Complex riverway geologic modeling method based on automatic pickup of multiple object source directions

Abstract

The invention discloses a complex riverway geologic modeling method based on automatic pickup of multiple object source directions, and belongs to the technical field of oil-gas exploration and development, and the method comprises the following steps: S1, obtaining the plane and longitudinal distribution characteristics of a riverway; s2, establishing a three-dimensional space distribution model of the river channel body in different periods, and processing boundary data of the river channel in different periods; s3, calling azimuth angle data to establish a sand body distribution model in the river channel in different periods; s4, single well reservoir types are divided, and reservoir type distribution models in the river channels in different periods are established; and S5, discretizing the porosity curve of the single well, calling azimuth angle data to establish porosity models under different reservoir types, and establishing permeability models under different reservoir types. According to the built river channel sand body and physical property parameter model, the situation that reservoir physical property distribution is controlled by river flow direction variability material sources is truly represented, the method is suitable for scenes with large work areas and complex river channel distribution, and modeling timeliness and accuracy can be improved.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration and development technology, and in particular to a method for modeling complex river geological features based on automatically picking multiple source directions. Background Technology

[0002] In meandering or braided river systems, tight sandstone reservoirs exhibit multiple superimposed channels vertically, complex physical property distribution, and strong heterogeneity. Accurately simulating the distribution of tight sandstone reservoirs within these complex channels is a core aspect of reservoir development research. It also serves as the geological basis for comprehensive reservoir evaluation, a necessary parameter for reservoir numerical simulation, and a direct basis for reservoir development adjustment schemes. Domestic geological modeling methods for meandering rivers generally rely on conventional geostatistical methods, where the distribution trends of these attribute data are closely related to the variogram. However, in areas with meandering or braided river deposits, multiple channel shifts and rerouting make it difficult to use conventional variograms to control the direction of the constrained data.

[0003] In geological modeling, traditional geostatistics relies primarily on a variogram model with a given source direction to determine relevant parameters during stochastic simulations. The primary source direction not only controls the azimuth of the variogram but also the overall pattern of the model; its accuracy directly impacts the simulation's precision. In actual geological sedimentary processes, most sedimentary systems are formed by the accumulation of sediments from multiple or varying sources. Traditional geological modeling, when simulating reservoir internal structure and property distribution, first sets a source direction, simulating the sand bodies and property distribution characteristics of each reservoir under the control of a single source. However, for multi-source sedimentary bodies, especially complex meandering or braided river deposits and high-straightness channels, the flow direction changes constantly with the flow streamlines. A single source control can no longer accurately reflect the changes in sand bodies and reservoir properties. Therefore, providing only a single source direction in reservoir geological modeling is insufficient to meet the required simulation accuracy.

[0004] Currently, two main approaches are commonly used to model fluvial reservoirs with multiple or varying sediment sources: One approach involves adding trend lines as zoning lines to divide the study area into several blocks, each representing a different sedimentary body. A variogram is then applied to each block using a grid filtering function, giving each sedimentary body an independent azimuth and variogram to simulate reservoir distribution under multiple source conditions. The other approach involves manually drawing azimuth lines based on the source direction to create a surface file with varying azimuth distribution, simulating different locations with different azimuths. The first approach is more suitable for situations with a small number of sedimentary bodies, as each body corresponds to an independent data analysis and stochastic interpolation simulation process. However, this method simulates sand bodies that are essentially separate and without overlap, neglecting the connectivity between different sedimentary bodies during modeling. Furthermore, when the river channel curvature changes rapidly and there are many channels in the study area, this method requires extensive manual zoning, making it unsuitable. The second approach uses the same variogram for the entire study area, ensuring connectivity between sand bodies and reflecting the control of different source directions on sand body distribution. However, when dealing with complex meandering rivers or braided channels, on the one hand, there are many types of channels, large differences in scale, and varied shapes, and manual drawing of the source direction is subject to a large degree of human error and has a high degree of ambiguity. On the other hand, when the work area is large and there are many channels, the timeliness of manual drawing of the source direction is low, and it is impossible to quickly, accurately, efficiently, and intelligently pick up the accurate azimuth information in each bend of different types of channels.

