Method and system for establishing natural fracture model of tight sandstone oil and gas reservoir

Abstract

This invention provides a method and system for establishing a natural fracture model in tight sandstone oil and gas reservoirs, belonging to the field of oil and gas exploration and development technology. The method includes: obtaining a geological information fracture development probability model based on fault distance fracture development probability models, bedding plane curvature fracture development probability models, and lithological component fracture development probability models; obtaining a geophysical information fracture development probability model based on three-dimensional fault zone fracture development probability models, three-dimensional fold zone fracture development probability models, and three-dimensional rock-controlled fracture development probability models; obtaining a fracture development probability model integrating geological and geophysical information; and obtaining a discrete fracture network model and a fracture attribute parameter distribution model based on drilling data and the integrated geological and geophysical information fracture development probability model. The discrete fracture network model and fracture attribute parameter distribution model more accurately characterize the spatial distribution of natural fractures.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration and development technology, specifically to a method for establishing a natural fracture model of a tight sandstone oil and gas reservoir, a system for establishing a natural fracture model of a tight sandstone oil and gas reservoir, a machine-readable storage medium, and an electronic device. Background Technology

[0002] Tight sandstone reservoirs develop multi-level pore-throat and fracture systems. Fractures can significantly improve reservoir permeability and provide a small amount of effective reservoir space, making them a crucial factor controlling hydrocarbon accumulation and production. Therefore, the detailed characterization and modeling of tight sandstone oil and gas reservoirs is a vital topic in the oil and gas field now and in the future. Fracture geological modeling is a three-dimensional quantitative model that reflects the characterization parameters and spatial distribution of fractures in fractured reservoirs. A high-quality fracture model needs to accurately reflect the distribution characteristics of fractures in the study area while also meeting the needs of reservoir engineering research.

[0003] Currently, the characterization and modeling of fractures in fractured oil and gas reservoirs are still in their early stages, and research focuses on describing and predicting fractures based on outcrops, drilling, and seismic data. Internationally, methods for modeling fractured reservoirs can generally be divided into two main categories: equivalent continuous fracture modeling methods and discrete network fracture modeling methods.

[0004] The equivalent continuity fracture modeling method constructs an equivalent medium fracture model that uses simplified fracture descriptions (such as isotropic, parallel plate-like fractures) to replace the actual fracture geometry and seepage characteristics. This includes dual-medium models, equivalent permeability models, and pipe network models, with the dual-medium model being the most representative. While this equivalent medium fracture model does not treat fractures separately, making it easy to simulate reservoirs, it struggles to accurately describe actual flow characteristics and cannot resolve the scale issue of data from different sources, leading to the loss of many realistic fracture details.

[0005] The Discrete Crack Network (DFN) model, constructed using the Discrete Network Crack Modeling Method, is an improvement upon the continuous dual-medium model. It represents each crack by displaying crack segments with different shapes, sizes, orientations, and inclinations in three-dimensional space. Multiple crack segments with consistent characteristics form crack groups, and multiple crack groups constitute a crack system. The DFN model is rooted in stochastic simulation, and the establishment of each crack follows these rules: the shape of the crack segment is a convex polygon (rectangular, elliptical, or more complex); the size of the crack segment conforms to a known distribution (such as a negative exponential distribution); the crack position follows a spatial distribution function; and the crack orientation is obtained by extracting a uniform or Fisher distribution. Essentially, this simulation method is goal-oriented, iteratively refining the final crack distribution to conform to given statistical characteristics.

[0006] Before modeling fractures, it is crucial to have a clear understanding of their developmental extent. Larger-scale fractures are mostly tectonic fractures, and their development is primarily controlled by geostress and rock properties. The former is related to the tectonic location and stratigraphic thickness, while the latter is related to lithofacies and rock physical properties. Therefore, modeling large-scale fractures can be directly obtained through geomechanical analysis.

[0007] Small-scale fracture distribution is difficult to identify and detect using seismic methods alone. Fractures at this scale primarily rely on fracture interpretation parameters at well points, utilizing the understanding of fracture distribution probabilities as well-to-well constraints. A fracture DFN (Digital-Dependent Network) model is constructed to accurately characterize their spatial distribution. Modeling small-scale discrete fractures first requires preparing the necessary data, including fracture attitude statistics such as fracture series, strike, dip, length, and aperture. This data can usually be obtained during drilling using imaging logging and core observation. However, the distribution and development level of fractures at well locations are often difficult to determine. Therefore, it is necessary to obtain a simulation of the fracture driving force from the well point to the wellbore based on the analysis of fracture development control factors.

[0008] The main inter-well control factors for fracture modeling include: ① Fracture development is controlled by sedimentation (lithology, minerals). In this case, geological constraints can be used, such as relevant data from matrix models, like matrix facies models and matrix porosity models. ② Fracture development can be analyzed using seismic-related attributes, such as wave impedance and curvature. ③ Fracture development is controlled by tectonic factors. In this case, the relationship between faults or folds and fracture development can be established to further constrain fracture distribution. ④ Fracture development is controlled by geostress. In this case, geostress characteristics can be analyzed first, a stress model can be established, and then fracture distribution can be constrained. ⑤ In addition, fracture development may also be related to stratigraphic conditions, such as stratigraphic thickness and sand-mud interbedded conditions.

[0009] The development patterns and controlling factors of fractures at different scales in tight sandstone are diverse (faults, folds, lithology, etc.). Fracture seismic responses are weak, making fracture description and inter-well fracture prediction challenging. Further research is needed to develop geological modeling techniques for multi-scale coupled, multi-genetic fracture systems. Due to the complexity and strong spatial heterogeneity of fractures in tight reservoirs, a single method is insufficient for a comprehensive and accurate understanding of fractures. The key to fracture modeling lies in how to reasonably extend the degree of fracture development from known well points to inter-well areas. To objectively interpret the distribution characteristics of subsurface fractures, various data are needed to obtain the fracture types and characteristic parameters of the reservoir. This not only provides preliminary geological information constraints for 3D geological modeling but also requires the use of geological modeling as a quantitative method to integrate information from multiple disciplines and predict the spatial distribution of fractures.

[0010] Therefore, accurately characterizing the spatial distribution of various parameters of natural fractures and providing a more reliable quantitative geological model basis for the efficient development of fractured oil and gas reservoirs such as tight sandstone is an urgent problem to be solved. Summary of the Invention

[0011] The purpose of this invention is to provide a method and system for establishing a natural fracture model of tight sandstone oil and gas reservoirs, so as to at least solve the problem of failing to accurately characterize the spatial distribution of various parameters of natural fractures.

[0012] To achieve the above objectives, the first aspect of the present invention provides a method for establishing a natural fracture model of a tight sandstone oil and gas reservoir, comprising: establishing, based on the stratigraphic geological data of the target reservoir, a probability model of fracture development based on fault distance, a probability model of fracture development based on bedding plane curvature, a probability model of fracture development based on lithological components, a probability model of fracture development based on three-dimensional fault zone, a probability model of fracture development based on three-dimensional fold zone, and a probability model of fracture development based on three-dimensional rock-controlled fracture; and obtaining geological information fracture development data corresponding to the target reservoir based on the probability models of fracture development based on fault distance, bedding plane curvature, and lithological components. The probability model for fracture development is derived by using three-dimensional fracture development probability models of three-dimensional fault zones, three-dimensional fold zones, and three-dimensional rock-controlled fracture development to obtain a geophysical information fracture development probability model for the target reservoir. Based on the geological information fracture development probability model and the geophysical information fracture development probability model for the target reservoir, a comprehensive geological and geophysical information fracture development probability model for the target reservoir is obtained. Based on drilling data and the comprehensive geological and geophysical information fracture development probability model for the target reservoir, a discrete fracture network model and a fracture attribute parameter distribution model for the target reservoir are obtained.

[0013] Optionally, the stratigraphic geological data of the target reservoir includes bedding plane data, fracture data, and drilling and logging data. Before establishing the probability volume models of fracture development based on fault distance, bedding plane curvature, lithological composition, three-dimensional fracture zone, three-dimensional fold zone, and three-dimensional rock-controlled fracture, the method further includes: establishing an initial three-dimensional structural stratigraphic model based on the bedding plane data and fracture data of the target reservoir; performing gridding processing on the initial three-dimensional structural stratigraphic model based on a preset modeling area and a preset grid step size; determining the heterogeneous distribution characteristics of the reservoir matrix attribute parameters based on drilling and logging data; wherein the drilling and logging data includes at least drilling core data, logging interpretation data, and seismic attribute data, and the reservoir matrix attribute parameters include at least the discrete attribute parameters of matrix porosity, permeability, sedimentary facies, and lithofacies; and obtaining a three-dimensional structural stratigraphic model based on the heterogeneous distribution characteristics of the reservoir matrix attribute parameters and the gridded initial three-dimensional structural stratigraphic model.

[0014] Optionally, the rules for establishing the above-mentioned fault distance fracture development probability volume model include: based on the target reservoir's bedding plane data, fracture data, and three-dimensional structural stratigraphic model, selecting fault models in the three-dimensional structural stratigraphic model; based on each selected fault model, establishing a distance trend volume model corresponding to each fault model; wherein, the distance trend volume model represents the vertical distance from each grid of the three-dimensional structural stratigraphic model to the corresponding fault; based on the range of the fault-controlled fracture zone, the distance trend volume model is sequentially truncated and normalized to obtain the fault distance fracture development probability volume model.

[0015] Optionally, the rules for establishing the above-mentioned probabilistic volume model of layer curvature fracture development include: determining the two-dimensional map of layer curvature distribution corresponding to each layer based on the three-dimensional structural stratigraphic model; normalizing the layer curvature according to the curvature relationship of fold-controlled fractures to convert the two-dimensional map of layer curvature distribution corresponding to each layer into a two-dimensional map of fold-controlled fracture development intensity coefficient corresponding to each layer; and using the two-dimensional map of fold-controlled fracture development intensity coefficient corresponding to each layer as a constraint condition to obtain the probabilistic volume model of layer curvature fracture development.

