A method for modeling initial velocity in complex exploration area based on compaction law

Abstract

This invention belongs to the field of oil and gas seismic exploration and discloses a method for initial velocity modeling in complex exploration areas based on compaction laws. The method includes the following steps: establishing a complex structural model and converting it into a parametric model; reconstructing well logging curves using AI to obtain full-depth velocity curves; segmenting the well logging curves according to formations and establishing the relationship between burial depth and velocity, i.e., statistical compaction laws; filling the statistical compaction laws into the parametric model to obtain the initial velocity model; and adjusting the velocity. This invention generates an initial velocity model that conforms to geological laws, improving the accuracy of velocity modeling, especially in areas with dual complexity. It significantly improves the migration effect in complex areas and also increases the efficiency of velocity generation. After establishing an initial velocity model for a dual complexity area, the target line deviation effect was significantly improved. This method is applicable to the establishment of initial velocity models, especially in areas with dual complexity.

Description

Technical Field

[0001] This invention belongs to the field of oil and gas seismic exploration and relates to a seismic signal processing and interpretation technology, specifically a method for modeling the initial velocity of complex exploration areas based on compaction laws. Background Technology

[0002] Pre-stack depth migration is a processing technique for spatially repositioning geological structures. It can address the problem of drastic lateral velocity variations and, compared to pre-stack time migration, provides more accurate repositioning and higher imaging precision. Furthermore, with the continuous advancement of exploration and the increasing computational power, pre-stack depth migration has become a routine processing technique. In the pre-stack depth migration process, the establishment of a velocity model is a crucial step, and the imaging effect is highly sensitive to the migration velocity field. The correctness or accuracy of the velocity model directly affects the migration imaging results.

[0003] Therefore, the core of pre-stack depth migration is the migration velocity. The establishment of the velocity model involves two stages: initial velocity modeling and iterative optimization of the velocity model. A high-precision initial velocity model can reduce the number of velocity iterations, shorten the velocity iteration cycle, and improve modeling efficiency. Traditional methods for generating velocity fields are based on the root mean square velocity field (RMS) using the CVI method to establish an initial velocity model, then migrating, modifying, and iteratively generating the velocity field layer by layer from shallow to deep. However, for complex exploration areas with both near-surface and subsurface structures, as well as areas with low signal-to-noise ratios, using traditional data-driven methods for velocity iteration often results in a velocity field that does not conform to geological patterns, and the migration effect is not ideal, especially for mid-to-deep layers where the migration effect is difficult to optimize. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention aims to provide a method for initial velocity modeling in complex exploration areas based on compaction laws, so as to conveniently and efficiently establish an initial velocity model and obtain a high-precision initial velocity model that conforms to geological laws.

[0005] To achieve the above objectives, the technical solution adopted by this invention is as follows: a method for modeling the initial velocity of complex exploration areas based on compaction laws, comprising the following steps:

[0006] S1. Establish a complex structural model and convert it into a parametric model;

[0007] S2. Based on AI, well curve reconstruction is performed to obtain the full-depth velocity curve;

[0008] S3. Divide the full-depth velocity curve into segments according to the strata and establish the relationship between burial depth and velocity, i.e., statistical compaction law;

[0009] S4. Fill the statistical compaction law into the parametric model to obtain the initial velocity model;

[0010] S5. Speed ​​adjustment: Smooth the initial speed model, offset the target line, adjust the relationship between burial depth and speed according to the offset effect, and repeat steps S4 and S5 until the track gather is flattened to obtain the final initial speed model.

[0011] As a limitation of the present invention, step S1 includes:

[0012] S11, Generate the construction model of the depth domain;

[0013] S12. Incorporate geologists' understanding of the region, make local modifications to the structural model, and generate a complex structural model in the depth domain.

[0014] S13. Convert the complex structural model into a parametric model.

[0015] As a limitation of the present invention: step S11 includes:

[0016] The construction model for generating the depth domain can be any of the following methods:

[0017] a. Use previous deep domain results data;

[0018] b. Convert the interpreted time-domain data to the depth domain using the time offset velocity.

[0019] As a limitation of the present invention, step S13 includes:

[0020] S131. Divide the complex structural model into smaller layers and generate a hierarchical structure model.

[0021] S132. Discretize the complex structural model to the grid position of the work area line, and express the depth and value of each small layer as different linear relationships to obtain the parametric model.

