Seismic waveform feature point driven horizon automatic tracking method, device and medium

Abstract

The application belongs to the technical field of seismic data geological interpretation, and specifically discloses a horizon automatic tracking method driven by seismic waveform feature points, equipment and a medium. The method is to determine the search range of adjacent record channels in the Z direction by using the coordinates of the waveform feature points at the seed points, and to perform recursive search in the X direction and the Y direction according to the local waveform similarity. In addition, the horizon closedness discrimination is used to reduce the phenomenon of layer stacking, so as to solve the problems of large calculation amount, large sample learning and training workload, insufficient tracking result precision and stability and the like of the existing horizon automatic tracking method. The equipment and the medium can store and execute the computer program based on the method. The application is suitable for horizon interpretation technology.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of seismic data geological interpretation, and relates to automatic horizon tracking, in particular to a method, device and medium for horizon automatic tracking driven by seismic waveform feature points. BACKGROUND

[0002] In seismic data geological interpretation, the accurate horizon information obtained through horizon interpretation is not only an important basis for reservoir prediction, but also is widely used in seismic processing methods such as velocity analysis and tomography. As a key link of horizon interpretation, the accuracy of horizon tracking directly affects the rationality of the final geological interpretation. However, the current manual horizon tracking technology not only consumes a lot of time and wastes human resources, but also is easily affected by subjective factors.

[0003] Therefore, it is crucial to develop an efficient and stable automatic horizon tracking technology to improve the effectiveness of seismic data processing and interpretation.

[0004] There are three main types of existing automatic horizon tracking methods: the first type is a horizon automatic tracking method based on waveform similarity, which mainly uses waveform cross-correlation algorithm to track the same phase axis by using the correlation between traces. This method has good noise resistance and stability, but has large amount of calculation, and when there are similar waveforms between adjacent horizons, it is easy to have layer stacking phenomenon, which reduces the reliability of the results; the second type is a horizon automatic tracking method based on artificial neural network, which has good horizon tracking effect, but needs a large amount of sample learning and training, which not only has large amount of work, but also cannot share the learning and training results between work areas due to different seismic data characteristics of different work areas; the third type is a horizon automatic tracking method based on image, which applies structure tensor to fault and horizon identification, uses a series of direction filters to extract feature directions, and then obtains the main direction of horizon development. Although this type of method has high calculation efficiency, when the horizon shape is complex, it will affect the accuracy and stability of the tracking results. SUMMARY

[0005] An object of the present application is to provide a horizon automatic tracking method driven by seismic waveform feature points, which determines the search range of adjacent traces in the Z direction using the coordinates of the waveform feature points at the seed points, and recursively searches in the X and Y directions according to the local waveform similarity, and also reduces the layer stacking phenomenon through horizon closure discrimination, thereby solving the problems of large amount of calculation, large amount of sample learning and training, and insufficient accuracy and stability of the tracking results of the existing horizon automatic tracking methods.

[0006] Another object of the present application is to provide a device and medium based on the above-mentioned horizon automatic tracking method driven by seismic waveform feature points.

[0007] To achieve the above object, the present application provides a horizon automatic tracking method driven by seismic waveform feature points, which comprises the following steps performed in sequence:

[0008] S1. Correcting the artificial interpretation seed points to the record trace waveform feature points in the Z direction to obtain corrected seed points;

[0009] S2. Sequencing the corrected seed points with the same X coordinate according to the Y coordinate from small to large to obtain sequence A;

[0010] Tracking the horizon of the corrected seed points at both ends of sequence A according to the local waveform similarity of adjacent traces to the corresponding boundary; tracking the horizon of the waveform in the range formed by the adjacent two corrected seed points in sequence A according to the local waveform similarity of adjacent traces and discriminating the closure to obtain the intermediate result of horizon tracking in sequence A;

[0011] S3. Sequencing the corrected seed points with the same Y coordinate according to the X coordinate from small to large to obtain sequence B;

[0012] Tracking the horizon of the corrected seed points at both ends of sequence B according to the local waveform similarity of adjacent traces to the corresponding boundary; tracking the horizon of the waveform in the range formed by the adjacent two corrected seed points in sequence B according to the local waveform similarity of adjacent traces and discriminating the closure to obtain the intermediate result of horizon tracking in sequence B;

[0013] S4. Discriminating the closure of the intermediate results of horizon tracking in sequence A and sequence B to obtain the current horizon automatic tracking result;

[0014] S5. Taking the current horizon automatic tracking result as the corrected seed point, repeating steps S2-S4 to obtain the horizon automatic tracking result.