[0016] Optionally, the rules for establishing the above-mentioned lithological component fracture development probability model include: establishing a quartz content model based on the lithofacies model; wherein, the lithofacies model is determined based on the three-dimensional tectonic stratigraphic model; based on the range of the fault-controlled fracture zone and the lower limit threshold of quartz content controlling fracture development, the quartz content model is truncated and normalized in sequence to obtain the lithological component fracture development probability model.

[0017] Optionally, the above-mentioned quartz content model based on the lithofacies model includes: statistically analyzing the vertical distribution ratio and multiple characteristic values ​​of quartz content in different lithofacies; wherein, the multiple characteristic values ​​of quartz content include the maximum value, minimum value, average value, variance, and / or coefficient of variation of quartz content; determining the quartz content variability function parameters based on the vertical distribution ratio and multiple characteristic values ​​of quartz content in different lithofacies; determining the spatial distribution data of quartz content in each lithofacies based on the lithofacies model; and obtaining the quartz content model using sequential Gaussian simulation based on the quartz content variability function parameters and the spatial distribution data of quartz content in each lithofacies.

[0018] Optionally, the rules for establishing the above-mentioned three-dimensional fracture development probability model of the fracture zone include: obtaining multiple first attribute volumes based on three-dimensional seismic data; wherein, the first attribute volume represents the attribute volume used to respond to the discontinuity characteristics of seismic wave reflection; comparing the spatial locations of each first attribute volume and fault data, determining the first attribute volume that best reflects fracture fragmentation as the fracture attribute volume; wherein, the spatial location of fault data is determined based on the bedding plane data, fault data, and three-dimensional structural stratigraphic model of the target reservoir; and after calibrating the maximum and minimum thresholds of the fracture attribute volume based on drilling data, normalizing the fracture attribute volume to obtain the three-dimensional fracture development probability model of the fracture zone.

[0019] Optionally, the above-mentioned calibration of the maximum and minimum thresholds of the fracture attribute body based on drilling data includes: calculating the interlayer average value of the fracture attribute body and obtaining the fracture interpretation density value at the well point with imaging logging; after performing cross-statistical analysis on the interlayer average value of the fracture attribute body and the fracture interpretation density value of imaging logging, the maximum and minimum values ​​of the interlayer average value of the fracture attribute body are extracted as the maximum and minimum thresholds of the fracture attribute body.

[0020] Optionally, the rules for establishing the above-mentioned three-dimensional fold zone fracture development probability model include: obtaining multiple second attribute bodies based on three-dimensional seismic data; wherein, the second attribute body characterizes the attribute body used to respond to the curvature characteristics of the rock strata reflected by seismic waves; comparing each second attribute body with the bedding morphology data, determining the second attribute body that best reflects the deformation intensity as the fold attribute body; wherein, the bedding morphology data is determined based on the bedding data, fracture data and three-dimensional structural stratigraphic model of the target reservoir; after normalizing the fold attribute body, calibrating the fold attribute body based on drilling data to obtain the three-dimensional fold zone fracture development probability model.

[0021] Optionally, the above-mentioned calibration of the fold attribute volume based on drilling data to obtain a three-dimensional fold zone fracture development probability volume model includes: calculating the interlayer average value of the fold attribute volume and obtaining the fracture interpretation density value at the well point with imaging logging; performing cross-sectional statistics on the interlayer average value of the fold attribute volume and the fracture interpretation density value of imaging logging, and fitting to obtain a correlation formula; and transforming the fold attribute volume based on the correlation formula to obtain a three-dimensional fold zone fracture development probability volume model.

[0022] Optionally, the rules for establishing the above-mentioned three-dimensional rock-controlled fracture development probability volume model include: calculating the wave impedance attribute volume based on three-dimensional seismic data; performing shale content inversion on the three-dimensional seismic data based on the wave impedance attribute volume and the correspondence between shale content and wave impedance data to obtain the shale content inversion attribute volume; wherein, the correspondence between shale content and wave impedance data is obtained from well logging interpretation data; and normalizing the shale content inversion attribute volume to obtain the three-dimensional rock-controlled fracture development probability volume model.

[0023] Optionally, the above-mentioned geological information fracture development probability model corresponding to the target reservoir, based on the fault distance fracture development probability model, bedding plane curvature fracture development probability model, and lithological component fracture development probability model corresponding to the target reservoir, includes: calculating the geological information fracture development probability model corresponding to the target reservoir using a weighted average method based on the fault distance fracture development probability model, bedding plane curvature fracture development probability model, lithological component fracture development probability model, and the geological influence ratio of fracture development; wherein, the geological influence ratio characterizes the quantitative degree of control of different geological factors on fracture development.

[0024] Optionally, the above-mentioned geological information fracture development probability model corresponding to the target reservoir is calculated using a weighted average method based on the fault distance fracture development probability volume model, bedding plane curvature fracture development probability volume model, lithological component fracture development probability volume model, and the geological influence ratio of fracture development. This includes: statistically analyzing the correlation between different geological factor parameters and fracture density to obtain the correlation coefficient between each geological factor parameter and fracture density; wherein, the geological factor parameters include the distance of fracture from the fault, folding coefficient, and quartz mineral content; calculating the weighted average of the correlation coefficients between different geological factor parameters and fracture density to obtain the influence ratio of each geological factor parameter on fracture development; calculating the product of the fault distance fracture development probability volume model, bedding plane curvature fracture development probability volume model, and lithological component fracture development probability volume model corresponding to the target reservoir with the influence ratio of the corresponding geological factor parameters on fracture development to obtain the contribution volume distribution model of each geological factor parameter on fracture development probability; and summing the contribution volumes of each geological factor parameter corresponding to fracture development probability in the same grid to obtain the geological information fracture development probability model corresponding to the target reservoir.

[0025] Optionally, the above-mentioned geophysical information fracture development probability model corresponding to the target reservoir, based on the three-dimensional fracture development probability model of the fracture zone, the three-dimensional fracture development probability model of the fold zone, and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir, includes: using the union method to perform model grid calculation based on the three-dimensional fracture development probability model of the fracture zone, the three-dimensional fracture development probability model of the fold zone, and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir, to obtain the geophysical information fracture development probability model corresponding to the target reservoir.

[0026] Optionally, the above-mentioned three-dimensional fracture development probability model of the target reservoir, based on the three-dimensional fracture development probability model of the fracture zone, the three-dimensional fracture development probability model of the fold zone, and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir, is obtained by performing model grid calculation using the union method. This includes: determining the number of fracture development probability data for each grid based on the three-dimensional fracture development probability model of the fracture zone, the three-dimensional fracture development probability model of the fold zone, and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir; for each grid, if the grid contains only one fracture development probability data, the fracture development probability data of that grid remains unchanged; if the grid contains at least two fracture development probability data, the fracture development probability data with the largest value is used as the fracture development probability data of that grid.

[0027] Optionally, the above-mentioned fracture development probability model based on the geological information fracture development probability model and geophysical information fracture development probability model corresponding to the target reservoir is obtained by using the Bayesian multivariate probability fusion method to perform model grid calculation based on the geological information fracture development probability model and geophysical information fracture development probability model corresponding to the target reservoir.

[0028] Optionally, the above-mentioned fracture development probability model based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir is obtained by performing model grid calculation using the Bayesian multivariate probabilistic fusion method. This includes: determining the number of fracture development probability data for each grid based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir; for each grid, if the grid contains only one fracture development probability data, the fracture development probability data of that grid remains unchanged; if the grid contains at least two fracture development probability data, the Bayesian multivariate probabilistic fusion method is used to fuse and calculate all the fracture development probability data of that grid to obtain the fracture development probability data of that grid.

[0029] Optionally, the aforementioned drilling data includes fracture occurrence data, fracture intensity data, and fracture aperture parameter distribution data; the aforementioned fracture development probability model based on drilling data and the comprehensive geological and geophysical information corresponding to the target reservoir, yielding the discrete fracture network model and fracture attribute parameter distribution model corresponding to the target reservoir, includes: a fracture development probability model based on the comprehensive geological and geophysical information corresponding to the target reservoir, using fracture occurrence data, fracture intensity data, and fracture aperture parameter distribution data as constraints, and employing the indicative point method to randomly simulate and obtain the discrete fracture network model and fracture attribute parameter distribution model corresponding to the target reservoir.

[0030] A second aspect of this invention provides a system for establishing a natural fracture model of a tight sandstone oil and gas reservoir, comprising: a model building module, used to establish, based on the stratigraphic geological data of the target reservoir, a fault distance fracture development probability model, a bedding plane curvature fracture development probability model, a lithological component fracture development probability model, a three-dimensional fault zone fracture development probability model, a three-dimensional fold zone fracture development probability model, and a three-dimensional rock-controlled fracture development probability model; and a model fusion module, used to obtain a geological information fracture development probability model corresponding to the target reservoir based on the fault distance fracture development probability model, the bedding plane curvature fracture development probability model, and the lithological component fracture development probability model. Based on the three-dimensional fracture development probability models of the target reservoir, the three-dimensional fracture development probability models of the fold zone, and the three-dimensional rock-controlled fracture development probability model, a geophysical information fracture development probability model corresponding to the target reservoir is obtained; the information integration module is used to obtain a comprehensive geological and geophysical information fracture development probability model corresponding to the target reservoir based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir; the fracture distribution determination module is used to obtain a discrete fracture network model and a fracture attribute parameter distribution model corresponding to the target reservoir based on drilling data and the comprehensive geological and geophysical information fracture development probability model corresponding to the target reservoir.

