Global seismic layer automatic tracking method, electronic equipment and medium

Abstract

The present application discloses a global seismic layer automatic tracking method, electronic device and medium. The method may include: step 1: obtaining seismic data and existing layers, and determining the existing layer segments; step 2: determining the unknown layer trend segments in the seismic data; step 3: modifying the unknown layer trend segments according to the existing layer segments; step 4: connecting the unknown layer trend segments into long lines or layers; step 5: calculating the similarity of seismic data at each location in the same layer; step 6: randomly modifying the unknown layer trend segments, repeating steps 4-5, and determining the layer tracking result with the highest similarity as the final layer tracking result. The present invention connects the trend into seismic layers through a global connection method, close to the real seismic data trend, and optimizes the tracking of seismic layers by determining the similarity with the real seismic data.

Description

Technical Field

[0001] The present invention relates to the field of geological exploration, and more specifically, to a global seismic layer automatic tracking method, electronic equipment and medium. Background Art

[0002] In exploration and development, seismic horizon interpretation is to identify and construct underground structural morphology based on relatively fixed seismic waveform, phase, amplitude and other information on the seismic profile, providing basic information and materials for structural modeling, paleogeomorphology restoration, sequence analysis, reservoir prediction, etc., and is one of the most important contents in seismic data interpretation. Conventional horizon interpretation is mainly carried out on important horizons, such as sequence interfaces or stratigraphic interfaces, and horizon information is widely used in the processing and interpretation of seismic data.

[0003] Seismic horizon interpretation is a basic work, and horizon identification and tracking is an important part of seismic horizon interpretation. Generally, horizon tracking is basically manual or automatic tracking of a single profile, which is time-consuming, labor-intensive, and restricted by limited vision. With the rapid growth of 3D seismic exploration workload, the existing seismic interpretation model based on manual interpretation cannot meet its accuracy and efficiency requirements, bringing challenges to seismic structural interpretation.

[0004] For a long time, layer picking and calibration have been carried out more by manual or machine-assisted methods, such as phase axis tracking, layer picking based on neural network and image edge extraction technology, correlation and neural network, etc., which usually have problems such as weak stability and low adaptability. The industry's seismic layer tracking methods focus on three-dimensional layer tracking and profile automatic tracking algorithms. There are defects such as low efficiency, the need for manual specification of seed points, and long training and tracking time. In addition, only one layer (surface) can be tracked at a time, and the relationship between layers cannot be explained.

[0005] In recent years, with the popularity of artificial intelligence, some methods have emerged that use machine learning technology to track seismic layers. For example, based on the basic idea of ​​ant colony algorithm, support vector machine (SVM) technology is introduced to classify and process seismic data. At the same time, various information such as seismic amplitude, instantaneous phase, layer dip, pheromone concentration, etc. are comprehensively considered and incorporated into the evaluation function of ant colony layer tracking. The improved ant colony search algorithm is used to realize automatic tracking of seismic layers.

[0006] In order to track all layers simultaneously, the stratigraphic model is constantly revised on the basis of multi-scale modeling, so that the layer tracking task can be completed quickly and accurately. Aiming at the bottleneck problem faced by the structural interpretation of large or ultra-large three-dimensional seismic data, based on the relative geological age body and drawing on the research results of predecessors, a relative geological age body analysis method based on the dynamic time bending algorithm - stratigraphic dynamic matching and tracking technology is proposed, and based on this, a set of efficient seismic structural interpretation methods and processes are established.

[0007] Existing technologies mainly rely on the similarity of adjacent seismic traces to track layers, and the tracking effect is weak globally, and the problem of not fitting the real layer direction as a whole often occurs. For technologies that track adjacent seismic traces to connect layers, the local line segments are close to the real layers, but there may be deviations in the scale of the entire layer line; and for technologies that track the entire layer line, since only one layer is tracked, the tracked layer can only fit the seismic data, and there are problems such as conflicts with other layer directions.

[0008] Therefore, it is necessary to develop a global seismic layer automatic tracking method, electronic equipment and medium to track all layers at the same time as a whole, and obtain the layer combination that is closest to the true trend of the seismic data as a whole, so as to achieve global optimization rather than local optimization, and realize tracking of continuous global adjustment of layer trends.

[0009] The information disclosed in the background technology section of the present invention is only intended to deepen the understanding of the general background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0010] The present invention proposes a global seismic layer automatic tracking method, electronic equipment and medium, which can connect the trends into seismic layers through a global connection method, close to the real seismic data trend, and optimize the tracking of seismic layers by determining the similarity with the real seismic data.

[0011] In a first aspect, an embodiment of the present disclosure provides a global seismic layer automatic tracking method, comprising:

[0012] Step 1: Obtain seismic data and existing horizons, and determine the existing horizon segments;

[0013] Step 2: determining the unknown layer strike line segment in the seismic data;

[0014] Step 3: modifying the unknown layer strike line segment according to the existing layer line segment;

[0015] Step 4: Connect the unknown layer strike line segments into long lines or layers;

[0016] Step 5: Calculate the similarity of seismic data at the same layer;

[0017] Step 6: Randomly modify the unknown layer strike line segment, repeat steps 4-5, and determine the layer tracking result with the highest similarity as the final layer tracking result.

[0018] Preferably, step 2 comprises:

[0019] Determining a long strike line segment of an unknown horizon in the seismic data;

[0020] The long strike line segment is cut to determine the strike line segment of the unknown layer.