[0015] As a limitation of the present application, in step S1, the corrected seed points are obtained by the following steps performed in sequence:

[0016] S11. Determining the Z-direction distribution range of the current horizon according to the depth distribution range of the coordinates of the artificial interpretation seed points in the Z direction;

[0017] S12. Identifying the waveform feature points of the seismic record waveform in the Z-direction distribution range of the current horizon;

[0018] S13. Identifying the waveform feature points of each record trace to obtain the record trace waveform feature points respectively;

[0019] S14. Correcting the artificial interpretation seed points to the corresponding record trace waveform feature points according to the nearest distance criterion to obtain the corrected seed points;

[0020] As the second limitation of the present application, in step S11, the current horizon Z-direction distribution range is determined by extending the depth distribution range of the artificial interpretation seed point in the Z-direction upward and downward by a fixed number of samples respectively.

[0021] As a further limitation of the present application, the waveform feature points are waveform positive and negative extreme points.

[0022] As the third limitation of the present application, the local waveform similarity of the adjacent trace is:

[0023] With the current correction seed point corresponding waveform feature point as the center, the current waveform is determined within the local window length range;

[0024] The same type of waveform feature points as the current correction seed point in the same local window length range in the adjacent trace of the current correction seed point are determined, and the adjacent trace waveform is determined in the same local window length range with the same type of waveform feature point as the center.

[0025] The cross-correlation coefficient of the current waveform and the adjacent trace waveform is calculated, and if the cross-correlation coefficient is greater than a set threshold, the horizon tracking is continued, otherwise the horizon tracking is interrupted.

[0026] As the fourth limitation of the present application, the horizon closure discrimination refers to selecting the tracking result with a larger waveform cross-correlation coefficient when the horizon tracking results of the adjacent two correction seed points are inconsistent.

[0027] The present application also provides a seismic waveform feature point driven horizon automatic tracking electronic device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the above-mentioned seismic waveform feature point driven horizon automatic tracking method when executing the computer program.

[0028] The present application also provides a computer readable storage medium storing a computer program for executing the above-mentioned seismic waveform feature point driven horizon automatic tracking method.

[0029] The present application has the advantages of:

[0030] The present application determines the Z-direction search range of the adjacent trace using the position of the waveform feature point at the artificial interpretation seed point, and recursively searches in the X and Y directions according to the local waveform similarity, thereby overcoming the defects of large calculation amount, many layer stringing phenomena and difficult closure of the conventional waveform similarity based horizon automatic tracking method, and realizing efficient horizon automatic interpretation driven by three-dimensional data with artificial interpretation as a sample.

[0031] The present application is suitable for horizon interpretation technology, reduces the phenomenon of layer stacking through horizon closure discrimination, and solves the problems of large calculation amount, large sample learning and training workload, insufficient tracking result precision and stability of existing horizon automatic tracking method. BRIEF DESCRIPTION OF DRAWINGS

[0032] The present application will be described in further detail below in conjunction with the accompanying drawings and specific embodiments.

[0033] Figure 1 A waveform feature point recognition result graph in Example 1 of the present application;

[0034] Figure 2 A result graph of correcting to corresponding recorded waveform feature points in Example 1 of the present application;

[0035] Figure 3 A result graph of corresponding profiles of artificial interpretation seed points after correction in Example 1 of the present application;

[0036] Figure 4 A three-dimensional space display graph of corresponding profiles of artificial interpretation seed points after correction in Example 1 of the present application;

[0037] Figure 5 A horizon tracking intermediate result graph in Sequence A in Example 1 of the present application;

[0038] Figure 6 A three-dimensional space display graph of horizon tracking intermediate result in Sequence A in Example 1 of the present application;

[0039] Figure 7 A horizon tracking intermediate result graph in Sequence B in Example 1 of the present application;

[0040] Figure 8 A three-dimensional space display graph of horizon tracking intermediate result in Sequence B in Example 1 of the present application;

[0041] Figure 9 A three-dimensional space display graph of current horizon automatic tracking result in Example 1 of the present application;

[0042] Figure 10 A three-dimensional space display graph of horizon automatic tracking result in Example 1 of the present application;

[0043] Figure 11 A horizon point fusion display result comparison graph of Example 1 and Comparative Example 1 of the present application, Figure 11 (a) is a horizon point fusion display result graph obtained by the method of Comparative Example 1 of the present application, Figure 11 (b) is a horizon point fusion display result graph obtained by the method of Example 1 of the present application. DETAILED DESCRIPTION

[0044] The application will be further described in connection with the following examples. However, those skilled in the art will understand that the application is not limited to the following examples, and any improvement and equivalent change made on the basis of the specific examples of the application are within the scope of protection of the claims of the application.