[0031] In a third aspect of the invention, a machine-readable storage medium is provided, on which instructions are stored, which, when executed by a processor, cause the processor to be configured to perform the above-described method for establishing a natural fracture model of a tight sandstone oil and gas reservoir.

[0032] In a fourth aspect of the present invention, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for establishing a natural fracture model of a tight sandstone oil and gas reservoir.

[0033] The above technical solution provides a method and system for establishing a natural fracture model for tight sandstone oil and gas reservoirs. Based on the stratigraphic geological data of the target reservoir, a fracture development probability model is established, reflecting the relationship between various geological factors (including faults, folds, and lithofacies) and fracture development probability. This model includes models for fault distance fracture development, bedding plane curvature fracture development, lithological composition fracture development, three-dimensional fault zone fracture development, three-dimensional fold zone fracture development, and three-dimensional rock-controlled fracture development. The model comprehensively considers the prediction of fracture development probability based on geological information to obtain a fracture probability model based on geological concepts (i.e., a geological information fracture development probability model). Simultaneously, it comprehensively considers the prediction of fracture development probability based on geophysical information to obtain a fracture probability model based on geophysical prediction (i.e., a geophysical information fracture development probability model). Finally, by combining the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir, a comprehensive geological and geophysical information fracture development probability model corresponding to the target reservoir is obtained. Combined with drilling data as conditional parameters, a discrete fracture network model and a fracture attribute parameter distribution model corresponding to the target reservoir are obtained. This method, through multi-level constraints and multi-probability fusion of faulting, folding, and lithological factors, achieves the construction of a fracture development probability model capable of supporting multi-source data fusion and multi-factor driven development. Furthermore, by using a discrete fracture network model and fracture attribute parameter distribution model corresponding to the target reservoir, it more accurately characterizes the spatial distribution of natural fractures, providing a more reliable geological model foundation for the efficient development of tight sandstone oil and gas reservoirs. This method and system fully consider the controlling factors of fracture development, fully explore and apply various fracture detection information such as geological and seismic data, forming a comprehensive geological model and geophysical prediction method for fracture model establishment. This more accurately characterizes the spatial distribution of various parameters of natural fractures, providing a more reliable quantitative geological model foundation for the efficient development of fractured oil and gas reservoirs such as tight sandstone.

[0034] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0035] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings:

[0036] Figure 1 This is a flowchart of a method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to one embodiment of the present invention;

[0037] Figure 2This is a block diagram of a system for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to one embodiment of the present invention;

[0038] Figure 3 This is a flowchart of another method for establishing a natural fracture model of a tight sandstone oil and gas reservoir provided by one embodiment of the present invention;

[0039] Figure 4 This is a detailed implementation roadmap of a method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to one embodiment of the present invention;

[0040] Figure 5 This is a fracture probability model based on fault distance constructed using a fault model in an experimental area, provided by one embodiment of the present invention.

[0041] Figure 6 This invention provides a crack probability model based on fold deformation, constructed using the curvature of the surface in the test area, according to one embodiment of the present invention.

[0042] Figure 7 This invention provides a crack probability model based on brittle minerals constructed using a lithofacies model in an experimental area, according to one embodiment of the present invention.

[0043] Figure 8 This is a crack probability model based on Fault Likelihood seismic attributes in a test area provided by one embodiment of the present invention;

[0044] Figure 9 This is a crack probability model based on the seismic properties of the fold coefficient in the test area provided by one embodiment of the present invention;

[0045] Figure 10 This invention provides a crack probability model for the test area based on seismic inversion of clay content, according to one embodiment of the present invention.

[0046] Figure 11 This invention provides a crack probability model for a test area based on geological information fusion, according to one embodiment of the present invention.

[0047] Figure 12 This is a crack probability model based on geophysical information fusion for the test area provided by one embodiment of the present invention;

[0048] Figure 13 This is a comprehensive fracture probability model that integrates geological and geophysical information of the test area, provided by one embodiment of the present invention.

[0049] Figure 14 This is a DFN model diagram of a multi-scale natural crack in the test area provided by one embodiment of the present invention;

[0050] Figure 15This is a model diagram of the properties of natural cracks in the test area provided by one embodiment of the present invention;

[0051] Figure 16 This is a comparison diagram of a new well and a dual-medium model in the test area provided by one embodiment of the present invention. Detailed Implementation

[0052] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0053] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0054] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0055] Example 1

[0056] Figure 1 This is a flowchart illustrating a method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to one embodiment of the present invention. Figure 1 As shown, this invention provides a method for establishing a natural fracture model of a tight sandstone oil and gas reservoir, comprising:

[0057] S110: Based on the stratigraphic geological data of the target reservoir, establish the following probability models for the development of fractures: fault distance, bedding curvature, lithological components, three-dimensional fault zone, three-dimensional fold zone, and three-dimensional rock-controlled fracture.

[0058] S120: Based on the fault distance fracture development probability model, the bedding curvature fracture development probability model and the lithological component fracture development probability model corresponding to the target reservoir, the geological information fracture development probability model corresponding to the target reservoir is obtained. Based on the three-dimensional fault zone fracture development probability model, the three-dimensional fold zone fracture development probability model and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir, the geophysical information fracture development probability model corresponding to the target reservoir is obtained.

[0059] S130: Based on the geological information fracture development probability model and geophysical information fracture development probability model corresponding to the target reservoir, the comprehensive geological information and geophysical information fracture development probability model corresponding to the target reservoir are obtained.

[0060] S140: Based on drilling data and comprehensive geological and geophysical information corresponding to the target reservoir, a fracture development probability model is obtained, resulting in a discrete fracture network model and a fracture attribute parameter distribution model corresponding to the target reservoir.

[0061] Specifically, the characterization and geological modeling of tight sandstone oil and gas reservoirs should focus on the detailed depiction and modeling of natural fractures. This method establishes a fracture development probability model (including fault distance fracture development probability model, bedding plane curvature fracture development probability model, lithological component fracture development probability model, three-dimensional fault zone fracture development probability model, three-dimensional fold zone fracture development probability model, and three-dimensional rock-controlled fracture development probability model) based on the stratigraphic geological data of the target reservoir, reflecting the relationship between various geological factors (including faults, folds, and lithofacies) and fracture development probability. It also comprehensively considers the prediction of fracture development probability by geological information to obtain a fracture probability model based on geological concepts (i.e., geological information fracture development probability model). At the same time, it comprehensively considers the prediction of fracture development probability by geophysical information to obtain a fracture probability model based on geophysical prediction (i.e., geophysical information fracture development probability model). Finally, it integrates the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir to obtain a comprehensive geological information and geophysical information fracture development probability model corresponding to the target reservoir. Combining drilling data as conditional parameters, it obtains a discrete fracture network model and a fracture attribute parameter distribution model corresponding to the target reservoir. This method, through multi-level constraints and multi-probability fusion of faulting, folding, and lithological factors, achieves the construction of a fracture development probability model capable of satisfying multi-source data fusion and multi-factor driven development. Furthermore, by using a discrete fracture network model and fracture attribute parameter distribution model corresponding to the target reservoir, it more accurately characterizes the spatial distribution of natural fractures, providing a more reliable geological model foundation for the efficient development of tight sandstone oil and gas reservoirs. This method fully considers the controlling factors of fracture development, fully explores and applies various fracture detection information such as geological and seismic data, forming a comprehensive geological model and geophysical prediction method for fracture model establishment. This more accurately characterizes the spatial distribution of various parameters of natural fractures, providing a more reliable quantitative geological model foundation for the efficient development of fractured oil and gas reservoirs such as tight sandstone.

[0062] In the above implementation process, this method considers the understanding of fracture control by faulting, folding, and lithofacies factors, improving the accuracy of the discrete fracture network model and fracture attribute parameter distribution model obtained in step S140. Furthermore, based on the dominant geological factors controlling fracture development, this method applies multiple seismic fracture attributes, more realistically reflecting the differences in fracture development across different tectonic regions. Through various fusion methods, this method effectively integrates the three geological factors—fracture, folding, and lithofacies—as well as multiple fracture development probabilities driven by geology and geophysics, resulting in a more accurate natural fracture model (i.e., the discrete fracture network model and fracture attribute parameter distribution model), providing a more reliable geological model guarantee for the quantitative study of tight sandstone oil and gas reservoirs.

[0063] Please refer to Figure 3 and Figure 4 , Figure 3 This is a flowchart of another method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to one embodiment of the present invention. Figure 4 This is a detailed implementation roadmap of a method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to one embodiment of the present invention. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir includes the following steps:

[0064] Step 1: Based on the bedding plane and fault data of the target reservoir obtained from the interpretation of 3D seismic data, establish an initial 3D structural stratigraphic model (i.e., a structural stratigraphic framework geological model). Define the modeling area and grid step size, perform gridding on the initial 3D structural stratigraphic model, and integrate data from drilling cores, well logging interpretation, and seismic attributes to clarify the heterogeneous distribution characteristics of the reservoir matrix attribute parameters, thus establishing a 3D structural stratigraphic model (i.e., a geological model with matrix reservoir attribute parameter information).

[0065] In step 1, the geological model of the stratigraphic framework should be established in accordance with the technical content of fault model and bedding plane model as specified in the industry standard (SY / T 7378). Specifically, fault data can be determined based on the bedding plane data and fault data of the target reservoir.

[0066] In step 1, the matrix reservoir property parameters include, but are not limited to, discrete property parameters such as sedimentary facies and lithofacies, as well as continuous property parameters such as matrix porosity and permeability. The specific implementation steps can be carried out with reference to the lithofacies model and reservoir property model technical content in standard SY / T 7378.

