A method, device, computer equipment and storage medium for identifying seismic phases

Abstract

This article relates to the field of oil and gas exploration, and in particular to a method, device, computer equipment and storage medium for seismic phase identification. The method includes obtaining a seismic data body; selecting part of the seismic data from the seismic data body as training samples, and obtaining the seismic phase labels of the training samples; using the training samples and their seismic phase labels to train a seismic phase identification model; using the seismic phase identification model to predict the seismic data in the seismic data body to determine the initial seismic phase identification result; using geological space feature constraints to adjust the initial seismic phase identification result to determine the final seismic phase identification result. This scheme uses geological space feature constraints to adaptively adjust the predetermined geological space range, achieve a dynamic balance between seismic data and geological space feature constraints, provide good technical support for conducting refined and efficient seismic data interpretation, and achieve adaptive seismic phase identification and interpretation of seismic data.

Description

Technical Field

[0001] The present invention relates to the field of oil and gas exploration, and in particular to a method, device, computer equipment and storage medium for seismic phase identification. Background Art

[0002] With the continuous improvement of oil and gas exploration and the continuous development of seismic acquisition technology, the identification of seismic phase categories plays an important role in exploring underground geological environment and geological structure, high-resolution processing, seismic inversion, reservoir prediction, structural interpretation, etc. The manual identification in existing technologies can no longer meet the current practical requirements, and must be analyzed with the help of computers combined with relevant seismic processing and interpretation technologies. Therefore, intelligent seismic data processing and intelligent seismic data interpretation have become new hot spots in the field of geophysical exploration.

[0003] Existing technologies for identifying seismic phase categories include: (1) researchers combining experience with visual observation, which requires a lot of manpower and is prone to loss of detailed information; (2) using neural network models to construct complex nonlinear intrinsic relationships between seismic data and seismic phase categories to classify and predict seismic phases. On the one hand, this method requires the use of a large amount of manually interpreted seismic data for training, and then predicts seismic data that has not been manually interpreted. On the other hand, the model training is complex, the number of network parameters is large, the computational efficiency is low, and the generalization ability is poor.

[0004] In view of the problems that the current seismic phase identification method is prone to losing details and the model training is complicated, a seismic phase identification method, device, computer equipment and storage medium are needed. Summary of the invention

[0005] To solve the above problems in the prior art, the embodiments of this document provide a method, apparatus, computer device and storage medium for seismic phase identification, which solve the problems in the prior art.

[0006] The embodiment of the present invention provides a method for seismic phase identification, including: obtaining a seismic data body; selecting part of the seismic data from the seismic data body as training samples, and obtaining seismic phase labels of the training samples; using the training samples and their seismic phase labels to train a seismic phase identification model; using the seismic phase identification model to predict the seismic data in the seismic data body to determine an initial seismic phase identification result; and adjusting the initial seismic phase identification result using geological space feature constraints to determine a final seismic phase identification result.

[0007] According to one aspect of the embodiments of this document, the method includes: selecting part of the seismic data from the seismic data body; numbering the seismic data in the seismic data body; and screening out the seismic data whose numbers satisfy an arithmetic progression from the seismic data body.

[0008] According to one aspect of the embodiments of this invention, the difference of the arithmetic progression accounts for 5% to 10% of all seismic data in the seismic data volume. According to one aspect of the embodiments of this invention, adjusting the initial seismic phase identification result using the geological spatial feature constraint to determine the final seismic phase identification result includes performing the following constraint processing on the initial seismic phase identification result of each pixel point in the seismic data: calculating the distribution of the initial seismic phase identification results within a predetermined geological space range centered on each pixel point; and using the seismic phase identification result with the largest distribution as the final seismic phase identification result of the pixel point.

[0009] According to one aspect of the embodiments of the present invention, before performing constraint processing on the initial seismic phase identification result of each pixel point in the seismic data, it also includes: determining whether the pixel point belongs to the training sample, if so, the initial seismic phase identification result of the pixel point is the final seismic phase identification result of the pixel point, if not, performing constraint processing on the initial seismic phase identification result of each pixel point in the seismic data.

[0010] According to one aspect of the embodiments of this invention, the process of determining the predetermined geological space range includes: calculating the distance between the pixel point and the training sample closest to the pixel point; and determining the predetermined geological space range according to the distance and the relationship between the distance and the geological space range. According to one aspect of the embodiments of this invention, the distance between the pixel point and the training sample is: the vertical distance between the pixel point and the training sample.

