A method, apparatus, and device for seismic inversion suitable for non-vertical wells

Abstract

The application provides a seismic inversion method, device and equipment suitable for non-vertical wells, and belongs to the field of oil and gas exploration.The inversion method comprises the following steps: obtaining seismic data, well trajectory data and logging data in a target region, determining the range of a target layer section for inversion, and performing standardization processing on the corresponding logging data; calibrating the position of the well trajectory on the seismic data body, determining a constraint well and a verification well, and generating the spatial coordinates of the seismic data corresponding to the well trajectory; determining the value window of the seismic data and the value window of the logging parameter, and making a seismic-logging parameter matching database; according to the principle of the matching tracking method, extracting the logging parameter in the data pair with the largest matching degree to form the logging parameter data body for inversion.The application utilizes the spatial position coordinates of the well trajectory on the seismic data, and uses the information of the inclined well section and the horizontal well section, so that the shortcoming of the conventional inversion method, i.e., slow inversion speed, is overcome.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration, and more specifically, to a seismic inversion method suitable for non-vertical wells, an apparatus suitable for seismic inversion of non-vertical wells, a computer device for implementing the seismic inversion method for non-vertical wells, and a computer-readable storage medium for implementing the seismic inversion method for non-vertical wells. Background Technology

[0002] As oil and gas exploration and development progresses, the geological conditions of exploration targets are becoming increasingly demanding, reducing the applicability of some traditional technologies. Simultaneously, advancements in computer performance and digital simulation technology have spurred the development of numerous new methods. Seismic inversion plays a significant role in the interpretation of seismic data in oil and gas exploration, used to interpret stratigraphic lithology, reservoir distribution, and differences in physical properties. The fundamental purpose of seismic inversion is to utilize the propagation patterns of seismic waves in the subsurface medium to image the structure and physical properties of subsurface rock strata through data acquisition, processing, and interpretation, thereby providing crucial information for exploration and development.

[0003] The two main categories of existing post-stack seismic data inversion methods are well-constrained model inversion methods and well-unconstrained inversion methods. Each category has further developed into various sub-methods based on different algorithms. Well-unconstrained methods convert seismic waveform data into wave impedance data or other physical property parameters through analytical expressions, such as band-limited recursive inversion and sparse pulse inversion. Well-constrained model inversion methods incorporate well data constraints during the inversion process. For example, before inversion, an initial model of the entire area is generated using well logging data and seismic horizon interpolation. The final inversion result is obtained by modifying the initial model to a certain extent, such as broadband constrained inversion and geostatistical inversion; or by learning the mapping relationship between seismic data and well data and then applying this relationship to the entire seismic data to obtain the inversion result, such as neural network inversion.

[0004] Currently, accurate and rapid seismic inversion processes for non-vertical wells are difficult to achieve. For example, patent document CN103033846A, published on April 10, 2013, describes a seismic inversion system and method controlled by geological facies, including a geological facies establishment module to establish the geological facies of the target layer at the well location; a seismic facies establishment module to establish the seismic facies on the plane of the target layer; a geological facies constraint establishment module to establish geological facies constraints; a random inversion module to select the number of constraint wells and perform random inversion controlled by geological facies; an analysis module to analyze and interpret the inversion data volume and further subdivide the geological facies; and an output module to iterate repeatedly to obtain and output the three-dimensional inversion data volume. This invention overcomes the shortcomings of existing inversion methods that do not consider the geological sedimentary background and patterns within the inversion area, and has the advantage of accurately describing the reservoir. However, this invention has poor practicality for non-vertical wells and fails to consider the role of well trajectory and seismic data in the inversion process. A patent document published on February 18, 2022, entitled "A Method and Apparatus for Automatically Inverting Focal Mechanism Solutions," with publication number CN114063153A, describes a method and apparatus for automatically inverting focal mechanism solutions. The method involves: calculating the theoretical P-wave arrival time and theoretical S-wave arrival time for each station based on an earthquake catalog and a one-dimensional velocity-structure model; extracting the truncated earthquake waveform data from the earthquake waveform data based on the earthquake catalog, theoretical P-wave arrival time, and theoretical S-wave arrival time; extracting the phase information from the truncated earthquake waveform data using a phase identification model and an initial motion polarity identification model to obtain the phase category and the initial motion polarity of the P-wave; and inverting the focal mechanism based on the earthquake catalog, phase category, and initial motion polarity of the P-wave to obtain the focal mechanism solution. This invention, by utilizing a phase identification model and an initial motion polarity identification model, can significantly reduce the workload of inverting the focal mechanism and improve the inversion speed of the focal mechanism solution. However, this method only utilizes seismic data and does not use well logging data, resulting in a significant difference between the inversion results and the actual situation.