[0005] Chinese patent document CN107895076A, published on April 10, 2018, discloses a target-based non-penetrating river channel modeling method, characterized by including: S1. Generate several river centerline nodes, connect several river centerline nodes to form a river centerline, and the river centerline runs through a preset three-dimensional research area. S2. Establish a three-dimensional river model within the three-dimensional study area with the river centerline as the center; S3. After determining the river boundary line, if the river boundary line is an irregular curve, then the river boundary line is discretized into several dense river boundary points. The function of the spline curve connecting several river boundary points is obtained by using cubic spline interpolation, which is the curve function corresponding to the river boundary line. The river boundary line truncates the three-dimensional river model in the three-dimensional study area. S4. The three-dimensional study area is meshed to form the first model. The grid through which the river boundary line passes in the first model is the boundary grid. The truncated river model is identified. The truncated river model is the part of the three-dimensional river model from the source of the river to the boundary grid. A meshed second model is established. The size of the second model is the same as that of the first model. The truncated river model and the river boundary line in the first model are mapped to the second model. The truncated river model and the river boundary line are in the same position in the first model and the second model. S5. Repeat steps S1-S4 to generate different three-dimensional river models and determine the river boundary lines of the three-dimensional river models. Simulate each truncated river model and its corresponding river boundary lines in the second model.

[0006] The patent document discloses a target-based non-penetrating river modeling method. This method establishes a river model, constructs irregular boundary lines, quantitatively describes the boundary lines using spline functions, and assigns the truncated river model and boundary lines to a mesh model to achieve river truncation. This creates a geological model that meets practical needs and improves oil and gas recovery rates. However, when there are many rivers, this method requires extensive manual partitioning, making it unsuitable and affecting the timeliness and accuracy of modeling. Summary of the Invention

[0007] In order to overcome the shortcomings of the prior art, this invention provides a method for modeling complex river geological features based on automatic picking of multiple source directions. The river sand body and physical property parameter model established by this invention realistically reproduces the situation where the distribution of reservoir physical properties is controlled by the variable source of river flow direction. It is suitable for scenarios with large work areas and complex river distribution, and can improve the timeliness and accuracy of modeling.

[0008] This invention is achieved through the following technical solution: A method for modeling complex river channel geology based on automatically picking multiple material source directions, characterized by the following steps: S1. Use seismic attributes and well logging data to characterize the river boundaries of different periods, correct the river range interpreted by seismic analysis, and make the river range interpreted by seismic analysis consistent with the river range interpreted by actual drilling. Input the river boundary data of different periods into geological modeling software to obtain the horizontal and vertical distribution characteristics of the river. S2. Establish a three-dimensional spatial distribution model of the river body in different phases and process the river boundary data in different phases; S3. Under the three-dimensional spatial distribution model of the river channel, based on the single-well sand group identification results, the extracted seismic attributes are used as trend constraints, and azimuth data is used to establish sand body distribution models in different periods of the river channel. S4. Classify the reservoir types of single wells, discretize the reservoir type curves of single wells, and establish a reservoir type distribution model in different phases of the channel. S5. Discretize the porosity curve of a single well. Under the reservoir type control, call the azimuth data to establish porosity models under different reservoir types. Finally, use the relationship between permeability and porosity to establish permeability models under different reservoir types.

[0009] In S1, depicting the boundaries of different channel phases refers to dividing the depositional phases of a single-well channel based on core and thin section observations from cored wells, combined with well logging curves and sedimentary characteristic analysis, and then depicting the planar distribution map and top and bottom surfaces of the channel for each phase based on seismic attribute response characteristic analysis.

[0010] In S1, the correction of the channel range in the seismic interpretation refers to picking the top and bottom positions of the channel in the actual drilled well, and using the root mean square amplitude attribute of the seismic earthquake as the thickness trend for correction, so that the top and bottom surfaces of the channel in the seismic interpretation are consistent with the top surface position of the actual drilled channel.

[0011] In S2, processing river boundary data of different periods refers to automatically picking it along the river direction.

[0012] The automatic river direction picking specifically refers to obtaining the azimuth line segments of the river flow direction distribution, converting the line segments into azimuth angles through planar drawing, and interpolating them to form surface data representing the azimuth, which is used as the azimuth angle data for modeling.

[0013] In S3, the single-well sand group identification result refers to the discrete data of the sand body identified by a single well.