[0067] Step 2: Based on the three-dimensional structural stratigraphic model obtained in Step 1, fault model screening is performed, and a distance trend body model with fault as the target is established. According to the range of the fracture zone controlled by the fault, the distance trend body model is truncated and normalized according to the 0-1 interval to obtain the fracture development probability body geological model based on fault distance (i.e., fault distance fracture development probability body model), denoted as P(F1).

[0068] In step 2, fault model screening refers to retaining faults with large extension lengths and displacements, as well as faults that significantly impact both hydrocarbon accumulation and development, based on fault data from the target oil and gas reservoir. Normalization calculation means that the greater the vertical distance from a spatial grid to a nearby fault, the larger the corresponding value in the distance trend model, and the lower the probability of fracture development for that grid. Conversely, the closer the vertical distance from a spatial grid to a nearby fault, the smaller the corresponding value in the distance trend model, and the higher the probability of fracture development for that grid. Faults that significantly impact both hydrocarbon accumulation and development can be defined as those revealed by the dynamics of oil well production that have a significant influence on oil and gas development.

[0069] In step 2, the distance trend volume model reflects the vertical distance from a certain grid in space to a nearby fault.

[0070] In step 2, the extent of the fault-controlled fracture zone can be obtained from the statistical relationship between the number of fractures in the well and the distance from the well location to the fracture, or from the actual field outcrop measurement data.

[0071] In step 2, if the target oil and gas reservoir has faults of different sizes, properties and orientations, the range of the fracture zone controlled by the faults of different sizes, properties and orientations may be different. In this case, it is necessary to perform truncation and normalization calculations on the distance trend volume model corresponding to each fault.

[0072] Step 3: Based on the three-dimensional structural stratigraphic model obtained in Step 1, calculate the curvature data distribution of key layers. According to the curvature relationship of folds controlling fractures, normalize the layer curvature in the 0-1 interval and convert it into the fold-controlled fracture development intensity coefficient to obtain the fracture development probability model based on layer curvature (i.e., layer curvature fracture development probability volume model), denoted as P(F2).

[0073] In step 3, the curvature of the layer refers to the measure of the degree of folding of the three-dimensional structural layer. The greater the curvature, the more severe the folding and deformation of the layer, and the greater the possibility of rock strata fracturing, i.e., crack development.

[0074] In step 3, the curvature relationship of the fold-controlled fracture can be obtained from the statistical relationship between the number of fractures and the ratio of fold height and fold width in the drilling statistics, or from the actual field outcrop measurement data.

[0075] In step 3, if the vertical undulation of the target oil and gas reservoir varies greatly, the planar distribution of folds and curvatures in different layers may be different. In this case, it is necessary to calculate the curvature values ​​of each layer separately, and use the fold control fracture development intensity coefficient of each layer as a constraint to obtain a fracture development probability model based on the layer curvature.

[0076] Step 4: Based on the lithofacies model obtained in Step 1 as a constraint, establish a quartz content model. Based on the range of the fault-controlled fracture zone obtained in Step 2, and combined with the lower limit of quartz content controlling fracture development, truncate the quartz content model and normalize it according to the 0-1 interval to obtain a geological model of fracture development probability volume based on lithological components (i.e., lithological component fracture development probability volume model), denoted as P(F3).

[0077] In step 4, the specific steps for constructing the quartz content model are as follows: a) Statistically determine the characteristic values ​​of quartz content in different lithofacies (such as maximum, minimum, average, variance, coefficient of variation, etc.) and the data distribution such as the vertical distribution ratio; b) Analyze and obtain the quartz content variation function parameters; c) Use the lithofacies model to constrain the spatial data distribution of quartz content in each lithofacies, and use sequential Gaussian stochastic simulation to obtain a three-dimensional geological model of quartz content.

[0078] In step 4, the lower limit of quartz content controlling fracture development can be obtained from the statistical relationship between the number of fractures in the drilling statistics and the quartz content interpreted by the well logging, or it can be obtained from the actual field outcrop measurement data.

[0079] Step 5: Based on the 3D seismic data, extract various attribute volumes that can respond to the discontinuity characteristics of seismic wave reflection. Compare the matching relationship between each attribute volume and the spatial location of the fault data in Step 1, and select the attribute volume that best reflects the fracture as the fracture attribute volume. According to the fracture information obtained from actual drilling, the maximum and minimum thresholds of the attribute volume are calibrated, and normalized according to the 0-1 interval to form a 3D fracture zone fracture probability volume (i.e., a 3D fracture zone fracture development probability volume model), denoted as P(F1').

[0080] It should be noted that the attribute volume in step 5 is a data volume that reflects the discontinuities in underground rock strata, derived from the seismic wave reflection signal. Depending on the calculation method, there are many types of attribute volumes, such as variance volumes and coherence volumes. If there is a discontinuity in the underground rock strata, the data at that location will be anomaly. If the location, degree, and longitudinal extent of these anomalies match the spatial location of the fault data (the fault data determined in step 1), then the attribute volume is considered relatively accurate.

[0081] Three-dimensional seismic data can be obtained through actual underground exploration.

[0082] In step 5, the fracture attribute volume is a quantitative description of the degree of fracture and its associated crack development at each sample point in space.

[0083] In step 5, the specific implementation steps for threshold calibration of fracture attribute bodies are as follows: a) Calculate the interlayer average value of each fracture attribute body; b) Extract the attribute average value at the well point with imaging logging; c) Perform cross-statistical analysis on the fracture attribute average value at the well point and the fracture interpretation density value of imaging logging, and extract the maximum and minimum values ​​of the attribute bodies as the fracture probability body of the fracture zone of the oil and gas reservoir.

[0084] Step 6: Based on the 3D seismic data, extract various attribute volumes that can respond to the curvature characteristics of rock strata reflected by seismic waves. Compare the matching relationship between each attribute volume and the bending morphology of the layer data in Step 1, and select the attribute volume that best reflects the deformation intensity as the fold attribute volume. Normalize the attribute volume in the 0-1 interval. According to the fracture intensity obtained from actual drilling, calibrate the fold attribute volume to form a 3D fold fracture probability volume (3D fold fracture development probability volume model), denoted as P(F2').

[0085] It should be noted that the attribute volume in step 6 is a data volume reflecting the bending deformation of underground rock strata, obtained from the seismic wave reflection signal, such as the maximum curvature volume and the minimum curvature volume. If the underground rock strata are bent, the data at the bending points will show anomalies. The location and degree of these anomalies are compared with the bending morphology of the bedding data (the bedding data determined in step 1). The specific judgment rule can be a quantitative expression using the folding coefficient (bending height / bending width). These attribute volumes are compared with the folding coefficients at the more obvious bending deformation points of the bedding plane; the better the correlation, the better the attribute volume.

[0086] In step 6, the fold attribute volume is a quantitative description of the manner and extent of tectonic deformation such as strata bending at each sample point in space.

[0087] In step 6, the specific implementation steps for calibrating the fold attribute body are as follows: a) Calculate the interlayer average value of each fold attribute body; b) Extract the attribute average value at the well point with imaging logging; c) Perform cross-statistical analysis on the fold attribute average value at the well point and the fracture interpretation density value of imaging logging, and fit the correlation formula; d) Use the correlation formula to transform the fold attribute body as the fracture probability body of the fold zone of the oil and gas reservoir.

[0088] Step 7: Based on the 3D seismic data, calculate the wave impedance attribute volume. According to the correspondence between the clay content and wave impedance data obtained from the well logging interpretation in Step 1, perform clay content inversion on the 3D seismic data to obtain the clay content inversion attribute volume. Normalize the clay content inversion attribute volume in the 1-0 interval to form the 3D rock-controlled fracture probability volume (i.e., the 3D rock-controlled fracture development probability volume model), denoted as P(F3').

[0089] Step 8: Based on the fracture development probability models P(F1), P(F2), and P(F3) obtained in Steps 2, 3, and 4, calculate a new probability model using the weighted average method according to the geological influence ratio of fracture development, forming a fracture development probability model based on geological information (i.e., the geological information fracture development probability model), denoted as P(FGeology).

[0090] In step 8, the geological influence ratio refers to the quantitative measure of the relative degree of control of different geological factors on fracture development. It can be obtained from the statistical relationship between the number of fractures in the drilling statistics and the geological factor parameters interpreted by the well logging, or from the actual field outcrop measurement data.

[0091] In step 8, the specific implementation steps for obtaining the fracture development probability model based on geological information are as follows: a) Statistically analyze the correlation between geological factor parameters such as distance from the fault, fold coefficient, and quartz mineral content and fracture density, and obtain their respective correlation coefficients; b) Calculate the weighted average of the correlation coefficients between different geological factor parameters and fracture density to obtain the influence ratio of each parameter on fracture development, with the sum of the ratios being 100%; c) Calculate the product of the fracture development probability volume model P(F1), P(F2), and P(F3) values ​​and their corresponding influence ratios to obtain the volume distribution model of the contribution of the geological factor to the fracture development probability; d) Add together the contribution volumes of each geological factor to the fracture development probability in the same grid to obtain the fracture development probability model P(F / Geology) based on geological information.

[0092] Step 9: Based on the fracture development probability models P(F1'), P(F2'), and P(F3') obtained in Steps 5, 6, and 7, perform model grid calculation using the union method to form a fracture development probability model based on geophysical information (i.e., geophysical information fracture development probability model), denoted as P(F / Geophysics).

[0093] In step 9, the specific implementation steps for calculating the model mesh using the union method are as follows: if the model mesh has only one crack development probability data, the mesh value remains unchanged; if the model mesh contains two or more crack development probability data, the data with the larger value is taken as the crack development probability data for that mesh.

[0094] Step 10: Based on the geological information fracture development probability model P(F / Geology) obtained in Step 8 and the geophysical information fracture development probability model P(F / Geophysics) obtained in Step 9, the Bayesian multivariate probability fusion method is used to perform model grid calculation to form a fracture development probability model that integrates geological information and geophysical information, denoted as P(F / Geology, Geophysics).

