Natural fracture modeling and fracturing simulation method, device, computer and storage medium

Abstract

The present invention provides a natural fracture modeling and fracturing simulation method, device, computer and storage medium, the method comprising establishing a three-dimensional geological grid model of a preset work area; calculating and obtaining post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and importing the post-stack attributes and pre-stack inversion attributes into the three-dimensional geological grid model; on the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, using a neural network algorithm to predict the spatial distribution of reservoir attributes, and obtaining a natural fracture model of the preset work area; on the basis of the natural fracture model, extracting the average fracture surface density attribute of the target layer, generating equivalent natural fracture information; and performing fracturing simulation according to the natural fracture parameters. By utilizing the pre-stack attributes and post-stack attributes of seismic data, combined with geological, lithological and logging information, and using a neural network algorithm, a fracture development distribution model, i.e., a natural fracture model, is established, thereby reducing the multi-solution of fracture reservoir prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of seismic exploration and development, and in particular to a natural fracture modeling and fracturing simulation method, device, computer and storage medium. Background Art

[0002] The characteristics of shale gas natural fractured oil and gas reservoirs are that their favorable reservoirs are complex and irregularly distributed. Existing technologies mainly use a single technology to predict and describe the distribution law of fractures. However, a single technology often has great uncertainty and multiple solutions, and it is impossible to use seismic data for fracturing simulation based on the material point method.

[0003] In the development of unconventional shale gas horizontal wells, the fracturing effects of each well vary greatly, and the test production also varies greatly; in the shale gas development engineering stage, completion engineers lack understanding of the development of underground natural fractures, and the simple treatment of multi-stage fracturing during completion easily leads to very little oil and gas production in the fracturing section; at the same time, due to the heterogeneity of shale gas reservoirs and the inhomogeneity of fracturing transformation, it is urgent to realize the integration of geology, geophysical exploration and engineering in order to maximize the potential of the reservoir and achieve the goal of economical and efficient exploration and development. Summary of the invention

[0004] Based on this, it is necessary to provide a natural fracture modeling and fracturing simulation method, device, computer and storage medium to address the above technical problems.

[0005] A natural fracture modeling and fracturing simulation method, comprising:

[0006] Establish a three-dimensional geological grid model of the preset work area;

[0007] Based on the seismic data of the preset work area, calculate and obtain post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model;

[0008] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area;

[0009] Based on the natural fracture model, extract the average fracture surface density attribute of the target layer to generate equivalent natural fracture information;

[0010] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and fracturing simulation is performed according to the natural fracture parameters.

[0011] In one embodiment, the step of obtaining the natural fracture model of the preset work area by using a neural network algorithm to predict the spatial distribution of reservoir properties based on the three-dimensional geological grid model and the roughened curve information on the well comprises:

[0012] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is adopted to train with the well fracture information as learning samples to predict the distribution of fractures in space and obtain the natural fracture model of the preset work area.

[0013] In one embodiment, the spatial distribution of the reservoir attributes includes at least one of the following attributes: lithology, reservoir porosity, permeability, and fracture density.

[0014] In one embodiment, based on the three-dimensional geological grid model, based on the roughened curve information on the well, using the neural network algorithm, taking the well fracture information as a learning sample for training, predicting the distribution of fractures in space, and obtaining the natural fracture model of the preset work area includes:

[0015] Based on the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is used to train with the well fracture information as a learning sample to predict the distribution of fractures in space and obtain multiple prediction data of the distribution of fractures in space;

[0016] Obtain actual data on the distribution of cracks in space;

[0017] Comparing the plurality of predicted data with the actual data, and extracting the preset data whose deviation from the actual data is within a preset range;

[0018] Based on the extracted preset data, a natural fracture model of the preset work area is generated.

[0019] In one embodiment, the step of establishing a three-dimensional geological grid model of a preset work area includes:

[0020] Obtaining a first horizontal axis coordinate value, a second horizontal axis coordinate value, and a depth coordinate value of the preset work area;

[0021] The three-dimensional geological grid model of the preset work area is established based on the first horizontal axis coordinate value, the second horizontal axis coordinate value and the depth coordinate value of the preset work area.

[0022] In one embodiment, the step of using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and performing fracturing simulation according to the natural fracture parameters includes:

[0023] Obtain the modern stress field conditions, the plane mechanical properties of the target layer rock, the fracturing construction time and the bottom hole pressure during fracturing construction;

[0024] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and fracturing simulation is performed according to the natural fracture parameters, the modern stress field conditions, the target layer rock mechanical plane properties, the fracturing construction time and the bottom hole pressure during the fracturing construction.

[0025] In one embodiment, the step of using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and performing fracturing simulation according to the natural fracture parameters includes:

[0026] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and a material point method is used to perform fracturing simulation according to the natural fracture parameters.

[0027] A natural fracture modeling and fracturing simulation device, comprising:

[0028] The grid model building module is used to build a three-dimensional geological grid model of the preset work area;

[0029] An attribute import module, used for calculating and obtaining post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures based on the seismic data of the preset work area, and importing the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model;

[0030] A natural fracture model acquisition module is used to obtain a natural fracture model of the preset work area by using a neural network algorithm to predict the spatial distribution of reservoir attributes based on the three-dimensional geological grid model and the roughened curve information on the well;

[0031] An equivalent natural fracture information generation module is used to extract the average fracture surface density attribute of the target layer based on the natural fracture model to generate equivalent natural fracture information;

[0032] The fracturing simulation module is used to use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and perform fracturing simulation according to the natural fracture parameters.

[0033] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the following steps when executing the computer program:

[0034] Establish a three-dimensional geological grid model of the preset work area;

[0035] Based on the seismic data of the preset work area, calculate and obtain post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model;

[0036] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area;

[0037] Based on the natural fracture model, extract the average fracture surface density attribute of the target layer to generate equivalent natural fracture information;

[0038] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and fracturing simulation is performed according to the natural fracture parameters.

[0039] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0040] Establish a three-dimensional geological grid model of the preset work area;

[0041] Based on the seismic data of the preset work area, calculate and obtain post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model;

[0042] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area;

[0043] Based on the natural fracture model, extract the average fracture surface density attribute of the target layer to generate equivalent natural fracture information;

[0044] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and fracturing simulation is performed according to the natural fracture parameters.

[0045] The above-mentioned natural fracture modeling and fracturing simulation method, device, computer and storage medium, by utilizing the pre-stack attributes and post-stack attributes of seismic data, combining geological, lithological and logging information, and using a neural network algorithm, establish a fracture development distribution model, i.e., a natural fracture model, thereby reducing the multi-solution nature of fracture reservoir prediction.