[0014] In S4, classifying a single well reservoir type refers to determining the reservoir type by comprehensively considering lithofacies type, physical properties, and pore throat size.

[0015] In S4, establishing a reservoir type distribution model within the channel at different stages refers to using the extracted seismic attributes as trend surface constraints and calling azimuth data for random modeling under the sand body distribution model.

[0016] In S5, calling azimuth data to establish porosity models for different reservoir types means using sequential Gaussian simulation, with porosity as the first variable and the porosity inversion result as the second variable for simulation.

[0017] In S5, establishing permeability models for different reservoir types using the relationship between permeability and porosity means using the established porosity model as the second variable constraint of the permeability model and employing sequential Gaussian simulation.

[0018] The basic principle of this invention is as follows: In stochastic simulations, the relevant parameters are primarily determined by a variogram model with a given source direction. The primary source direction not only controls the azimuth of the variogram but also the overall pattern of the model; its accuracy directly affects the simulation's accuracy. For complex channel deposits, traditional modeling methods assign a single primary source direction (i.e., a single azimuth), resulting in sand body deposition along a single direction. This leads to poor continuity of sand bodies in narrower channels, which contradicts sedimentological principles.

[0019] This invention proposes a method for automatically picking up variable azimuth angles based on the shape of the river channel. Based on the depicted river channel boundary data volume, the data editing function is used to extract line segments representing the direction of river channel deposition, which are used as azimuth angle data to generate azimuth surface data. This data serves as an important control parameter to constrain river channel deposition. The sand body distribution established based on this method is distributed along the direction of river flow, which is more consistent with sedimentological laws than the traditional model established by a single azimuth angle.

[0020] In addition, traditional models built with a single azimuth angle only require a specific source direction to be input during the modeling process based on regional sedimentary research results. That is, the azimuth angle data is used as the control parameter for the interpolation direction of the entire model. For example, if the source direction is due north, then the azimuth angle of 0° is input in the variation function module; if the source direction is northeast, then the azimuth angle of 45° is input.

[0021] In this invention, an azimuth surface data established by a line segment representing the direction of river sedimentation is used as the control parameter. It is not limited by the shape, number, or complexity of the river channel and is applicable to any type of river channel, thereby improving the accuracy and efficiency of geological modeling of complex river channels.

[0022] The beneficial effects of this invention are mainly reflected in the following aspects: 1. Compared with the prior art, the model of river sand bodies and physical property parameters established by this invention truly reproduces the situation where the distribution of reservoir physical properties is controlled by the variable source of river flow direction. It is suitable for scenarios with large working areas and complex river distribution, and can improve the timeliness and accuracy of modeling.

[0023] 2. Compared with traditional geological models established based on a single source direction, this invention, by establishing a model of channel sand bodies and physical property parameters, is more in line with sedimentological laws and geological understanding.

[0024] 3. In the process of obtaining the source direction of the variogram in complex meandering rivers or braided channels, this invention reduces the artificial multiple solutions caused by manually drawing the source direction. When the work area is large and the river distribution is complex, it can further improve the timeliness and accuracy of modeling.

[0025] 4. This invention is applicable to complex meandering river or braided channel sedimentary reservoirs, and is also applicable to sedimentary geological bodies with multiple or varying sediment sources.

[0026] 5. This invention comprehensively utilizes sedimentology and statistics, conforming to sedimentary patterns and meeting the requirements of geostatistics, thereby improving the level of geological modeling.

[0027] 6. This invention can provide a strong geological basis for comprehensive evaluation of oil and gas reservoirs, numerical simulation, scheme formulation and well location deployment.

[0028] 7. This invention has been applied in multiple gas fields in the Sichuan Basin and Ordos Basin, which can improve the quality of geological models and guide well site deployment and reservoir stimulation. Attached Figure Description

[0029] The present invention will now be further described in detail with reference to the accompanying drawings and specific embodiments: Figure 1 This is a flowchart of the present invention; Figure 2 This is a schematic diagram of the extraction of river boundary lines in this invention; Figure 3 This invention uses line segments to represent the direction of river flow. Figure 4 This is the azimuth surface data map generated by the present invention; Figure 5 A river sand body sedimentary model diagram established for the automatic picking of complex river geological models with multiple source directions in this invention; Figure 6 A diagram of a channel sand body sedimentary model established for a traditional single source direction. Detailed Implementation