[0005] Both of the above methods are commonly used inversion methods at present. However, they are complex inversion processes, cannot quickly and accurately obtain relatively precise inversion results, and cannot meet the timeliness requirements of field exploration. Summary of the Invention

[0006] The purpose of this invention is to address at least one of the aforementioned shortcomings of the prior art. For example, this invention provides a seismic inversion method applicable to non-vertical wells to overcome the defects of the prior art, such as complex seismic inversion processes for non-vertical wells, large amounts of required seismic data, and inaccurate results.

[0007] To achieve the above objectives, this invention provides a seismic inversion method applicable to non-vertical wells. The method includes the following steps: acquiring seismic data, well trajectory data, and logging data from multiple wells within a target area; determining the target layer range for inversion and inversion parameters; performing regional standardization on the logging data corresponding to the inversion parameters; marking the well trajectory positions of all wells on the seismic data volume, simultaneously identifying constraint wells and verification wells, and generating the spatial coordinates of the seismic data corresponding to the well trajectories. The constraint wells include vertical, inclined, and horizontal wells within the target area; determining the seismic data value window and the logging parameter value window; extracting the seismic data and logging parameters traversed by the well trajectory along the target layer of the constraint well according to the seismic data value window and the logging parameter value window, and creating a seismic-logging parameter matching database; and, according to the matching pursuit principle, matching the seismic data to be inverted point by point with the seismic-logging parameter data pairs in the seismic-logging parameter matching database, extracting the logging parameters from the data pair with the highest matching degree to form the inverted logging parameter data volume.

[0008] In an exemplary embodiment of a seismic inversion method applicable to non-vertical wells according to the present invention, the seismic inversion method may further include: after creating a seismic-well logging parameter matching database, given one or more interpretation horizons of the seismic data volume as constraints, dividing the seismic-well logging parameter matching database into multiple seismic-well logging parameter pair matching sub-databases corresponding to each horizon based on these horizons and the spatial location of the seismic-well logging parameter data pairs.

[0009] In an exemplary embodiment of a seismic inversion method applicable to non-vertical wells according to the present invention, the seismic data can be seismic waveform data or wave impedance data, and the seismic waveform data can include three-dimensional time-domain seismic waveform data and three-dimensional depth-domain seismic waveform data.

[0010] In an exemplary embodiment of a seismic inversion method applicable to non-vertical wells according to the present invention, for three-dimensional depth domain seismic waveform data, the wells involved can be directly calibrated to the corresponding positions in the seismic waveform data based on the well trajectory; for three-dimensional time domain seismic waveform data, if a velocity field exists, the three-dimensional time domain seismic waveform data is converted into three-dimensional depth domain seismic waveform data, and then the well trajectory is calibrated; if no velocity field exists, time-depth conversion is performed based on the sonic logging curve to generate a seismic composite record, and the composite record is compared with the three-dimensional time domain seismic waveform data to adjust the time-depth relationship and calibrate the well trajectory on the three-dimensional time domain seismic waveform data.

[0011] In an exemplary embodiment of a seismic inversion method for non-vertical wells according to the present invention, the regional standardization process includes: removing wells with significantly abnormal value ranges, wherein the number of abnormal wells removed is less than 10%, and statistically analyzing the value range of the logging curves of the remaining wells [L].min L max This method corrects curves that are outside the range to the range by modifying the mean or the range of maximum and minimum values.

[0012] In an exemplary embodiment of a seismic inversion method applicable to non-vertical wells according to the present invention, for three-dimensional time-domain seismic waveform data, the seismic data value window is determined to be a time domain of 0.01s to 0.1s; for three-dimensional depth-domain seismic waveform data, the seismic data value window is determined to be a formation thickness of 20m to 200m; and the well logging parameter value window is determined to be a depth range of 0.125m to 10m.

[0013] In an exemplary embodiment of the seismic inversion method applicable to non-vertical wells according to the present invention, the method for determining the data pair with the highest degree of matching is as follows: the seismic data in the seismic-well logging parameter matching database is normalized using the L2 norm normalization method to obtain the first seismic vector. Centered on the seismic data points to be inverted, a seismic data segment of a predetermined time length is obtained as the second seismic vector. The first vector was normalized using the L2 norm normalization method to obtain the third seismic vector. Calculate the dot product of the first and third earthquake vectors, and determine the data pair with the largest dot product result as the data pair with the highest matching degree.