[0045] Example 1: Seismic waveform feature point driven horizon automatic tracking method

[0046] The present embodiment is a seismic waveform feature point driven horizon automatic tracking method, which comprises the following steps in turn:

[0047] S1. Correct the artificial interpretation seed points to the record trace waveform feature points in the Z direction to obtain the corrected seed points, and the specific method is:

[0048] S11. Extend the depth distribution range of all artificial interpretation seed points in the Z direction in the work area by 200 fixed sampling numbers upward and downward respectively to determine the current horizon Z direction distribution range.

[0049] S12. Identify the waveform feature points for the seismic record waveform in the above-mentioned current horizon Z direction distribution range, and identify two types of waveform feature points: waveform positive extreme value points (i.e. wave peaks) and waveform negative extreme value points (i.e. wave troughs);

[0050] Figure 1 The figure is the waveform feature point identification result, in which the curve is the seismic record waveform, and the circle point is the waveform positive extreme value point; the asterisk is the waveform positive extreme value point.

[0051] S13. Perform the waveform feature point identification of step S12 for each record trace within the current horizon Z direction distribution range in the whole work area to obtain the record trace waveform feature points respectively.

[0052] S14. Correct the artificial interpretation seed points to the corresponding record trace waveform feature points according to the nearest distance criterion to obtain the corrected seed points.

[0053] Figure 2 The figure is the result of correcting to the corresponding record trace waveform feature points, in which the white point is the original seed point position; and the gray point is the seed point position after being corrected to the waveform feature point position.

[0054] After correction, the corresponding profile result of the corrected seed points is as shown in Figure 3 , and the three-dimensional space display is as shown in Figure 4 .

[0055] S2. Sort the corrected seed points with the same X coordinate according to the Y coordinate from small to large to obtain sequence A.

[0056] (I) For the corrected seed points at both ends of sequence A, perform horizon tracking to the corresponding boundary according to the local waveform similarity of adjacent traces:

[0057] Among them, the sequence A has 6 correction seed points with the same X coordinate and sequentially increasing Y coordinate, and the two end correction seed points are the first seed point and the sixth seed point.

[0058] The horizon tracking is performed according to the similarity of local waveforms of adjacent channels, and specifically:

[0059] (1) determining a current waveform: when the current correction seed point is the first seed point, the local window length is set to 20, the local window length range is that the Y coordinate size is in the range of 10-30, and the current waveform is the waveform in the range of 10-30 of the current channel.

[0060] (2) determining a waveform of an adjacent channel: determining the waveform feature points in the Y coordinate size range of 10-30 (i.e. the local window length range) in the adjacent channel, which are both troughs (the same type) as the first seed point (i.e. the current correction seed point), and determining the waveform of the adjacent channel in the Y coordinate size range of 10-30 with the trough as the center.

[0061] (3) calculating the cross-correlation coefficients of the current waveform and the waveform of the adjacent channel, and if the two cross-correlation coefficients are greater than a set threshold, the horizon tracking is continued, otherwise the horizon tracking is interrupted.

[0062] The two end correction seed points are horizon tracked to the corresponding boundaries according to the above method.

[0063] (II) the waveforms in the range formed by the remaining two adjacent correction seed points in sequence A are horizon tracked and closed according to the similarity of local waveforms of adjacent channels, and the intermediate results of the horizon tracking of sequence A are obtained:

[0064] Among them, the range formed by the remaining two adjacent correction seed points includes: the range formed by the first seed point and the second seed point, the range formed by the second seed point and the third seed point, the range formed by the third seed point and the fourth seed point, the range formed by the fourth seed point and the fifth seed point, and the range formed by the fifth seed point and the sixth seed point.