[0011] The embodiments of the present invention also provide a seismic phase identification device, including: a unified seismic data acquisition unit, used to acquire a seismic data body; a seismic phase label acquisition unit, used to select part of the seismic data from the seismic data body as training samples, and acquire the seismic phase labels of the training samples; a seismic phase identification model training unit, used to train a seismic phase identification model using the training samples and their seismic phase labels; an initial seismic phase identification result determination unit, used to predict the seismic data in the seismic data body using the seismic phase identification model to determine an initial seismic phase identification result; and a final seismic phase identification result determination unit, used to adjust the initial seismic phase identification result using geological space feature constraints to determine a final seismic phase identification result.

[0012] The embodiments of the present invention further provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0013] The embodiments of the present invention also provide a computer-readable storage medium on which computer instructions are stored. When the computer instructions are executed by a processor, the above method is implemented.

[0014] This scheme uses geological spatial feature constraints to adaptively adjust the predetermined geological spatial range, further adjust the seismic phase identification results, and achieve a dynamic balance between seismic data and geological spatial feature constraints, providing good technical support for the development of refined and efficient seismic data interpretation, and achieving efficient adaptive identification and interpretation of multiple categories of seismic phases that integrate high-precision and high-efficiency seismic data with geological features. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of this article or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of this article. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 The figure shows a structural schematic diagram of a seismic phase identification system according to an embodiment of this invention;

[0017] Figure 2 The flowchart of a seismic phase identification method according to an embodiment of this invention is shown;

[0018] Figure 3 The figure is a schematic diagram of a method for determining training samples according to an embodiment of this invention;

[0019] Figure 4 It is a schematic diagram of determining the final seismic phase identification result according to an embodiment of this invention;

[0020] Figure 5 The figure is a flow chart of a method for determining a predetermined geological space range according to an embodiment of the present invention;

[0021] Figure 6 The figure shows a schematic diagram of the structure of a seismic phase identification device according to an embodiment of this invention;

[0022] Figure 7 The figure shows a schematic diagram of the specific structure of the seismic phase identification device according to the embodiment of this article;

[0023] Fig. 8A Shown is a schematic diagram of seismic data of an embodiment of this article;

[0024] Figure 8B It is a schematic diagram of a seismic phase label corresponding to seismic data in an embodiment of this article;

[0025] Fig. 9 It is a schematic diagram of a seismic phase identification model training and seismic phase identification result prediction process in the embodiment of this article;

[0026] Fig.10 Shown is a schematic diagram of the structure of a computer device according to an embodiment of this invention.

[0027] Description of the accompanying symbols:

[0028] 101. Server;

[0029] 102. User terminal;

[0030] 601. Seismic data volume acquisition unit;

[0031] 6011, sending module;

[0032] 6012, data processing module;

[0033] 602. Seismic phase label acquisition unit;

[0034] 6021, receiving module;

[0035] 603. Seismic phase identification model training unit;

[0036] 6031, training sample screening module;

[0037] 604. Initial seismic phase identification result determination unit;

[0038] 6041, judgment module;

[0039] 6042. Geological spatial feature constraint determination module;

[0040] 6043. Initial seismic phase identification result adjustment module;

[0041] 605. Final seismic phase identification result determination unit;

[0042] 6051, distance calculation module;

[0043] 6052. Module for determining the predetermined geological space range;

[0044] 1002. Computer equipment;

[0045] 1004. Processor;

[0046] 1006. Memory;

[0047] 1008. Driving mechanism;

[0048] 1010, input / output module;

[0049] 1012. Input device;

[0050] 1014. Output device;

[0051] 1016. Presentation equipment;

[0052] 1018. Graphical user interface;

[0053] 1020. Network interface;

[0054] 1022. Communication link;

[0055] 1024. Communication bus. DETAILED DESCRIPTION

[0056] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0057] It should be noted that the terms "first", "second", etc. in the specification and claims of this article and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this article described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0058] This specification provides method operation steps as described in the embodiments or flow charts, but more or fewer operation steps may be included based on routine or non-creative work. The order of steps listed in the embodiments is only one way of executing the order of many steps and does not represent the only execution order. When the system or device product is executed in practice, it can be executed in the order of the method shown in the embodiments or the drawings or in parallel.

[0059] It should be noted that the seismic phase identification method in this article can be used in the field of oil and gas exploration, and can also be used in other fields except oil and gas exploration. This article does not limit the application field of the seismic phase identification method and device.