[0014] In another aspect, the present invention provides a seismic inversion device suitable for non-vertical wells. The inversion device may include: a parameter standardization unit, a data calibration unit, a database construction unit, and a matching inversion unit. The parameter standardization unit is configured to acquire seismic data, well trajectory data, and logging data from multiple wells within a target area, determine the target layer range for inversion, and perform regional standardization processing on the logging data corresponding to the inversion parameters. The data calibration unit is connected to the parameter standardization unit and is configured to calibrate the well trajectory positions of all wells within the target area on the seismic data volume, simultaneously determine constraint wells and verification wells, and generate the seismic data spatial coordinates corresponding to the well trajectories. The constraint wells include… The system includes vertical, deviated, and horizontal wells in the target area. The database construction unit, connected to the data calibration unit, is configured to determine the seismic data value window and the well logging parameter value window. Along the target layer segment of the constrained well trajectory, it extracts the seismic data and well logging parameters traversed by the well trajectory according to the seismic data value window and the well logging parameter value window, respectively, and creates a seismic-well logging parameter matching database. The matching inversion unit, connected to the database construction unit, is configured to match the seismic data to be inverted point by point with the seismic-well logging parameter data pairs in the seismic-well logging parameter matching database according to the matching pursuit method principle, and extract the well logging parameters from the data pair with the highest degree of matching to form the inverted well logging parameter data body.

[0015] In another aspect, the present invention provides a computer device, which may include: a processor; a memory storing a computer program, wherein when the computer program is executed by the processor, the seismic inversion method applicable to non-vertical wells as described above is implemented.

[0016] In another aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the seismic inversion method applicable to non-vertical wells as described above.

[0017] Compared with the prior art, the beneficial effects of the present invention may include at least one of the following:

[0018] (1) By obtaining the spatial coordinates of the well trajectory on the specific seismic data, this invention can utilize the information of the inclined well section and the horizontal well section to establish the constraint information for inversion, thus overcoming the shortcomings of conventional inversion methods that only use the information of the vertical well section.

[0019] (2) This invention does not need to consider the inversion wavelet problem and the initial model construction problem, and its algorithm is simpler and easier to operate than conventional inversion methods;

[0020] (3) Under the premise of accurate well trajectory calibration, the present invention can also include the logging curves of horizontal wells and highly deviated wells in the inversion, which improves the accuracy of the inversion results in locations with well constraints and improves the consistency between the inversion results and the wells. Attached Figure Description

[0021] The above and other objects and / or features of the present invention will become clearer from the following description taken in conjunction with the accompanying drawings, in which:

[0022] Figure 1 This diagram illustrates a method for obtaining "seismic-logging parameter" data pairs along an inclined well trajectory, representing an exemplary embodiment of a seismic inversion method applicable to non-vertical wells according to the present invention.

[0023] Figure 2 The diagram illustrates an exemplary embodiment of the seismic inversion method for non-vertical wells according to the present invention, showing an extrapolation matching database query and value retrieval.

[0024] Figure 3 A flowchart illustrating an exemplary embodiment of the seismic inversion method for non-vertical wells according to the present invention is shown.

[0025] Figure 4 A GR inversion well profile is shown as an exemplary embodiment of a seismic inversion method for non-vertical wells according to the present invention.

[0026] Figure 5A schematic diagram of an exemplary embodiment of an apparatus for seismic inversion of non-vertical wells according to the present invention is shown.

[0027] Figure 6 A schematic diagram of a computer device structure is shown, illustrating an exemplary embodiment of a seismic inversion device for non-vertical wells according to the present invention.

[0028] Explanation of reference numerals in the attached figures:

[0029] 100 - Seismic inversion device suitable for non-vertical wells; 110 - Parameter standardization unit; 120 - Data calibration unit; 130 - Database construction unit; 140 - Matching inversion unit; 200 - Computer equipment; 210 - Memory; 220 - Processor. Detailed Implementation

[0030] The following will describe in detail, with reference to the accompanying drawings and specific embodiments, a seismic inversion method, apparatus and device applicable to non-vertical wells according to the present invention.

[0031] This invention provides a seismic inversion method applicable to non-vertical wells.

