generate a crack network.

[0137] It is understandable that in this embodiment, a set of attribute values ​​such as crack size (including horizontal extension range and vertical extension range), strike (azimuth angle), and inclination angle are randomly extracted according to the distribution of the cracks. First, the coordinates of the four vertices of the crack slice are calculated based on the attribute values ​​such as the crack size to complete the establishment of the initial crack slice. Then, according to the strike and inclination values ​​of the crack slice, its strike and inclination angle are changed to obtain the final crack slice. For cracks with curved shapes in practice, they can be approximated as multiple straight lines. Therefore, multiple crack slices are connected front and back to simulate curved cracks. For cracks with conjugate shapes in practice, two adjacent cracks can be divided into a group. The two crack slices share the same core point and use different strike angles to simulate conjugate cracks. For crack combination patterns of arbitrary shapes, such as "lightning cracks", the cracks can be considered as tree structures formed by continuous growth and splitting. By defining the bifurcation, extension length, number of bifurcations, and bifurcation angles of the tree, the reproduction of complex cracks can be achieved.

[0138] Because this method uses a grouping approach, neighboring fractures are grouped together to simulate real fracture morphology, resulting in a more realistic fracture model. Each group of fracture slices represents a single fracture. By traversing all groups of fracture slices in the grid, the simulation of all fractures is completed, forming a fracture network.

[0139] An embodiment of the present invention provides a modeling method based on an improved discrete fracture network, belonging to the field of oil and gas exploration and development technology. It enables the generation of fractures of arbitrary shapes and combinations. The method can be integrated with the commercial Petrel software, supporting Petrel's fracture saving format and importing fracture simulation results into Petrel to facilitate subsequent research. Compared with existing technologies, the present invention improves upon and builds on existing discrete fracture network modeling algorithms, aiming to address the following deficiencies of existing discrete fracture network modeling algorithms: simple morphology and inability to characterize curved fractures; inability to consider combinations of multiple fractures and spatial correlations; and inability to simultaneously simulate fractures at multiple scales. Using geological data such as well data, seismic data, core data, and imaging logging data, users construct a 3D geological model of the study area, integrating fracture attributes such as strength, azimuth, and dip, along with parameters such as fracture extension and size range. The model is then refined into a grid volume, and a corresponding number of independent fracture slices are assigned to each grid cell. By grouping the individual fracture slices and adding shape attributes to each, various fracture types can be constructed, including isolated fractures with a single fracture slice, fractures with a curved shape formed by connecting multiple fracture slices, conjugate fractures formed by intersecting two fracture slices, and lightning-shaped fractures with multiple fracture slices arranged in a binary tree pattern. For arbitrary fracture shapes, a circle approximation is employed in computer graphics. Any curved fracture can be approximated as multiple straight lines. These lines, when stitched together, represent the curved fracture morphology in a plane. On a cross-section, lines are spliced ​​together using a similar curvature concept to depict curved fractures in the cross-section, thereby characterizing any fracture shape. First, the curvature of the fracture and the length of the extended lines are determined. Then, its segmentation is defined and the number of separating lines is determined; finally, the changes in the orientation of different separating lines are determined, and the rotational splicing of the lines is completed to construct a crack with a curved shape. For combined cracks, such as the simulation of conjugate cracks, the conjugate crack is thrown as a whole. In the prediction of the point process, the indicative point determines whether the location is a conjugate crack. Once it is determined to be a conjugate crack, the conjugate crack can be generated and thrown. When the conjugate crack is generated, the length, orientation, inclination of the conjugate crack, and the position of the intersection between the two cracks can be obtained by statistics, so as to better characterize the combined form of the conjugate crack. By grouping adjacent cracks and defining the combined pattern of cracks within the group, such as lightning cracks, it is considered that the cracks are a tree-like structure formed by continuous growth and splitting. By defining the bifurcation, extension length, and number of bifurcation layers of the tree, the reproduction of such complex cracks can be achieved.

[0140] Since the present invention uses an improved discrete fracture network modeling method, not only does the established fracture model have different shapes, coordinates, sizes, orientations, and openings, but it also takes into account the reliability of sampling parameters such as the fracture shape, size, dip, and inclination, as well as the spatial correlation of fracture slices, thereby establishing a fracture model that is closer to the actual fracture pattern, improving the modeling accuracy of fractured gas reservoirs, more accurately predicting high-quality seepage channels, and providing reliable data for numerical simulation of fractured gas reservoirs, which can efficiently serve scientific research and oilfield geological research.