[0046] In addition, this application uses the MPM material point method to carry out fracturing simulation and fracturing effect evaluation, which can guide the deployment of horizontal wells and achieve the goal of economical and efficient exploration and development. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1A schematic diagram of a process of natural fracture modeling and fracturing simulation method in one embodiment;

[0048] Figure 2 A structural block diagram of a natural fracture modeling and fracturing simulation device in one embodiment;

[0049] Figure 3 is an internal structure diagram of a computer device in one embodiment;

[0050] Figure 4 is a schematic diagram of a three-dimensional geological grid model of a work area X in one embodiment;

[0051] Figure 5 is a schematic diagram of various seismic attributes of the X work area in one embodiment;

[0052] Figure 6 A schematic diagram of a natural fracture modeling result of an X work area in one embodiment;

[0053] Figure 7 It is a schematic diagram of the fracturing simulation results of the well area A in the X work area in one embodiment;

[0054] Figure 8 Schematic diagram of microseismic monitoring results of well area A in work area X in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] Embodiment 1

[0057] In this embodiment, Figure 1 As shown, a natural fracture modeling and fracturing simulation method is provided, which includes:

[0058] Step 110, establishing a three-dimensional geological grid model of a preset work area.

[0059] Specifically, the preset work area is a work area where natural fracture modeling and fracturing simulation need to be performed. In the following embodiment, the preset work area is the X work area.

[0060] In this step, a three-dimensional geological grid model is established using the collected seismic horizon interpretation results. The seismic horizon interpretation results are the interpretation results of the horizon and fault in the study area. For example, the horizon data and fault data of the study area are obtained, and the three-dimensional geological data of the preset work area are obtained. Based on the three-dimensional geological data of the preset work area, a three-dimensional geological grid model of the preset work area is established.

[0061] Step 120, based on the seismic data of the preset work area, calculate and obtain post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model.

[0062] Specifically, seismic data of a preset work area is obtained, and the seismic data is the seismic data of the preset work area. Post-stack attributes reflecting faults and natural fractures and pre-stack inversion attributes are calculated based on the seismic data of the preset work area.

[0063] In this embodiment, the post-stack attributes of natural fractures include: seismic curvature attributes and similarity attributes of natural fractures, etc.; the pre-stack inversion attributes include: longitudinal wave impedance, transverse wave impedance, density, porosity, Poisson's ratio, Young's modulus, etc.

[0064] In this embodiment, the post-stack attributes reflecting faults and natural fractures can be obtained by at least one of the following methods: coherence volume technology, curvature technology, ant volume technology, and discontinuity detection technology. It should be understood that the method of obtaining the post-stack attributes reflecting faults and natural fractures can also adopt other methods in the prior art, which are not described in this embodiment. In addition, in this application, the pre-stack inversion class attributes are obtained by the pre-stack inversion method.

[0065] Step 130, based on the three-dimensional geological grid model and based on the roughened curve information on the well, a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area.

[0066] In one embodiment, the spatial distribution of the reservoir attributes includes at least one of the following attributes: lithology, reservoir porosity, permeability, and fracture density.

[0067] In this step, the roughened curve information of the well can be first obtained, and based on the constructed three-dimensional geological grid model into which the post-stack attributes and the pre-stack inversion attributes have been imported, the spatial distribution of reservoir attributes can be predicted using a neural network algorithm based on the roughened curve information of the well, and the natural fracture model of the preset work area can be obtained. Specifically, the predicted spatial distribution of reservoir attributes includes lithology, reservoir porosity, permeability, fracture density, etc.

[0068] Specifically, different types of seismic attributes describe the possible development areas of natural fractures from different perspectives. A neural network method is used to comprehensively consider multiple seismic attributes and use wellbore fracture information as learning samples to predict the distribution of fractures in space.

[0069] It should be understood that in this step, the spatial distribution of reservoir properties can also be predicted by geostatistical methods to obtain the natural fracture model of the preset work area.

[0070] In this step, based on the three-dimensional geological grid model and the roughened curve information on the well, a geostatistical method or a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area, wherein the geostatistical method is the Kriging method.

[0071] Step 140: based on the natural fracture model, extract the average fracture surface density attribute of the target layer to generate equivalent natural fracture information.

[0072] Step 150: Using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and performing fracturing simulation according to the natural fracture parameters.

[0073] In this step, by simulating the modern stress field, the order of pressurization of the fracturing stages can be simulated according to the actual or designed simulation, and the estimated J integral on each fracturing stage can be simulated. The J integral result can show that the higher the energy at the fracturing stage, the more microseismic events there are. The J integral result can serve as a bridge between microseismicity and geomechanics. Through this application, completion engineers can predict which fracturing stage will show the highest or lowest microseismic activity during fracturing design.

[0074] In the above embodiment, by utilizing the pre-stack attributes and post-stack attributes of seismic data, combined with geological, lithological and logging information, a fracture development distribution model, i.e. a natural fracture model, is established through a neural network algorithm, thereby reducing the multi-solution nature of fracture reservoir prediction.

[0075] In one embodiment, the step of using a neural network algorithm to predict the spatial distribution of reservoir properties based on the roughened curve information on the well on the basis of the three-dimensional geological grid model to obtain the natural fracture model of the preset work area includes: based on the three-dimensional geological grid model, based on the roughened curve information on the well, using the neural network algorithm, using the well fracture information as a learning sample for training, predicting the spatial distribution of fractures, and obtaining the natural fracture model of the preset work area.

[0076] In this embodiment, a neural network algorithm is first used to train the wellbore fracture information as a learning sample to obtain a natural fracture model of the preset work area.

[0077] Specifically, based on the structural model, the neural network method is used to predict the spatial distribution of reservoir attributes, including lithology, reservoir porosity, permeability, fracture density, etc., based on the roughened curve information on the well. It should be understood that different seismic attributes reflect fractures from different angles. The neural network method can comprehensively consider multiple seismic attributes and use the fracture information on the well as a learning sample to predict the distribution of fractures in space. Through a variety of quality control methods, such as setting the learning training data ratio, setting the correlation threshold value of training data and test data, and observing the intersection diagram of calculated data and actual data, it is ensured that the prediction results of the neural network algorithm are consistent with the known data, so as to obtain a parameter distribution that is closer to the actual. At the same time, the neural network algorithm is a random algorithm. In order to reduce randomness, 10 realizations are calculated each time, and only 1-2 realizations are saved. More than 30 realizations need to be saved cumulatively. Multiple results are statistically analyzed, and the average of 30 realizations is taken as the average natural fracture model to obtain the natural fracture modeling results of the X work area.

[0078] In this embodiment, for the characteristics of the fracture indication curve, the various seismic attributes extracted based on the seismic data are classified into categories: amplitude attributes (frequency division results of different frequencies based on the matching pursuit algorithm, amplitude envelope and original seismic amplitude), frequency attributes (instantaneous attributes and frequency division derived attributes), and structural attributes, and fuzzy logic sorting is performed respectively. Preferably, the 1-2 most sensitive seismic attributes in each type of seismic attribute to the fracture surface density curve are used together for random modeling using the neural network algorithm.