[0060] Figure 1 The figure shows a schematic diagram of the structure of a seismic phase identification system according to an embodiment of the present invention, and describes a seismic phase identification method combining a server 101 and a user terminal 102. The server 101 and the user terminal 102 can exchange data.

[0061] In some embodiments of the present specification, the server 101 may be an electronic device with a network interaction function, or may be software running in the electronic device to provide business logic for data processing and network interaction. The server 101 may acquire a seismic data body through field acquisition, and the server 101 may send the acquired seismic data body to the user terminal 102 for the selection of training samples. The server 101 may obtain a seismic phase identification model based on the training samples determined by the user terminal 102; the server 101 may use the seismic phase identification model to predict the seismic data in the seismic data body and determine the initial seismic phase identification result; the server 101 may also determine the geological space feature constraints, and adjust the initial seismic phase identification result based on the geological space feature constraints to determine the final seismic phase identification result.

[0062] In some embodiments of the present specification, the user terminal 102 may be a desktop computer, a tablet computer, a laptop computer, a smart phone, a digital assistant, a smart wearable device, etc. The user terminal 102 is not limited to the above-mentioned electronic device with a certain entity, and it can also be software running in the above-mentioned electronic device. The user terminal 102 is used to select part of the seismic data from the seismic data body obtained by the server 101 as a training sample, and the user manually interprets the selected part of the seismic data and marks the seismic phase label for the part of the seismic data. The user terminal 102 is also used to send part of the seismic data and its corresponding seismic phase label to the server 101; the user terminal 102 can perform data cleaning, normalization and other processing on the seismic data body and part of the seismic data. The user terminal 102 can also display the part of the seismic data and the marked seismic phase label on the interactive display interface.

[0063] like Figure 2 The flowchart of a seismic phase identification method according to an embodiment of this invention is shown, which specifically includes the following steps:

[0064] Step 201, obtain seismic data volume. In this step, explosive source is used to excite seismic waves for field data collection, and a three-dimensional seismic data volume can be obtained. After the three-dimensional seismic data volume is stacked, a post-stack three-dimensional seismic data volume is formed. The three-dimensional seismic data volume can characterize the nature and morphology of underground geological structures. The three-dimensional seismic data volume combined with the seismic phase label can describe the seismic phase characteristics, interpret the formation attributes, and analyze the reservoir characteristics of different types of terrain, providing a basis for oil field exploration.

[0065] In the field of seismic exploration, the properties and morphology of underground rock formations can be inferred by utilizing the differences in elasticity and density of underground media and observing and analyzing the response of formations to seismic waves.

[0066] The seismic data body obtained in this step initially has no corresponding seismic phase label. The seismic data located in different areas and locations in the seismic data body have different corresponding seismic phase labels. Therefore, further processing is required to determine the precise seismic phase label corresponding to each seismic data in the seismic data body to complete efficient multi-category seismic phase interpretation.

[0067] This step also includes seismic data processing procedures such as data cleaning and data normalization of the seismic data collected in the field.

[0068] Step 202, select some seismic data from the seismic data body as training samples, and obtain seismic phase labels of the training samples. In this step, the seismic data is a small amount of seismic data that has been manually interpreted. Furthermore, the seismic data in the seismic data body can be numbered and seismic data whose numbers meet preset rules can be selected from the seismic data body as training samples. For a detailed description of determining training samples, see Figure 3 In this step, a two-dimensional seismic image (such as Fig. 8A The schematic diagram of seismic data shown in Figure 1 is seismic data. Some seismic data have corresponding manually annotated seismic phase labels (such as Figure 8B Schematic diagram of seismic phase labels shown). Seismic phase labels correspond to underground topography and reservoir resource distribution. For example, seismic phase labels include, but are not limited to, salt dome areas, inclined strong amplitude areas, low continuity areas, low amplitude deformation areas, inclined low amplitude areas, extremely strong amplitude areas, medium amplitude high continuity areas, chaotic fractured zone areas, background areas, etc., or any combination thereof. In some embodiments of the present specification, the manually interpreted seismic phase labels can be adjusted according to the actual conditions of the exploration area and topography. The present application does not limit the types and forms of seismic phase labels.

