A method for statistical curve constrained inversion of desert surface velocities
By fitting micrologging statistical curves to constrain inversion in desert regions, the problems of severe dune undulation and insufficient ray density were solved, resulting in a more accurate surface velocity model, improved static correction and pre-stack depth migration effects, and reduced costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA NAT PETROLEUM CORP
- Filing Date
- 2022-06-24
- Publication Date
- 2026-05-19
AI Technical Summary
In desert regions, the accuracy of first-arrival tomography inversion of desert surface velocity is insufficient due to the severe dune undulations and insufficient ray density. Furthermore, the qualitative constraint inversion accuracy is low when microlog data is unavailable, failing to meet the accuracy requirements of static correction and pre-stack depth migration.
By collecting micrologging data, fitting regional statistical fitting constraint curves, forming constraint files, using exponential polynomials to fit the velocity-depth relationship, establishing an initial velocity model, and applying constraints during the inversion process to eliminate outliers and form a reasonable surface velocity model.
It improves the accuracy of surface velocity models in desert regions, enhances static correction, improves the imaging accuracy of pre-stack depth migration, and reduces data acquisition costs.
Smart Images
Figure QLYQS_2 
Figure QLYQS_3 
Figure QLYQS_4
Abstract
Description
Technical Field
[0001] This invention belongs to the field of seismic data processing technology, and relates to first-arrival tomographic static correction technology, specifically a method for statistical curve-constrained inversion of desert surface velocity. Background Technology
[0002] Establishing a shallow velocity model is a crucial step in seismic data processing, as it relates to the calculation of static corrections to the base plane and subsequent pre-stack depth migration. For horizontal subsurface media, horizontal stacking can recover the subsurface structural morphology. However, for complex subsurface structures with undulations, horizontal stacking cannot reflect the true subsurface structural morphology. Pre-stack depth migration is an effective method for recovering subsurface morphology, and the imaging accuracy of depth migration is highly dependent on the accuracy of velocities, especially those in the shallow subsurface.
[0003] While elevation static correction methods can roughly calculate the static correction amount at the base surface, their accuracy is no longer sufficient for current production needs. Field near-surface velocity survey data can improve the calculation accuracy of static correction, but dense deployment is not feasible due to construction costs. First-arrival tomographic static correction, utilizing first-arrival information, can calculate the static correction amount at the base surface relatively accurately at a very low cost. Typically, the first-arrival tomographic inversion method is used to establish an accurate shallow surface velocity model.
[0004] However, in large desert regions, where sand dunes are highly undulating, the regular grid is insufficient to describe ray paths when using first-arrival tomography to invert near-surface models. Furthermore, the insufficient density of rays penetrating the surface results in low surface ray density coverage, leading to inadequate surface information and significantly impacting the accuracy of first-arrival tomography inversion of shallow velocities under these surface conditions. To address this issue, micrologging can be used to constrain the velocity model during inversion, thereby improving accuracy. When using micrologging for constraint, data from a specific point is typically embedded into the initial velocity model established during inversion, thus constraining the model and obtaining a more accurate velocity model for the vicinity of that point.
[0005] However, many logging lines lack relevant micrologging data due to issues such as the age of acquisition or cost, making it impossible to perform constrained inversion. For areas without micrologging data, a segmented inversion constraint method based on offsets is typically used to improve inversion accuracy. However, these methods only provide qualitative constraints, and their accuracy is far lower than that of micrologging technology-constrained inversion. Summary of the Invention
[0006] The purpose of this invention is to provide a method for statistical curve constraint inversion of surface velocity in desert areas, in order to solve the problem that the accuracy of shallow velocity models cannot be improved through quantitative constraints in areas lacking micrologging information.
[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0008] A method for statistically constrained inversion of desert surface velocities includes the following steps:
[0009] Collect micrologging results from the work area and adjacent areas, extract the time-depth relationship of micrologging, and fit to obtain the regional micrologging statistical fitting constraint curve;
[0010] A preliminary tomographic velocity model was established, a stable low-velocity layer velocity was selected, and the thickness of the dune layer was obtained.
[0011] Create a constraint file whose constraint depth does not exceed the thickness of the dune layer;
[0012] Perform a conventional tomographic velocity model inversion and use a constraint file to constrain the inversion to obtain a reasonable surface velocity model.