[0079] In one embodiment, based on the three-dimensional geological grid model, based on the roughened curve information on the well, using the neural network algorithm, taking the well fracture information as a learning sample for training, predicting the distribution of fractures in space, and obtaining the natural fracture model of the preset work area includes:

[0080] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is adopted, and the well fracture information is used as a learning sample for training to predict the distribution of fractures in space and obtain a plurality of predicted data of the distribution of fractures in space; the actual data of the distribution of fractures in space are obtained; the plurality of predicted data are compared with the actual data, and the preset data whose deviation from the actual data is within a preset range is extracted; based on the extracted preset data, a natural fracture model of the preset work area is generated.

[0081] In this embodiment, based on the structural model, based on the roughened curve information on the well, a neural network method is used to predict the spatial distribution of reservoir attributes, including lithology, reservoir porosity, permeability, fracture density, etc. Different seismic attributes reflect fractures from different angles. The neural network method can comprehensively consider multiple seismic attributes and use the well fracture information as a learning sample to predict the distribution of fractures in space. Through a variety of quality control methods, such as setting the learning training data ratio, setting the threshold value of the correlation between training data and test data, and observing the intersection diagram of predicted data and actual data, it is ensured that the prediction results of the neural network algorithm are consistent with the known data, so as to obtain a parameter distribution that is closer to the actual. At the same time, the neural network algorithm is a random algorithm. In order to reduce randomness, 10 realizations are calculated each time, and only 1-2 realizations are saved. More than 30 realizations need to be saved cumulatively. Multiple results are statistically analyzed, and 30 realizations are averaged as the average natural fracture model to obtain the natural fracture modeling result of the X work area. In this embodiment, each "realization" is equivalent to the result of a calculation, and each "realization" is a natural fracture model obtained by a calculation. By obtaining the realizations multiple times and averaging them, a natural fracture model can be obtained.

[0082] In one embodiment, the step of establishing a three-dimensional geological grid model of a preset work area includes: obtaining a first horizontal axis coordinate value, a second horizontal axis coordinate value and a depth coordinate value of the preset work area; and establishing the three-dimensional geological grid model of the preset work area based on the first horizontal axis coordinate value, the second horizontal axis coordinate value and the depth coordinate value of the preset work area.

[0083] In this embodiment, the stratum data and fault data of the study area are obtained to obtain the three-dimensional geological data of the preset work area, which are the first horizontal axis coordinate value, the second horizontal axis coordinate value and the depth coordinate value. In this embodiment, the first horizontal axis coordinate value is the X-axis value, the second horizontal axis coordinate value is the Y-axis value, and the depth coordinate value is the vertical axis value, that is, the Z-axis value, wherein the first horizontal axis coordinate value and the second horizontal axis coordinate value are the values ​​of the coordinate axis along the horizontal direction. The three-dimensional geological grid model of the preset work area can be established through the above-mentioned X-axis value, Y-axis value and depth Z-axis value.

[0084] In one embodiment, the step of using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and performing fracturing simulation according to the natural fracture parameters includes:

[0085] Acquire modern stress field conditions, target layer rock mechanical plane properties, fracturing construction time and bottom hole pressure during fracturing construction; use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and perform fracturing simulation based on the natural fracture parameters, the modern stress field conditions, the target layer rock mechanical plane properties, the fracturing construction time and the bottom hole pressure during fracturing construction.

[0086] In this embodiment, based on the natural fracture model, the average fracture surface density attribute of the target layer is extracted to generate equivalent natural fracture information as the natural fracture parameters of the differential stress field simulation, and the following parameters are pre-set: 1) modern stress field conditions; 2) rock mechanics plane attributes such as Poisson's ratio and Young's modulus of the target layer; 3) fracturing construction time and bottom hole pressure during construction. The material point method (MPM) is used to perform fracturing simulation through the natural fracture parameters simulated by the differential stress field and the above pre-set parameters. Among them, the rock mechanics plane attributes include Poisson's ratio and Young's modulus of the target layer, and the data of the rock mechanics plane attributes come from the attribute body of pre-stack inversion.

[0087] In one embodiment, the step of using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and performing fracturing simulation according to the natural fracture parameters includes:

[0088] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and a material point method is used to perform fracturing simulation according to the natural fracture parameters.

[0089] In this embodiment, the material point method (MPM) is used to solve the coupling between fractures (fault zones) and stress under modern stress field conditions to obtain the differential stress field properties. At the same time, it can also simulate the strain of the surrounding formations during multi-stage fracturing under the pressure and duration of each fracturing stage of the horizontal well, and describe the possible fracturing transformation range under the influence of the local stress field.

[0090] Specifically, the material point method (MPM) belongs to the mechanics solution of momentum equation, belongs to the meshless algorithm, and is the third generation of mechanics solution of momentum equation. When it comes to extremely large deformation and crushing of materials, the meshless material point method uses material points that carry all the information of the material to discretize the material region to characterize the motion and deformation state of the material region, and avoids the processing of convection terms. The material point method uses a regular Euler background grid to calculate the spatial derivatives and momentum equations, thereby realizing the interaction and connection between particles and avoiding the problem of grid distortion. Based on the natural fracture density model, the average fracture surface density attributes of the target layer are extracted to generate equivalent natural fracture information as the natural fracture parameters for MPM differential stress field simulation.

[0091] By simulating the modern stress field, the sequence of pressurization of the fracturing stages can be simulated according to the actual or designed simulation, and the estimated J integral on each fracturing stage can be simulated. The J integral results can show that the higher the energy at the fracturing stage, the more microseismic events there are. The J integral results can serve as a bridge between microseismicity and geomechanics. Through this application, completion engineers can predict which fracturing stage will show the highest or lowest microseismic activity during fracturing design.

[0092] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0093] Embodiment 2

[0094] This application uses seismic data to model natural fractures and simulate fracturing, realizing the integration of geology, geophysical exploration and engineering. It can maximize the reservoir potential, improve the fracturing effect of shale gas horizontal wells, and achieve the goal of economical and effective exploration and development. It lays the foundation for the development of shale gas horizontal wells and is innovative.

[0095] In this embodiment, a method for natural fracture modeling and fracturing simulation using seismic data is provided. A comprehensive fracture modeling based on seismic-geology-well logging-rock mechanics is used to identify favorable reservoir (sweet spot) development areas. On this basis, artificial fracturing simulations are carried out for completed horizontal well locations and predicted horizontal well locations to verify and predict fracturing effects.

[0096] Specifically, in this embodiment, it is necessary to first establish a structural model of the study area, and use the collected seismic horizon interpretation results to establish a three-dimensional geological grid. On the basis of the structural model, based on the roughened curve information on the well, geostatistics or neural network methods are used to predict the spatial distribution of reservoir attributes, including lithology, reservoir porosity, permeability, fracture density, etc. Different types of seismic attributes describe the possible development areas of natural fractures from different angles. The neural network method is used to comprehensively consider multiple seismic attributes, and the well fracture information is used as a learning sample to predict the distribution of fractures in space.