[0022] As a limitation of the present invention, step S2 specifically involves: using GeoEast software, taking the processed velocity as the background velocity, and using an AI method to reconstruct the velocity curve in the logging curve based on existing VSP logging data and seismic data to obtain the full-depth velocity curve.

[0023] As a limitation of the present invention, step S3 includes:

[0024] S31. Using the parametric model level as the boundary, the full-depth velocity curve is divided into layers;

[0025] S32. Filter and smooth the segmented data;

[0026] S33. Extract a portion of the data in equal proportions or equal thicknesses, and statistically analyze the relationship between the well VSP data segmented along the layers and the burial depth in different fault blocks using cross plots. That is, fit the variation function of velocity and burial depth as the formula for the initial velocity of each layer, input and calculate the velocity.

[0027] As a limitation of the present invention, step S4 includes:

[0028] S41. Using the parameter model obtained in step S132, the layer and fault block of each location in space are obtained, and the burial depth of each location in space is calculated.

[0029] S42. Substitute the burial depth into the velocity-burial depth variation function of each layer in step S33 to obtain the velocity value at that point. Calculate and record the velocity values ​​of all points to generate the initial velocity model.

[0030] As a limitation of the present invention, step S5 includes:

[0031] S51. Use GeoEast software to smooth the generated initial velocity model;

[0032] S52. Locate the target line that the user is interested in or that has obvious features, and use GeoEast software to offset the target line;

[0033] S53. Analyze the target line offset results to determine whether the gather is flattened, thereby analyzing whether the speed of each layer is high or low.

[0034] S54. Based on the analysis results, adjust the relationship between velocity and burial depth in step S33, and repeat steps S4 and S51-S53.

[0035] S55. Once the track assembly is roughly leveled, this can be determined as the final initial velocity model. The final velocity model is then used to offset the entire work area.

[0036] By adopting the above technical solution, the beneficial effects achieved by the present invention compared with the prior art are as follows:

[0037] This invention addresses the challenge of conventional approaches to initial velocity modeling in complex dual-region areas. It proposes a method based on geological understanding, using a sequence structure model as a foundation, and generating an initial velocity model under multiple constraints based on compaction laws. This generates an initial velocity model that conforms to geological laws, improving the accuracy of velocity modeling, especially in complex dual-region areas. This significantly improves the migration effect in complex areas and also increases the efficiency of velocity generation. After establishing an initial velocity model for a complex dual-region area, the target line deviation effect was significantly improved. Attached Figure Description

[0038] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0039] Figure 1 This is a flowchart of an embodiment of the present invention;

[0040] Figure 2 This is a graph showing the relationship between the exploration speed and burial depth in an embodiment of the present invention;

[0041] Figure 3 This is a schematic diagram illustrating the distance from a spatial location to the Earth's surface in an embodiment of the present invention;

[0042] Figure 4 This is a schematic diagram of the initial velocity model of the exploration area in an embodiment of the present invention, wherein... Figure 4 (a) is a three-dimensional view of the initial velocity model. Figure 4 (b) is a cross-sectional view of the initial velocity model;

[0043] Figure 5 This is a depth offset effect diagram of the exploration area in an embodiment of the present invention, wherein... Figure 5 (a) is a superimposed cross-section. Figure 5 (b) is the cross section of the Dao Collection. Detailed Implementation

[0044] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the initial velocity modeling method for complex exploration areas based on compaction laws described herein is a preferred embodiment and is only used for illustration and explanation of the present invention, and does not constitute a limitation thereof.

[0045] This embodiment illustrates a method for modeling the initial velocity of complex exploration areas based on compaction laws. It uses a dual-complexity exploration area with both complex near-surface and subsurface structures as an example. Figure 1 As shown, it includes the following steps.

[0046] S1. Establish a high-precision structural model; this step is completed using existing functions in GeoEast software, specifically including:

[0047] S11. Generate the construction model for the depth domain; in the absence of a velocity model, the construction model for generating the depth domain may be any of the following methods.

[0048] a. Use previous deep domain results data;

[0049] b. Convert the interpreted time-domain data to the depth domain using the time offset velocity;

[0050] S12. Generate a complex structural model in the depth domain. When generating the structural model, incorporate the geologist's understanding of the region and make local modifications to the structural model, such as fine-tuning fault and stratigraphic data, modifying or adding / deleting anomalies such as volcanoes and caves, and finally generating a complex structural model.