[0095] In step 10, Bayesian multivariate probability is a statistical method based on Bayes' theorem, which calculates an optimal posterior probability distribution by using the prior probability P(F) and multiple probability data that have a mapping relationship with each other.

[0096] In step 10, the Bayesian multivariate probability can be calculated using, but is not limited to, the following functional expression:

[0097]

[0098] In step 10, the specific implementation steps for model mesh calculation using the Bayesian multivariate probability fusion method are as follows: if the model mesh has only one crack development probability data, the mesh value remains unchanged; if the model mesh contains two crack development probability data, the posterior probability value is calculated using Bayesian multivariate probability to obtain the crack development probability data of the mesh.

[0099] Step 11: Based on the fracture development probability model obtained from the comprehensive geological and geophysical information in Step 10, and using the distribution of fracture occurrence, fracture intensity, and fracture aperture parameters obtained from drilling as conditional data, the indicative point method is used for random simulation to obtain the discrete fracture network (DFN) model and the fracture attribute parameter distribution model.

[0100] In the aforementioned implementation process, the method for establishing a natural fracture model for tight sandstone oil and gas reservoirs, based on the main controlling factors of natural fracture development in tight sandstone oil and gas reservoirs, starts from three main controlling factors: faulting, folding, and lithology. It integrates the joint constraints of fault distance, fold deformation, and lithology and mineralization, obtaining both a geological information-driven fracture development probability model and a geophysical information-driven fracture development probability model. Through three fusion methods—weighted averaging, union calculation, and Bayesian multivariate probability—a comprehensive representation of multimodal and multi-data types of natural fracture information is achieved. The discrete fracture network (DFN) model and fracture attribute parameter distribution model established using this method more accurately reflect the probability of rock mass discontinuity in space, while maintaining the control relationship between tectonic, sedimentary, and other geological factors on fracture development, providing a more reliable geological model guarantee for the quantitative study of tight sandstone oil and gas reservoirs. This method provides a fracture model construction method that integrates geological fracture development laws and geophysical fracture distribution prediction. It integrates the joint constraints of fault distance, fold deformation, and lithology and mineralization based on the main controlling factors of natural fracture development in tight sandstone oil and gas reservoirs. A fracture probabilistic volume model driven by geological information was constructed by weighted averaging based on fault distance, bedding plane folds, and quartz content, according to drilling correlation. A fracture probabilistic volume model driven by geophysical information was constructed based on fracture attribute volumes, fold attribute volumes, and clay content inversion, using union calculations. Through Bayesian PR multivariate probabilistic fusion, a quantitative geological model of the fracture volume was formed, integrating the geological model and geophysical prediction results, and this model was used to further control fracture DFN simulation and attribute parameter simulation.

[0101] Example 2

[0102] The test area is a large uplift zone in region A, and the target reservoir is a reservoir in region A. Since its deposition, it has undergone multiple tectonic movements, and the current overall structure is a long anticline trending approximately east-west, with a steep southern flank and a gentle northern flank. The test area has 3D seismic data, including pre-stack and post-stack data, which can provide relevant information for rock physical parameters in stress simulation and fracture probability volumes (or density volumes) in fracture simulation. More than 40 wells have been drilled, containing logging curves, logging data, well test production data, etc. Among them, 12 core wells and 15 electrical imaging logging wells can provide fracture occurrence information such as fracture dip angle and fracture inclination for drilling interpretation.

[0103] The target reservoir has proven natural gas reserves of hundreds of billions of cubic meters. The reservoir in this area is characterized by low porosity, low permeability, and low daily production per well, making gas reservoir development technically challenging. There is an urgent need to improve the recovery rate of low-permeability, tight sandstone gas reservoirs. The presence of natural fractures in this area not only increases gas storage space but also improves the seepage characteristics of the reservoir. Statistical analysis shows that wells with high total gas-water production and a large proportion of water production are mostly located in areas with high fracture density, while wells with low total gas-water production and a small proportion of water production are mostly located in areas with low fracture density, indicating that fractures have a significant effect on production control. Therefore, accurately and quantitatively characterizing the spatial distribution of natural fractures and establishing an accurate natural fracture model is of great practical significance for the further development and production optimization of this gas field.

[0104] Please refer to Figure 3 , Figure 3 This is a flowchart of another method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to one embodiment of the present invention. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir using this method includes the following steps:

[0105] (1) By utilizing the time-depth conversion relationship of multiple seismic interpretation levels of the entire target reservoir, the stratigraphic interpretation data of the depth domain of the target reservoir is obtained through time-depth conversion. A refined stratigraphic framework built from drilling is applied, and the actual drilling stratigraphic information is used to further correct the top stratigraphic data of the seismically interpreted sand groups, obtaining the stratigraphic data and structural maps of each sand group. The vertical stratigraphic data includes 10 sand groups and 11 stratigraphic levels, from TX2-1 to TX2-10. The fault depth of each fault at different locations is meticulously analyzed and characterized, the fault attitude is refined, and the combination relationships between faults are depicted.

[0106] The obtained three-dimensional structural stratigraphic model was meshed with a planar grid precision of 100m×100m and a vertical grid of approximately 1-2m to establish a structural-stratigraphic framework model. The total number of grids in the model is nearly 32 million, meeting the accuracy requirements. Integrating data from drilling cores, well logging interpretations, and seismic attributes, the distribution of oil and gas reservoir matrix sand bodies and the planar heterogeneous distribution characteristics of lithology, physical properties, and petrophysical parameters under the sequence stratigraphic framework were clarified. A sedimentary microfacies and lithofacies model was established using multi-point geostatistical methods, a brittle mineral (quartz) content model was established using a multi-level facies control constraint method, and a petrophysical parameter model was established using a mineral synergistic method.

[0107] (2) Structural fractures can be classified into low-angle fractures, oblique fractures, and high-angle fractures according to their dip angle. Multiple fractures are developed in the study area, all of which are reverse faults. Through core fracture description and well logging fracture interpretation, and statistical analysis of the distances to the fractures, it was found that the relationship between fracture development and faults is mainly reflected in three aspects: ① The development of fractures with a dip angle greater than 30 degrees is mainly controlled by the distance between the well location and the fracture; ② The area within 400m of the hanging wall of the fault is the fracture-controlled fracture development zone; ③ The orientation of high-angle fractures in the fracture-controlled fracture development zone is basically perpendicular to the fracture orientation.

[0108] In summary, the relationship between fracture development and fault distance can be simply described. It mainly occurs in the 400m region of the hanging wall, with fractures trending nearly vertically, primarily in a WE direction, and mainly high-angle fractures. Based on the geological information of the fault-controlled fractures, a geological model P(F1) based on the fault distance for fracture development probability is established, such as... Figure 5 As shown, Figure 5 This invention provides a fracture probability model based on fault distance, constructed using a fault model in an experimental area, according to one embodiment of the present invention. The specific application process in the modeling step is as follows: ① Extract the previously established fault models, screen them, and prioritize retaining larger faults and small faults with fault-barrier characteristics; ② Establish a trend model based on the distance parameters of the faults; ③ Process the trend model, retaining the trend values ​​within a 400m region on the hanging wall of the fault and performing 1-0 normalization; ④ Establish a fracture development probability model based on fault distance, using the fault distance distribution of each sand group as a constraint.

[0109] (3) The local structural factors affecting the development of different types of fractures mainly include stratigraphic attitude and faults. In fold-controlled fracture systems, fracture development is solely controlled by the structural undulations and deformations caused by folds. Tension joint fractures usually form in anticlines, and fractures also develop in synclines, but not as densely as in anticlines. They are mostly formed by the expansion and contraction deformation of rock strata. In this fracture system, fractures are mainly developed in the anticline hub, with divergent fracture attitudes, generally medium to high angle fractures. Vertically, fractures usually do not cross strata, and most develop in rigid, dense sandstone layers. Fracture development is rarely seen in mudstone.

[0110] Based on the above geological understanding, and using the three-dimensional structural stratigraphic model obtained in Step 1, the curvature data distribution of key bedding planes was calculated, resulting in a two-dimensional planar distribution map of the curvature of each bedding plane. According to the statistical relationship between the number of fractures and the ratio of fold height to fold width in the corresponding sand groups during drilling, this relationship was equated to the curvature relationship of fold-controlled fractures. The bedding plane curvature was normalized in the 0-1 interval, and then converted into a two-dimensional planar map of the fold-controlled fracture development intensity coefficient. Using the two-dimensional map of the fold-controlled fracture development intensity coefficient for each bedding plane of the 10 sand groups in the test area as constraints, a three-dimensional fracture development probability model P(F2) based on bedding plane curvature was obtained, as follows... Figure 6 As shown, Figure 6 This invention provides a crack probability model based on fold deformation, constructed using the curvature of the surface in the test area, according to one embodiment of the invention.

[0111] (4) The control of fracture development by lithology is essentially due to the influence of rock composition, grain size, and grain arrangement on the rock's mechanical properties, which in turn control the fracture development characteristics of different lithologies under tectonic stress. Statistical analysis of fracture parameters and lithofacies in wells revealed that coarse sandstone had the highest degree of effective fracture development (fracture dip angles mostly greater than 30 degrees), while medium and fine sandstone had roughly equal proportions of fracture-developed and non-fracture-developed sections, with some sections lacking fractures. Furthermore, statistical analysis of the correlation between fracture development and mineral content showed that within fractured sections, when the quartz content exceeded 60%, higher quartz content resulted in more developed effective fractures and larger fracture dip angles.