[0097] According to the characteristics of the fracture indication curve, the various seismic attributes extracted from seismic data are classified into categories: amplitude attributes (frequency division results of different frequencies based on the matching pursuit algorithm, amplitude envelope and original seismic amplitude), frequency attributes (instantaneous attributes and frequency division derived attributes), and structural attributes, and fuzzy logic sorting is performed respectively. The 1-2 most sensitive seismic attributes in each type of seismic attribute are selected for random modeling using the neural network algorithm.

[0098] The fracture surface density curve is established based on the fractures identified by imaging logging. It can only represent the fracture development within a few meters around the wellbore and does not completely match the geological grid surface element. Therefore, the correlation between the training result and the original data should not be too high, and a correlation of 70% is sufficient. At the same time, the verification result only needs to reach 65% (mainly reflecting the trend of fracture development, and cannot be required to be completely consistent). At the same time, due to the high uncertainty of fractures, more realizations are required. Usually, each test operation selects to save up to 2 realizations with good quality control effects, and the operation is repeated many times to obtain a total of several realizations. The average of several realizations is taken as the average natural fracture model, and the natural fracture modeling is completed.

[0099] The present invention uses the material point method (MPM) to solve the coupling between cracks (fault zones) and stress under modern stress field conditions and obtain the differential stress field properties. At the same time, it can also simulate the strain of the surrounding formations during multi-stage fracturing under the pressure and duration of each fracturing stage of the horizontal well, and describe the possible fracturing transformation range under the influence of the local stress field.

[0100] The material point method (MPM) belongs to the solution of momentum equation in mechanics, belongs to meshless algorithm, and is the third generation of mechanics solution of momentum equation. When it comes to large deformation and crushing of materials, the meshless material point method uses material points that carry all the information of the material to discretize the material region to characterize the motion and deformation state of the material region, and avoids the processing of convection terms. The material point method uses a regular Euler background grid to calculate the spatial derivatives and momentum equations, thereby realizing the interaction and connection between particles and avoiding the problem of grid distortion. Based on the natural fracture density model, the average fracture surface density attributes of the target layer are extracted to generate equivalent natural fracture information as the natural fracture parameters for MPM differential stress field simulation.

[0101] By simulating the modern stress field, the sequence of pressurizing the fracturing stages can be simulated according to the actual or designed conditions, and the estimated J-integral on each fracturing stage can be simulated. The J-integral results can show that the higher the energy at the fracturing stage, the more microseismic events there are. The J-integral results can serve as a bridge between microseismicity and geomechanics. Through the present invention, completion engineers can predict which fracturing stage will show the highest or lowest microseismic activity during fracturing design.

[0102] Compared with the existing technology, this application has the following advantages:

[0103] (1) Ability to comprehensively utilize multiple attribute information of 3D seismic data to model natural fractures;

[0104] (2) Three-dimensional seismic data can be used to simulate artificial fracturing, help optimize well locations before drilling, increase single well production, and guide horizontal well design, optimize fracturing design, and achieve cost savings.

[0105] Embodiment 3

[0106] First, if Figure 4 As shown, a three-dimensional geological grid model of the X work area is established. In this embodiment, the preset work area is the X work area. Based on the seismic data of the X work area, the post-stack attributes such as the seismic curvature attributes and similarity attributes reflecting the faults and natural fractures and the pre-stack inversion attributes are calculated and imported. After the import, as shown in FIG. Figure 5 shown.

[0107] The post-stack attributes reflecting faults and natural fractures can be obtained by at least one of the following methods: coherence volume technology, curvature technology, ant volume technology, and discontinuity detection technology. It should be understood that the method of obtaining the post-stack attributes reflecting faults and natural fractures can also adopt other methods in the prior art, which are not listed one by one in this embodiment. In addition, in this application, the pre-stack inversion class attributes are obtained by the pre-stack inversion method.

[0108] On the basis of the structural model, based on the roughened curve information on the well, the neural network method is used to predict the spatial distribution of reservoir attributes, including lithology, reservoir porosity, permeability, fracture density, etc. Different seismic attributes reflect fractures from different angles. The neural network method can comprehensively consider multiple seismic attributes and use the fracture information on the well as a learning sample to predict the distribution of fractures in space. Through a variety of quality control methods, such as setting the ratio of learning training data, setting the threshold value of the correlation between training data and test data, and observing the intersection diagram of calculated data and actual data, it is ensured that the prediction results of the neural network algorithm are consistent with the known data, so as to obtain a parameter distribution that is closer to the actual. At the same time, the neural network algorithm is a random algorithm. In order to reduce randomness, 10 realizations are calculated each time, and only 1-2 realizations are saved. A total of more than 30 realizations need to be saved. Statistical analysis is performed on multiple results, and the average of 30 realizations is taken as the average natural fracture model to obtain the natural fracture modeling results of the X work area. The modeled natural fracture model is as follows: Figure 6 shown.

[0109] Based on the natural fracture model, the average fracture surface density attribute of the target layer is extracted to generate equivalent natural fracture information as the natural fracture parameters for differential stress field simulation, and the following parameters are set: 1) modern stress field conditions; 2) rock mechanics plane attributes such as Poisson's ratio and Young's modulus of the target layer (data comes from the attribute body of prestack inversion); 3) fracturing construction time and bottom hole pressure during construction. The material point method (MPM) is used for fracturing simulation to obtain Figure 7 Fracturing simulation results of well area A in X work area.

[0110] From the fracturing simulation results of Well A ( Figure 7 ) and microseismic monitoring results ( Figure 8) It can be seen that the overall morphology of the two is similar, which generally reflects that there are fewer microseismic events at the front fracturing level; the middle fracturing level is affected by the slightly longer segment spacing, and the microseismic events are slightly sparse; the terminal fracturing level has more microseismic events, and at the same time, there are more microseismic events on the east side than on the west side, and the extension distance is longer. Through comparative analysis, the degree of consistency between the fracturing simulation results and the microseismic monitoring results is high, which verifies the reliability of the method of the present invention.

[0111] The present application provides a method for natural fracture modeling and fracturing simulation using seismic data. Due to the complexity of favorable reservoir development and irregular distribution of shale natural fracture oil and gas reservoirs, the application of any single technology to predict and describe the distribution law of fractures has great uncertainty and multiple solutions. The present invention comprehensively utilizes geological, well logging, seismic pre-stack, post-stack attributes and other data to establish a fracture development distribution model under the comprehensive optimization of artificial intelligence technology, and finely characterizes the fracture distribution and distribution characteristics in unconventional shale gas reservoirs. On this basis, based on the natural fracture model, the material point method is used to simulate the interaction between the fracture and the natural fracture during the fracturing process, and artificial multi-stage fracturing simulation is carried out to predict the fracturing effect of shale gas horizontal wells, thereby improving economic benefits.