[0051] S13. Convert the complex structural model into a parametric model; including the following steps:

[0052] S131. Divide the complex structural model into smaller layers and generate a hierarchical structure model.

[0053] S132. Discretize the complex structural model to the grid position of the work area line, and express the depth and value of each small layer as different linear relationships to obtain the parametric model, where equal scale is the scale value, parallel bottom is the distance to the virtual bottom of the large layer, and parallel top is the distance to the virtual top of the large layer.

[0054] S2. Reconstruct the logging curves using AI methods;

[0055] This invention requires statistical compaction patterns, which in turn require statistical well VSP (Vertical Seismic Profile) data. However, velocity modeling often faces challenges such as insufficient VSP data and inadequate logging and micrologging data. To address this issue, this step utilizes existing functions in GeoEast software. The processed velocity is first used as the background velocity. Then, using an AI method based on existing VSP logging and seismic data, the velocity curves in the logging curves are reconstructed to obtain full-depth velocity curves, providing a solid foundation for velocity modeling.

[0056] S3. Divide the full-depth velocity curve into layers according to the strata and establish the relationship between burial depth and velocity, i.e., statistical compaction law.

[0057] The compaction patterns of different layers are generally different, while the compaction patterns of the same layer are roughly the same. Therefore, it is more reasonable to perform stratified statistical analysis when analyzing compaction patterns. This requires first processing the well logging curves into layers. This step specifically includes:

[0058] S31. Using the parametric model level as the boundary, the full-depth velocity curve is divided into layers;

[0059] S32. Filter and smooth the layered data;

[0060] S33. Extract a portion of the data proportionally or at equal thicknesses, and statistically analyze the relationship between the well VSP data used in different fault blocks and the burial depth along the strata using cross-plots. This involves fitting a velocity-depth variation function, which serves as the formula for the initial velocity of each stratum. Input and calculate the velocity. For example... Figure 2As shown, the fitting curves of multiple logging data from the same layer were tested, and the curve with the closest correlation coefficient was found and used as the formula for the initial velocity of each layer.

[0061] It should be noted that, generally, wells are not divided into blocks. However, if a large fault exists, the relationship between velocity and burial depth in the hanging wall and footwall sections of the same fault layer will typically differ significantly. In such cases, wells need to be classified by fault block, and then statistical analysis should be performed layer by layer and block by block. Specifically, the hanging wall and footwall are analyzed separately, which is equivalent to two constraints. First, the hanging wall and footwall are divided. Then, the wells in the hanging wall area are analyzed layer by layer using statistical formulas, and the same applies to the footwall. This avoids data that is too scattered, with each block containing too few wells, thus losing statistical regularity.

[0062] S4. Incorporate statistical compaction patterns into complex structural models.

[0063] Generating a velocity model involves substituting the statistical compaction law function point by point into the velocity volume. Since the law is statistically analyzed in layers and segments, it is first necessary to locate the layer and fault block where the point to be calculated is located before substituting it. This step specifically includes:

[0064] S41. Using the parametric model obtained in step S132, the stratum and fault block of each location in space are obtained. Specifically, the parametric model treats the Earth's surface as a special stratum, so that the distance from each location in the computational space to the Earth's surface can be represented by h ("Z-Surface" in Table 1), such as... Figure 3 As shown;

[0065] S42. Substitute the distance h into the fitted relation of each layer in step S33 to obtain the velocity value of that point. Calculate and record the velocity values ​​of all points, generating a result like... Figure 4 The initial velocity model is shown.

[0066] S5. Speed ​​Adjustment; This step specifically includes:

[0067] S51. The migration algorithm requires a relatively smooth velocity body, but the initial velocity model generated in step S42 changes rather sharply at the boundary between layers and faults. Therefore, GeoEast software is used in this step to smooth the generated initial velocity model.

[0068] S52. Locate the target line that the user is interested in or that has obvious features, and then use GeoEast software to offset the target line.

[0069] S53. Analyze the target line offset results to determine whether the gather is flattened, thereby analyzing whether the speed of each layer is high or low.

[0070] S54. Based on the analysis results, adjust the relationship between velocity and burial depth in step S33, and repeat steps S4 and S51-S53. Table 1 shows the adjusted relationship between velocity and burial depth.