[0030] Example 1 See Figure 1 A method for modeling complex river channel geology based on automatically picking multiple material source directions includes the following steps: S1. Use seismic attributes and well logging data to characterize the river boundaries of different periods, correct the river range interpreted by seismic analysis, and make the river range interpreted by seismic analysis consistent with the river range interpreted by actual drilling. Input the river boundary data of different periods into geological modeling software to obtain the horizontal and vertical distribution characteristics of the river. S2. Establish a three-dimensional spatial distribution model of the river body in different phases and process the river boundary data in different phases; S3. Under the three-dimensional spatial distribution model of the river channel, based on the single-well sand group identification results, the extracted seismic attributes are used as trend constraints, and azimuth data is used to establish sand body distribution models in different periods of the river channel. S4. Classify the reservoir types of single wells, discretize the reservoir type curves of single wells, and establish a reservoir type distribution model in different phases of the channel. S5. Discretize the porosity curve of a single well. Under the reservoir type control, call the azimuth data to establish porosity models under different reservoir types. Finally, use the relationship between permeability and porosity to establish permeability models under different reservoir types.

[0031] This embodiment is the most basic implementation method. Compared with the prior art, the established channel sand body and physical property parameter model truly reproduces the situation where the distribution of reservoir physical properties is controlled by the variable source of river flow direction. It is suitable for scenarios with large working areas and complex channel distribution, and can improve the timeliness and accuracy of modeling.

[0032] Example 2 See Figure 1 A method for modeling complex river channel geology based on automatically picking multiple material source directions includes the following steps: S1. Use seismic attributes and well logging data to characterize the river boundaries of different periods, correct the river range interpreted by seismic analysis, and make the river range interpreted by seismic analysis consistent with the river range interpreted by actual drilling. Input the river boundary data of different periods into geological modeling software to obtain the horizontal and vertical distribution characteristics of the river. S2. Establish a three-dimensional spatial distribution model of the river body in different phases and process the river boundary data in different phases; S3. Under the three-dimensional spatial distribution model of the river channel, based on the single-well sand group identification results, the extracted seismic attributes are used as trend constraints, and azimuth data is used to establish sand body distribution models in different periods of the river channel. S4. Classify the reservoir types of single wells, discretize the reservoir type curves of single wells, and establish a reservoir type distribution model in different phases of the channel. S5. Discretize the porosity curve of a single well. Under the reservoir type control, call the azimuth data to establish porosity models under different reservoir types. Finally, use the relationship between permeability and porosity to establish permeability models under different reservoir types.

[0033] Preferably, in S1, depicting the boundaries of different channel phases refers to dividing the depositional phases of a single-well channel based on core and thin section observations from cored wells, combined with well logging curves and sedimentary characteristic analysis, and depicting the planar distribution map and top and bottom surfaces of the channel for each phase based on seismic attribute response characteristic analysis.

[0034] In S1, the correction of the channel range in the seismic interpretation refers to picking the top and bottom positions of the channel in the actual drilled well, and using the root mean square amplitude attribute of the seismic earthquake as the thickness trend for correction, so that the top and bottom surfaces of the channel in the seismic interpretation are consistent with the top surface position of the actual drilled channel.

[0035] This embodiment is a preferred implementation method. Compared with the traditional geological model established by a single source direction, the model established by the channel sand body and physical property parameters is more in line with sedimentological laws and geological understanding.

[0036] Example 3 See Figure 1 A method for modeling complex river channel geology based on automatically picking multiple material source directions includes the following steps: S1. Use seismic attributes and well logging data to characterize the river boundaries of different periods, correct the river range interpreted by seismic analysis, and make the river range interpreted by seismic analysis consistent with the river range interpreted by actual drilling. Input the river boundary data of different periods into geological modeling software to obtain the horizontal and vertical distribution characteristics of the river. S2. Establish a three-dimensional spatial distribution model of the river body in different phases and process the river boundary data in different phases; S3. Under the three-dimensional spatial distribution model of the river channel, based on the single-well sand group identification results, the extracted seismic attributes are used as trend constraints, and azimuth data is used to establish sand body distribution models in different periods of the river channel. S4. Classify the reservoir types of single wells, discretize the reservoir type curves of single wells, and establish a reservoir type distribution model in different phases of the channel. S5. Discretize the porosity curve of a single well. Under the reservoir type control, call the azimuth data to establish porosity models under different reservoir types. Finally, use the relationship between permeability and porosity to establish permeability models under different reservoir types.