[0021] Preferably, determining the long strike line segment of the unknown horizon in the seismic data comprises:

[0022] Dividing the seismic data into seismic sections according to inline and xline directions;

[0023] A seismic section is divided into multiple tracking units, and each tracking unit tracks a long line segment representing the direction of the unit layer;

[0024] Traverse the directions of all long-line segments, compare the similarities between the time window segments of each seismic trace near the current trend segment in turn, add up the differences between all adjacent seismic segments, that is, the seismic data differences corresponding to the current long-line segment, and then get the direction that best matches the seismic data trend.

[0025] Preferably, step 3 comprises:

[0026] For each seismic trace, obtaining all existing stratigraphic line segments whose left endpoints are located in the seismic trace;

[0027] Obtain all unknown layer strike line segments whose left endpoints are located in the seismic trace among the unknown layer strike line segments;

[0028] Delete the unknown layer strike line segments of the seismic trace that intersect with the existing layer line segments, then merge the unknown layer strike line segments with the existing layer line segments and sort them according to the depth values ​​of the left endpoints.

[0029] Preferably, for a two-dimensional seismic profile, step 4 comprises:

[0030] Connect all unknown layer strike line segments in the current section into a long line;

[0031] According to the depth of the current stratigraphic line, search among all unknown stratigraphic trend line segments on the right, find the nearest line segment and extend it parallel to it.

[0032] Preferably, step 5 comprises:

[0033] Calculate the seismic data difference value between each two adjacent seismic trace segments from left to right and add them up in a weighted manner;

[0034] The difference values ​​of all the horizon lines are added together to obtain the similarity between the horizon tracing result of the current seismic data volume and the seismic data.

[0035] Preferably, for three-dimensional space, step 4 includes:

[0036] Select a point in the seismic network, determine the corresponding seismic trace, traverse it at intervals of the tracking unit depth, extend and diffuse each tracking unit in all directions to obtain the depths of other network points, and the depths obtained from all the network points together constitute the same seismic layer;

[0037] Starting from the midpoint of the tracking unit, the diffusion is spread out in all directions. The diffusion method in each direction is the same as the line segment extension method in the two-dimensional seismic profile.

[0038] Preferably, step 5 comprises:

[0039] Calculate the difference values ​​of seismic data between all seismic trace segments and perform weighted addition;

[0040] The difference values ​​of all the horizon planes are added together to obtain the similarity between the horizon tracing result of the current seismic data volume and the seismic data.

[0041] As a specific implementation method of the embodiment of the present disclosure,

[0042] In a second aspect, an embodiment of the present disclosure further provides an electronic device, the electronic device comprising:

[0043] A memory storing executable instructions;

[0044] A processor runs the executable instructions in the memory to implement the global seismic layer automatic tracking method.

[0045] In a third aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the global seismic layer automatic tracking method is implemented.

[0046] Its beneficial effects are:

[0047] 1. Tracking the long segment of the layer trend of seismic data is more accurate than looking at the trend of two adjacent seismic traces;

[0048] 2. Connect the strike line segments into long lines, so that the direction of each segment fits the seismic data trend at the location, and extend from multiple locations to both sides to make the traced stratigraphic lines more complete;

[0049] 3. The method of connecting the strike line segments into a layer plane solves the problem of inconsistent depth values ​​of the measuring points obtained by diffusing different paths through averaging and smoothing. At the same time, starting from multiple positions and extending to both sides, the traced layer plane is more complete;

[0050] 4. Randomly adjust the stratigraphic trend line segments. Repeated cycles can obtain stratigraphic tracking results that are closer to the actual seismic data trend.

[0051] The methods and apparatus of the present invention have other features and advantages that will be apparent from, or will be described in detail in, the accompanying drawings and subsequent detailed descriptions incorporated herein, which together serve to explain the specific principles of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present invention.

[0053] Figure 1 A flow chart showing the steps of a global seismic layer automatic tracking method according to an embodiment of the present invention.

[0054] Figure 2 A schematic diagram showing the segmentation of existing layers according to an embodiment of the present invention is shown.

[0055] Figure 3 A schematic diagram of a long line segment of a stratigraphic trend traced by seismic data according to an embodiment of the present invention is shown.

[0056] Figure 4 A schematic diagram showing a long line segment running inside a tracking unit according to an embodiment of the present invention is shown.

[0057] Figure 5 A schematic diagram showing unknown layer strike line segments connected into long lines according to an embodiment of the present invention is shown.

[0058] Figure 6a , Figure 6b , Figure 6c , Figure 6d , Figure 6e , Figure 6f A schematic diagram showing a long line extending from different positions to both ends according to an embodiment of the present invention is shown.

[0059] Figure 7 A schematic diagram showing the layer trend tracking effect according to an embodiment of the present invention.

[0060] Figure 8 A schematic diagram showing connections from a certain starting point to surrounding areas according to an embodiment of the present invention is shown.

[0061] Figure 9a , Figure 9b , Fig.9c , Figure 9d A schematic diagram showing layer bit depth divergence processing at the same point according to an embodiment of the present invention is shown.

[0062] Fig.10 A schematic diagram of randomly modifying an unknown layer strike line segment according to an embodiment of the present invention is shown.