[0069] Step 203, using the training samples and their seismic phase labels, train to obtain a seismic phase identification model. In this step, a small number of training samples with manually interpreted seismic phase labels can be used to train the seismic phase identification model, and the training samples are used to train the lightweight deep learning data-driven model to obtain the seismic phase identification model. Specifically, the training samples with seismic phase labels are input into the lightweight deep learning data-driven model for supervised training to obtain the seismic phase identification model. Among them, the network structure and parameters of the lightweight deep learning data-driven model are much smaller than those of conventional deep neural networks. In some embodiments of the present specification, the seismic phase identification model is only a model trained by a small number of training samples, and the accuracy of the model in identifying seismic data is not high to a certain extent. Therefore, it is necessary to further combine the methods of steps 204 and 205 to determine the final seismic phase identification result with higher accuracy.

[0070] Step 204, using the seismic phase recognition model to predict the seismic data in the seismic data volume to determine the initial seismic phase recognition result. In this step, the seismic phase recognition model trained in step 203 is used to predict the seismic phase division result of the seismic data in the seismic data volume, and the initial seismic phase recognition result of the seismic data is relatively rough. There may be a certain gap between the initial seismic phase recognition result and the actual seismic phase of the seismic data, that is, the initial seismic phase recognition result is relatively rough and the accuracy is not high.

[0071] In some embodiments of the present specification, the following formula may be used to express the initial seismic phase identification result obtained by prediction of the seismic phase identification model:

[0072] PrediCt=ADFC[MOdel[Data,Label]]#

[0073] Wherein, Model is the seismic phase recognition model trained in step 203, Data and Label are the seismic data input into the seismic phase recognition model and its corresponding seismic phase label. ADFC (Adaptive geological feature constraint) is the process of using geological spatial features to constrain the seismic phase recognition results output by the seismic phase recognition model. This process will be Figure 4 Detailed description in.

[0074] Step 205, using geological space feature constraints to adjust the initial seismic phase identification results to determine the final seismic phase identification results. In this step, geological space feature constraints can be understood as: feature constraints on the seismic phases of the seismic data based on the geological space features of the seismic data. Specifically, for the seismic data in the seismic data body of the seismic phase to be identified, by giving the range of geological space feature constraints, the KNN machine learning algorithm is used to calculate the seismic phase division within the geological space feature constraint range and the probability of occurrence of various seismic phases, and the division of other seismic phases within the constraint range is combined to adjust the seismic phase category of the geological body, so as to obtain a fine seismic phase division result, realize the dual constraints of geological data characteristics and geological space features, and achieve a dynamic balance between algorithm accuracy and model size.

[0075] Figure 3 FIG. 1 is a schematic diagram of a method for determining training samples according to an embodiment of this invention.

[0076] Step 301: number the seismic data in the seismic data volume.

[0077] In this step, the seismic data body is a three-dimensional seismic data body, and there is a strong similarity between adjacent post-stack profiles of the seismic data body. The seismic data body can be segmented along the longitudinal direction of the stratum and the horizontal direction of the ground, respectively, to obtain a plurality of two-dimensional profile seismic images that are continuously distributed in the longitudinal direction of the stratum and continuously distributed in the horizontal direction of the ground. In some embodiments of the present specification, the seismic data body can be segmented along the longitudinal direction of the stratum, and the plurality of two-dimensional profile seismic images that are continuously distributed in the longitudinal direction of the stratum can be numbered; the seismic data body can also be segmented along the horizontal direction of the ground, and the plurality of two-dimensional profile seismic images that are continuously distributed in the horizontal direction of the ground can be numbered. For example, the seismic data body is segmented along the longitudinal direction of the stratum to obtain a plurality of two-dimensional profile seismic images that are continuously distributed in the longitudinal direction of the stratum, and the plurality of segmented two-dimensional profile seismic images are numbered sequentially from 1 to obtain five hundred two-dimensional profile seismic images numbered 1-500.