[0065] The horizon tracking of the two end correction seed points is performed by the same method, and when the horizon tracking results between the two adjacent artificial interpretation seed points are inconsistent, the tracking result with the larger waveform cross-correlation coefficient is selected, and the intermediate results of the horizon tracking of sequence A are obtained, as shown in Figure 5 , and the three-dimensional space display of the intermediate results of the horizon tracking of sequence A is as shown in Figure 6 .

[0066] The horizon tracking and the closedness discrimination ensure the uniqueness of the horizon tracking results.

[0067] S3. The correction seed points with the same Y coordinate are sorted according to the X coordinate from small to large, and sequence B is obtained.

[0068] For the correction seed points at both ends of sequence B, layer tracing is performed to the corresponding boundaries using the adjacent channel local waveform similarity method as described in step S2; for the waveforms within the range of the remaining two adjacent correction seed points in sequence B, layer tracing and closure determination are performed according to the adjacent channel local waveform similarity method, resulting in intermediate layer tracing results for sequence B, as follows. Figure 7 As shown, the three-dimensional spatial display of the intermediate results of sequence B layer tracing is as follows: Figure 8 .

[0069] S4. Closedness detection is performed on the intermediate results of layer tracing in sequence A and sequence B to obtain the current layer's automatic tracing result, which is displayed in three-dimensional space as follows: Figure 9 .

[0070] S5. Using the current automatic stratum tracking result as the correction seed point, repeat steps S2-S4 to obtain the automatic stratum tracking result, which is displayed in three-dimensional space as follows. Figure 10 .

[0071] In steps S2 and S3, X and Y are processed in the horizontal direction; S1 is in the vertical depth direction. Only in S1 is the layer depth range determined based on the depth range of all seed points, instead of determining waveform feature points at all depths, thus significantly improving efficiency.

[0072] Comparative Example 1: A conventional automatic layer tracking method based on waveform envelope similarity

[0073] This comparative example demonstrates a conventional automatic layer tracking method based on waveform envelope similarity, a technique well-known in the field. Using this method, automatic layer tracking was performed on the same work area as in Example 1, yielding conventional automatic layer tracking results based on waveform envelope similarity.

[0074] The seismic waveform image is fused with the layer location obtained from conventional automatic layer tracking based on waveform envelope similarity, and the result is as follows: Figure 11 (a) The seismic waveform diagram is fused with the layer location obtained from the automatic layer tracking results in Example 1, and the result is as follows: Figure 11 (b) Comparing the rectangular areas in the figure, it can be seen that the layer fusion display results obtained by the conventional automatic layer tracking method based on waveform envelope similarity have obvious layer crossover phenomenon, while the layer fusion obtained by the method of Embodiment 1 of the present invention is good, and the layers are relatively in the same layer without layer crossover phenomenon.

[0075] The above results show that the automatic tracking method of the present invention overcomes the shortcomings of conventional automatic layer tracking methods based on waveform envelope similarity, such as large computational load, many layer crossover phenomena, and difficulty in closure, and the tracking results are more reasonable.

[0076] Example 2: A computer device

[0077] The embodiment provides a computer device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor to implement the seismic waveform feature point driven horizon automatic tracking method in the embodiment 1.

[0078] The memory is used for storing non-transient computer readable instructions. Specifically, the memory can comprise one or more computer program products, which can comprise various forms of computer readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, comprise a random access memory (RAM) and / or a cache memory, etc. The non-volatile memory may, for example, comprise a read-only memory (ROM), a hard disk, a flash memory, etc.

[0079] The processor can be a central processing unit (CPU) or other forms of processing units with data processing capability and / or instruction execution capability, and can control other components in the electronic device to perform desired functions. The processor is used for running the computer readable instructions stored in the memory.

[0080] Those skilled in the art should understand that, in order to solve the technical problem of how to obtain a good user experience effect, the embodiment can also include well-known structures such as a communication bus, an interface, etc., which should also be included in the protection scope of the present disclosure.

[0081] The detailed description of the embodiment can refer to the corresponding description in the foregoing embodiments, which will not be described here. Embodiment 3 A computer readable storage medium

[0082] The embodiment provides a computer readable storage medium, which stores a computer program. When the computer program is executed by a processor, the seismic waveform feature point driven horizon automatic tracking method in the embodiment 1 is implemented.

[0083] The computer readable storage medium stores non-transient computer readable instructions. When the non-transient computer readable instructions are run by the processor, all or part of the steps of the method in the foregoing embodiments are executed.