[0037] In S1, depicting the boundaries of different channel phases refers to dividing the depositional phases of a single-well channel based on core and thin section observations from cored wells, combined with well logging curves and sedimentary characteristic analysis, and then depicting the planar distribution map and top and bottom surfaces of the channel for each phase based on seismic attribute response characteristic analysis.

[0038] In S1, the correction of the channel range in the seismic interpretation refers to picking the top and bottom positions of the channel in the actual drilled well, and using the root mean square amplitude attribute of the seismic earthquake as the thickness trend for correction, so that the top and bottom surfaces of the channel in the seismic interpretation are consistent with the top surface position of the actual drilled channel.

[0039] More preferably, in S2, processing river boundary data of different periods refers to automatically picking it through the river direction.

[0040] The automatic river direction picking specifically refers to obtaining the azimuth line segments of the river flow direction distribution, converting the line segments into azimuth angles through planar drawing, and interpolating them to form surface data representing the azimuth, which is used as the azimuth angle data for modeling.

[0041] In S3, the single-well sand group identification result refers to the discrete data of the sand body identified by a single well.

[0042] This embodiment is another preferred implementation method. In the process of obtaining the source direction of the variogram of complex meandering rivers or braided channels, it reduces the artificial multiple solutions caused by manually drawing the source direction. When the work area is large and the river distribution is complex, it can further improve the timeliness and accuracy of modeling.

[0043] Example 4 See Figure 1 A method for modeling complex river channel geology based on automatically picking multiple material source directions includes the following steps: S1. Use seismic attributes and well logging data to characterize the river boundaries of different periods, correct the river range interpreted by seismic analysis, and make the river range interpreted by seismic analysis consistent with the river range interpreted by actual drilling. Input the river boundary data of different periods into geological modeling software to obtain the horizontal and vertical distribution characteristics of the river. S2. Establish a three-dimensional spatial distribution model of the river body in different phases and process the river boundary data in different phases; S3. Under the three-dimensional spatial distribution model of the river channel, based on the single-well sand group identification results, the extracted seismic attributes are used as trend constraints, and azimuth data is used to establish sand body distribution models in different periods of the river channel. S4. Classify the reservoir types of single wells, discretize the reservoir type curves of single wells, and establish a reservoir type distribution model in different phases of the channel. S5. Discretize the porosity curve of a single well. Under the reservoir type control, call the azimuth data to establish porosity models under different reservoir types. Finally, use the relationship between permeability and porosity to establish permeability models under different reservoir types.

[0044] In S1, depicting the boundaries of different channel phases refers to dividing the depositional phases of a single-well channel based on core and thin section observations from cored wells, combined with well logging curves and sedimentary characteristic analysis, and then depicting the planar distribution map and top and bottom surfaces of the channel for each phase based on seismic attribute response characteristic analysis.

[0045] In S1, the correction of the channel range in the seismic interpretation refers to picking the top and bottom positions of the channel in the actual drilled well, and using the root mean square amplitude attribute of the seismic earthquake as the thickness trend for correction, so that the top and bottom surfaces of the channel in the seismic interpretation are consistent with the top surface position of the actual drilled channel.

[0046] In S2, processing river boundary data of different periods refers to automatically picking it along the river direction.

[0047] The automatic river direction picking specifically refers to obtaining the azimuth line segments of the river flow direction distribution, converting the line segments into azimuth angles through planar drawing, and interpolating them to form surface data representing the azimuth, which is used as the azimuth angle data for modeling.

[0048] In S3, the single-well sand group identification result refers to the discrete data of the sand body identified by a single well.

[0049] In S4, classifying a single well reservoir type refers to determining the reservoir type by comprehensively considering lithofacies type, physical properties, and pore throat size.

[0050] In S4, establishing a reservoir type distribution model within the channel at different stages refers to using the extracted seismic attributes as trend surface constraints and calling azimuth data for random modeling under the sand body distribution model.

