A combined deep domain modeling method for buried hill interior

Abstract

The application provides a buried hill inside combined deep domain modeling method: shallow layer gather coverage times body cannot accurately invert near-surface velocity by using reflection wave information, and shallow layer high-precision near-surface velocity model is established by using full-area first arrival information and well information. Secondly, the layered inversion iterative model based on data driving is used on the basis of the shallow layer velocity model. Finally, the local deep layer stratum with weak lateral velocity change is further optimized to the base depth domain velocity model by referring to the pre-stack time migration velocity model, the high signal-to-noise ratio and low signal-to-noise ratio velocity difference is eliminated, and the small-scale global inversion is carried out again, the high-frequency depth domain velocity model is further optimized, and finally a complete buried hill inside combined depth domain modeling method is formed.

Description

Technical Field

[0001] This invention relates to the field of oil and gas geophysical seismic data processing, and in particular to a joint depth domain modeling method for buried hill interiors. Background Technology

[0002] As oil and gas exploration and development enters a stage of refined potential tapping, the requirements for imaging the complex interior of buried hills have increased, making the establishment of detailed depth-domain velocity models crucial. The interior of buried hills is mostly composed of old strata with poor stratification. Velocity characteristics are characterized by significant differences in velocity between the overlying strata and the surrounding rocks. Furthermore, the presence of faults leads to multiple reflections and refractions of seismic rays during excitation between strata and faults, resulting in extremely complex ray paths. These factors all contribute to low signal-to-noise ratios in seismic data, difficulty in velocity analysis, and challenges in migration imaging. Therefore, detailed depth-domain velocity modeling of buried hill interiors is a vital step in seismic data imaging.

[0003] To address the complex structural features of buried hills and other imaging characteristics, scholars both domestically and internationally have proposed numerous pre-stack depth-domain velocity modeling methods. Currently, the main depth-domain velocity modeling methods include:

[0004] The wellpoint direct interpolation method primarily utilizes abundant well information within the exploration area to establish a depth-domain velocity model through techniques such as kriging interpolation or thin-plate spline interpolation. This technique is suitable for areas with stable geological conditions, relatively gentle structures, and small lateral velocity variations. Secondly, based on the initial velocity model, common imaging point gathers (CIPs) generated through pre-stack depth migration are used. Velocity adjustments are then made to these CIPs using a data-driven tomographic inversion imaging method, thereby eliminating and correcting the velocity model. Finally, the depth-domain velocity model is established through multiple iterations and repeated adjustments.

[0005] The Dix formula is used to convert the pre-stack time-migration velocity model into a depth-domain layer velocity model, which is applicable to moderately complex geological targets. However, this is limited to horizontally layered strata without abrupt velocity changes between strata; otherwise, direct time-depth conversion will introduce uncertainties and cause significant errors. Secondly, common imaging point gathers generated by depth migration are used based on the initial velocity model. Isotropic tomographic velocity inversion and anisotropic tomographic velocity inversion techniques are then used to iteratively update and correct the initial velocity model of the shallow-to-deep integrated depth domain model, resulting in a high-precision depth-domain velocity model and further improving the imaging quality of depth migration.

[0006] Existing depth migration velocity modeling methods suffer from two main shortcomings. Firstly, they rely on existing well data and stratigraphic information to establish depth-domain velocity models. However, the limited exploration depth and scarcity of well data within buried hills make it difficult to establish accurate depth-domain velocity models based solely on well and formation information. Secondly, current velocity modeling primarily employs data-driven tomographic inversion, which places certain demands on data quality. Factors such as steep dip angles in buried hill formations, low signal-to-noise ratios, velocity inversion issues in deep formations, and low velocity model accuracy directly lead to low migration imaging gather quality, affecting the acquisition of residual curvature. This results in low reliability of data-driven tomographic inversion iterative techniques, making it impossible to obtain accurate high-precision velocity models. Summary of the Invention

[0007] In view of the above problems, the present invention is proposed to provide a joint depth domain modeling method for buried hill interiors that overcomes or at least partially solves the above problems.

[0008] According to one aspect of the present invention, a joint depth domain modeling method for buried hill interiors is provided, comprising:

[0009] Preprocess the collected Dao collection data to obtain preprocessed Dao collection data;

[0010] Based on the preprocessed gather data, an inversion model is performed to obtain a very shallow, high-precision velocity model.

[0011] The aforementioned ultra-shallow high-precision velocity model is used to establish a shallow depth domain velocity model based on well logging information within the work area;

[0012] Using the shallow velocity model mentioned above, geological and stratum velocity information of the work area are referenced, and a mid-depth velocity model is established by filling the mid-deep layers with regional constant velocity, thus obtaining the initial buried hill internal depth domain velocity model.