[0078] Step 302, filter out the seismic data whose numbers satisfy the arithmetic progression from the seismic data body. In some embodiments of the present specification, some seismic data are filtered out from the seismic data body according to a certain rule for training the seismic phase identification model. As described in step 202, the part of the seismic data has a pre-manually labeled seismic phase label, which accounts for a small number of all the two-dimensional seismic data in the seismic data body. From the seismic data after the numbering in step 201, filter out some seismic data according to a certain arithmetic progression, so as to achieve uniform acquisition of seismic data from all seismic data, and provide great help to the stability of the training seismic phase identification model. Specifically, the difference of the arithmetic progression accounts for 5% to 10% of all seismic data in the seismic data body. For example, the seismic data body is cut and numbered in sequence along the longitudinal direction of the stratum, and a plurality of continuously distributed two-dimensional profile seismic images are obtained, namely, the first image, the second image... to the 500th image, and the difference along the arithmetic progression accounts for 10% of the 500 two-dimensional seismic images, that is, the difference of the arithmetic progression is 50. Based on this rule, selecting part of the seismic data from the seismic data body is: selecting the 1st image, the 51st image, the 101st image, the 151st image, the 201st image, the 251st image, the 301st image, the 351st image, the 401st image, and the 451st image, a total of 10 seismic data. In some other embodiments of this specification, the difference of the arithmetic sequence can also be other values, such as 8%, 12%, 16%, 20%, 25%, etc. The difference of the arithmetic sequence in this step can be any other reasonable data, and this application does not limit the difference of the arithmetic sequence.

[0079] like Figure 4Shown is a schematic diagram of determining the final seismic phase identification result of an embodiment of this invention. In the seismic phase identification method of the present application, underground sedimentary movement occurs continuously, and the underground geological structure has a strong continuity, so the seismic data that are very close in spatial position have the same seismic phase category to a certain extent, and there are almost no frequent mutations. Based on this geological characteristic, the seismic data with similar spatial positions in the three-dimensional seismic data body have basically the same seismic phase category division results during the seismic data interpretation process. Further, using this law, seismic phase identification work is performed according to the geological spatial characteristics of the seismic data in the same seismic data body. That is, the seismic phase of the seismic data within a certain range around a certain geological body can be determined as a certain same category, which is used as the seismic phase identification result of the underground geological result adjusted by the geological characteristic constraint.

[0080] Step 401, calculate the distribution of initial seismic phase identification results within a predetermined geological space range centered on each pixel point. In this step, the predetermined geological space range is a cubic space range with adjustable space size. Specifically, a certain pixel point is taken as the center, and a certain preset length is used as the side length of the cubic space to form a cubic space. Further, the preset length of the predetermined geological space range can be determined according to the characteristics of the pixel point at the center of the predetermined geological space, for example, the side length of the predetermined geological space range is determined according to the distance between the pixel point and the training sample.

[0081] According to the description of determining the initial seismic phase identification result by using the seismic phase identification model in step 204, each pixel point in the two-dimensional profile seismic image has an initial seismic phase identification result determined by the seismic phase identification model, and each pixel point has its corresponding seismic phase identification result. Therefore, each pixel point can be used as the center point of the predetermined geological space, and then the size of the predetermined geological space is determined according to the distance characteristics of the pixel points.

[0082] In this step, before the initial seismic phase identification result of each pixel is subjected to constraint processing, it also includes: judging whether the pixel belongs to the training sample, if so, the initial seismic phase identification result of the pixel is the final seismic phase identification result of the pixel, if not, the initial seismic phase identification result of each pixel in the seismic data is subjected to constraint processing. Specifically, by obtaining the serial number of the two-dimensional profile seismic image where the pixel is located and the serial number of the training sample, it is judged whether the pixel belongs to the training sample. If the pixel belongs to at least one of the training samples, the seismic phase division result of the pixel is the result of pre-manual annotation. For example, the serial number of the two-dimensional profile seismic image where a certain pixel in the seismic data is located is the 51st image in the seismic data body, and the training samples are the 1st image, the 51st image, the 101st image, the 151st image, the 201st image, the 251st image, the 301st image, the 351st image, the 401st image, and the 451st image. If the pixel point belongs to one of the training samples, the initial seismic phase recognition result predicted by the seismic phase recognition model for the pixel point is used as the final seismic phase recognition result for the pixel point. If the pixel point does not belong to the training sample, the initial seismic phase recognition result of each pixel point in the seismic data is constrained.