[0063] Fig.11 A schematic diagram of collectively moving unknown layer strike line segments according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0064] The preferred embodiments of the present invention will be described in more detail below. Although the preferred embodiments of the present invention are described below, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0065] In the process of seismic layer tracking, if you want to track multiple layers at the same time, you need to design a global layer tracking algorithm, otherwise tracking each layer individually cannot meet the overall seismic data. When tracking seismic layers in each sub-area, it may not be accurate to use only adjacent seismic traces to track the layers, and multiple seismic traces are needed to track the overall trend. In the process of obtaining the overall trend, there are multiple possibilities for which direction each part of the same layer adopts, and how to obtain the optimal direction is also a problem that needs to be solved. In the process of seismic layer tracking, since there are multiple seismic layer tracking results, how to judge which ones are closer to the true stratigraphic direction is of great significance to seismic layer tracking.

[0066] If multiple adjacent seismic traces are put together during the seismic layer tracking process, the trend of the seismic data can be seen more clearly, rather than only looking at two adjacent seismic traces, which makes it difficult to see the stratigraphic trend. After obtaining the seismic data trend of each small area, a global connection method is used to connect these trends into a real batch of seismic layers, so that this batch of seismic layers are close to the real seismic data trend. After obtaining the global seismic layer, it is optimized by determining the similarity with the real seismic data to optimize the tracked seismic layer.

[0067] Figure 1 A flow chart showing the steps of a global seismic layer automatic tracking method according to an embodiment of the present invention.

[0068] The present invention provides a global seismic layer automatic tracking method, comprising:

[0069] Step 1: Obtain seismic data and existing horizons, and determine the existing horizon segments.

[0070] Specifically, seismic data can be viewed as a three-dimensional array, with inline and xline in the xy direction and seismic traces in the z direction. For displaying seismic data, you can slice the seismic data volume in the inline or xline direction to view the seismic profile. Seismic horizon data is a batch of depth values, indicating the depth of the stratum at each point in the seismic network.

[0071] First, the seismic network is divided into seismic sections in the inline and xline directions. For each seismic section, the projection of each layer on it is obtained, and these layers are segmented. Each segment records two depths on the left and right. Each layer is divided according to the seismic traces in the section to form a batch of layer segments.

[0072] Step 2: Determine the strike line segment of the unknown layer in the seismic data; in one example, step 2 includes: determining the strike long line segment of the unknown layer in the seismic data; cutting the strike long line segment to determine the strike line segment of the unknown layer. In one example, determining the strike long line segment of the unknown layer in the seismic data includes: dividing the seismic data into seismic profiles according to the inline and xline directions; dividing a seismic profile into multiple tracking units, each tracking unit tracks a long line segment representing the strike of the layer of this unit; traversing the directions of all the strike long line segments, comparing the similarities between the time window segments of each seismic trace near the current strike line segment in turn, adding the differences between all adjacent seismic segments, that is, the seismic data differences corresponding to the current strike long line segment, and then obtaining the direction that best fits the seismic data trend.

[0073] Specifically, the 3D seismic data is tracked for the stratigraphic trend. For a seismic profile, it is divided into multiple tracking units, and each unit tracks a long line segment representing the stratigraphic trend of this unit. The direction of the long line segment is traversed to obtain the direction that best matches the seismic data trend. The method for judging whether the direction matches the seismic data is to compare the similarities between the time window segments of each seismic trace near the current trend segment in turn, and add the differences between all adjacent seismic segments, that is, the seismic data difference corresponding to the current trend long line segment. Interleaving is not allowed for seismic stratigraphic layers, so when the upper and lower long line segments are interleaved, the interleaving process must be performed.

[0074] The long stratigraphic strike line of each seismic section is cut into unknown stratigraphic strike line segments of length 1, and each line segment records two depths on the left and right. According to the depth values ​​at the left and right ends of the stratigraphic strike long line segment, the depth values ​​of several points in the middle are interpolated in sequence.

[0075] Step 3: Modify the unknown layer trend line segment according to the existing layer line segment; in an example, step 3 includes: for each seismic trace, obtain all existing layer line segments whose left endpoints are located in the seismic trace; obtain all unknown layer trend line segments whose left endpoints are located in the seismic trace; delete the unknown layer trend line segments of the seismic trace that intersect with the existing layer line segments, and then merge the unknown layer trend line segments with the existing layer line segments and sort them according to the depth values ​​of the left endpoints.

[0076] Specifically, for each seismic trace, first obtain all existing layer line segments whose left endpoints are this seismic trace in the existing layer line segments, and then obtain all line segments whose left endpoints are this seismic trace in the unknown layer trend line segments obtained in the previous step. Then discard the line segments that intersect with the former in the latter, and then merge the two groups of line segments and sort them according to the depth values ​​of the left endpoints.

[0077] Step 4: Connect the unknown layer strike line segments into long lines or layers; Step 5: Calculate the similarity of seismic data at each location in the same layer.

[0078] In one example, for a two-dimensional seismic profile, step 4 includes: connecting all unknown stratigraphic strike line segments in the current profile into a long line; searching all unknown stratigraphic strike line segments on the right according to the depth of the current stratigraphic line, finding the nearest line segment to extend parallel to it. In one example, step 5 includes: calculating the difference value of seismic data between each adjacent seismic trace segment from left to right and weighted addition; adding the difference values ​​of all stratigraphic lines to obtain the similarity between the stratigraphic tracking result of the current seismic data volume and the seismic data.

[0079] In one example, for three-dimensional space, step 4 includes: selecting a point in the seismic network, determining the corresponding seismic trace, traversing at intervals of the tracking unit depth, extending and diffusing each tracking unit in all directions to obtain the depths at other network points, and all the depths obtained from all the passing network points together constitute the same seismic horizon; starting from the midpoint of the tracking unit and diffusing in all directions, the diffusion method in each direction is the same as the line segment extension method in the two-dimensional seismic profile. In one example, step 5 includes: calculating the difference values ​​of seismic data between all seismic trace segments and weighted addition; adding the difference values ​​of all horizon planes to obtain the similarity between the horizon tracking result of the current seismic data volume and the seismic data.