[0032] In an exemplary embodiment of a seismic inversion method applicable to non-vertical wells according to the present invention, the identification method may include the following steps:

[0033] S100. Acquire seismic data, well trajectory data, and well logging data from multiple wells within the target area, determine the target layer range for inversion, and inversion parameters, and perform regional standardization processing on the well logging data corresponding to the inversion parameters.

[0034] Specifically, the seismic data can be seismic waveform data or wave impedance data. The seismic waveform data includes three-dimensional time-domain seismic waveform data and three-dimensional depth-domain seismic waveform data. For three-dimensional depth-domain seismic waveform data, the relevant wells are directly calibrated onto the corresponding positions in the seismic waveform data based on the well trajectory. For three-dimensional time-domain seismic waveform data, if a velocity field exists, the three-dimensional time-domain seismic waveform data is converted into three-dimensional depth-domain seismic waveform data, and then the well trajectory is calibrated. If no velocity field exists, time-depth conversion is performed based on the sonic logging curve to generate a synthetic seismic record. The synthetic record is compared with the three-dimensional time-domain seismic waveform data, the time-depth relationship is adjusted, and the well trajectory is calibrated onto the three-dimensional time-domain seismic waveform data.

[0035] Specifically, regional standardization processing may include: removing wells with significantly abnormal value ranges, with the number of abnormal wells removed being less than 10%, and statistically analyzing the value ranges of the logging curves of the remaining wells [L]. min L max This method corrects curves that are outside the range to the range by modifying the mean or the range of maximum and minimum values.

[0036] Specifically, the seismic inversion method also includes: after creating a seismic-well logging parameter matching database, given one or more interpretation horizons of the seismic data volume as constraints, and based on these horizons and the spatial location of the seismic-well logging parameter data pairs, dividing the seismic-well logging parameter matching database into multiple seismic-well logging parameter data pair matching sub-databases corresponding to each horizon.

[0037] S200. Mark the well trajectory positions of all wellheads in the target area on the seismic data volume, and simultaneously determine the constraint wells and verification wells, and generate the seismic data spatial coordinates corresponding to the well trajectories. The constraint wells include vertical wells, inclined wells and horizontal wells in the target area.

[0038] S300. Determine the seismic data acquisition window and the well logging parameter acquisition window. Along the target layer segment of the constrained well trajectory, extract the seismic data and well logging parameters traversed by the well trajectory according to the seismic data acquisition window and the well logging parameter acquisition window, and create a seismic-well logging parameter matching database.

[0039] Specifically, for three-dimensional time-domain seismic waveform data, the seismic data value window is determined to be a time domain of 0.01s to 0.1s; for three-dimensional depth-domain seismic waveform data, the seismic data value window is determined to be a formation thickness of 20m to 200m; and the well logging parameter value window is determined to be a depth range of 0.125m to 10m.

[0040] S400. According to the matching pursuit method, the seismic data to be inverted is matched point by point with the seismic-well logging parameter data pairs in the seismic-well logging parameter matching database, and the well logging parameters in the data pair with the highest matching degree are extracted to form the inverted well logging parameter data body.

[0041] Specifically, the method for determining the data pair with the highest degree of matching is as follows:

[0042] The seismic data in the seismic-well logging parameter matching database were normalized using the L2 norm normalization method to obtain the first seismic vector.

[0043] Centered on the seismic data points to be inverted, a seismic data segment of a predetermined time length is obtained as the second seismic vector. The first vector was normalized using the L2 norm normalization method to obtain the third seismic vector.

[0044] Calculate the dot product of the first and third earthquake vectors, and determine the data pair with the largest dot product result as the data pair with the highest matching degree.

[0045] Another aspect of the present invention provides a seismic inversion device suitable for non-vertical wells.

[0046] In an exemplary embodiment of the seismic inversion device for non-vertical wells of the present invention, the seismic inversion device includes: a parameter standardization unit, a data calibration unit, a database construction unit, and a matching inversion unit.

[0047] The parameter standardization unit is configured to acquire seismic data, well trajectory data, and well logging data from multiple wells within the target area, determine the target layer range for inversion, and perform regional standardization processing on the well logging data corresponding to the inversion parameters.

[0048] The data calibration unit is connected to the parameter standardization unit and is configured to calibrate the well trajectory positions of all wellheads in the target area on the seismic data volume, while determining the constraint wells and verification wells, and generating the seismic data spatial coordinates corresponding to the well trajectories. The constraint wells include vertical wells, inclined wells and horizontal wells in the target area.