[0083] In this step, after determining the predetermined geological space range centered on each pixel point, the distribution of the initial seismic phase identification results within the predetermined geological space range is calculated as follows: the initial seismic phase identification results of all pixel points within the predetermined geological space range are calculated. Specifically, there are multiple pixel points within the predetermined geological space range, corresponding to multiple initial seismic phase identification results. The number of each type of seismic phase identification results is different, and the number distribution of each type of seismic phase identification results is obtained by calculation. For example, in a seismic data volume of 500 pixels × 500 pixels × 500 pixels, the predetermined geological space range of a certain pixel point is determined to be 9 pixels × 9 pixels × 9 pixels, then in the geological space, except for the pixel point located at the center of the geological space, there are 728 pixels. The 728 pixels have corresponding 728 initial seismic phase identification results. These 728 initial seismic phase identification results can be classified into several categories, for example, salt dome area, inclined strong amplitude area, low continuity area, low amplitude deformation area, inclined low amplitude area, extremely strong amplitude area, medium amplitude high continuity area, chaotic fracture zone area, background area and other categories. The distribution of the results of the categories to which these 728 initial seismic phase identification results belong is further calculated. For example, there are 172 seismic phase results belonging to the inclined low-amplitude area, 130 seismic phase results belonging to the salt dome area, 210 seismic phase results belonging to the medium-amplitude high-continuity area, and 216 seismic phase identification results belonging to the low-amplitude deformation area.

[0084] Step 402, the most distributed seismic phase identification result is used as the final seismic phase identification result of the pixel point. In view of the strong continuity of the underground geological structure, the seismic data in the same seismic data volume are classified using the KNN mechanism. In the same seismic phase category, the similarity of seismic data is high; in different seismic phase categories, the similarity of seismic data is low.

[0085] For example, in step 401, among the classifications of the four initial seismic phase identification results, the seismic phase identification results of the low-amplitude deformation area are distributed in the largest number, and this type of seismic phase identification result is used as the final seismic phase identification result of the pixel point.

[0086] Figure 5 Shown is a flow chart of a method for determining a predetermined geological space range according to an embodiment of this invention.

[0087] Step 501, calculate the distance between the pixel point and the training sample closest to the pixel point. In this step, the distance between the pixel point and the training sample is: the vertical distance between the pixel point and the training sample. It can also be understood as the distance between the sample where the pixel point is located and the training sample. That is, the distance between the pixel point (or the two-dimensional profile seismic image of the sample where the pixel point is located) and the plane where the two-dimensional profile seismic image of the training sample is located. In this step, the distance can be understood as the distance between each pixel point and the training sample closest to the two-dimensional profile seismic image where the pixel point is located. Specifically, the distance can be determined by the number corresponding to each seismic data in the seismic data body. For example, a pixel point is a pixel point in the 53rd image in the seismic data body, and the sample data in the seismic data body are the 1st image, the 51st image, the 101st image, the 151st image, the 201st image, the 251st image, the 301st image, the 351st image, the 401st image, and the 451st image, among which the 53rd two-dimensional seismic image where the pixel point is located is closest to the 51st image in the sample data. By calculating the distance between the pixel point and the training sample closest to the pixel point as 2 pixels, the size of the predetermined geological space range can be further determined.

[0088] Step 502, according to the distance and the correlation between the distance and the geological space range, determine the predetermined geological space range. In this step, the correlation between the distance and the geological space range means that in step 501, the distance between different pixels and training samples has different correlations with the geological space range. Specifically, the smaller the distance between the pixel and the training sample, the closer the pixel is to the training sample, that is, the seismic phase of the pixel is similar to the seismic phase of the training sample, and a smaller geological space range can be set, so that the distribution probability of the seismic phase of the training sample in the geological space range tends to be larger; the larger the distance between the pixel and the training sample, the farther the pixel is from the training sample, that is, the seismic phase of the pixel is less similar to the seismic phase of the training sample, and then a larger geological space range can be set, so that the geological space includes a larger number of other seismic data, and the probability of other seismic data appearing is more uniform.

[0089] In some embodiments of the present specification, the geological space range can be determined according to a certain distance threshold range. For example, when the distance between a pixel point and the training sample closest to the pixel point is less than the first preset distance, the side length of the geological space is determined to be the first preset length; when the distance between a pixel point and the training sample closest to the pixel point is greater than or equal to the first preset distance and less than the second preset distance, the side length of the geological space is determined to be the second preset length. In some embodiments of the present specification, the first preset distance is 5, and the first preset length is 3 pixels; the second preset distance is 18, and the second preset length is 9 pixels. For example, if the distance between a sample where a certain pixel point is located and the training sample closest to the pixel point is 4, which is less than the first preset distance, the side length of the geological space range can be set to the first preset length, that is, 3 pixels; for another example, if the distance between a sample where a certain pixel point is located and the training sample closest to the pixel point is 7, which exceeds the first preset distance and is less than the second preset distance, the side length of the geological space range can be set to the second preset length, that is, 9 pixels. In the present application, the values ​​of the first preset distance, the first preset length, the second preset distance, and the second preset length may have other variations, which will not be elaborated in this application.