[0080] Specifically, these line segments are connected to form a horizon. If you only view the horizon line in a two-dimensional seismic profile, you only need to connect all the unknown horizon strike line segments in the current profile into long lines. According to the depth of the current horizon line, search among all the unknown horizon strike line segments on the right, find the nearest line segment and extend it parallel to it. The overall method of connecting all the unknown horizon strike line segments of a seismic profile is to start from a certain track, connect long lines to both sides of each grid from top to bottom, and then find another one after completion. For the grids in the track where no horizon line passes, start with the unknown horizon strike line segment in the grid and extend it to both sides until all tracks are selected. In three-dimensional space, if you want to get the seismic horizon in space, you need to connect all the line segments of all profiles into horizon surfaces. Select a point in the seismic network, get the corresponding seismic track, traverse it at intervals according to the depth of the tracking unit, and extend and spread it to the surroundings for each tracking unit to get the depth of other network points. The depths obtained by all the network points passing through together constitute the same seismic horizon. Starting from the midpoint of the tracking unit, the depth is diffused in all directions. The diffusion method in each direction is the same as the line extension method in the profile. In three-dimensional space, there may be multiple paths from one measuring point to another. For the same measuring point, the depth values ​​diffused from different directions may be inconsistent, and average and smoothing are required.

[0081] Determine whether the currently tracked seismic horizon is close to the real seismic data trend. For two-dimensional horizon lines, calculate the seismic data difference values ​​between each adjacent seismic trace segment from left to right and add them up by weight. For three-dimensional horizon planes, calculate the seismic data difference values ​​between all seismic trace segments and add them up by weight. Add the difference values ​​of all horizon lines or planes to obtain the overall similarity between the current seismic data volume horizon tracking result and the seismic data.

[0082] Step 6: Randomly modify the unknown layer trend line segment, repeat steps 4-5, and determine the layer tracking result with the highest similarity as the final layer tracking result.

[0083] Specifically, after obtaining the seismic data similarity of the current layer tracking result, the unknown layer strike line segments in each profile are randomly modified, and then the line segments are reconnected to calculate the similarity between the new layer and the seismic data. After reaching the set number of cycles, the layer tracking result with the highest similarity is retained as the final layer tracking result.

[0084] The present invention also provides an electronic device, which includes: a memory storing executable instructions; and a processor, which runs the executable instructions in the memory to implement the above-mentioned global seismic layer automatic tracking method.

[0085] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned global seismic layer automatic tracking method is implemented.

[0086] To facilitate understanding of the solutions and effects of the embodiments of the present invention, three specific application examples are given below. Those skilled in the art should understand that the examples are only for facilitating understanding of the present invention, and any specific details thereof are not intended to limit the present invention in any way.

[0087] Example 1

[0088] In this example, the inline range is 284 to 475, with a total of 192 lines, and the xline direction is 414 to 606, with a total of 190 lines, that is, 192 seismic profiles in the inline direction and 190 seismic profiles in the xline direction, a total of 382 seismic profiles, which contain all seismic data. In this example, 7 existing layers are obtained, namely h23, h31x, h33, h33x, h34, h41x, and h43, which represent a batch of layers drawn manually before. If there is no existing layer, the step of obtaining the existing layer can also be omitted.

[0089] Figure 2 A schematic diagram showing the segmentation of existing layers according to an embodiment of the present invention is shown.

[0090] For each seismic profile, obtain the projection of each layer on it. The method is to use the inline number and xline number of each seismic trace in this seismic profile, and find the depth value of the position with the same inline number and xline number in the layer data, which is the depth value of this layer at this seismic trace position. Figure 2 Shown is a schematic diagram of three layers in a seismic section.

[0091] These layers are segmented, and each segment records two depths, such as Figure 2 As shown on the right, each layer is divided according to the seismic traces in the profile to form a group of layer line segments.

[0092] The overall method for tracking the stratigraphic trend of 3D seismic data is to track the stratigraphic line trends of the 382 seismic profiles in the inline and xline directions mentioned above, and obtain a series of unknown stratigraphic trend line segments. The subsequent steps will connect these line segments into stratigraphic lines.

[0093] Figure 3 A schematic diagram of a long line segment of a stratigraphic trend traced by seismic data according to an embodiment of the present invention is shown.

[0094] For a seismic profile, the method of tracing the long line segment of its layer direction is to group 5 seismic traces into a group, and each group has a depth interval of 50 as a tracing unit, and for each tracing unit, a long line segment representing the layer direction of this unit is traced. Figure 3 As shown in the figure, the black box represents a tracking unit, which has 5 seismic traces horizontally and 50 seismic amplitude data vertically. The group size 5 and the depth interval 50 are both adjustable parameters. When the seismic data trend is not obvious, the seismic trace group size can be increased; when the similarity between seismic data traces is low, the depth interval can be increased. In this example, the tracking unit size is 5×50.

[0095] The length of the long line segment refers to the size of the seismic trace group. The reason why multiple seismic traces are used instead of two adjacent seismic traces is that it is not accurate to see the trend from two adjacent seismic traces. It is necessary to look back several traces to show the general trend. The depth interval indicates the depth interval at which the horizon is tracked, representing a basic interval for tracking the horizon.

[0096] Figure 4 A schematic diagram showing a long line segment running inside a tracking unit according to an embodiment of the present invention is shown.