[0013] As a limitation, the extraction of the time-depth relationship of micrologging specifically includes:
[0014] Using time as the variable, a data space is generated with depth as the observation value:
[0015] [dp(t1),dp(t2),……,dp(tn)]
[0016] Where t1, t2, ..., tn are the observation times, and dp(tn) is the depth corresponding to time tn.
[0017] As a further limitation, the fitting to obtain the regional micrologging statistical fitting constraint curve specifically includes:
[0018] Transform time-depth relationship data into velocity-depth relationship data, based on...
[0019]
[0020] Outliers with speeds lower than the center speed are removed;
[0021] according to
[0022]
[0023] Outliers with speeds greater than the center speed are removed.
[0024] Among them vel hi This refers to the micrologging velocity information that needs to be judged; the right side of the inequality represents the center velocity; ε1 and ε2 are threshold parameters, with ε1 being negative and ε2 being positive; n is the total number of micrologging data at this depth; vel j It is the j-th micrologging velocity information at that depth;
[0025] We use exponential polynomials to fit the velocity-depth statistics and establish a system of equations:
[0026]
[0027] Where A is the coefficient of the quadratic term, B is the coefficient of the linear term, and C is the constant term; v1, v2, ... v n For the observed velocities, dp1, dp2, ..., dp n For observation depth;
[0028] Solving the coefficients A, B, and C of this system of equations yields the statistical fitting constraint curve for the regional micrologging.
[0029] As a second limitation, the establishment of the preliminary chromatography velocity model specifically includes:
[0030] A preliminary tomographic velocity model was established using conventional tomographic inversion methods.
[0031] As a further limitation, the selection of a stable low-speed layer velocity specifically includes:
[0032] In the preliminary chromatography velocity model, a stable low-velocity layer with a velocity of 600–1000 m / s is extracted.
[0033] As a third limitation, the thickness of the dune layer obtained specifically includes:
[0034] The depth interface is formed based on the selected stable low-velocity layer velocity, and the distance from the deepest low-velocity layer interface to the ground surface is the thickness of the dune layer.
[0035] As a further limitation, the constraint file whose constraint depth does not exceed the thickness of the dune layer specifically includes:
[0036] Based on the dune layer thickness, the portion of the statistical fitting constraint curve of the regional micrologging that exceeds the dune layer thickness is removed, thus obtaining the constraint file where the constraint depth does not exceed the dune layer thickness.
[0037] By adopting the above technical solution, the technical progress achieved by this invention compared with the prior art is as follows:
[0038] This invention provides a method for statistically constrained inversion of surface velocities in desert areas. Through long-term work and data research, the inventors discovered that in desert regions, a curve representing the near-surface velocity variation within a certain area can be obtained using statistical fitting. Applying this fitted curve to constrain near-surface velocity inversion effectively solves the surface inversion constraint problem in areas lacking micrologging data, significantly reducing drilling and micrologging data acquisition costs. Alternatively, the fitted curve can be used to replace anomalous micrologging information, thereby more accurately inverting the surface velocity model and obtaining a more precise and reasonable surface velocity model, providing a good guarantee for subsequent static correction and pre-stack depth migration work.
[0039] This invention provides a method for statistical curve constraint inversion of desert surface velocity, which is applicable to constraining the velocity model established by near-surface analysis in desert areas, thereby obtaining a more accurate surface velocity model. Attached Figure Description
[0040] Figure 1 This is a velocity-depth statistical chart formed by micrologging data in the work area and adjacent areas in the embodiment;
[0041] Figure 2 This is a velocity-depth statistical chart generated from micrologging data after removing outliers in the embodiment;
[0042] Figure 3 This is the statistical fitting constraint curve of the regional micrologging obtained after fitting in the embodiment;
[0043] Figure 4 The thickness of the dune layer is obtained from a stable low-velocity layer velocity in the embodiment.