[0013] Fusion velocity model to establish full-depth overall initial depth domain velocity model;

[0014] The full-depth overall initial depth domain velocity model is subjected to offset inverse transformation to obtain a high signal-to-noise ratio data depth domain velocity model.

[0015] The high signal-to-noise ratio data depth domain velocity model is optimized by using the dip angle time difference correction method for the deep-domain velocity model of the deep-domain strata inside the buried hill with low signal-to-noise ratio data to obtain the steep-dip angle depth domain velocity model of the buried hill.

[0016] Based on the steep inclination angle depth domain velocity model of the buried hill and the location of the local deep imaging difference, the depth domain layer velocity model is obtained;

[0017] The depth domain velocity model is then subjected to a small-scale global inversion to obtain a complete joint depth domain velocity model for the underground chamber of a buried hill.

[0018] Optionally, the step of inverting and modeling based on the preprocessed gather data to obtain a high-precision velocity model for extremely shallow layers specifically includes:

[0019] Based on the preprocessed gather data, obtain the initial arrival information for the entire region;

[0020] An initial velocity model is established based on the initial arrival information for the entire region.

[0021] Calculate the difference between the initial arrival time and the forward travel time;

[0022] The initial velocity model is updated based on the differences.

[0023] A high-precision velocity model for extremely shallow layers is obtained based on the initial velocity model and the ray density iterative inversion.

[0024] Optionally, the step of using the ultra-shallow high-precision velocity model to establish a shallow depth domain velocity model from the logging information in the work area specifically includes:

[0025] Obtain well logging information within the work area;

[0026] The well logging information is converted into velocity curves using the aforementioned ultra-shallow high-precision velocity model.

[0027] A shallow depth domain velocity model is established using the Kriging difference method to represent the velocity curve.

[0028] Optionally, the fusion velocity model, establishing a full-depth overall initial depth domain velocity model, specifically includes:

[0029] The ultra-shallow high-precision velocity model, the shallow depth domain velocity model, and the mid-depth domain velocity model are fused to obtain a fused model;

[0030] The velocity differences between layers in the fusion model are eliminated by horizontal and vertical small smoothing to remove velocity outliers, the depth domain velocity model is optimized, and a full-depth overall initial depth domain velocity model is established.

[0031] Optionally, the step of performing an offset inverse transformation on the overall initial depth domain velocity model to obtain a high signal-to-noise ratio data depth domain velocity model specifically includes:

[0032] Perform large-scale mesh depth offset on the full-depth overall initial depth domain velocity model;

[0033] The remaining curvature of the depth domain gather after offset is picked up, and the dip field constraint is applied in combination with the overall initial depth domain velocity model of the full depth. The velocity model is then updated by travel time layer inversion.

[0034] Obtain a high signal-to-noise ratio data depth domain velocity model.

[0035] Optionally, the optimization of the high signal-to-noise ratio data depth domain velocity model using the dip angle time-difference correction method for the deep-dip strata within the buried hill with low signal-to-noise ratio data to obtain the steep-dip depth domain velocity model specifically includes:

[0036] The dip angle time difference correction method is used to correct the dip angle of the deep domain velocity model of the high signal-to-noise ratio data for the deep domain strata inside the buried hill with low signal-to-noise ratio.

[0037] By combining geological strata and stratum velocity information, the depth domain velocity model of the buried hill with steep dip angle is optimized, and the depth domain velocity model of the buried hill with steep dip angle is obtained.

[0038] Optionally, obtaining the depth-domain layer velocity model based on the steep-dip depth-domain velocity model of the buried hill and the location of the local deep imaging difference specifically includes:

[0039] Based on the accuracy of the velocity model of steeply dipping strata inside the buried hill, the time-domain velocity model is modified by referring to the pre-stack time migration method to obtain the optimized time-domain velocity model for the location of local deep imaging differences.

[0040] The layer velocity model is converted to a depth domain layer velocity model using the DIX formula. Pre-stack depth migration and stacking checks are performed to determine if the model meets the requirements for gather leveling and stacking imaging improvement. If it does, the layer velocity model is considered accurate. If not, the time domain velocity model is modified and converted to a depth domain layer velocity model.

[0041] Optionally, the step of performing a small-scale global inversion of the depth domain velocity model to obtain a complete joint depth domain velocity model of the buried hill interior specifically includes:

[0042] The established depth domain velocity model is then subjected to small-scale global inversion to further optimize the interlayer local high-frequency depth domain layer velocity, thereby obtaining an optimized depth domain velocity model.