[0141] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0142] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0144] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0145] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0146] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0147] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A modeling method based on an improved discrete fracture network, characterized in that: include: S1, grid the study area and obtain data representing the intensity, density, morphology and occurrence of each level of fractures based on the geological data of the study area; The morphology of cracks includes: isolated, curved, conjugate or binary tree, and the occurrence of cracks includes: extension, inclination and orientation; S2, calculating the number of crack slices in a single grid according to the intensity value or density value of the cracks in each grid, and generating the coordinates of the center point of each crack slice; S3, based on the shape of the cracks, groups all adjacent crack slices in the grid into the same group. Based on the crack shape, occurrence, and the coordinates of the center points of each group of crack slices, the scale and posture of each crack slice are adjusted to simulate complex cracks and generate a crack network. This includes: According to the shape of the crack to be simulated and the coordinates of the center points of each crack slice, all adjacent crack slices in the grid are divided into the same group to obtain several groups of crack slices; Randomly extract any group of fracture slices according to the distribution of the fractures to be simulated, and obtain the horizontal extension value, vertical extension value, azimuth angle value and inclination angle value of the current group of fracture slices extracted according to the fracture occurrence data or set data; Based on the center point coordinates of the crack slice, the coordinates of the four vertices of each crack slice in the current group of crack slices extracted are calculated according to the horizontal extension value and the vertical extension value of the crack slice to establish an initial crack slice; the azimuth angle and inclination angle of the initial crack slice are adjusted according to the obtained azimuth angle value and inclination angle value to obtain a final crack slice; Generate corresponding types of cracks according to the shape of the crack to be simulated; Traverse all groups of crack slices in all grids to generate a crack network; S4, evaluating whether the fracture network is reasonable based on the geological data of the study area. If it is not reasonable, re-execute step S2.

2. A modeling method based on an improved discrete fracture network according to claim 1, characterized in that: Step S1: grid the study area and obtain data representing the strength, density, morphology, and occurrence of fractures of various levels based on the geological data of the study area; The morphology of cracks includes: isolated, curved, conjugate or binary tree, and the occurrence of cracks includes: extension, inclination and orientation; specifically, they include: Construct a model body corresponding to the scope of the study area, divide the study area into grids according to a preset grid size, and record the model body data through the grid index. The sedimentary facies map, well-connected profile map and manually interpreted well sedimentary facies data in the geological data are digitized into the gridded model volume to establish a sedimentary facies model corresponding to the study area; Core sampling, logging, seismic and production dynamic data from the study area are used to identify and describe core fractures and logging fractures, so as to obtain data bodies characterizing the intensity, density, morphology, occurrence and distribution of fractures of various orders in the study area. The morphology includes: isolated, curved, conjugate or binary tree, and the occurrence includes: extension, inclination and orientation.

3. A modeling method based on an improved discrete fracture network according to claim 1 or 2, characterized in that: Step S1 also includes constructing a fracture-related attribute model based on geological data of the study area, wherein the attribute model includes at least a fracture range model, a fracture intensity model, a fracture orientation model, and a fracture inclination model; wherein, Constructing a fracture range model, including: constructing a sedimentary facies model using a geological modeling algorithm based on lithofacies, bedding planes, and fault data. The sedimentary facies model is used to determine whether a grid belongs to the area to be simulated, thereby constraining the fracture range; Establishing a fracture strength model includes: constructing fracture data of different scales through core, well logging, seismic, dynamic data and literature research data, extracting strength curve data from the fracture data, coarsening the strength curve data into the established sedimentary facies model, using the coarsened strength curve data as conditional point data, using fault distance and structural curvature as constraints, and using Gaussian random simulation to establish a fracture strength model for the study area to characterize fracture density attributes; Establish a fracture orientation model and a fracture inclination model, including: for the well location area of ​​the study area, quantitatively describe and calculate the fracture parameters of the single well target layer imaging logging to obtain the fracture dip, azimuth, opening, and extension length; for the non-well location area of ​​the study area, extract the dip and azimuth attributes through the ant volume data in the seismic data to form a fracture inclination model and a fracture orientation model to constrain the generated fracture dip and azimuth.