[0044] Figure 5 This is the initial velocity model with added regional micrologging statistical fitting constraint curves in the embodiment;
[0045] Figure 6 This is the velocity model obtained by constraint inversion in the embodiment;
[0046] Figure 7 illustrates the calculation of static correction using a velocity model in this embodiment. Figure 7a For single-shot records of velocity models without micrologging constraint inversion, Figure 7b Single-shot records for static corrections of velocity models derived using constrained inversion via regional micrologging statistical fitting constraint curves;
[0047] Figure 8 illustrates the application of a velocity model for pre-stack depth migration in this embodiment. Figure 8a This is a depth migration profile of a velocity model that was not inverted using micrologging constraints. Figure 8bThis is a depth migration profile for a velocity model constrained by a constraint curve obtained using regional micrologging statistical fitting. Detailed Implementation
[0048] The present invention will be further described in detail below through specific embodiments. It should be understood that the described embodiments are only for explaining the present invention and do not limit the present invention.
[0049] The embodiment uses the statistical curve constraint inversion method to establish a velocity model for desert surface velocity.
[0050] (I) Establishing a velocity model
[0051] This invention was applied in a work area in the Tarim Basin. Combining data from 873 micro-logging wells in adjacent areas, constraint curves were statistically fitted to establish a shallow constraint model. This model was then applied to the calculation of shallow velocity models with a total logging line length of 1348 km. The method accurately established a near-surface velocity model. The specific steps are as follows:
[0052] S1. Collect the logging results of 873 micrologging wells in a certain work area and adjacent areas in the Tarim Basin, extract the time-depth relationship of the micrologging wells, and generate a data space with time as the variable and depth as the observation value:
[0053] [dp(t1),dp(t2),……,dp(tn)]
[0054] Where t1, t2, ..., tn are the observation times, and dp(tn) is the depth corresponding to time tn;
[0055] S2. Transform the time-depth relationship data into velocity-depth relationship data using a differencing method, forming a structure like... Figure 1 The velocity-depth statistics shown are based on
[0056]
[0057] Outliers with speeds significantly lower than the center speed are removed.
[0058] according to
[0059]
[0060] Outliers with speeds significantly exceeding the center speed are removed, resulting in the following: Figure 2 The velocity-depth statistics are shown below;
[0061] Among them vel hiThis refers to the micro-logging velocity information that needs to be judged; the right side of the inequality represents the central velocity; ε1 and ε2 are threshold parameters. A smaller value is chosen when the overall velocity convergence is good, and a larger value is chosen when the overall velocity convergence is large. The convergence is determined by the average value of the overall velocity and the variance of each point. ε1 is negative, and ε2 is positive. The larger the absolute value, the more sample points are retained; the smaller the absolute value, the fewer sample points are retained. In this embodiment, ε1 is -0.2 and ε2 is 0.2; n is the total number of micro-logging information at this depth. In this embodiment, n is 873; vel j It is the j-th micrologging velocity information at that depth;
[0062] S3. Using exponential polynomials Figure 2 By fitting the velocity-depth statistics, a system of equations is established:
[0063]
[0064] Where A is the coefficient of the quadratic term, B is the coefficient of the linear term, and C is the constant term; v1, v2, ... v n For the observed velocities, dp1, dp2, ..., dp n For observation depth;
[0065] Solve the system of equations, where A is -0.0025, B is 6, and C is 400. The fitted result is as follows: Figure 3 The area shown is a statistical fitting constraint curve for micrologging.
[0066] S4. Using conventional tomographic inversion methods, establish a preliminary tomographic velocity model, such as... Figure 4 As shown in the figure, different colors represent different velocity ranges. A stable low-velocity layer is extracted, meaning a velocity layer whose color does not change abruptly at the same depth. Within different work areas, the low-velocity layer velocity is determined based on the thickness of the thin sand layer. The thicker the sand layer, the higher its maximum velocity. Generally, the maximum velocity is 600–1000 m / s and should not exceed 1000 m / s. In this embodiment, the selected low-velocity layer velocity is 784 m / s. A depth interface is formed based on the selected stable low-velocity layer velocity, as shown... Figure 4 As shown, the distance from the deepest interface of the low-velocity layer to the surface is the thickness of the dune layer, which is approximately 30–40 m.