[0043] The drilling depth was used to fine-tune the optimized depth domain velocity model, and finally a complete combined depth domain velocity model for buried hill interior was obtained.

[0044] This invention provides a method for joint depth domain modeling of buried hill interiors, comprising: preprocessing collected gather data to obtain preprocessed gather data; performing inversion modeling based on the preprocessed gather data to obtain a very shallow high-precision velocity model; using the very shallow high-precision velocity model to establish a shallow depth domain velocity model based on well logging information within the work area; using the shallow velocity model and referencing geological and layer velocity information of the work area, establishing a mid-depth depth domain velocity model for mid-deep layers using regional constant velocity filling, to obtain an initial buried hill interior depth domain velocity model; and fusing the velocity models to establish a comprehensive full-depth model. An initial depth-domain velocity model is generated. This model is then migrated and inverted to obtain a high signal-to-noise ratio (SNR) depth-domain velocity model. This high SNR model is further optimized using dip-time time-of-flight correction for low SNR deep-hill strata with steep dip angles, resulting in a deep-domain velocity model for steep dip angles within the buried hill. Based on this steep dip angle deep-domain velocity model and the location of local deep-layer imaging differences, a depth-domain layer velocity model is obtained. Finally, a small-scale global inversion is performed on this depth-domain velocity model to obtain a complete joint depth-domain velocity model for the buried hill interior. After global tomographic inversion, the final pre-stack depth migration profile shows good agreement with the structural trends of the depth-domain velocity model. This establishes a complete method for joint depth-domain modeling of buried hill interiors.

[0045] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 A flowchart illustrating a joint depth domain modeling method for buried hill interiors provided in this embodiment of the invention;

[0048] Figure 2 This invention provides a high-precision depth-domain velocity model for extremely shallow layers in an embodiment of the invention.

[0049] Figure 3 A shallow multi-well combined depth domain velocity model;

[0050] Figure 4 To integrate and optimize the depth domain velocity model;

[0051] Figure 5Pre-stack depth offset stacking;

[0052] Figure 6 The velocity spectrum after differential correction at tilt angle;

[0053] Figure 7 The optimized depth-domain velocity model;

[0054] Figure 8 The final pre-stack depth migration profile (left) and depth domain velocity model (right). Detailed Implementation

[0055] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0056] The terms "comprising" and "having," and any variations thereof, in the specification, embodiments, claims, and drawings of this invention are intended to cover non-exclusive inclusion, such as including a series of steps or units.

[0057] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0058] This invention proposes a joint depth domain modeling method for the interior of buried hills:

[0059] First, due to the shallow gather coverage, it is impossible to accurately invert near-surface velocity using reflected wave information. Therefore, a high-precision near-surface velocity model for shallow layers is established using first arrival information and well information from the entire area.

[0060] Secondly, a data-driven layered inversion iterative model is adopted based on the shallow velocity model. High signal-to-noise ratio data are used to obtain a more accurate depth-domain velocity model through inversion iteration. Low signal-to-noise ratio steep-dip data inside buried hills are used to enhance the signal-to-noise ratio of steep-dip data through dip angle time-difference correction method. At the same time, the depth-domain velocity model of steep-dip buried hills is optimized by combining layer velocity and geological information.

[0061] Finally, the basement depth domain velocity model was further optimized for the pre-stack time migration velocity model of the strata with weak local deep lateral velocity variation. To eliminate the velocity difference between high and low signal-to-noise ratios, a small-scale global inversion was performed to further optimize the high-frequency depth domain velocity model, ultimately forming a complete method for joint depth domain modeling of buried hill interiors.

[0062] like Figure 1 As shown, this invention is a method for joint deep domain modeling of buried hill interiors. The processing flow includes the following steps:

[0063] Step 1: Import the preprocessed gather data, establish an initial velocity model using first-arrival information for the entire area, update the velocity model by the difference between first-arrival travel time and forward travel time, and obtain a high-precision velocity model for the very shallow layer based on the velocity model and ray density iterative inversion. Figure 2 As shown.

[0064] Step 2: Based on the extremely shallow high-precision velocity model, after converting the well logging information within the work area into velocity curves, the discrete velocity curves are interpolated using Kriging interpolation to establish a shallow velocity model, such as... Figure 3 As shown.

[0065] Step 3: Based on the shallow depth domain velocity model, and combined with geological and layer velocity information of the work area, a depth domain velocity model is established for the middle and deep layers by filling with regional constant velocities, and finally the initial buried hill internal depth domain velocity model is obtained.