[0071] Table 1. Relationship between velocity and burial depth in a complex exploration area.

[0072]

[0073] S55, such as Figure 5 As shown in (b), once the track gather is roughly leveled, this can be determined as the final initial velocity model. The entire work area is then offset using this final initial velocity model. Figure 5 (a) It can be seen that the figure has certain data axes, that is, there are continuous seismic traces with similar amplitudes at the same depth, which is a significant improvement compared to other previous migration results.

[0074] This invention provides a high-precision velocity model for migration imaging that is more consistent with geological patterns, which can effectively reduce the number of iterations of subsequent velocity models and significantly improve the efficiency and accuracy of velocity modeling.

Claims

1. A method for modeling the initial velocity of complex exploration areas based on compaction laws, characterized in that, Includes the following steps: S1. Establish a complex structural model and convert it into a parametric model; S2. Reconstruct the logging curves using AI methods to obtain the full-depth velocity curves; S3. Divide the full-depth velocity curve into segments according to the strata and establish the relationship between burial depth and velocity, i.e., statistical compaction law; S4. Fill the statistical compaction law into the parametric model to obtain the initial velocity model; S5. Speed ​​adjustment: Smooth the initial speed model, offset the target line, adjust the relationship between burial depth and speed according to the offset effect, and repeat steps S4 and S5 until the track gather is flattened to obtain the final initial speed model.

2. The method for modeling the initial velocity of complex exploration areas based on compaction laws according to claim 1, characterized in that, Step S1 includes: S11, Generate the construction model of the depth domain; S12. Incorporate geologists' understanding of the region, make local modifications to the structural model, and generate a complex structural model in the depth domain. S13. Convert the complex structural model into a parametric model.

3. The method for modeling the initial velocity of complex exploration areas based on compaction laws according to claim 2, characterized in that, Step S11 includes: The construction model for generating the depth domain can be any of the following methods: a. Use previous deep domain results data; b. Convert the interpreted time-domain data to the depth domain using the time offset velocity.

4. The method for modeling the initial velocity of complex exploration areas based on compaction laws according to claim 3, characterized in that, Step S13 includes: S131. Divide the complex structural model into smaller layers and generate a hierarchical structure model. S132. Discretize the complex structural model to the grid position of the work area line, and express the depth and value of each small layer as different linear relationships to obtain the parametric model.

5. The method for modeling the initial velocity of complex exploration areas based on compaction laws according to claim 4, characterized in that, Step S2 specifically involves using GeoEast software to take the processed velocity as the background velocity and, through an AI method, reconstructing the velocity curve in the logging curve based on existing VSP logging data and seismic data to obtain the full-depth velocity curve.

6. The method for modeling the initial velocity of complex exploration areas based on compaction laws according to claim 5, characterized in that, Step S3 includes: S31. Using the parametric model level as the boundary, the full-depth velocity curve is divided into layers; S32. Filter and smooth the segmented data; S33. Extract a portion of the data in equal proportions or equal thicknesses, and statistically analyze the relationship between the well VSP data segmented along the layers and the burial depth in different fault blocks using cross plots. That is, fit the variation function of velocity and burial depth as the formula for the initial velocity of each layer, input and calculate the velocity.

7. The method for modeling the initial velocity of complex exploration areas based on compaction laws according to claim 6, characterized in that, Step S4 includes: S41. Using the parameter model obtained in step S132, the layer and fault block of each location in space are obtained, and the burial depth of each location in space is calculated. S42. Substitute the burial depth into the velocity-burial depth variation function of each layer in step S33 to obtain the velocity value at that point. Calculate and record the velocity values ​​of all points to generate the initial velocity model.

8. The method for modeling the initial velocity of complex exploration areas based on compaction laws according to claim 7, characterized in that, Step S5 includes: S51. Use GeoEast software to smooth the generated initial velocity model; S52. Locate the target line that the user is interested in or that has obvious features, and use GeoEast software to offset the target line; S53. Analyze the target line offset results to determine whether the gather is flattened, thereby analyzing whether the speed of each layer is high or low. S54. Based on the analysis results, adjust the relationship between velocity and burial depth in step S33, and repeat steps S4 and S51-S53. S55. Once the track assembly is roughly leveled, this can be determined as the final initial velocity model. The final velocity model is then used to offset the entire work area.

Images