[0049] The database construction unit is connected to the data calibration unit and is configured to determine the seismic data value window and the well logging parameter value window. Along the target layer of the well trajectory of the constrained well, the seismic data and well logging parameters traversed by the well trajectory are extracted according to the seismic data value window and the well logging parameter value window, respectively, and a seismic-well logging parameter matching database is created.

[0050] The matching inversion unit is connected to the database construction unit and is configured to match the seismic data to be inverted point by point with the seismic-well logging parameter data pairs in the seismic-well logging parameter matching database according to the matching pursuit method principle, and extract the well logging parameters from the data pair with the highest degree of matching to form the inverted well logging parameter data body.

[0051] An exemplary embodiment of another aspect of the present invention also provides a computer-readable storage medium storing a computer program. The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to execute the seismic inversion method applicable to non-vertical wells according to the present invention. The computer-readable recording medium is any data storage device capable of storing data read by a computer system. Examples of computer-readable recording media include: read-only memory, random access memory, read-only optical disk, magnetic tape, floppy disk, optical data storage device, and carrier waves (such as data transmission via the Internet through wired or wireless transmission paths).

[0052] An exemplary embodiment of another aspect of the present invention also provides a computer device. The computer device includes a processor and a memory. The memory stores a computer program. The computer program is executed by the processor, causing the processor to execute the computer program according to the present invention, which is applicable to a seismic inversion method for non-vertical wells.

[0053] To better understand the present invention, the following description, in conjunction with the accompanying drawings and examples, further clarifies the content of the present invention, but the content of the present invention is not limited to the following embodiments.

[0054] Example 1

[0055] This example illustrates a seismic inversion method applicable to non-vertical wells, used when the work area contains vertical, deviated, or horizontal wells, and the curves of these wells are required to participate in the inversion.

[0056] In this example, the well trajectory locations are first marked on the seismic data volume to determine the required inversion data. Values ​​from the seismic data volume traversed by the well trajectory and the corresponding well logging curves are extracted within a specific window. These pairs are then used to construct a "seismic-well logging parameter" set. We can refer to each "seismic-well logging parameter" data pair as an "atom," and the collection of multiple atoms is called an "atom matching database." During extrapolation, based on the matching pursuit method, the data from the input data volume is queried point by point in the database. The matching well logging parameters are extracted from the database based on the degree of matching, ultimately forming the inverted parameter data volume. This processing method allows well logging curves from horizontal and highly deviated wells to participate in the inversion, improving the accuracy of the inversion results at locations with well constraints. When a sufficient number of wells participate in the inversion, good results are also achieved at locations without well constraints.

[0057] This example demonstrates the application of this invention in the P oilfield in Ecuador. The target layer in the study area is the TS layer, and the exploration objective is the channel sand body within the layer. Due to the low gamma (GR) value of the sand body, seismic data was used to invert the GR data volume.

[0058] This example may specifically include the following steps:

[0059] Step 1: Prepare the input seismic data volume and well logging data, determine the data range of the target layer for inversion, and the parameters that need to be inverted.

[0060] Specifically, the input data volume in this example is three-dimensional data, but the input data can also be other types of data such as wave impedance data or seismic waveform data volumes, as well as other model data.

[0061] In this example, the input data volume is three-dimensional time-domain seismic waveform data, and the target layer range for inversion is from TA layer to TS layer.

[0062] Step 2: Standardize the region of the well curve corresponding to the parameters to be inverted.

[0063] Specifically, the standardization method for well curve regionalization can employ commonly used standardization methods, by statistically analyzing the value range of most curves [L]. minL max [Or data distribution, curves that deviate significantly from this range or distribution can be corrected to the range of most curve values ​​by modifying the mean or the range of maximum and minimum values.]

[0064] Step 3: Mark the required wells and their trajectories on the data volume.

[0065] Specifically, the 3D time-domain seismic waveform data in this example is time-domain seismic data without a velocity field. A well-seismic calibration method is used, employing sonic logging curves (DT) for time-depth conversion to generate a synthetic seismic record. The synthetic record is then compared with the time-domain seismic data, and the time-depth relationship is adjusted to calibrate the well trajectory onto the seismic data. Sixteen wells were selected as constraint wells, including both deviated and vertical wells. Well PRH-25 is a validation well and is not involved in the inversion.

[0066] Specifically, when the input data is depth-domain seismic data, the relevant wells can be directly calibrated onto the corresponding locations in the seismic data based on the well trajectory. For time-domain seismic data, if a velocity field is available, the seismic data can be converted to depth-domain data using a time-depth conversion, and then the well trajectory can be calibrated. If no velocity field is available, the sonic logging curve DT can be used for time-depth conversion to generate a synthetic seismic record. By comparing the synthetic record with the time-domain seismic data, the time-depth relationship can be adjusted, and the well trajectory can be calibrated onto the seismic data.