[0066] Step 4: The fused velocity model includes the extremely shallow velocity from the initial arrival inversion, the shallow velocity established from well-connected information, and the mid-to-deep velocity model filled with constant velocity. These three are fused to establish a comprehensive initial depth-domain velocity model. The velocity differences between layers in the fused model are eliminated by horizontal and vertical smoothing to remove velocity outliers, further optimizing the depth-domain velocity model, such as... Figure 4 As shown.

[0067] Step 5: The optimized velocity model is first subjected to large-scale grid depth migration. By picking up the residual curvature of the depth domain gathers after migration, and combining it with the velocity model for dip field constraints, the velocity model is updated using travel-time layered inversion. For high signal-to-noise ratio data, migration imaging is improved after 2-3 iterations using a data-driven tomographic inversion mode. Figure 5 As shown, the speed model update for high signal-to-noise ratio data in the deep domain is accurate.

[0068] Step 6: Based on the velocity model retrieved from the high signal-to-noise ratio data, a dip angle time-difference correction method is applied to the steeply dipped strata within the buried hill with low signal-to-noise ratio data. This improves the signal-to-noise ratio of the steeply dipped strata while simultaneously obtaining accurate velocity spectrum information for these strata. Figure 6 As shown, the velocity model in the depth domain of the steep dip angle of the buried hill is optimized by combining geological strata and stratum velocity information.

[0069] Step 7: Based on the accuracy of the steeply dipping formation velocity model within the buried hill, and targeting locations with poor local deep imaging, modify the time-domain velocity model using pre-stack time migration methods. The optimized time-domain velocity model is then converted to a depth-domain layer velocity model using the DIX formula. Pre-stack depth migration gather and stack checks are performed to determine if the model meets the requirements for gather leveling and stack imaging improvement. If it does, the layer velocity model is considered accurate; otherwise, continue modifying the time-domain velocity model to convert it back to a depth-domain layer velocity model. Figure 7 As shown.

[0070] Step 8: Perform small-scale global inversion on the established depth-domain velocity model to further optimize the inter-layer local high-frequency depth-domain layer velocities. Then, fine-tune the velocity model locally using drilling depth to finally obtain a complete combined depth-domain velocity model for buried hill interiors. After global tomographic inversion, the final pre-stack depth migration profile and the structural trend of the depth-domain velocity model show good agreement. Figure 8 As shown.

[0071] Beneficial effects:

[0072] 1. An initial near-surface model is established using first arrival data from the entire area. A high-precision near-surface model is then established through iterative inversion using cyclotron wave tomography. Simultaneously, a shallow well-connected velocity model is established by combining well information within the work area to ensure the accuracy of the shallow velocity model.

[0073] 2. Based on the accuracy of the shallow velocity model, a data-driven layered inversion method was adopted for the entire region's data. High signal-to-noise ratio (SNR) data were used to obtain a relatively accurate depth-domain velocity model through inversion iterations. Low SNR data were used to enhance the SNR of steeply dipping strata data through dip angle time-difference correction. At the same time, the depth-domain velocity model of steeply dipping strata was optimized by combining layer velocity information and geological information.

[0074] 3. Further optimize the basement depth domain velocity model for the pre-stack time migration velocity model of strata with weak local deep lateral velocity variations.

[0075] 4. To eliminate the velocity difference between high signal-to-noise ratio and low signal-to-noise ratio data, a global inversion is performed again to further optimize the high-frequency velocity model, ultimately forming a complete method for joint deep domain modeling of buried hill interiors.

[0076] The above specific embodiments further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A joint depth domain modeling method for buried hill interiors, characterized in that, The modeling method includes: Preprocess the collected Dao collection data to obtain preprocessed Dao collection data; Based on the preprocessed gather data, an inversion model is performed to obtain a very shallow, high-precision velocity model. The aforementioned ultra-shallow high-precision velocity model is used to establish a shallow depth domain velocity model based on well logging information within the work area; Using the shallow depth domain velocity model, the geological and layer velocity information of the work area is referenced, and a mid-depth depth domain velocity model is established by filling the mid-deep layers with regional constant velocity, thus obtaining the initial buried hill internal depth domain velocity model. Fusion velocity model to establish full-depth overall initial depth domain velocity model; The full-depth overall initial depth domain velocity model is subjected to offset inverse transformation to obtain a high signal-to-noise ratio data depth domain velocity model. The high signal-to-noise ratio data depth domain velocity model is optimized by using the dip angle time difference correction method for the deep-domain velocity model of the deep-domain strata inside the buried hill with low signal-to-noise ratio data to obtain the steep-dip angle depth domain velocity model of the buried hill. Based on the steep inclination angle depth domain velocity model of the buried hill and the location of the local deep imaging difference, the depth domain layer velocity model is obtained; The depth domain velocity model is then subjected to a small-scale global inversion to obtain a complete joint depth domain velocity model for the underground chamber of a buried hill.