[0090] In some other embodiments of this specification, the size of the predetermined geological space range can also be determined according to a certain ratio based on the distance between the pixel point and the training sample. The association between the distance and the geological space range in this application can be modified in other forms, and this application does not limit this.

[0091] like Figure 6The figure shows a schematic diagram of the structure of a seismic phase identification device according to an embodiment of the present invention. The figure describes the basic structure of the seismic phase identification device, wherein the functional units and modules can be implemented by software, or by general chips or specific chips. Part or all of the functional units and modules can be on a server, or part of them can also be on a user terminal, to implement seismic phase identification. The device specifically includes:

[0092] A seismic data volume acquisition unit 601 is used to acquire a seismic data volume;

[0093] A seismic phase label acquisition unit 602 is used to select part of seismic data from the seismic data volume as training samples and acquire seismic phase labels of the training samples;

[0094] A seismic phase identification model training unit 603 is used to train a seismic phase identification model using the training samples and their seismic phase labels;

[0095] An initial seismic phase identification result determining unit 604 is used to predict the seismic data in the seismic data volume using the seismic phase identification model to determine an initial seismic phase identification result;

[0096] The final seismic phase identification result determining unit 605 is used to adjust the initial seismic phase identification result by using geological space feature constraints to determine the final seismic phase identification result.

[0097] The predetermined geological space range is adaptively adjusted by utilizing the constraints of geological space characteristics, and the seismic phase identification results are further adjusted to achieve a dynamic balance between seismic data and geological space characteristic constraints, providing good technical support for the refined and efficient interpretation of seismic data, and realizing efficient adaptive identification and interpretation of multiple categories of seismic phases that integrate high-precision and high-efficiency seismic data with geological characteristics.

[0098] As an embodiment of this article, you can also refer to Figure 7 Schematic diagram of the specific structure of the seismic phase identification device of this embodiment is shown.

[0099] As an embodiment of this invention, the seismic data volume acquisition unit 601 further includes:

[0100] A sending module 6011 is used to send the initial seismic data to a user terminal;

[0101] A data processing module 6012 is used to perform data cleaning and normalization processing on the initial seismic data;

[0102] As an embodiment of this article, the seismic phase label acquisition unit 602 further includes:

[0103] Receiving module 6021, used for receiving seismic phase labels of part of seismic data sent by the user end;

[0104] As an embodiment of this invention, the seismic phase identification model training unit 603 further includes:

[0105] A training sample screening module 6031 is used to select part of the seismic data from the seismic data volume as training samples;

[0106] As an embodiment of this invention, the initial seismic phase identification result determination unit 604 further includes:

[0107] A judgment module 6041 is used to judge whether a pixel point belongs to a training sample;

[0108] A geological spatial feature constraint determination module 6042 is used to perform constraint processing on the initial seismic phase identification result of each pixel point;

[0109] An initial seismic phase identification result adjustment module 6043 is used to adjust the initial seismic phase identification result according to the geological spatial feature constraint;

[0110] As an embodiment of this invention, the final seismic phase identification result determination unit 605 further includes:

[0111] A distance calculation module 6051 is used to calculate the distance between a pixel point and the training sample closest to the pixel point;

[0112] The predetermined geological space range determination module 6052 is used to determine the predetermined geological space range according to the distance and the correlation relationship between the distance and the geological space range.

[0113] like Fig. 9 The figure shows a schematic diagram of a seismic phase identification model training and seismic phase identification result prediction process of an embodiment of this invention. As can be seen in the figure, 10 seismic data are selected from the seismic data body and manually labeled so that the 10 seismic data have accurate seismic phase labels. The 10 seismic data are used as training samples and input into the lightweight deep learning data-driven model for training to obtain a seismic phase identification model. The seismic phase identification model is used to predict a large amount of seismic data in the seismic data body to obtain an initial seismic phase identification result, and then the KNN machine learning algorithm is used to constrain geological spatial features, and the initial seismic phase identification result is adjusted to obtain a fine seismic phase prediction result, that is, the final seismic phase identification result. This embodiment reduces the complexity and parameter amount of the model on the basis of ensuring the performance of the seismic phase automatic identification model, improves the performance of the lightweight deep learning data-driven model, and can improve the recognition accuracy of the model for multiple categories of seismic phases.