[0112] Embodiment 4

[0113] In this embodiment, Figure 2 As shown, a natural fracture modeling and fracturing simulation device is provided, comprising:

[0114] A grid model building module 210 is used to build a three-dimensional geological grid model of a preset work area;

[0115] The attribute import module 220 is used to calculate the post-stack attributes and pre-stack inversion attributes reflecting the faults and natural fractures based on the seismic data of the preset work area, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model;

[0116] The natural fracture model acquisition module 230 is used to predict the spatial distribution of reservoir attributes based on the three-dimensional geological grid model and the roughened curve information on the well by using a neural network algorithm to obtain the natural fracture model of the preset work area;

[0117] An equivalent natural fracture information generating module 240 is used to extract the average fracture surface density attribute of the target layer based on the natural fracture model to generate equivalent natural fracture information;

[0118] The fracturing simulation module is used to use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and perform fracturing simulation according to the natural fracture parameters.

[0119] In one embodiment, the natural fracture model acquisition module is also used to, based on the three-dimensional geological grid model and based on the roughened curve information on the well, adopt the neural network algorithm, use the well fracture information as a learning sample for training, predict the distribution of fractures in space, and obtain the natural fracture model of the preset work area.

[0120] In one embodiment, the spatial distribution of the reservoir attributes includes at least one of the following attributes: lithology, reservoir porosity, permeability, and fracture density.

[0121] In one embodiment, the natural fracture model acquisition module includes:

[0122] A prediction data acquisition unit is used to predict the distribution of fractures in space based on the three-dimensional geological grid model and the roughened curve information on the well, adopt the neural network algorithm, and use the fracture information on the well as a learning sample for training, so as to obtain a plurality of prediction data of the distribution of fractures in space;

[0123] An actual data acquisition unit, used to acquire actual data of the distribution of cracks in space;

[0124] A data extraction unit, used for comparing the plurality of predicted data with the actual data, and extracting the preset data whose deviation from the actual data is within a preset range;

[0125] A natural fracture model generating unit is used to generate a natural fracture model of the preset work area based on the extracted preset data.

[0126] In one embodiment, the grid model building module includes:

[0127] A coordinate value acquisition unit, used to acquire a first horizontal axis coordinate value, a second horizontal axis coordinate value and a depth coordinate value of the preset work area;

[0128] The three-dimensional geological grid model establishing unit is used to establish the three-dimensional geological grid model of the preset work area based on the first horizontal axis coordinate value, the second horizontal axis coordinate value and the depth coordinate value of the preset work area.

[0129] In one embodiment, the fracturing simulation module includes:

[0130] A simulation parameter acquisition unit is used to obtain modern stress field conditions, target rock mechanical plane properties, fracturing construction time and bottom hole pressure during fracturing construction;

[0131] A fracturing simulation unit is used to use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and to perform fracturing simulation according to the natural fracture parameters, the modern stress field conditions, the target layer rock mechanical plane properties, the fracturing construction time and the bottom hole pressure during the fracturing construction.

[0132] In one embodiment, the fracturing simulation module is further used to use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and to perform fracturing simulation according to the natural fracture parameters using a material point method.

[0133] For the specific definition of the natural fracture modeling and fracturing simulation device, please refer to the definition of the natural fracture modeling and fracturing simulation method above, which will not be repeated here. Each unit in the above-mentioned natural fracture modeling and fracturing simulation device can be implemented in whole or in part by software, hardware and their combination. The above-mentioned units can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned units.

[0134] Embodiment 5

[0135] In this embodiment, a computer device is provided. Its internal structure diagram can be shown as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program, and the non-volatile storage medium is deployed with a database, which is used to store seismic data and natural fracture models. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with other computer devices that deploy application software. When the computer program is executed by the processor, a natural fracture modeling and fracturing simulation method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0136] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0137] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0138] Step 110: Establish a three-dimensional geological grid model of the preset work area.

[0139] Specifically, the preset work area is a work area where natural fracture modeling and fracturing simulation need to be performed. In the following embodiment, the preset work area is the X work area.

[0140] In this step, a three-dimensional geological grid model is established using the collected seismic horizon interpretation results. The seismic horizon interpretation results are the interpretation results of the horizon and fault in the study area. For example, the horizon data and fault data of the study area are obtained, and the three-dimensional geological data of the preset work area are obtained. Based on the three-dimensional geological data of the preset work area, a three-dimensional geological grid model of the preset work area is established.

[0141] Step 120, based on the seismic data of the preset work area, calculate and obtain post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model.

[0142] Specifically, seismic data of a preset work area is obtained, and the seismic data is the seismic data of the preset work area. Post-stack attributes reflecting faults and natural fractures and pre-stack inversion attributes are calculated based on the seismic data of the preset work area.

[0143] In this embodiment, the post-stack attributes of natural fractures include: seismic curvature attributes and similarity attributes of natural fractures, etc.; the pre-stack inversion attributes include: longitudinal wave impedance, transverse wave impedance, density, porosity, Poisson's ratio, Young's modulus, etc.

[0144] In this embodiment, the post-stack attributes reflecting faults and natural fractures can be obtained by at least one of the following methods: coherence volume technology, curvature technology, ant volume technology, and discontinuity detection technology. It should be understood that the method of obtaining the post-stack attributes reflecting faults and natural fractures can also adopt other methods in the prior art, which are not described in this embodiment. In addition, in this application, the pre-stack inversion class attributes are obtained by the pre-stack inversion method.

[0145] Step 130, based on the three-dimensional geological grid model and based on the roughened curve information on the well, a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area.

[0146] In one embodiment, the spatial distribution of the reservoir attributes includes at least one of the following attributes: lithology, reservoir porosity, permeability, and fracture density.

[0147] In this step, the roughened curve information of the well can be first obtained, and based on the constructed three-dimensional geological grid model into which the post-stack attributes and the pre-stack inversion attributes have been imported, the spatial distribution of reservoir attributes can be predicted using a neural network algorithm based on the roughened curve information of the well, and the natural fracture model of the preset work area can be obtained. Specifically, the predicted spatial distribution of reservoir attributes includes lithology, reservoir porosity, permeability, fracture density, etc.

[0148] Specifically, different types of seismic attributes describe the possible development areas of natural fractures from different perspectives. A neural network method is used to comprehensively consider multiple seismic attributes and use wellbore fracture information as learning samples to predict the distribution of fractures in space.

[0149] It should be understood that in this step, the spatial distribution of reservoir properties can also be predicted by geostatistical methods to obtain the natural fracture model of the preset work area.