[0067] Step 4: Given the size parameters of the atoms, extract data from the data volume and well logging data according to the inverted layers and pair them to form atoms, and create a "seismic-well logging parameter" atom matching database (i.e., seismic-well logging parameter matching database).

[0068] Specifically, the seismic endpoint of the atomic library can be represented as Normalization was performed using the L2 norm normalization method to obtain normalized atomic seismic endpoint data. At the atomic logging data terminal, the data within a 4-meter depth range of point p are compressed into a single value using the averaging method.

[0069] Specifically, the size parameter of an atom is given as follows: For the seismic data side of the atom, if the data is in the time domain, the size is the longitudinal time length W, and the value range (i.e., the seismic data value window) is generally [0.01, 0.1], in seconds; if the data is in the depth domain, the size is the formation thickness D, and the value range (i.e., the seismic data value window) is generally [20, 200], in meters. For the well logging data side of the atom, the size is the depth range, and the range (i.e., the well logging parameter value window) is generally [0.125, 10], in meters.

[0070] Figure 1The diagram illustrates the method for obtaining "seismic-logging parameter" data pairs along the trajectory of a deviated well according to the present invention. In the diagram, W represents the value range of the seismic data value window, D represents the value range of the logging parameter value status, thin dashed lines represent seismic traces, thick dashed lines represent constrained well trajectories, thin wavy lines represent seismic waveforms, thick wavy lines represent seismic waveforms taken from the seismic data value window, and rectangular boxes represent logging parameter value windows.

[0071] Specifically, the method for creating the "seismic-well logging parameter" atom matching database is as follows: Along the target layer of the well trajectory, a discrete data sequence of "depth (md) - data volume spatial coordinates (p)" is generated, where p is represented by (line number, trace number, time / depth). In this sequence, the same spatial coordinate should appear only once to ensure no duplicate sampling. Then, based on this coordinate sequence and the given atom size, the seismic and well logging data at each coordinate point are extracted to form atoms.

[0072] Step 5: Given one or more interpretation layers of the data volume as constraints, divide the atom matching database into multiple sub-matching databases according to these layers and the location of the atoms, to correspond to each layer.

[0073] Specifically, the top boundary of the TS layer can be used as the constraint layer, the segment to be inverted can be divided into two segments, and then according to the relationship between the time of the center point of the atomic space coordinate p in the database and the constraint layer, they can be divided into the set corresponding to the two segments.

[0074] Step 6: Extrapolate using the input data volume and the atomic matching database. During extrapolation, based on the matching pursuit method, the data in the input data volume is queried point by point in the database. The matching logging parameters in the database are extracted according to the degree of matching, and finally the inverted parameter data volume is formed.

[0075] Specifically, the method for calculating the degree of matching may include: during extrapolation, for the data point in the seismic data that needs to be extrapolated, taking that point as the center, obtaining a seismic data segment with a length of 50 milliseconds, represented by a vector. Similarly, L2 norm normalization is used to obtain Based on the relationship between its center point's time and the TS layer, determine its layer family and which subset of the atomic database to search for matching atoms. Then, based on the matching pursuit algorithm, calculate... The seismic endpoint of each atom in the subset The dot product. Find the... The atom at the seismic endpoint with the largest dot product is used as the inversion result for that point, taking the value from its logging endpoint.

[0076] Figure 2This diagram illustrates the extrapolation matching database query and value retrieval methods of the present invention. Figure 2 In the middle, from left to right, the curves represent seismic traces. The seismic traces are sent to the 'seismic-well logging parameter' matching database for matching queries. Matching data pairs are found, and finally, the well logging data is extracted to form the final inversion result.

[0077] Step 7: Traverse all data points that need to be inverted and repeat step 6 to obtain the final inverted data volume.

[0078] Based on the above steps, draw the flowchart as follows: Figure 3 As shown.

[0079] After completing steps 1-7 above, the GR data volume of the target layer was calculated. Figure 4 The images show the cross-sectional views of wells PRH-25 and PRH-27. Darker colors represent high GR values, while lighter colors represent low GR values, ranging from 0 to 150 API.