[0097] For a tracking unit, determine the long line segment, such as Figure 4 As shown, the maximum time window offset Δd is set. In this example, the Δd value is 10. The direction of the long line segment is offset within positive and negative Δd. From -Δd to Δd, 1 is added each time, and the long line segments in various directions are traversed in turn to obtain the direction that best matches the seismic data.

[0098] Methods for judging whether the trend is consistent with the seismic data are as follows: Figure 4 As shown on the right side of the figure, according to the previously set depth interval size or another specified value as the time window, which is 50 in this example, the similarities between the time window segments of each seismic trace near the current strike segment are compared in turn, that is, the seismic segments within 25 above and below the current strike segment. Figure 4 This group of seismic traces has five seismic traces from 1 to 5. The seismic data segment in the quadrilateral box of each seismic trace is the seismic data that needs to be compared. Each of the two adjacent seismic segments contains 50 seismic data. The seismic data are subtracted in sequence. The sum of the absolute values ​​of the differences is the difference between the two seismic segments. The differences between all adjacent seismic segments are added together, that is, the difference in seismic data corresponding to the current strike long segment. The one with the smallest difference in seismic data in the directions from -Δd to Δd is the strike long segment of the current tracking unit. The left endpoint of the strike long segment is the midpoint of the depth of the current tracking unit or the depth position with the largest amplitude in the current tracking unit. In this example, the midpoint is taken. The depth of the right endpoint is the depth position obtained by extending rightward from the left endpoint in the direction of the strike long segment.

[0099] For each group of seismic traces in the profile, determine the strike long line segments in each tracking unit from top to bottom according to depth. In some cases, for example, when the value of Δd is set to exceed half of the time window size, the upper and lower strike long line segments may be interlaced, and interlacing is not allowed for seismic layers, so interlacing is required at this time. The method is to first check whether the strike long line segment with the second smallest difference in seismic data in the current tracking unit is interlaced with the upper one. If it is no longer interlaced, take this strike long line segment as the strike long line segment of the current tracking unit. If it is still interlaced, do not change it first. When the strike long line segments of all tracking units in this group of seismic traces are determined, traverse each strike long line segment in turn. When a strike long line segment intersects with the upper or lower strike long line segment, determine whether the depth distance between the right end points of the strike long line segment above and the strike long line segment below is greater than the time window. If it is greater, change the position of the right end point of the current strike long line segment to the center point of the right end points of the upper and lower strike long segments. If it is not greater, discard the current strike long segment.

[0100] Cut the long stratigraphic trend segments of each seismic profile into unknown stratigraphic trend segments, and cut each stratigraphic trend long segment into a segment with a length of 1. In this example, the length of the stratigraphic trend long segment is 5, that is, each long segment is cut into 5 segments, and each segment records two depths on the left and right. According to the depth values ​​at the left and right ends of the stratigraphic trend long segment, the depth values ​​of several points in the middle are interpolated in sequence, thereby obtaining 5 unknown stratigraphic trend segments.

[0101] Modify the unknown layer trend line segment by the obtained existing layer line segment, and take the existing layer line segment as the standard. For each seismic trace, first obtain all the existing layer line segments whose left endpoint is this seismic trace in the existing layer line segment, and then obtain all the line segments whose left endpoint is this seismic trace in the unknown layer trend line segment obtained in the previous step. Then discard the line segments in the latter that intersect with the former, and then merge the two groups of line segments and sort them according to the depth value of the left endpoint.

[0102] After processing each seismic trace of each seismic profile, the unknown stratigraphic strike line segments of all seismic profiles are obtained, which are the unknown stratigraphic strike line segments of 192 profiles in the inline direction and 190 profiles in the xline direction in this example.

[0103] Connect the existing stratigraphic line segments to form stratigraphic lines. If stratigraphic lines are only viewed in a two-dimensional seismic profile, all unknown stratigraphic trend line segments in the current profile only need to be connected into long lines. If a three-dimensional seismic stratigraphic line is to be obtained, all line segments of all inline and xline profiles need to be connected into stratigraphic surfaces. For the case of viewing stratigraphic lines in a two-dimensional seismic profile, the previous steps do not need to obtain the unknown stratigraphic trend line segments of all inline and xline profiles. It is only necessary to track the unknown stratigraphic trend line segments of the current seismic profile. That is, the steps only obtain the seismic and stratigraphic data of this profile, segment the stratigraphic lines of the obtained profile, only track and cut the stratigraphic trend long line segments of this profile, cut the stratigraphic data line of this profile, and then use the existing stratigraphic line segments to modify the tracked unknown stratigraphic trend line segments.

[0104] Figure 5 A schematic diagram showing unknown layer strike line segments connected into long lines according to an embodiment of the present invention is shown.

[0105] After obtaining the unknown stratigraphic strike line segments of the modified 2D seismic profile, these unknown stratigraphic strike line segments are connected into long lines to obtain the stratigraphic line of this profile. Figure 5 As shown in the figure, a stratigraphic line starts from left to right. When it is connected to the current position, according to the depth of the current stratigraphic line, it is searched in all the unknown stratigraphic trend line segments on the right. It is found that it is located between the two segments S2 and S3, and is closer to S2 and the distance is less than the tracking unit size. Then the stratigraphic line is extended from the current direction to the right in parallel with S2. If the depth of the current position is greater than the tracking unit size from both the upper and lower line segments, it means that it cannot be extended to the right and the connection is terminated. If the current stratigraphic line position is above the first unknown stratigraphic trend line segment and the distance is less than the tracking unit size, it is extended to the right in parallel with the first unknown stratigraphic trend line segment; if the current stratigraphic line position is below the last unknown stratigraphic trend line segment and the distance is less than the tracking unit size, it is extended to the right in parallel with the last unknown stratigraphic trend line segment. If the depth of the endpoint after extension is less than the minimum depth of the seismic profile, the minimum depth is taken; if the depth of the endpoint after extension is greater than the maximum depth of the seismic profile, the maximum depth is taken.