[0051] This embodiment is another preferred implementation method, applicable to complex meandering river or braided channel sedimentary reservoirs, and also applicable to sedimentary geological bodies with multiple or varying sources.

[0052] Example 5 See Figures 1-4 A method for modeling complex river channel geology based on automatically picking multiple material source directions includes the following steps: S1. Use seismic attributes and well logging data to characterize the river boundaries of different periods, correct the river range interpreted by seismic analysis, and make the river range interpreted by seismic analysis consistent with the river range interpreted by actual drilling. Input the river boundary data of different periods into geological modeling software to obtain the horizontal and vertical distribution characteristics of the river. S2. Establish a three-dimensional spatial distribution model of the river body in different phases and process the river boundary data in different phases; S3. Under the three-dimensional spatial distribution model of the river channel, based on the single-well sand group identification results, the extracted seismic attributes are used as trend constraints, and azimuth data is used to establish sand body distribution models in different periods of the river channel. S4. Classify the reservoir types of single wells, discretize the reservoir type curves of single wells, and establish a reservoir type distribution model in different phases of the channel. S5. Discretize the porosity curve of a single well. Under the reservoir type control, call the azimuth data to establish porosity models under different reservoir types. Finally, use the relationship between permeability and porosity to establish permeability models under different reservoir types.

[0053] In S1, depicting the boundaries of different channel phases refers to dividing the depositional phases of a single-well channel based on core and thin section observations from cored wells, combined with well logging curves and sedimentary characteristic analysis, and then depicting the planar distribution map and top and bottom surfaces of the channel for each phase based on seismic attribute response characteristic analysis.

[0054] In S1, the correction of the channel range in the seismic interpretation refers to picking the top and bottom positions of the channel in the actual drilled well, and using the root mean square amplitude attribute of the seismic earthquake as the thickness trend for correction, so that the top and bottom surfaces of the channel in the seismic interpretation are consistent with the top surface position of the actual drilled channel.

[0055] In S2, processing river boundary data of different periods refers to automatically picking it along the river direction.

[0056] The automatic river direction picking specifically refers to obtaining the azimuth line segments of the river flow direction distribution, converting the line segments into azimuth angles through planar drawing, and interpolating them to form surface data representing the azimuth, which is used as the azimuth angle data for modeling.

[0057] In S3, the single-well sand group identification result refers to the discrete data of the sand body identified by a single well.

[0058] In S4, classifying a single well reservoir type refers to determining the reservoir type by comprehensively considering lithofacies type, physical properties, and pore throat size.

[0059] In S4, establishing a reservoir type distribution model within the channel at different stages refers to using the extracted seismic attributes as trend surface constraints and calling azimuth data for random modeling under the sand body distribution model.

[0060] More preferably, in step S5, calling azimuth data to establish porosity models for different reservoir types means using sequential Gaussian simulation, with porosity as the first variable and the porosity inversion result as the second variable for simulation.

[0061] In S5, establishing permeability models for different reservoir types using the relationship between permeability and porosity means using the established porosity model as the second variable constraint of the permeability model and employing sequential Gaussian simulation.

[0062] This embodiment represents the optimal implementation method, which comprehensively utilizes sedimentology and statistics. It conforms to sedimentary patterns and meets the requirements of geostatistics, thereby improving the level of geological modeling.

[0063] It can provide a strong geological foundation for comprehensive evaluation of oil and gas reservoirs, numerical simulation, scheme formulation and well location deployment.

[0064] It has been applied in multiple gas fields in the Sichuan Basin and Ordos Basin, which can improve the quality of geological models and guide well site deployment and reservoir stimulation.

[0065] The automatic river direction picking process is as follows: Export the river boundary data and edit it using Excel's data editing function. The first and second columns are the X and Y coordinates of the river boundary line, respectively, and do not modify them. Start editing from the third row of the third column, where the line number is the line number. Enter "=C2+1", enter "=C3" in the fourth row of the third column, and so on until all the line numbers corresponding to the coordinate values ​​have been processed, resulting in an arrangement format of 1, 1, 2, 2, 3, 3, 4, 4...

[0066] The data is then input into the Petrel software, and the connecting lines at the beginning and end are removed to obtain the azimuth angle representing the direction of river flow. Then, azimuth surface files of the direction of river flow in different periods within the work area are generated as control parameters for the variation function.