[0080] Among them, the verification well PRH-25 is a vertical well, and the constraint well PRH-27 is an inclined well. Figure 4 The GR curve on the wellbore was plotted along the well trajectory, with color bars consistent with the profile. The bottom of the profile is the top interface of the Bmts layer, and the second layer from the bottom is the top interface of the TS layer. Our target is the sand body between these two layer interfaces, i.e., the lighter-colored area in the inversion is the favorable target. Figure 4 The white arrow points to the mudstone layer in the target formation. It can be seen that, besides the GR profile next to the PRH-27 well trajectory corresponding to the GR values ​​of the logging curve, the high GR bands next to the PRH-25 well trajectory also correspond well with the well logging data. Furthermore, the overall appearance of the profile is consistent with geological understanding, with natural lateral variations and high resolution; the inversion results also show good correspondence in other formations.

[0081] The results of this example demonstrate that the present invention can create an atomic database and perform inversion using time-domain seismic waveform data, vertical well data, and deviated well data; and the conformity of the verified wells demonstrates the effectiveness of the extrapolation method.

[0082] Example 2

[0083] like Figure 5 As shown, the seismic inversion device 100 suitable for non-vertical wells includes: a parameter standardization unit 110, a data calibration unit 120, a database construction unit 130, and a matching inversion unit 140.

[0084] The parameter standardization unit 110 is configured to acquire seismic data, well trajectory data and logging data of multiple wells within the target area, determine the target layer range for inversion, and perform regional standardization processing on the logging data corresponding to the inversion parameters.

[0085] The data calibration unit 120 is connected to the parameter standardization unit 110 and is configured to calibrate the well trajectory positions of all wells on the seismic data volume, while determining the constraint wells and verification wells, and generating the seismic data spatial coordinates corresponding to the well trajectories. The constraint wells include vertical wells, inclined wells and horizontal wells in the target area.

[0086] The database construction unit 130 is connected to the data calibration unit 120 and is configured to determine the seismic data value window and the well logging parameter value window. Along the target layer of the well trajectory of the constrained well, the seismic data and well logging parameters traversed by the well trajectory are extracted according to the seismic data value window and the well logging parameter value window, respectively, and a seismic-well logging parameter matching database is created.

[0087] The matching inversion unit 140 is connected to the database construction unit 130 and is configured to match the seismic data to be inverted point by point with the seismic-well logging parameter data pairs in the seismic-well logging parameter matching database according to the matching pursuit method principle, and extract the well logging parameters from the data pair with the highest matching degree to form the inverted well logging parameter data body.

[0088] Example 3

[0089] like Figure 6 As shown, a computer device 200 includes a memory 210 and a processor 220. The memory 210 stores a computer program, which, when executed by the processor, implements the seismic inversion method applicable to non-vertical wells as described above.

[0090] Although the present invention has been described above in conjunction with exemplary embodiments and accompanying drawings, those skilled in the art should understand that various modifications can be made to the above embodiments without departing from the spirit and scope of the claims.

Claims

1. A seismic inversion method applicable to non-vertical wells, characterized in that, The earthquake inversion method includes the following steps: Acquire seismic data, well trajectory data, and well logging data from all wellheads within the target area; determine the target layer range for inversion and the inversion parameters; and perform regional standardization processing on the well logging data corresponding to the inversion parameters. The well trajectory positions of all wellheads within the target area are marked on the seismic data volume, and the constraint wells and verification wells are determined. The seismic data spatial coordinates corresponding to the well trajectories are generated. The constraint wells include vertical wells, inclined wells and horizontal wells in the target area. Determine the seismic data acquisition window and the well logging parameter acquisition window. Along the target layer segment of the constrained well trajectory, extract the seismic data and well logging parameters traversed by the well trajectory according to the seismic data acquisition window and the well logging parameter acquisition window respectively, and create a seismic-well logging parameter matching database. According to the matching pursuit method, the seismic data to be inverted is matched point by point with the seismic-well logging parameter data pairs in the seismic-well logging parameter matching database, and the well logging parameters in the data pair with the highest matching degree are extracted to form the inverted well logging parameter data body. The method for determining the data pair with the highest matching degree is as follows: The seismic data in the seismic-well logging parameter matching database were normalized using the L2 norm normalization method to obtain the first seismic vector. ; Centered on the seismic data points to be inverted, a seismic data segment of a predetermined time length is obtained as the second seismic vector. The second vector was normalized using the L2 norm normalization method to obtain the third seismic vector. ; Calculate the dot product between the first and third earthquake vectors, and determine the data pair with the largest dot product result as the data pair with the highest matching degree; The seismic inversion method further includes: after creating a seismic-well logging parameter matching database, given one or more interpretation horizons of the seismic data volume as constraints, and based on these horizons and the spatial location of the seismic-well logging parameter data pairs, dividing the seismic-well logging parameter matching database into multiple seismic-well logging parameter data pair matching sub-databases corresponding to each horizon.