[0150] In this step, based on the three-dimensional geological grid model and the roughened curve information on the well, a geostatistical method or a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area, wherein the geostatistical method is the Kriging method.

[0151] Step 140: based on the natural fracture model, extract the average fracture surface density attribute of the target layer to generate equivalent natural fracture information.

[0152] Step 150: Using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and performing fracturing simulation according to the natural fracture parameters.

[0153] In this step, by simulating the modern stress field, the order of pressurization of the fracturing stages can be simulated according to the actual or designed simulation, and the estimated J integral on each fracturing stage can be simulated. The J integral result can show that the higher the energy at the fracturing stage, the more microseismic events there are. The J integral result can serve as a bridge between microseismicity and geomechanics. Through this application, completion engineers can predict which fracturing stage will show the highest or lowest microseismic activity during fracturing design.

[0154] In the above embodiment, by utilizing the pre-stack attributes and post-stack attributes of seismic data, combined with geological, lithological and logging information, a fracture development distribution model, i.e. a natural fracture model, is established through a neural network algorithm, thereby reducing the multi-solution nature of fracture reservoir prediction.

[0155] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0156] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is adopted to train with the well fracture information as learning samples to predict the distribution of fractures in space and obtain the natural fracture model of the preset work area.

[0157] In this embodiment, a neural network algorithm is first used to train the wellbore fracture information as a learning sample to obtain a natural fracture model of the preset work area.

[0158] Specifically, based on the structural model, the neural network method is used to predict the spatial distribution of reservoir attributes, including lithology, reservoir porosity, permeability, fracture density, etc., based on the roughened curve information on the well. It should be understood that different seismic attributes reflect fractures from different angles. The neural network method can comprehensively consider multiple seismic attributes and use the fracture information on the well as a learning sample to predict the distribution of fractures in space. Through a variety of quality control methods, such as setting the learning training data ratio, setting the correlation threshold value of training data and test data, and observing the intersection diagram of calculated data and actual data, it is ensured that the prediction results of the neural network algorithm are consistent with the known data, so as to obtain a parameter distribution that is closer to the actual. At the same time, the neural network algorithm is a random algorithm. In order to reduce randomness, 10 realizations are calculated each time, and only 1-2 realizations are saved. More than 30 realizations need to be saved cumulatively. Multiple results are statistically analyzed, and the average of 30 realizations is taken as the average natural fracture model to obtain the natural fracture modeling results of the X work area.

[0159] In this embodiment, for the characteristics of the fracture indication curve, the various seismic attributes extracted based on the seismic data are classified into categories: amplitude attributes (frequency division results of different frequencies based on the matching pursuit algorithm, amplitude envelope and original seismic amplitude), frequency attributes (instantaneous attributes and frequency division derived attributes), and structural attributes, and fuzzy logic sorting is performed respectively. Preferably, the 1-2 most sensitive seismic attributes in each type of seismic attribute to the fracture surface density curve are used together for random modeling using the neural network algorithm.

[0160] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0161] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is adopted, and the well fracture information is used as a learning sample for training to predict the distribution of fractures in space and obtain a plurality of predicted data of the distribution of fractures in space; the actual data of the distribution of fractures in space are obtained; the plurality of predicted data are compared with the actual data, and the preset data whose deviation from the actual data is within a preset range is extracted; based on the extracted preset data, a natural fracture model of the preset work area is generated.

[0162] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0163] Acquire a first horizontal axis coordinate value, a second horizontal axis coordinate value, and a depth coordinate value of the preset work area; and establish the three-dimensional geological grid model of the preset work area based on the first horizontal axis coordinate value, the second horizontal axis coordinate value, and the depth coordinate value of the preset work area.

[0164] In this embodiment, the stratum data and fault data of the study area are obtained to obtain the three-dimensional geological data of the preset work area, which are the first horizontal axis coordinate value, the second horizontal axis coordinate value and the depth coordinate value. In this embodiment, the first horizontal axis coordinate value is the X-axis value, the second horizontal axis coordinate value is the Y-axis value, and the depth coordinate value is the vertical axis value, that is, the Z-axis value, wherein the first horizontal axis coordinate value and the second horizontal axis coordinate value are the values ​​of the coordinate axis along the horizontal direction. The three-dimensional geological grid model of the preset work area can be established through the above-mentioned X-axis value, Y-axis value and depth Z-axis value.

[0165] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0166] Acquire modern stress field conditions, target layer rock mechanical plane properties, fracturing construction time and bottom hole pressure during fracturing construction; use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and perform fracturing simulation based on the natural fracture parameters, the modern stress field conditions, the target layer rock mechanical plane properties, the fracturing construction time and the bottom hole pressure during fracturing construction.

[0167] In this embodiment, based on the natural fracture model, the average fracture surface density attribute of the target layer is extracted to generate equivalent natural fracture information as the natural fracture parameters of the differential stress field simulation, and the following parameters are pre-set: 1) modern stress field conditions; 2) rock mechanics plane attributes such as Poisson's ratio and Young's modulus of the target layer; 3) fracturing construction time and bottom hole pressure during construction. The material point method (MPM) is used to perform fracturing simulation through the natural fracture parameters simulated by the differential stress field and the above pre-set parameters. Among them, the rock mechanics plane attributes include Poisson's ratio and Young's modulus of the target layer, and the data of the rock mechanics plane attributes come from the attribute body of pre-stack inversion.

[0168] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0169] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and a material point method is used to perform fracturing simulation according to the natural fracture parameters.

[0170] In this embodiment, the material point method (MPM) is used to solve the coupling between fractures (fault zones) and stress under modern stress field conditions to obtain the differential stress field properties. At the same time, it can also simulate the strain of the surrounding formations during multi-stage fracturing under the pressure and duration of each fracturing stage of the horizontal well, and describe the possible fracturing transformation range under the influence of the local stress field.

[0171] Specifically, the material point method (MPM) belongs to the mechanics solution of momentum equation, belongs to the meshless algorithm, and is the third generation of mechanics solution of momentum equation. When it comes to extremely large deformation and crushing of materials, the meshless material point method uses material points that carry all the information of the material to discretize the material region to characterize the motion and deformation state of the material region, and avoids the processing of convection terms. The material point method uses a regular Euler background grid to calculate the spatial derivatives and momentum equations, thereby realizing the interaction and connection between particles and avoiding the problem of grid distortion. Based on the natural fracture density model, the average fracture surface density attributes of the target layer are extracted to generate equivalent natural fracture information as the natural fracture parameters for MPM differential stress field simulation.

[0172] By simulating the modern stress field, the sequence of pressurization of the fracturing stages can be simulated according to the actual or designed simulation, and the estimated J integral on each fracturing stage can be simulated. The J integral results can show that the higher the energy at the fracturing stage, the more microseismic events there are. The J integral results can serve as a bridge between microseismicity and geomechanics. Through this application, completion engineers can predict which fracturing stage will show the highest or lowest microseismic activity during fracturing design.