[0106] Figure 6a , Figure 6b , Figure 6c , Figure 6d , Figure 6e , Figure 6f A schematic diagram showing a long line extending from different positions to both ends according to an embodiment of the present invention is shown.

[0107] The method of connecting all unknown layer strike segments of a seismic section as a whole is as follows: Figure 6a-6f As shown, Figure 6aIt is the trend line segment of all unknown layers in the current seismic profile. The grid in the figure represents the interval of the tracking unit depth. The seismic profile is divided into blocks according to the depth of the tracking unit. Each grid in the horizontal direction represents a seismic trace, and each grid in the vertical direction represents the depth of a tracking unit. There are some empty grids in the figure, indicating that these segments have been abandoned because of the staggered.

[0108] Specific tracking methods include Figure 6b As shown, starting from a certain road, connect each grid from top to bottom to both sides, according to Figure 5 For a grid, the first line segment of the layer line is the line segment in this grid, and then extends to the left and right sides. The right extension method is Figure 5 The method shown in the left extension method is the same as Figure 5 The method in the middle is symmetrical. If the extension process encounters a situation where the layer line is intersected with the previously generated layer line, the extension is stopped. Figure 6b The results of extending each grid line segment to both ends are as follows Figure 6c As shown, several layer lines are obtained. Figure 6c The right end of the second layer line in the middle is interrupted because it intersects with the first layer, and the left end of the second-to-last layer line is interrupted because it intersects with the layer line above. In addition, it can be found that there are many grid areas where no layer line passes. At this time, press Figure 6d As shown in , find another path, and for the grids where no stratigraphic line passes, start from the unknown stratigraphic strike line segment in the grid and extend it to both sides. Figure 6d The path found in the figure has no layer line passing through the 1st, 3rd, and 8th grids from top to bottom, so we start from the layer line segments in these three grids and trace the long lines to both sides. The connection result is as follows: Figure 6e As shown in the figure, some empty cells after the first connection have layer lines passing through them. Next, another line is selected to start connecting the lines to both sides, such as Figure 6f As shown, and so on.

[0109] Figure 7 A schematic diagram showing the layer trend tracking effect according to an embodiment of the present invention.

[0110] When connecting the stratigraphic lines of the profile, start from the middle track and connect to both sides; then connect from the rightmost track to both sides, that is, connect to the left; then connect from the leftmost track to both sides, that is, connect to the right. Then randomly select the remaining tracks and connect to both sides until all tracks have been selected. In this example, when connecting each track, traverse each tracking unit from top to bottom to connect. Due to the non-interlaced nature of the stratigraphic units, when connecting to both ends, the current position must be below the previous stratigraphic unit. Therefore, when looking for the two line segments between the current position at the other end, there is no need to traverse all the line segments at the other end. You only need to continue searching downwards based on the two line segments corresponding to the previous stratigraphic position, thereby increasing the search speed. The stratigraphic direction tracking effect is as follows: Figure 7 shown.

[0111] In three-dimensional space, if you want to get the seismic horizon in space, you need to connect all the line segments of all inline and xline sections into a layer surface. Similar to the previous two-dimensional section, first select a line and select a point in the seismic network to get the corresponding seismic trace. There are two seismic sections in the inline and xline directions through this seismic trace, so for this seismic trace, it is necessary to spread out to the surroundings from top to bottom instead of connecting to the left and right as before.

[0112] For a seismic trace, it is traversed at intervals of the tracking unit depth. For each tracking unit, it is extended and diffused in all directions to obtain the depths of other measuring points. When all measuring points have been diffused or the diffusion is terminated due to a fault area, the depths obtained by all the measuring points passed during the diffusion process together constitute the same seismic layer.

[0113] The diffusion method starts from a tracking unit, and obtains the depth of the four network points around it according to the unknown stratigraphic strike line segments of the seismic profile passing through the tracking unit in the inline and xline directions. Then each of these four points diffuses to four points around it, skipping the points that have been diffused, until all points are diffused. There are some points where unknown stratigraphic strike line segments cannot be found around them, such as some unknown stratigraphic strike line segments that are discarded because of interlacing. When the distance between the unknown stratigraphic strike line segment closest to the current point in depth and the depth of this point in a certain direction is greater than the depth of the tracking unit, it means that this point cannot find an extension direction in this direction, and the next point in this direction will not be added to the subsequent diffusion process.

[0114] Figure 8 A schematic diagram showing connections from a certain starting point to surrounding areas according to an embodiment of the present invention is shown.

[0115] The starting method for connecting three-dimensional layers is different from that for the two-dimensional sections. In two-dimensional, the starting point is the unknown layer trend line segment in the tracking unit, while in three-dimensional, the starting point is the midpoint of the tracking unit. Figure 8 As shown in the figure, there are two directional sections, inline and xline, through this unit, with a total of 4 unknown layer strike line segments. The first point of this layer is the midpoint of the depth direction of this tracking unit, and then it spreads to the surroundings. The diffusion method in each direction is the same as Figure 5 The method of extending the line segment in the middle section is the same, that is, find the unknown layer trend line segment closest to it in the diffusion direction, and then connect a line segment parallel to it. The depth of the other end of the line segment is the depth of the next point in the diffusion direction. If there is an intersection with the previously traced layer, it will be discarded.