[0084] The computer readable storage medium includes but is not limited to: an optical storage medium (for example: CD-ROM and DVD), a magneto-optical storage medium (for example: MO), a magnetic storage medium (for example: magnetic tape or a mobile hard disk), a medium with a built-in rewritable non-volatile memory (for example: a memory card) and a medium with a built-in ROM (for example: a ROM cartridge).

[0085] It should be noted that the above only describes the preferred embodiments of the present application and is not used to limit the present application. Although the present application is described in detail with reference to the above embodiments, those skilled in the art can modify the technical solutions recorded in the above embodiments or make equivalent replacements to some technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of protection of the claims of the present application.

Claims

1. A seismic waveform feature point driven horizon automatic tracking method, characterized in that, The method comprises the following steps in sequence: S1. Correcting the artificial interpretation seed points to the record trace waveform feature points in the Z direction to obtain corrected seed points; S2. Sequencing the corrected seed points with the same X coordinate from small to large according to the Y coordinate to obtain sequence A; For the corrected seed points at both ends of sequence A, layer position tracking is performed to the corresponding boundaries according to the local waveform similarity of adjacent traces; for the waveforms in the range formed by the remaining adjacent two corrected seed points in sequence A, layer position tracking and closure discrimination are performed according to the local waveform similarity of adjacent traces to obtain the intermediate results of the layer position tracking of sequence A; S3. Sequencing the corrected seed points with the same Y coordinate from small to large according to the X coordinate to obtain sequence B; For the corrected seed points at both ends of sequence B, layer position tracking is performed to the corresponding boundaries according to the local waveform similarity of adjacent traces; for the waveforms in the range formed by the remaining adjacent two corrected seed points in sequence B, layer position tracking and closure discrimination are performed according to the local waveform similarity of adjacent traces to obtain the intermediate results of the layer position tracking of sequence B; S4. Performing closure discrimination on the intermediate results of the layer position tracking of sequence A and the intermediate results of the layer position tracking of sequence B to obtain the current layer position automatic tracking result; S5. Taking the current layer position automatic tracking result as the corrected seed point, repeating steps S2-S4 to obtain the layer position automatic tracking result; The closure discrimination refers to that when the layer position tracking results of the adjacent two corrected seed points are inconsistent, the tracking result with a larger waveform cross-correlation coefficient is selected.

2. The seismic waveform feature point driven horizon automatic tracking method according to claim 1, characterized in that, In step S1, the corrected seed points are obtained by the following steps in sequence: S11. Determining the current layer position Z direction distribution range according to the depth distribution range of the artificial interpretation seed point coordinates in the Z direction; S12. Identifying the waveform feature points of the seismic record waveforms in the current layer position Z direction distribution range; S13. Identifying the waveform feature points of each record trace to obtain the record trace waveform feature points respectively; S14. Correcting the artificial interpretation seed points to the corresponding record trace waveform feature points according to the nearest distance criterion to obtain the corrected seed points.

3. The seismic waveform feature point driven horizon automatic tracking method according to claim 2, characterized in that, In step S11, the current layer position Z direction distribution range is determined by respectively extending the depth distribution range of the artificial interpretation seed points in the Z direction upward and downward by a fixed number of samples.

4. The seismic waveform feature point driven horizon automatic tracking method according to claim 3, characterized in that, The waveform feature points are the waveform positive extreme points and the waveform negative extreme points.

5. The seismic waveform feature point driven horizon automatic tracking method according to claim 1, characterized in that, The local waveform similarity of adjacent traces is as follows: Taking the waveform feature point corresponding to the current corrected seed point as the center, the current waveform is determined within a local window length range; Taking the waveform feature point of the same type as the current corrected seed point in the local window length range of the adjacent trace as the center, the adjacent trace waveform is determined within the same local window length range; The cross-correlation coefficient of the current waveform and the adjacent trace waveform is calculated, and if the cross-correlation coefficient is greater than a set threshold, the layer position tracking is continued, otherwise the layer position tracking is interrupted.

6. A seismic waveform feature point driven horizon automatic tracking electronic device, characterized in that, A computer program product comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the seismic waveform feature point driven layer position automatic tracking method of any one of claims 1-5 when executing the computer program.

7. A computer readable storage medium characterized in that, The computer readable storage medium stores a computer program for performing the seismic waveform feature point driven horizon automatic tracking method of any one of claims 1-5.

Images