[0173] Embodiment 6

[0174] In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0175] Step 110, establishing a three-dimensional geological grid model of a preset work area.

[0176] Specifically, the preset work area is a work area where natural fracture modeling and fracturing simulation need to be performed. In the following embodiment, the preset work area is the X work area.

[0177] In this step, a three-dimensional geological grid model is established using the collected seismic horizon interpretation results. The seismic horizon interpretation results are the interpretation results of the horizon and fault in the study area. For example, the horizon data and fault data of the study area are obtained, and the three-dimensional geological data of the preset work area are obtained. Based on the three-dimensional geological data of the preset work area, a three-dimensional geological grid model of the preset work area is established.

[0178] Step 120, based on the seismic data of the preset work area, calculate and obtain post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model.

[0179] Specifically, seismic data of a preset work area is obtained, and the seismic data is the seismic data of the preset work area. Post-stack attributes reflecting faults and natural fractures and pre-stack inversion attributes are calculated based on the seismic data of the preset work area.

[0180] In this embodiment, the post-stack attributes of natural fractures include: seismic curvature attributes and similarity attributes of natural fractures, etc.; the pre-stack inversion attributes include: longitudinal wave impedance, transverse wave impedance, density, porosity, Poisson's ratio, Young's modulus, etc.

[0181] In this embodiment, the post-stack attributes reflecting faults and natural fractures can be obtained by at least one of the following methods: coherence volume technology, curvature technology, ant volume technology, and discontinuity detection technology. It should be understood that the method of obtaining the post-stack attributes reflecting faults and natural fractures can also adopt other methods in the prior art, which are not described in this embodiment. In addition, in this application, the pre-stack inversion class attributes are obtained by the pre-stack inversion method.

[0182] Step 130, based on the three-dimensional geological grid model and based on the roughened curve information on the well, a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area.

[0183] In one embodiment, the spatial distribution of the reservoir attributes includes at least one of the following attributes: lithology, reservoir porosity, permeability, and fracture density.

[0184] In this step, the roughened curve information of the well can be first obtained, and based on the constructed three-dimensional geological grid model into which the post-stack attributes and the pre-stack inversion attributes have been imported, the spatial distribution of reservoir attributes can be predicted using a neural network algorithm based on the roughened curve information of the well, and the natural fracture model of the preset work area can be obtained. Specifically, the predicted spatial distribution of reservoir attributes includes lithology, reservoir porosity, permeability, fracture density, etc.

[0185] Specifically, different types of seismic attributes describe the possible development areas of natural fractures from different perspectives. A neural network method is used to comprehensively consider multiple seismic attributes and use wellbore fracture information as learning samples to predict the distribution of fractures in space.

[0186] It should be understood that in this step, the spatial distribution of reservoir properties can also be predicted by geostatistical methods to obtain the natural fracture model of the preset work area.

[0187] In this step, based on the three-dimensional geological grid model and the roughened curve information on the well, a geostatistical method or a neural network algorithm is used to predict the spatial distribution of reservoir properties to obtain a natural fracture model of the preset work area, wherein the geostatistical method is the Kriging method.

[0188] Step 140: based on the natural fracture model, extract the average fracture surface density attribute of the target layer to generate equivalent natural fracture information.

[0189] Step 150: Using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and performing fracturing simulation according to the natural fracture parameters.

[0190] In this step, by simulating the modern stress field, the order of pressurization of the fracturing stages can be simulated according to the actual or designed simulation, and the estimated J integral on each fracturing stage can be simulated. The J integral result can show that the higher the energy at the fracturing stage, the more microseismic events there are. The J integral result can serve as a bridge between microseismicity and geomechanics. Through this application, completion engineers can predict which fracturing stage will show the highest or lowest microseismic activity during fracturing design.

[0191] In the above embodiment, by utilizing the pre-stack attributes and post-stack attributes of seismic data, combined with geological, lithological and logging information, a fracture development distribution model, i.e. a natural fracture model, is established through a neural network algorithm, thereby reducing the multi-solution nature of fracture reservoir prediction.

[0192] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0193] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is adopted to train with the well fracture information as learning samples to predict the distribution of fractures in space and obtain the natural fracture model of the preset work area.

[0194] In this embodiment, a neural network algorithm is first used to train the wellbore fracture information as a learning sample to obtain a natural fracture model of the preset work area.

[0195] Specifically, based on the structural model, the neural network method is used to predict the spatial distribution of reservoir attributes, including lithology, reservoir porosity, permeability, fracture density, etc., based on the roughened curve information on the well. It should be understood that different seismic attributes reflect fractures from different angles. The neural network method can comprehensively consider multiple seismic attributes and use the fracture information on the well as a learning sample to predict the distribution of fractures in space. Through a variety of quality control methods, such as setting the learning training data ratio, setting the correlation threshold value of training data and test data, and observing the intersection diagram of calculated data and actual data, it is ensured that the prediction results of the neural network algorithm are consistent with the known data, so as to obtain a parameter distribution that is closer to the actual. At the same time, the neural network algorithm is a random algorithm. In order to reduce randomness, 10 realizations are calculated each time, and only 1-2 realizations are saved. More than 30 realizations need to be saved cumulatively. Multiple results are statistically analyzed, and the average of 30 realizations is taken as the average natural fracture model to obtain the natural fracture modeling results of the X work area.

[0196] In this embodiment, for the characteristics of the fracture indication curve, the various seismic attributes extracted based on the seismic data are classified into categories: amplitude attributes (frequency division results of different frequencies based on the matching pursuit algorithm, amplitude envelope and original seismic amplitude), frequency attributes (instantaneous attributes and frequency division derived attributes), and structural attributes, and fuzzy logic sorting is performed respectively. Preferably, the 1-2 most sensitive seismic attributes in each type of seismic attribute to the fracture surface density curve are used together for random modeling using the neural network algorithm.

[0197] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0198] On the basis of the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is adopted, and the well fracture information is used as a learning sample for training to predict the distribution of fractures in space and obtain a plurality of predicted data of the distribution of fractures in space; the actual data of the distribution of fractures in space are obtained; the plurality of predicted data are compared with the actual data, and the preset data whose deviation from the actual data is within a preset range is extracted; based on the extracted preset data, a natural fracture model of the preset work area is generated.

[0199] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0200] Acquire a first horizontal axis coordinate value, a second horizontal axis coordinate value, and a depth coordinate value of the preset work area; and establish the three-dimensional geological grid model of the preset work area based on the first horizontal axis coordinate value, the second horizontal axis coordinate value, and the depth coordinate value of the preset work area.