[0116] Figure 9a , Figure 9b , Fig.9c , Figure 9d A schematic diagram showing layer bit depth divergence processing at the same point according to an embodiment of the present invention is shown.

[0117] Unlike the stratum lines extending left and right in the two-dimensional profile, in three-dimensional space, there may be multiple paths from one survey point to another. For the same survey point, the depth values ​​obtained from different paths may be inconsistent, so an average and smoothing process is required. Figure 9a-9d As shown, Figure 9a Represents a measurement grid. The depth of the measurement point p may be obtained by extending and diffusing from different paths in the grid. The example in the figure shows two paths, path 1 and path 2. The depths obtained by diffusing from these two paths may not be the same. In this case, average processing is required. Figure 9b As shown, the two depths move h and h' respectively and average to the same depth. The ratio of h and h' is determined by the number of times the depth of point p has been calculated before. Assuming that the number of times point p has been diffused in the diffusion process before is n, that is, the depth has been calculated n times before, then the ratio of the distance between the old depth and the new depth is 1:n. For example, if the depth of point p has been calculated once before, then the old depth of point p and the depth obtained by diffusion of the current point p are averaged to be the new depth of point p. If the depth of point p has been calculated twice before, then the distance between the old depth of point p and the depth obtained by diffusion of the current point p is 1:2. That is, each time the depth of point p needs to be changed, the more times the depth of point p has been diffused and calculated before, the less likely the depth of point p will change. When changing the depth of point p, it is necessary to simultaneously change the depth of the measuring points around point p to achieve a smoothing effect, such as Fig.9c As shown in the figure, there are four points abcd around the measuring point p. Assume that acd is the measuring point whose depth has been obtained in the previous diffusion, b is the measuring point diffused to p this time, and the depth change of point p is that the old depth moves down by h, and the new depth moves up by h'. Then the depths of the four measuring points abcd need to be changed synchronously to achieve smoothness. Figure 9d As shown in the figure, the new and old depths are weighted averaged to get the new depth of the measuring point p, then the three measuring points of acd are decreased by h / 2, and the depth of point b is increased by h' / 2 to achieve smoothing. Of course, there may be points in acd that have not been diffused before, that is, they have no depth yet, so there is no need to change their depths. If the average and smoothing process will intersect with the traced layers, their depths will not be changed.

[0118] and Figure 6a-6fSimilar to the long lines formed from different positions at both ends, in the acquisition of three-dimensional layers, it is necessary to start from different positions of the survey network and spread to the surrounding areas. In this example, it first spreads from the center point of the survey network, then from the four corner points, and finally randomly selects survey network points for diffusion until every tracking unit of each seismic channel through which the stratigraphic strike line segment passes has been traversed.

[0119] The geological properties in the same layer are similar, that is, the data in the seismic trace segments above and below each point in the same layer should be similar. Based on this property, it can be judged whether the currently tracked seismic layer is close to the actual seismic data trend. In the same layer, the higher the similarity of the data in the seismic trace segments above and below each point, the more accurate the layer is. By obtaining the seismic data of the upper and lower half of the tracking unit at the depth position of each measuring point through which the seismic layer passes, and then following Figure 4 A method for calculating the difference between two seismic segments in a long line segment tracing the stratigraphic trend is used to calculate the data differences of each seismic trace segment.

[0120] For two-dimensional layer lines, the seismic data difference values ​​between each two adjacent seismic trace segments are calculated from left to right and weighted addition is performed. For three-dimensional layer planes, the seismic data difference values ​​between all seismic trace segments are calculated and weighted addition is performed. When there are many network points, the network points can be extracted sparsely at fixed intervals of inline and xline for comparison. The weight for addition is the inverse of the distance between the two seismic trace segments on the inline and xline planes. The larger the difference value after addition, the smaller the similarity, and vice versa. The difference values ​​of all layer lines or planes are added to obtain the overall similarity between the layer tracking result of the current seismic data volume and the seismic data.

[0121] After obtaining the seismic data similarity of the current layer tracking result, the unknown layer trend segments in each profile are randomly modified, and then the segments are reconnected to calculate the similarity between the new layer and the seismic data. After multiple cycles, the layer tracking result with the highest similarity is retained.

[0122] Fig.10 A schematic diagram of randomly modifying an unknown layer strike line segment according to an embodiment of the present invention is shown.

[0123] Specific random modification methods are as follows: Fig.10As shown, traverse each unknown layer trend line segment, and for the unknown layer trend line segment s currently traversed, search among all the unknown layer trend line segments on the right, and find that it is located between the two segments s1 and s2, and its distance from the two is a and b respectively. For s, change the line segment direction with a certain probability, and the new right end depth position distance becomes b and a, so that when the following line segments are connected into a long line, different line segments will be selected to extend. In this example, when the maximum value of a and b is greater than the tracking unit size, the change probability is 10%, and when it is less than the tracking unit size, the change probability is 0.1%.

[0124] Fig.11 A schematic diagram of collectively moving unknown layer strike line segments according to an embodiment of the present invention is shown.