2. The joint depth domain modeling method for buried hill interiors according to claim 1, characterized in that, The step of inverting and modeling based on the preprocessed gather data to obtain a very shallow, high-precision velocity model specifically includes: Based on the preprocessed gather data, obtain the initial arrival information for the entire region; An initial velocity model is established based on the initial arrival information for the entire region. Calculate the difference between the initial arrival time and the forward travel time; The initial velocity model is updated based on the differences. A high-precision velocity model for extremely shallow layers is obtained based on the initial velocity model and the ray density iterative inversion.

3. The joint depth domain modeling method for buried hill interiors according to claim 1, characterized in that, The specific steps of establishing a shallow depth domain velocity model from well logging information within the work area using the ultra-shallow high-precision velocity model include: Obtain well logging information within the work area; The well logging information is converted into velocity curves using the aforementioned ultra-shallow high-precision velocity model. A shallow depth domain velocity model is established using the Kriging difference method to represent the velocity curve.

4. The joint depth domain modeling method for buried hill interiors according to claim 1, characterized in that, The fusion velocity model, which establishes a full-depth overall initial depth domain velocity model, specifically includes: The ultra-shallow high-precision velocity model, the shallow depth domain velocity model, and the mid-depth domain velocity model are fused to obtain a fused model; The velocity differences between layers in the fusion model are eliminated by horizontal and vertical small smoothing to remove velocity outliers, the depth domain velocity model is optimized, and a full-depth overall initial depth domain velocity model is established.

5. The joint depth domain modeling method for buried hill interiors according to claim 1, characterized in that, The step of performing an offset inverse transformation on the full-depth overall initial depth domain velocity model to obtain a high signal-to-noise ratio data depth domain velocity model specifically includes: Perform large-scale mesh depth offset on the full-depth overall initial depth domain velocity model; The remaining curvature of the depth domain gather after offset is picked up, and the dip field constraint is applied in combination with the overall initial depth domain velocity model of the full depth. The velocity model is then updated by travel time layer inversion. Obtain a high signal-to-noise ratio data depth domain velocity model.

6. The joint depth domain modeling method for buried hill interiors according to claim 1, characterized in that, The optimization of the high signal-to-noise ratio data depth domain velocity model for low signal-to-noise ratio buried hill interior steep-dip strata using the dip angle time-difference correction method to obtain a steep-dip strata depth domain velocity model for buried hills specifically includes: The dip angle time difference correction method is used to correct the dip angle of the deep domain velocity model of the high signal-to-noise ratio data for the deep domain strata inside the buried hill with low signal-to-noise ratio. By combining geological strata and stratum velocity information, the depth domain velocity model of the buried hill with steep dip angle is optimized, and the depth domain velocity model of the buried hill with steep dip angle is obtained.

7. The joint depth domain modeling method for buried hill interiors according to claim 1, characterized in that, The specific steps of obtaining the depth-domain layer velocity model based on the steep inclination angle depth-domain velocity model of the buried hill and the location of local deep imaging differences include: Based on the accuracy of the velocity model of steeply dipping strata inside the buried hill, the time-domain velocity model is modified by referring to the pre-stack time migration method to obtain the optimized time-domain velocity model for the location of local deep imaging differences. The layer velocity model is converted to a depth domain layer velocity model using the DIX formula. Pre-stack depth migration and stacking checks are performed to determine if the model meets the requirements for gather leveling and stacking imaging improvement. If it does, the layer velocity model is considered accurate. If not, the time domain velocity model is modified and converted to a depth domain layer velocity model.

8. The joint depth domain modeling method for buried hill interiors according to claim 1, characterized in that, The step of performing a small-scale global inversion of the depth-domain velocity model to obtain a complete joint depth-domain velocity model for the interior of a buried hill specifically includes: The established depth domain velocity model is then subjected to small-scale global inversion to further optimize the interlayer local high-frequency depth domain layer velocity, thereby obtaining an optimized depth domain velocity model. The drilling depth was used to fine-tune the optimized depth domain velocity model, and finally a complete combined depth domain velocity model for buried hill interior was obtained.

Images