2. The seismic inversion method according to claim 1, characterized in that, The seismic data is seismic waveform data or wave impedance data, and the seismic waveform data includes three-dimensional time-domain seismic waveform data and three-dimensional depth-domain seismic waveform data.

3. The seismic inversion method according to claim 2, characterized in that, For three-dimensional depth domain seismic waveform data, the relevant wells are directly calibrated onto the corresponding positions in the seismic waveform data based on the well trajectory. For three-dimensional time domain seismic waveform data, if a velocity field exists, the three-dimensional time domain seismic waveform data is converted into three-dimensional depth domain seismic waveform data, and then the well trajectory is calibrated. If no velocity field exists, time-depth conversion is performed based on the sonic logging curve to generate a seismic composite record. The composite record is compared with the three-dimensional time domain seismic waveform data, the time-depth relationship is adjusted, and the well trajectory is calibrated onto the three-dimensional time domain seismic waveform data.

4. The seismic inversion method according to claim 1, characterized in that, The regional standardization process includes: removing wells with significantly abnormal value ranges, with the number of abnormal wells removed being less than 10%, and then statistically analyzing the value range of the logging curves of the remaining wells [L]. min L max This method corrects curves that are outside the range to the range by modifying the mean or the range of maximum and minimum values.

5. The seismic inversion method according to claim 4, characterized in that, For three-dimensional time-domain seismic waveform data, the seismic data value window is determined to be 0.01s to 0.1s in the time domain; for three-dimensional depth-domain seismic waveform data, the seismic data value window is determined to be 20m to 200m of formation thickness; and the well logging parameter value window is determined to be 0.125m to 10m of depth measurement range.

6. A seismic inversion device suitable for non-vertical wells, characterized in that, The seismic inversion device includes: a parameter standardization unit, a data calibration unit, a database construction unit, and a matching inversion unit, wherein, The parameter standardization unit is configured to acquire seismic data, well trajectory data and logging data from multiple wells within the target area, determine the target layer range for inversion, and perform regional standardization processing on the logging data corresponding to the inversion parameters. The data calibration unit is connected to the parameter standardization unit and is configured to calibrate the well trajectory positions of all wellheads in the target area on the seismic data volume, while determining the constraint wells and verification wells, and generating the seismic data spatial coordinates corresponding to the well trajectories. The constraint wells include vertical wells, inclined wells and horizontal wells in the target area. The database construction unit is connected to the data calibration unit and is configured to determine the seismic data value window and the well logging parameter value window. Along the target layer of the well trajectory of the constrained well, the seismic data and well logging parameters traversed by the well trajectory are extracted according to the seismic data value window and the well logging parameter value window, respectively, and a seismic-well logging parameter matching database is created. The matching inversion unit is connected to the database construction unit and is configured to match the seismic data to be inverted point by point with the seismic-well logging parameter data pairs in the seismic-well logging parameter matching database according to the matching pursuit method principle, and extract the well logging parameters from the data pair with the highest degree of matching to form the inverted well logging parameter data body; The method for determining the data pair with the highest matching degree in the matching inversion unit is as follows: The seismic data in the seismic-well logging parameter matching database were normalized using the L2 norm normalization method to obtain the first seismic vector. ; Centered on the seismic data points to be inverted, a seismic data segment of a predetermined time length is obtained as the second seismic vector. The second vector was normalized using the L2 norm normalization method to obtain the third seismic vector. ; Calculate the dot product between the first and third earthquake vectors, and determine the data pair with the largest dot product result as the data pair with the highest matching degree; The seismic inversion device further includes: after creating a seismic-well logging parameter matching database, given one or more interpretation horizons of the seismic data volume as constraints, and based on these horizons and the spatial location of the seismic-well logging parameter data pairs, dividing the seismic-well logging parameter matching database into multiple seismic-well logging parameter data pair matching sub-databases corresponding to each horizon.

7. A computer device, characterized in that, The computer device includes: processor; A memory containing a computer program that, when executed by a processor, implements the seismic inversion method applicable to non-vertical wells as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the seismic inversion method applicable to non-vertical wells as described in any one of claims 1 to 5.

Images