[0067] S5. Since the maximum depth of micrologging is 20m, which is much smaller than the thickness of the dune layer, a constraint file with a constraint depth not exceeding the thickness of the dune layer can be directly generated and loaded into inversion software such as KLSeis. Figure 5 As shown, regional micrologging statistical fitting curve constraints are added to the initial velocity model, and the inversion is constrained by the constraint file during the inversion process;
[0068] S6. Through constraint inversion, we obtain the following... Figure 6 The surface velocity model shown.
[0069] (II) Applying the velocity model to verify its rationality
[0070] The static correction was calculated using the velocity model obtained from the inversion, and the seismic data was statically corrected. The results are shown in Figure 7.
[0071] As shown in Figure 7, compared with the traditional unconstrained inversion method, the surface velocity model obtained by using the constrained curve of the regional micrologging statistical fitting is more reasonable. The surface velocity model obtained by applying this method to calculate the static correction amount greatly improves the static correction effect in desert areas.
[0072] The velocity model was embedded into the overall offset velocity model, and pre-stack depth migration was performed. The result is shown in Figure 8.
[0073] As shown in Figure 8, embedding the constrained inversion near-surface velocity model into the depth migration velocity profile yields a new velocity field. Using this new velocity field for depth migration improves the imaging accuracy of the depth migration and provides more effective data for subsequent seismic data processing.
Claims
1. A method for statistically curve-constrained inversion of desert surface velocity, characterized in that, Includes the following steps: Collect micrologging results from the work area and adjacent areas, extract the time-depth relationship of micrologging, and fit to obtain the regional micrologging statistical fitting constraint curve; A preliminary tomographic velocity model was established, a stable low-velocity layer velocity was selected, and the thickness of the dune layer was obtained. Create a constraint file whose constraint depth does not exceed the thickness of the dune layer; Perform a conventional tomographic velocity model inversion and use a constraint file to constrain the inversion, thus obtaining a reasonable surface velocity model. The fitting process to obtain the regional micrologging statistical fitting constraint curve specifically includes: Transform time-depth relationship data into velocity-depth relationship data, based on... Outliers with speeds lower than the center speed are removed; according to Outliers with speeds greater than the center speed are removed. Among them vel hi This refers to the micrologging velocity information that needs to be judged; the right side of the inequality represents the center velocity; ε1 and ε2 are threshold parameters, with ε1 being negative and ε2 being positive; n is the total number of micrologging data at this depth; vel j It is the j-th micrologging velocity information at that depth; We use exponential polynomials to fit the velocity-depth statistics and establish a system of equations: Where A is the coefficient of the quadratic term, B is the coefficient of the linear term, and C is the constant term; v1, v2, ... v n For the observed velocities, dp1, dp2, ..., dp n For observation depth; Solving the coefficients A, B, and C of this system of equations yields the statistical fitting constraint curve for the regional micrologging.
2. The method for statistical curve-constrained inversion of desert surface velocity according to claim 1, characterized in that, The extraction of the time-depth relationship of micrologging specifically includes: Using time as the variable, a data space is generated with depth as the observation value: [dp(t1),dp(t2),……,dp(tn)] Where t1, t2, ..., tn are the observation times, and dp(tn) is the depth corresponding to time tn.
3. The method for statistically constrained inversion of desert surface velocity according to claim 1 or 2, characterized in that, The establishment of the preliminary tomographic velocity model specifically includes: A preliminary tomographic velocity model was established using conventional tomographic inversion methods.
4. The method for statistical curve-constrained inversion of desert surface velocity according to claim 3, characterized in that, The selection of a stable low-speed layer velocity specifically includes: In the preliminary chromatography velocity model, a stable low-velocity layer with a velocity of 600~1000 m / s is extracted.
5. A method for statistically constrained inversion of desert surface velocity according to any one of claims 1, 2, and 4, characterized in that, The thickness of the dune layer obtained specifically includes: The depth interface is formed based on the selected stable low-velocity layer velocity, and the distance from the deepest low-velocity layer interface to the ground surface is the thickness of the dune layer.
6. The method for statistically constrained inversion of desert surface velocity according to claim 5, characterized in that, The constraint file that forms a constraint depth not exceeding the thickness of the dune layer specifically includes: Based on the dune layer thickness, the portion of the statistical fitting constraint curve of the regional micrologging that exceeds the dune layer thickness is removed, thus obtaining the constraint file where the constraint depth does not exceed the dune layer thickness.