[0112] In summary, the relationship between fracture development and lithofacies can be simply described. Within fracture-developed areas, fracture development is second only to lithofacies (grain size) and is controlled by lithology (mainly quartz content). The higher the quartz content, the more developed the fractures. Based on the geological information of rock-controlled fractures, a volumetric geological model P(F3) based on the fracture development probability of quartz content is established, such as... Figure 7 As shown, Figure 7 This invention provides a fracture probability model based on brittle minerals, constructed using a lithofacies model in an experimental area, according to one embodiment of the present invention. The specific application process in the modeling procedure is as follows: ① Based on the grain size lithofacies model, a quartz content distribution model is established using a phase-controlled modeling approach; ② The quartz content model of the fracture zone is processed, retaining the portion with a quartz content higher than 60%; ③ The model values ​​are normalized to the 0-1 interval to form a fracture development probability model based on lithology.

[0113] (5) Seismic responses contain fracture reflection information, providing a possibility for interpreting fractures in reservoir space. The Fault Likelihood attribute (hereinafter referred to as the FL attribute) reflects the probability of fracture existence, and its calculation is based on the fracture identification-oriented similarity attribute Semblance (S attribute). The FL attribute improves the contrast between different points, enhances the detection capability of traditional similarity attributes, responds to larger fractures, and can characterize small-scale cracks. Compared with other response fracture attributes, the FL attribute can better reflect the development trend of fractures at different dip angles, presenting fracture anomalies at different scales in the profile, especially reflecting the longitudinal and lateral variation trends of fracture development degree in areas with dense fracture development.

[0114] The cross-plotting of fracture density and FL attributes for fractures with dip angles greater than 30° (i.e., effective fractures) in statistical logging fracture interpretation showed correlation factors of 0.78 and 0.86 for the TX2-2 and TX2-4 sand groups, respectively, indicating that the FL attribute can accurately predict the spatial distribution of fractures. Based on the FL attribute extraction results for each sand group, the planar distribution of the FL attribute data for each sand group was obtained, representing the fracture distribution of that sand group. Combining wellpoint fracture interpretation and evaluation statistics, the effective fracture development density curve for dip angles greater than 30° was calculated and obtained. Using wellpoint fracture density as conditional data, each sand group used the adjusted FL attribute as a fracture distribution constraint, and a fracture density model was established using the sequential Gaussian simulation method, serving as the three-dimensional fracture probability volume P(F1') based on geophysical information, as shown below. Figure 8 As shown, Figure 8 This is a crack probability model based on Fault Likelihood seismic attributes for a test area provided by one embodiment of the present invention.

[0115] (6) Folding causes stratigraphic deformation, resulting in fractures. The fracture density is highest at the anticline hub, exhibiting tensional joint fractures. Fractures decrease in the fold limbs, and while fractures also develop in the syncline, their development is much less pronounced than in the anticline. Because the fault thrusting in the test area caused severe folding and faulting of the upper structural layers, stratigraphic curvature and fracture attitude control the degree of curvature, meaning that different layers at the same longitudinal location exhibit varying degrees of fold curvature. To achieve the goal of regulating fracture development through the fold curvature spatial volume, this invention utilizes intelligent algorithms for spatial interpolation based on the local structural attributes of folds extracted from multiple layers, resulting in a relatively smooth, multi-layered controlled local fold structure. This structure is normalized to a range of 0–2, with values ​​of 1–2 for the anticline region and 0–1 for the syncline region.

[0116] In the fractured reservoir of the dome, the maximum curvature and perturbation data volumes reveal the intensity variations of folds and fractures. The magnitudes of curvature and perturbation vary significantly throughout the reservoir. Specifically, high curvature primarily occurs at the top of the anticline, where extensional fractures caused by upward bending of the reservoir are most likely to develop. A three-dimensional model is formed by meshing the maximum curvature seismic attribute volume. The local fold structures are used as adjustment factors for multiplication calculations to enhance the anticline and weaken the syncline. The final results are normalized to the 0-1 interval to obtain the three-dimensional fracture probability volume P(F2') based on fold control, as shown below. Figure 9 As shown, Figure 9 This invention provides a crack probability model for a test area based on the seismic properties of the fold coefficient, according to one embodiment of the present invention.

[0117] (7) Cross-plotting of clay content and acoustic impedance data revealed that the quartz deposits disclosed in the well logging generally had a clay content of less than 10% and an acoustic impedance greater than 1.3 × 10⁻⁶. 7To construct the three-dimensional volume of clay content, we started with clay content inversion to obtain a higher quality and more reliable spatial distribution of clay content. In this step, we selected multiple seismic attributes and used neural networks to learn the nonlinear relationship between clay content and seismic data, thus obtaining the clay content inversion data volume.

[0118] The selected neural network belongs to the supervised neural network method, namely the Multilayer Perceptron (MLP). The application of MLP in clay content prediction involves two steps: training and prediction. During training, the input data to the neural network includes low-frequency, mid-frequency, and high-frequency data, instantaneous phase, instantaneous frequency, instantaneous amplitude, amplitude integral, and relative impedance—seismic attributes that can express lithological changes. The output is a spatial inversion volume of clay content. The output is then normalized to 1-0 to represent the distribution of brittle minerals such as quartz, forming a three-dimensional rock-controlled fracture probability volume, denoted as P(F3'). Figure 10 , Figure 10 This is a crack probability model based on seismic inversion of mud content in the test area provided by one embodiment of the present invention.

[0119] (8) Based on the statistical relationship between the number of fractures in the drilling statistics and the geological factor parameters interpreted by the well logging, the correlation between the degree of fracture development and the distance from the fault, the folding coefficient, and the quartz mineral content are 0.92, 0.63, and 0.78, respectively. That is, the contribution of the distance of the fracture influence, the formation bending deformation, and the sedimentary lithology to the generation of natural fractures in tight reservoirs is about 39.5%, 27.1%, and 33.4%, respectively.

[0120] Based on steps two, three, and four, the established fracture development probability volume models P(F1), P(F2), and P(F3) are multiplied by the corresponding geological factor contribution ratios to obtain the volume distribution model of the contribution of each geological factor to the fracture development probability. The contribution volumes of each geological factor to the fracture development probability in the same grid are then summed to obtain the fracture development probability model P(F / Geology) based on geological information. Figure 11 , Figure 11 This is a crack probability model based on geological information fusion for the test area provided by one embodiment of the present invention.

[0121] (9) Based on the fracture development probability models P(F1'), P(F2'), and P(F3') obtained in steps five, six, and seven, the probability models for FL (characterizing fracture bodies), curvature (characterizing fold bodies), and lithology-inverted fracture bodies are normalized proportionally according to their maximum and minimum values ​​to obtain the value range of each probability body. The model grid is calculated using the union method, fusing the three probability bodies. The fusion principle is: each grid value is unique, with larger values ​​taking priority, followed by medium and smaller values. If a model grid contains only one fracture development probability data point, that grid value remains unchanged; if the model grid contains two or more fracture development probability data points, they are fused according to the above principle to form a fracture development probability model based on geophysical information, denoted as P(F / Geophysics). Figure 12 , Figure 12 This is a crack probability model based on geophysical information fusion for the test area provided by one embodiment of the present invention.

[0122] (10) Based on the crack development probability model P(F / Geology) obtained in step eight and the crack development probability model P(F / Geophysics) obtained in step nine, the model grid is calculated using the Bayesian multivariate probability fusion method.

[0123] Bayesian multivariate probability is a statistical optimization method based on Bayes' theorem. Its generalization expression is: P(A|B)∝P(B|A)·P(A);

[0124] Wherein, P(A|B) is the posterior probability, representing the probability calculation result based on observational data and geological knowledge; P(B|A) is the likelihood probability, responsible for establishing the mapping relationship between observational data and the optimization objective; and P(A) is the prior probability, representing the geological knowledge information based on previous geological studies. This invention uses the loss function as the final optimization object, actual observational data such as well point data as constraint data, and combines prior information such as geostress simulation and corresponding calculation results, employing iterative optimization to perform probability calculations on the calculation results. During the calculation process, the Bayesian optimizer first adjusts the model parameters in real time based on the loss function calculation result in the current iteration step, and then absorbs the modeling result after parameter adjustment in the next iteration step and recalculates the loss function. If the loss function result tends to decrease, it indicates that the optimization is effective, and the optimizer will continue to adjust the parameters according to the previous model parameter adjustment strategy, making the probability calculation result increasingly realistic and reasonable. Conversely, when the loss function result increases, it indicates that the parameter adjustment strategy is no longer suitable, and the parameter adjustment strategy needs to be changed or the optimization process stopped. When the loss function tends to stabilize or the loss function result is below a preset threshold, the probability reaches the global optimum.

[0125] In this trial, Bayesian multivariate probability fusion was used, and the calculation was performed using the following function expression:

[0126]

[0127] Among them, in the calculation of this test area, the global probability is optimal when the prior probability P(F) is 0.2.

[0128] In the two fracture development probability models based on geological and geophysical information, if the model grid contains only one fracture development probability data point, the grid value remains unchanged. If the model grid contains two fracture development probability data points, the posterior probability value of the fracture development probability data for that grid is calculated using the aforementioned Bayesian multivariate probability method, resulting in the final fracture development probability volume model P(F / Geology, Geophysics) for the experimental area. Figure 13 , Figure 13 This invention provides a comprehensive fracture probability model that integrates geological and geophysical information of the test area, according to one embodiment of the present invention. Figure 13 a is a three-dimensional model diagram of the crack development probability volume; Figure 13 b is a plane diagram showing the crack probability of the two key sand groups; Figure 13 c is the cross-sectional view of the corresponding two-dimensional model in the three-dimensional model of crack probability.