[0201] In this embodiment, the stratum data and fault data of the study area are obtained to obtain the three-dimensional geological data of the preset work area, which are the first horizontal axis coordinate value, the second horizontal axis coordinate value and the depth coordinate value. In this embodiment, the first horizontal axis coordinate value is the X-axis value, the second horizontal axis coordinate value is the Y-axis value, and the depth coordinate value is the vertical axis value, that is, the Z-axis value, wherein the first horizontal axis coordinate value and the second horizontal axis coordinate value are the values ​​of the coordinate axis along the horizontal direction. The three-dimensional geological grid model of the preset work area can be established through the above-mentioned X-axis value, Y-axis value and depth Z-axis value.

[0202] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0203] Acquire modern stress field conditions, target layer rock mechanical plane properties, fracturing construction time and bottom hole pressure during fracturing construction; use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation, and perform fracturing simulation based on the natural fracture parameters, the modern stress field conditions, the target layer rock mechanical plane properties, the fracturing construction time and the bottom hole pressure during fracturing construction.

[0204] In this embodiment, based on the natural fracture model, the average fracture surface density attribute of the target layer is extracted to generate equivalent natural fracture information as the natural fracture parameters of the differential stress field simulation, and the following parameters are pre-set: 1) modern stress field conditions; 2) rock mechanics plane attributes such as Poisson's ratio and Young's modulus of the target layer; 3) fracturing construction time and bottom hole pressure during construction. The material point method (MPM) is used to perform fracturing simulation through the natural fracture parameters simulated by the differential stress field and the above pre-set parameters. Among them, the rock mechanics plane attributes include Poisson's ratio and Young's modulus of the target layer, and the data of the rock mechanics plane attributes come from the attribute body of pre-stack inversion.

[0205] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0206] The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and a material point method is used to perform fracturing simulation according to the natural fracture parameters.

[0207] In this embodiment, the material point method (MPM) is used to solve the coupling between fractures (fault zones) and stress under modern stress field conditions to obtain the differential stress field properties. At the same time, it can also simulate the strain of the surrounding formations during multi-stage fracturing under the pressure and duration of each fracturing stage of the horizontal well, and describe the possible fracturing transformation range under the influence of the local stress field.

[0208] Specifically, the material point method (MPM) belongs to the mechanics solution of momentum equation, belongs to the meshless algorithm, and is the third generation of mechanics solution of momentum equation. When it comes to extremely large deformation and crushing of materials, the meshless material point method uses material points that carry all the information of the material to discretize the material region to characterize the motion and deformation state of the material region, and avoids the processing of convection terms. The material point method uses a regular Euler background grid to calculate the spatial derivatives and momentum equations, thereby realizing the interaction and connection between particles and avoiding the problem of grid distortion. Based on the natural fracture density model, the average fracture surface density attributes of the target layer are extracted to generate equivalent natural fracture information as the natural fracture parameters for MPM differential stress field simulation.

[0209] By simulating the modern stress field, the sequence of pressurization of the fracturing stages can be simulated according to the actual or designed simulation, and the estimated J integral on each fracturing stage can be simulated. The J integral results can show that the higher the energy at the fracturing stage, the more microseismic events there are. The J integral results can serve as a bridge between microseismicity and geomechanics. Through this application, completion engineers can predict which fracturing stage will show the highest or lowest microseismic activity during fracturing design.

[0210] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0211] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0212] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A natural fracture modeling and fracturing simulation method, characterized in that: include: Establish a three-dimensional geological grid model of the preset work area; Based on the seismic data of the preset work area, calculate and obtain post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures, and import the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model; Based on the three-dimensional geological grid model, based on the roughened curve information on the well, a neural network algorithm is used to predict the spatial distribution of reservoir attributes, and a natural fracture model of the preset work area is obtained; including: based on the three-dimensional geological grid model, based on the roughened curve information on the well, the neural network algorithm is used to train with the well fracture information as a learning sample, predict the distribution of fractures in space, and obtain multiple predicted data of the distribution of fractures in space; obtain actual data of the distribution of fractures in space; compare the multiple predicted data with the actual data, and extract the preset data whose deviation from the actual data is within a preset range; based on the extracted preset data, generate a natural fracture model of the preset work area; Based on the natural fracture model, extract the average fracture surface density attribute of the target layer to generate equivalent natural fracture information; The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and fracturing simulation is performed according to the natural fracture parameters.

2. The method according to claim 1, characterized in that The spatial distribution of the reservoir attributes includes at least one of the following attributes: lithology, reservoir porosity, permeability, and fracture density.

3. The method according to claim 1, characterized in that The step of establishing a three-dimensional geological grid model of a preset work area comprises: Obtaining a first horizontal axis coordinate value, a second horizontal axis coordinate value, and a depth coordinate value of the preset work area; The three-dimensional geological grid model of the preset work area is established based on the first horizontal axis coordinate value, the second horizontal axis coordinate value and the depth coordinate value of the preset work area.

4. The method according to claim 1, characterized in that: The step of using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and performing fracturing simulation according to the natural fracture parameters comprises: Obtain the modern stress field conditions, the plane mechanical properties of the target layer rock, the fracturing construction time and the bottom hole pressure during fracturing construction; The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and fracturing simulation is performed according to the natural fracture parameters, the modern stress field conditions, the target layer rock mechanical plane properties, the fracturing construction time and the bottom hole pressure during the fracturing construction.

5. The method according to any one of claims 1 to 4, characterized in that: The step of using the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and performing fracturing simulation according to the natural fracture parameters comprises: The equivalent natural fracture information is used as the natural fracture parameters for differential stress field simulation, and a material point method is used to perform fracturing simulation according to the natural fracture parameters.

6. A natural fracture modeling and fracturing simulation device, characterized in that: include: The grid model building module is used to build a three-dimensional geological grid model of the preset work area; An attribute import module, used for calculating and obtaining post-stack attributes and pre-stack inversion attributes reflecting faults and natural fractures based on the seismic data of the preset work area, and importing the post-stack attributes and the pre-stack inversion attributes into the three-dimensional geological grid model; A natural fracture model acquisition module is used to predict the spatial distribution of reservoir attributes based on the three-dimensional geological grid model and the roughened curve information on the well using a neural network algorithm to obtain the natural fracture model of the preset work area; including: based on the three-dimensional geological grid model, based on the roughened curve information on the well, using the neural network algorithm, using the well fracture information as a learning sample for training, predicting the spatial distribution of fractures, and obtaining a plurality of predicted data of the spatial distribution of fractures; obtaining actual data of the spatial distribution of fractures; comparing a plurality of the predicted data with the actual data, extracting the preset data whose deviation from the actual data is within a preset range; generating the natural fracture model of the preset work area based on the extracted preset data; An equivalent natural fracture information generation module is used to extract the average fracture surface density attribute of the target layer based on the natural fracture model to generate equivalent natural fracture information; The fracturing simulation module is used to use the equivalent natural fracture information as natural fracture parameters for differential stress field simulation and perform fracturing simulation according to the natural fracture parameters.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images