[0067] from Figure 5 and Figure 6 It can be seen that the channel sand body sedimentary model established by automatically picking up complex channel geological models from multiple source directions according to this invention has sand bodies distributed along the direction of river flow, which conforms to sedimentary patterns. In contrast, the channel sand body sedimentary model established by using a traditional single source direction has sand bodies interpolated along one direction, which is discontinuous and does not conform to sedimentary patterns.

Claims

1. A method for modeling complex river geological features based on automatically picking multiple material source directions, characterized in that, Includes the following steps: S1. Use seismic attributes and well logging data to characterize the river boundaries of different periods, correct the river range interpreted by seismic analysis, and make the river range interpreted by seismic analysis consistent with the river range interpreted by actual drilling. Input the river boundary data of different periods into geological modeling software to obtain the horizontal and vertical distribution characteristics of the river. S2. Establish a three-dimensional spatial distribution model of the river body in different phases and process the river boundary data in different phases; S3. Under the three-dimensional spatial distribution model of the river channel, based on the single-well sand group identification results, the extracted seismic attributes are used as trend constraints, and azimuth data is used to establish sand body distribution models in different periods of the river channel. S4. Classify the reservoir types of single wells, discretize the reservoir type curves of single wells, and establish a reservoir type distribution model in different phases of the channel. S5. Discretize the porosity curve of a single well. Under the reservoir type control, call the azimuth data to establish porosity models under different reservoir types. Finally, use the relationship between permeability and porosity to establish permeability models under different reservoir types.

2. The method for complex river geological modeling based on automatic picking of multiple material source directions as described in claim 1, characterized in that: In S1, depicting the boundaries of different channel phases refers to dividing the depositional phases of a single-well channel based on core and thin section observations from cored wells, combined with well logging curves and sedimentary characteristic analysis, and then depicting the planar distribution map and top and bottom surfaces of the channel for each phase based on seismic attribute response characteristic analysis.

3. The method for complex river geological modeling based on automatic picking of multiple material source directions as described in claim 1, characterized in that: In S1, the correction of the channel range in the seismic interpretation refers to picking the top and bottom positions of the channel in the actual drilled well, and using the root mean square amplitude attribute of the seismic earthquake as the thickness trend for correction, so that the top and bottom surfaces of the channel in the seismic interpretation are consistent with the top surface position of the actual drilled channel.

4. The method for complex river geological modeling based on automatic picking of multiple material source directions as described in claim 1, characterized in that: In S2, processing river boundary data of different periods refers to automatically picking it along the river direction.

5. A method for modeling complex river geological features based on automatic picking of multiple material source directions, as described in claim 4, is characterized in that: The automatic river direction picking specifically refers to obtaining the azimuth line segments of the river flow direction distribution, converting the line segments into azimuth angles through planar drawing, and interpolating them to form surface data representing the azimuth, which is used as the azimuth angle data for modeling.

6. The method for complex river geological modeling based on automatic picking of multiple material source directions as described in claim 1, characterized in that: In S3, the single-well sand group identification result refers to the discrete data of the sand body identified by a single well.

7. The method for complex river geological modeling based on automatic picking of multiple material source directions as described in claim 1, characterized in that: In S4, classifying a single well reservoir type refers to determining the reservoir type by comprehensively considering lithofacies type, physical properties, and pore throat size.

8. The method for complex river geological modeling based on automatic picking of multiple material source directions as described in claim 1, characterized in that: In S4, establishing a reservoir type distribution model within the channel at different stages refers to using the extracted seismic attributes as trend surface constraints and calling azimuth data for random modeling under the sand body distribution model.

9. A method for modeling complex river geological features based on automatic picking of multiple material source directions, as described in claim 1, is characterized in that: In S5, calling azimuth data to establish porosity models for different reservoir types means using sequential Gaussian simulation, with porosity as the first variable and the porosity inversion result as the second variable for simulation.

10. A method for modeling complex river geological features based on automatic picking of multiple material source directions, as described in claim 1, is characterized in that: In S5, establishing permeability models for different reservoir types using the relationship between permeability and porosity means using the established porosity model as the second variable constraint of the permeability model and employing sequential Gaussian simulation.

Images