[0129] (11) Using the fracture development probability model P(F / Geology, Geophysics) obtained in step ten as the constraint variable for the degree of fracture development between wells, and based on the fracture orientation, fracture density, and fracture aperture data revealed by imaging logging, different parameter inputs are given. Using the indicator point method, small-scale fractures in near-WE (most developed fractures) and near-NS cluster systems are stochastically simulated by sand group to establish a discrete fracture network (DFN) model, such as... Figure 14 , Figure 14 This is a DFN model diagram of a multi-scale natural crack in the test area provided by one embodiment of the present invention.

[0130] Fracture aperture data with dip angles greater than 30° that could be responded to by geophysical calculations were selected. The average and variability of fracture aperture for each sand group were obtained. The average fracture aperture for sand groups TX2-2 and TX2-4 were 0.057 and 0.053, respectively, with variability factors of 0.0436 and 0.0418, respectively. Based on the obtained Discrete Fracture Network (DFN) model and combined with the fracture aperture data distribution obtained from drilling interpretation, a fracture aperture distribution field was simulated, and a three-dimensional distribution model of fracture porosity was calculated. Fracture permeability is related to fracture aperture and fracture porosity. Using the Oda calculation formula, the fracture permeability (in the i, j, and k directions) and fluid channeling coefficient (reflecting connectivity) models were calculated, as follows: Figure 15 , Figure 15 This is a model diagram of the properties of natural cracks in a test area provided by one embodiment of the present invention, wherein... Figure 15a is a diagram of the crack porosity model; Figure 15 b is a diagram of the crack permeability model.

[0131] Comparative analysis revealed that the fracture model established in this invention reflects the probability of rock mass discontinuity in space and maintains the control relationship between geological factors such as tectonic and sedimentary structures and fracture development. The model shows a consistency of over 85% with fracture development data revealed by drilling, and further evidence is obtained by comparing it with new wells (such as...). Figure 16 Well B-2 in the middle, Figure 16 This is a comparison diagram of a newly drilled well and a dual-medium model in a test area provided by one embodiment of the present invention, wherein... Figure 16 a is a cross-sectional view of the new well and the grain size facies model; Figure 16 (b is a cross-sectional view of the new drilling and fracture probability model) Verification of the fracture system encountered during drilling shows that the model can better characterize the spatial distribution of fractures. This invention provides a new modeling approach for the quantitative characterization and prediction of natural fractures in tight sandstone gas reservoirs, and provides a more accurate model foundation. The fracture model established by this invention has higher accuracy and reliability, and can also guarantee the spatial distribution of natural fracture parameters. Based on this fracture model, reservoir quality and distribution can be evaluated more specifically for oil and gas reservoirs, providing a guarantee for subsequent optimization of well locations, fracturing evaluation, and the establishment of post-fracturing fracture network models.

[0132] Example 3

[0133] Figure 2 This is a block diagram of a system for establishing a natural fracture model of a tight sandstone oil and gas reservoir, provided by one embodiment of the present invention. Figure 2 As shown, this invention provides a system for establishing a natural fracture model of a tight sandstone oil and gas reservoir, comprising:

[0134] The model building module is used to build probability models of fault distance fracture development, bedding curvature fracture development, lithological component fracture development, three-dimensional fault zone fracture development, three-dimensional fold zone fracture development, and three-dimensional rock-controlled fracture development based on the stratigraphic geological data of the target reservoir.

[0135] The model fusion module is used to obtain the geological information fracture development probability model of the target reservoir based on the fault distance fracture development probability model, the bedding curvature fracture development probability model and the lithological component fracture development probability model corresponding to the target reservoir, and to obtain the geophysical information fracture development probability model corresponding to the target reservoir based on the three-dimensional fault zone fracture development probability model, the three-dimensional fold zone fracture development probability model and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir.

[0136] The information integration module is used to obtain the fracture development probability model of the target reservoir based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir.

[0137] The fracture distribution determination module is used to obtain a discrete fracture network model and a fracture attribute parameter distribution model for the target reservoir based on the fracture development probability model of drilling data and comprehensive geological and geophysical information corresponding to the target reservoir.

[0138] Specifically, based on the stratigraphic geological data of the target reservoir, the system establishes a fracture development probability model reflecting the relationship between various geological factors (including faults, folds, and lithofacies) and fracture development probability. This model includes models for fault distance fracture development, bedding plane curvature fracture development, lithological composition fracture development, three-dimensional fault zone fracture development, three-dimensional fold zone fracture development, and three-dimensional rock-controlled fracture development. It also comprehensively considers the prediction of fracture development probability based on geological information to obtain a fracture probability model based on geological concepts (i.e., a geological information fracture development probability model). Simultaneously, it comprehensively considers the prediction of fracture development probability based on geophysical information to obtain a fracture probability model based on geophysical prediction (i.e., a geophysical information fracture development probability model). Finally, by combining the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir, a comprehensive geological and geophysical information fracture development probability model for the target reservoir is obtained. Finally, by combining drilling data as conditional parameters, a discrete fracture network model and a fracture attribute parameter distribution model corresponding to the target reservoir are obtained. This system, through multi-level constraints and multi-probability fusion of faulting, folding, and lithological factors, achieves the construction of a fracture development probability model capable of supporting multi-source data fusion and multi-factor driven development. Furthermore, by using a discrete fracture network model and fracture attribute parameter distribution model corresponding to the target reservoir, it more accurately characterizes the spatial distribution of natural fractures, providing a more reliable geological model foundation for the efficient development of tight sandstone oil and gas reservoirs. The system fully considers the controlling factors of fracture development, fully explores and applies various fracture detection information such as geological and seismic data, forming a comprehensive geological model and geophysical prediction method for fracture model establishment. This more accurately characterizes the spatial distribution of various parameters of natural fractures, providing a more reliable quantitative geological model foundation for the efficient development of fractured oil and gas reservoirs such as tight sandstone.

[0139] The optional embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above embodiments. Within the scope of the technical concept of the embodiments of the present invention, various simple modifications can be made to the technical solutions of the embodiments of the present invention, and these simple modifications all fall within the protection scope of the embodiments of the present invention.

[0140] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the embodiments of the present invention will not describe the various possible combinations separately.

[0141] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a microcontroller, chip, or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0142] Furthermore, various different implementations of the present invention can be combined arbitrarily, as long as they do not violate the spirit of the present invention, they should also be regarded as the content disclosed in the present invention.

Claims

1. A method for establishing a natural fracture model of a tight sandstone oil and gas reservoir, characterized in that, include: Based on the stratigraphic geological data of the target reservoir, we establish the following probability models for the development of fractures: fault distance, bedding curvature, lithological components, three-dimensional fault zone, three-dimensional fold zone, and three-dimensional rock-controlled fracture. Based on the fault distance fracture development probability model, the bedding curvature fracture development probability model and the lithological component fracture development probability model corresponding to the target reservoir, the geological information fracture development probability model corresponding to the target reservoir is obtained. Based on the three-dimensional fault zone fracture development probability model, the three-dimensional fold zone fracture development probability model and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir, the geophysical information fracture development probability model corresponding to the target reservoir is obtained. Based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir, a comprehensive geological information and geophysical information fracture development probability model corresponding to the target reservoir is obtained. Based on the fracture development probability model of drilling data and comprehensive geological and geophysical information corresponding to the target reservoir, a discrete fracture network model and fracture attribute parameter distribution model corresponding to the target reservoir are obtained.

2. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 1, characterized in that, The geological data of the target reservoir includes bedding plane data, fracture data, and drilling logging data of the target reservoir. Before establishing the probability volumetric models of fracture development based on fault distance, bedding plane curvature, lithological composition, three-dimensional fault zone, three-dimensional fold zone, and three-dimensional rock-controlled fracture development corresponding to the target reservoir, the method further includes: Based on the bedding and fracture data of the target reservoir, an initial three-dimensional structural stratigraphic model is established. Based on the preset modeling area and preset grid step size, the initial model of the three-dimensional structural strata is meshed; Based on drilling and logging data, the heterogeneous distribution characteristics of the matrix reservoir attribute parameters of oil and gas reservoirs are determined; wherein, the drilling and logging data includes at least drilling core data, logging interpretation data and seismic attribute data, and the matrix reservoir attribute parameters of oil and gas reservoirs include at least the discrete attribute parameters of matrix porosity, permeability, sedimentary facies and lithofacies. Based on the heterogeneous distribution characteristics of oil and gas reservoir matrix attribute parameters and the initial three-dimensional structural stratigraphic model after gridding, a three-dimensional structural stratigraphic model is obtained.

3. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The rules for establishing the fault distance fracture development probability volume model include: Based on the target reservoir's bedding plane data, fracture data, and three-dimensional structural stratigraphic model, fault models in the three-dimensional structural stratigraphic model are selected. Based on the selected fault models, a distance trend volume model corresponding to each fault model is established; wherein, the distance trend volume model represents the vertical distance from each grid of the three-dimensional structural stratigraphic model to the corresponding fault. Based on the range of the fault-controlled fracture zone, the distance trend volume model is sequentially truncated and normalized to obtain the fault distance fracture development probability volume model.

4. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The rules for establishing the probabilistic volume model of layer curvature crack development include: Based on the three-dimensional structural stratigraphic model, a two-dimensional map of the curvature distribution of each layer is determined. Based on the curvature relationship of fold-controlled cracks, the curvature of the layers is normalized to convert the two-dimensional map of the curvature distribution of each layer into a two-dimensional map of the development intensity coefficient of fold-controlled cracks corresponding to each layer. Using the two-dimensional diagram of the fold control crack development intensity coefficient corresponding to each layer as a constraint condition, a probabilistic volume model of layer curvature crack development is obtained.

5. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The rules for establishing the probabilistic volume model of fracture development in lithological components include: A quartz content model is established based on the lithofacies model; wherein the lithofacies model is determined based on a three-dimensional tectonic stratigraphic model. Based on the range of the fault-controlled fracture zone and the lower limit threshold of the quartz content controlling fracture development, the quartz content model is truncated and normalized sequentially to obtain a lithological component fracture development probability model.

6. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 5, characterized in that, The establishment of a quartz content model based on the lithofacies model includes: The vertical distribution ratio and various characteristic values ​​of quartz content in different rock facies were statistically analyzed; wherein, the various characteristic values ​​of quartz content include the maximum value of quartz content, the minimum value of quartz content, the average value of quartz content, the variance of quartz content and / or the coefficient of variation of quartz content. Based on the vertical distribution ratio of quartz content in different lithofacies and various characteristic values, the parameters of the quartz content variation function are determined. Based on the lithofacies model, the spatial distribution data of quartz content in each lithofacies were determined. Based on the quartz content variation function parameters and the spatial distribution data of quartz content in each lithofacies, a quartz content model was obtained using sequential Gaussian simulation.

7. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The rules for establishing the three-dimensional fracture zone fracture development probability volume model include: Based on three-dimensional seismic data, multiple first attribute volumes are obtained; among them, the first attribute volume represents the attribute volume used to respond to the discontinuity characteristics of seismic wave reflection. By comparing the spatial locations of each first attribute body and fault data, the first attribute body that best reflects the fracture fragmentation is determined as the fracture attribute body; wherein, the spatial location of the fault data is determined based on the bedding plane data, fracture data and three-dimensional structural stratigraphic model of the target reservoir; Based on drilling data, the maximum and minimum thresholds of the fracture attribute volume are calibrated, and then the fracture attribute volume is normalized to obtain a three-dimensional fracture zone fracture development probability volume model.

8. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 7, characterized in that, The calibration of the maximum and minimum thresholds of the fracture attribute volume based on drilling data includes: Calculate the interlayer average value of the fracture property volume and obtain the fracture interpretation density value at the well point with imaging logging; After statistically analyzing the interlayer average value of the fracture attribute body and the fracture interpretation density value of imaging logging, the maximum and minimum values ​​of the interlayer average value of the fracture attribute body are extracted as the maximum and minimum threshold values ​​of the fracture attribute body.

9. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The rules for establishing the three-dimensional fold zone crack development probability volume model include: Based on three-dimensional seismic data, a variety of second attribute volumes are obtained; among them, the second attribute volume represents the attribute volume used to respond to the curvature characteristics of the rock strata reflected by seismic waves; By comparing each second attribute body with the bedding morphology data, the second attribute body that best reflects the deformation intensity is determined as the fold attribute body; wherein, the bedding morphology data is determined based on the bedding data, fracture data and three-dimensional structural stratigraphic model of the target reservoir. After normalizing the fold attribute volume, the fold attribute volume is calibrated based on drilling data to obtain a three-dimensional fold zone fracture development probability model.

10. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 9, characterized in that, The calibration of the fold attribute volume based on drilling data yields a three-dimensional fold fracture development probability model, including: Calculate the interlayer average value of the fold property volume and obtain the fracture interpretation density value at the well point with imaging logging; The interlayer average value of the fold property volume and the fracture interpretation density value of imaging logging are intersected and statistically analyzed to obtain the correlation formula. Based on the correlation formula, the fold attribute volume is transformed to obtain a three-dimensional fold zone crack development probability volume model.

11. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The rules for establishing the three-dimensional rock-controlled fracture development probabilistic volume model include: Calculate wave impedance property volume based on 3D seismic data; Based on the acoustic impedance attribute volume and the correspondence between clay content and acoustic impedance data, clay content inversion is performed on 3D seismic data to obtain the clay content inversion attribute volume; wherein, the correspondence between clay content and acoustic impedance data is obtained from well logging interpretation data; The mud content inversion attribute volume is normalized to obtain a three-dimensional rock-controlled fracture development probability model.

12. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The geological information fracture development probability model corresponding to the target reservoir is obtained based on the fault distance fracture development probability model, the bedding plane curvature fracture development probability model, and the lithological component fracture development probability model. This includes: Based on the fault distance fracture development probability model, bedding plane curvature fracture development probability model, and lithological component fracture development probability model corresponding to the target reservoir, and the geological influence ratio of fracture development, a weighted average method is used to calculate the geological information fracture development probability model corresponding to the target reservoir; among which... The geological influence ratio represents a quantitative measure of the relative degree of control exerted by different geological factors on fracture development.

13. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 12, characterized in that, The geological information fracture development probability model corresponding to the target reservoir is calculated using a weighted average method based on the fault distance fracture development probability model, bedding plane curvature fracture development probability model, lithological component fracture development probability model, and the geological influence ratio of fracture development. This model includes: The correlation between different geological parameters and fracture density was statistically analyzed to obtain the correlation coefficient between each geological parameter and fracture density; wherein, the geological parameters include the distance of the fracture from the fault, the folding coefficient, and the quartz mineral content; The weighted average of the correlation coefficients between different geological parameters and fracture density was calculated to obtain the influence ratio of each geological parameter on fracture development. The volumetric distribution model of the contribution of each geological factor parameter to the fracture development probability is obtained by multiplying the volumetric model of the fault distance fracture development probability model, the volumetric model of the bedding curvature fracture development probability model, and the volumetric model of the lithological component fracture development probability model corresponding to the target reservoir with the corresponding geological factor parameters on fracture development. Based on the volume distribution model of the contribution of various geological factors to fracture development probability, the contribution volumes of fracture development probability corresponding to various geological factors in the same grid are added together to obtain the geological information fracture development probability model corresponding to the target reservoir.

14. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The geophysical information fracture development probability model corresponding to the target reservoir is obtained based on the three-dimensional fracture development probability model of the fault zone, the three-dimensional fracture development probability model of the fold zone, and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir, including: Based on the three-dimensional fracture development probability models of the fracture zone, the three-dimensional fracture development probability models of the fold zone, and the three-dimensional rock-controlled fracture development probability models corresponding to the target reservoir, the model grid calculation is performed using the union method to obtain the geophysical information fracture development probability model corresponding to the target reservoir.

15. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 14, characterized in that, The probability model of fracture development in the target reservoir, based on the three-dimensional fracture zone fracture development probability model, the three-dimensional fold fracture zone fracture development probability model, and the three-dimensional rock-controlled fracture development probability model, is obtained by performing model mesh calculation using the union method, including: Based on the three-dimensional fracture development probability volume model, the three-dimensional fold fracture development probability volume model, and the three-dimensional rock-controlled fracture development probability volume model corresponding to the target reservoir, the number of fracture development probability data for each grid is determined. For each grid, if a grid contains only one crack development probability data, then the crack development probability data of that grid remains unchanged. If a grid contains at least two crack development probability data, the crack development probability data with the largest value shall be used as the crack development probability data for that grid.

16. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 2, characterized in that, The fracture development probability model based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir is obtained by combining the geological information and geophysical information of the target reservoir, including: Based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir, the Bayesian multivariate probability fusion method is used to perform model grid calculation to obtain the comprehensive geological information and geophysical information fracture development probability model corresponding to the target reservoir.

17. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 16, characterized in that, The fracture development probability model based on geological information and geophysical information corresponding to the target reservoir is obtained by performing model grid calculation using a Bayesian multivariate probability fusion method to obtain a comprehensive geological and geophysical information fracture development probability model corresponding to the target reservoir, including: Based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir, the number of fracture development probability data for each grid is determined. For each grid, if a grid contains only one crack development probability data, then the crack development probability data of that grid remains unchanged. If a grid contains at least two crack development probability data points, then the Bayesian multivariate probability fusion method is used to fuse all crack development probability data points of the grid to obtain the crack development probability data of the grid.

18. The method for establishing a natural fracture model of a tight sandstone oil and gas reservoir according to claim 1, characterized in that, The drilling data includes fracture occurrence data, fracture strength data, and fracture aperture parameter distribution data. The fracture development probability model based on drilling data and comprehensive geological and geophysical information corresponding to the target reservoir yields a discrete fracture network model and a fracture attribute parameter distribution model corresponding to the target reservoir, including: Based on the fracture development probability model of the comprehensive geological and geophysical information corresponding to the target reservoir, and using fracture occurrence data, fracture intensity data, and fracture aperture parameter distribution data as constraints, the discrete fracture network model and fracture attribute parameter distribution model corresponding to the target reservoir are obtained by random simulation using the indicative point method.

19. A system for establishing a natural fracture model of a tight sandstone oil and gas reservoir, characterized in that, include: The model building module is used to build probability models of fault distance fracture development, bedding curvature fracture development, lithological component fracture development, three-dimensional fault zone fracture development, three-dimensional fold zone fracture development, and three-dimensional rock-controlled fracture development based on the stratigraphic geological data of the target reservoir. The model fusion module is used to obtain the geological information fracture development probability model of the target reservoir based on the fault distance fracture development probability model, the bedding curvature fracture development probability model and the lithological component fracture development probability model corresponding to the target reservoir, and to obtain the geophysical information fracture development probability model corresponding to the target reservoir based on the three-dimensional fault zone fracture development probability model, the three-dimensional fold zone fracture development probability model and the three-dimensional rock-controlled fracture development probability model corresponding to the target reservoir. The information integration module is used to obtain the fracture development probability model of the target reservoir based on the geological information fracture development probability model and the geophysical information fracture development probability model corresponding to the target reservoir. The fracture distribution determination module is used to obtain a discrete fracture network model and a fracture attribute parameter distribution model for the target reservoir based on the fracture development probability model of drilling data and comprehensive geological and geophysical information corresponding to the target reservoir.

20. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform the method for establishing a natural fracture model of a tight sandstone oil and gas reservoir as described in any one of claims 1 to 18.

21. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for establishing a natural fracture model of a tight sandstone oil and gas reservoir as described in any one of claims 1 to 18.

Images