4. A modeling method based on an improved discrete fracture network according to claim 3, characterized in that: Step S2, calculating the number of crack slices in a single grid according to the intensity value or density value of the cracks in each grid, and generating the coordinates of the center points of each crack slice; including: Select a grid according to the grid index, obtain the coordinate range of the current grid, and determine whether the current grid has crack strength constraints: If there is a crack strength constraint, the strength value of the crack in the current grid is obtained, and the number of crack slices in the current grid is calculated based on the crack strength value; If there is no crack strength constraint, set the crack density value and the density multiplier value in the three-dimensional direction (X, Y, Z) of the model body to determine the number of crack slices in the current grid; Using the coordinate range of the current grid as a constraint, randomly generate the center point coordinates of each crack slice; Traverse all the grids in the area to be simulated and obtain the coordinates of the center points of the crack slices in each grid.

5. The modeling method based on an improved discrete fracture network according to claim 4, characterized in that: If there is no crack intensity constraint, and the crack density of some areas to be simulated is different from the overall crack density, then a separate crack density value is set for the area where the crack density is different from the overall crack density; or, The phase control method is used to set different phase types for different density areas, and a mapping relationship between phase types and fracture density values ​​is established, thereby setting different fracture density values ​​for different areas.

6. The improved discrete fracture network modeling method according to claim 4, characterized in that: The number of crack slices in the current grid is calculated based on the crack strength value and is calculated using the following formula (1): （1）， In formula (1), Sum_lf is the total number of cracks, n is the total number of grids in the study area, V i For the i The volume of the grid, Q i For the i The strength property value of each mesh.

7. A modeling method based on an improved discrete fracture network according to any one of claims 4 to 6, characterized in that: If there is no crack strength constraint, then the crack density value and the density multiplier value of the model body in the three-dimensional direction (X, Y, Z) are set to determine the number of crack slices in the current grid, including: For a single grid, obtain its grid size based on the three-dimensional directions X, Y, and Z of the model body, which are L_x, L_y, and L_z respectively. Assume that the overall initial density is Den_lf, then the initial number of crack points Num_x, Num_y, and Num_z in the three directions of X, Y, and Z are calculated by the following formula (2): （2）； Set the density magnifications T_x, T_y, and T_z of the model body in the three-dimensional directions X, Y, and Z respectively. According to the density magnifications T_x, T_y, and T_z in the X, Y, and Z directions, calculate the final number of crack points Fin_x, Fin_y, and Fin_z in the X, Y, and Z directions using the following formula (3): （3）； The total number of cracks in the grid Sum_lf is calculated by the following formula (4): （4）。 8. A modeling method based on an improved discrete fracture network according to any one of claims 4 to 6, characterized in that: In step S1, based on the geological data of the study area, data bodies representing the occurrence and distribution of fractures of various levels are obtained, including: In the observation of core and imaging logging fracture parameters in the study area, the occurrence data volume of each fracture is obtained. The occurrence data volume includes: Horizontal extension range, which represents the maximum and minimum horizontal extension ranges of the cracks in the study area and the corresponding horizontal horizontal extension range; Vertical extension range, which represents the maximum and minimum vertical extension ranges of the cracks in the study area and the corresponding average vertical extension range; Azimuth angle range, which represents the strike range of the fractures rotating around the vertical direction in the study area and the corresponding average azimuth angle range; The tilt angle range characterizes the strike range of the fractures rotating around the horizontal plane and the corresponding average tilt angle range in the study area; The distribution mode of the horizontal extension range, vertical extension range, azimuth angle range and inclination angle range of the cracks in the study area is obtained, and the distribution mode is random distribution, uniform distribution or normal distribution.

9. The improved discrete fracture network modeling method according to claim 8, characterized in that: The step of generating a corresponding type of crack according to the shape of the crack to be simulated includes: If the shape of the crack to be simulated is an isolated crack, the final crack slice is used as the crack simulation result; If the shape of the crack to be simulated is a curved crack, the adjacent crack slices are spliced ​​end to end in sequence according to the positional relationship of the crack slices to simulate the curved crack; If the shape of the crack to be simulated is a conjugate crack, then based on the positional relationship of the crack slices, two adjacent crack slices are divided into a group and share a common center coordinate point. The two crack slices use different azimuth angles and / or tilt angles to simulate the conjugate crack. If the shape of the crack to be simulated is a binary tree crack, the parameters of the binary tree are defined: number of branches, extension length, branch length, and branch angle, and the crack slices in the same group are spliced ​​according to the defined binary tree parameters to simulate the binary tree crack.

Images