[0125] When moving the unknown layer trend line segment, in order to avoid intersecting with the existing unknown layer trend line segment, when the maximum value of a and b is smaller than the tracking unit size, the unknown layer trend line segment needs to be moved collectively, such as Fig.11 As shown in the figure, if a certain unknown layer trend line segment moves downward by a distance of d, then the unknown layer trend line segment below it moves downward by distances of d-1, d-2, d-3, and so on, until the moving distance is less than 0, or the distance between the line segment to be moved and the line segment behind it is greater than the tracking unit, and the movement of the subsequent trend line segment is stopped. The same is true for moving upward, and the moving distance decreases in sequence until the moving distance is 0 or the distance between the currently moving unknown layer trend line segment and the subsequent unknown layer trend line segment is greater than the tracking unit size. After randomly changing the unknown layer trend line segment, re-enter the line segment connection step until the number of cycles is reached. In this example, the number of cycles is 1000, and finally the layer tracking result with the highest similarity is retained.

[0126] Example 2

[0127] The present disclosure provides an electronic device, which includes: a memory storing executable instructions; a processor running the executable instructions in the memory to implement the above-mentioned global seismic layer automatic tracking method.

[0128] An electronic device according to an embodiment of the present disclosure includes a memory and a processor.

[0129] The memory is used to store non-temporary computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0130] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of the present disclosure, the processor is used to run the computer-readable instructions stored in the memory.

[0131] Those skilled in the art should be able to understand that in order to solve the technical problem of how to obtain a good user experience, the present embodiment may also include well-known structures such as a communication bus and an interface, and these well-known structures should also be included in the protection scope of the present disclosure.

[0132] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0133] Example 3

[0134] An embodiment of the present disclosure provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the global seismic layer automatic tracking method is implemented.

[0135] According to the computer-readable storage medium of the embodiment of the present disclosure, non-transitory computer-readable instructions are stored thereon. When the non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the method of each embodiment of the present disclosure are executed.

[0136] The above-mentioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or mobile hard disk), media with built-in rewritable non-volatile memory (e.g., memory card) and media with built-in ROM (e.g., ROM box).

[0137] Those skilled in the art should understand that the purpose of the above description of the embodiments of the present invention is only to exemplarily illustrate the beneficial effects of the embodiments of the present invention, and is not intended to limit the embodiments of the present invention to any given examples.

[0138] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A global seismic layer automatic tracking method, characterized in that: include: Step 1: Obtain seismic data and existing horizons, and determine the existing horizon segments; Step 2: determining the unknown layer strike line segment in the seismic data; Step 3: modifying the unknown layer strike line segment according to the existing layer line segment; Step 4: Connect the unknown layer strike line segments into long lines or layers; Step 5: Calculate the similarity of seismic data at the same layer; Step 6: randomly modify the unknown layer strike line segment, repeat steps 4-5, and determine the layer tracking result with the highest similarity as the final layer tracking result; Wherein, for the two-dimensional seismic profile, step 4 includes: Connect all unknown layer strike line segments in the current section into a long line; According to the depth of the current stratigraphic line, search among all unknown stratigraphic trend line segments on the right, find the nearest line segment and extend it parallel to it; Wherein, for the three-dimensional space, step 4 includes: Select a point in the seismic network, determine the corresponding seismic trace, traverse it at intervals of the tracking unit depth, extend and diffuse each tracking unit in all directions to obtain the depths of other network points, and the depths obtained from all the network points together constitute the same seismic layer; Starting from the midpoint of the tracking unit, the diffusion is spread out in all directions. The diffusion method in each direction is the same as the line segment extension method in the two-dimensional seismic profile.

2. The global seismic layer automatic tracking method according to claim 1, wherein: Step 2 includes: Determining a long strike line segment of an unknown horizon in the seismic data; The long strike line segment is cut to determine the strike line segment of the unknown layer.

3. The global seismic layer automatic tracking method according to claim 2, wherein: Determining the long strike line segment of the unknown horizon in the seismic data includes: Dividing the seismic data into seismic sections according to inline and xline directions; A seismic section is divided into multiple tracking units, and each tracking unit tracks a long line segment representing the direction of the unit layer; Traverse the directions of all long-line segments, compare the similarities between the time window segments of each seismic trace near the current trend segment in turn, add up the differences between all adjacent seismic segments, that is, the seismic data differences corresponding to the current long-line segment, and then get the direction that best matches the seismic data trend.

4. The global seismic layer automatic tracking method according to claim 1, wherein: Step 3 includes: For each seismic trace, obtaining all existing stratigraphic line segments whose left endpoints are located in the seismic trace; Obtain all unknown layer strike line segments whose left endpoints are located in the seismic trace among the unknown layer strike line segments; Delete the unknown layer strike line segments of the seismic trace that intersect with the existing layer line segments, then merge the unknown layer strike line segments with the existing layer line segments and sort them according to the depth values ​​of the left endpoints.

5. The global seismic layer automatic tracking method according to claim 1, wherein: Step 5 includes: Calculate the seismic data difference value between each two adjacent seismic trace segments from left to right and add them up in a weighted manner; The difference values ​​of all the horizon lines are added together to obtain the similarity between the horizon tracing result of the current seismic data volume and the seismic data.

6. The global seismic layer automatic tracking method according to claim 1, wherein: Step 5 includes: Calculate the difference values ​​of seismic data between all seismic trace segments and perform weighted addition; The difference values ​​of all the horizon planes are added together to obtain the similarity between the horizon tracing result of the current seismic data volume and the seismic data.

7. An electronic device, characterized in that: The electronic device comprises: A memory storing executable instructions; A processor, wherein the processor runs the executable instructions in the memory to implement the global seismic layer automatic tracking method described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the global seismic layer automatic tracking method described in any one of claims 1-6 is implemented.

Images