[0114] like Fig.10As shown, it is a structural diagram of a computer device of an embodiment of this invention, and the computer device 1002 may include one or more processors 1004, such as one or more central processing units (CPUs), and each processing unit may implement one or more hardware threads. The computer device 1002 may also include any memory 1006, which is used to store any kind of information such as code, settings, data, etc. Non-limitingly, for example, the memory 1006 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any memory may use any technology to store information. Further, any memory may provide volatile or non-volatile retention of information. Further, any memory may represent a fixed or removable component of the computer device 1002. In one case, when the processor 1004 executes an associated instruction stored in any memory or a combination of memories, the computer device 1002 may perform any operation of the associated instruction. The computer device 1002 also includes one or more drive mechanisms 1008 for interacting with any memory, such as a hard disk drive mechanism, an optical disk drive mechanism, etc.

[0115] The computer device 1002 may also include an input / output module 1010 (I / O) for receiving various inputs (via input devices 1012) and for providing various outputs (via output devices 1014). A specific output mechanism may include a presentation device 1016 and an associated graphical user interface (GUI) 1018. In other embodiments, the input / output module 1010 (I / O), the input device 1012, and the output device 1014 may not be included, and the computer device 1002 may be used as a computer device in a network. The computer device 1002 may also include one or more network interfaces 1020 for exchanging data with other devices via one or more communication links 1022. One or more communication buses 1024 couple the components described above together.

[0116] The communication link 1022 may be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 1022 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

[0117] Corresponds to Figures 1 to 5 The method in this article also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are executed.

[0118] The embodiment of the present invention also provides a computer readable instruction, wherein when the processor executes the instruction, the program therein causes the processor to execute the following Figures 1 to 5 The method shown.

[0119] It should be understood that in the various embodiments of this document, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0120] It should also be understood that in the embodiments of this article, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0121] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this article.

[0122] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0123] In the several embodiments provided herein, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0124] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of this article.

[0125] In addition, each functional unit in each embodiment of this invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional unit.

[0126] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this article is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of this article. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0127] Specific embodiments are used in this article to illustrate the principles and implementation methods of this article. The description of the above embodiments is only used to help understand the methods and core ideas of this article. At the same time, for general technicians in this field, according to the ideas of this article, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on this article.

Claims

1. A method for identifying seismic phases, characterized in that: The method comprises: Acquire seismic data volume; Selecting part of the seismic data from the seismic data body as training samples, and obtaining seismic phase labels of the training samples, wherein selecting part of the seismic data from the seismic data body comprises: numbering the seismic data in the seismic data body; screening out seismic data whose numbers satisfy an arithmetic progression from the seismic data body; Using the training samples and their seismic phase labels, a seismic phase identification model is trained; Predicting the seismic data in the seismic data volume using the seismic phase identification model to determine an initial seismic phase identification result; The initial seismic phase identification result is adjusted using geological space feature constraints to determine the final seismic phase identification result, including performing the following constraint processing on the initial seismic phase identification result of each pixel point in the seismic data: calculating the distribution of the initial seismic phase identification results within a predetermined geological space range centered on each pixel point; and taking the seismic phase identification result with the largest distribution as the final seismic phase identification result of the pixel point.

2. The seismic phase identification method according to claim 1, characterized in that: The difference values ​​of the arithmetic progression account for 5% to 10% of all seismic data in the seismic data volume.

3. The seismic phase identification method according to claim 2, characterized in that: Before the constraint processing is performed on the initial seismic phase identification result of each pixel point in the seismic data, the following steps are also included: Determine whether the pixel point belongs to the training sample. If so, the initial seismic phase identification result of the pixel point is the final seismic phase identification result of the pixel point. If not, constraint processing is performed on the initial seismic phase identification result of each pixel point in the seismic data.

4. The seismic phase identification method according to claim 3, characterized in that: The process of determining the predetermined geological space range includes: Calculating the distance between the pixel point and the training sample closest to the pixel point; The predetermined geological space range is determined according to the distance and the correlation between the distance and the geological space range.

5. The seismic phase identification method according to claim 4, characterized in that: The distance between the pixel point and the training sample is: the vertical distance between the pixel point and the training sample.

6. A seismic phase identification device, characterized in that: The device comprises: A seismic data volume acquisition unit, used for acquiring a seismic data volume; A seismic phase label acquisition unit is used to select part of the seismic data from the seismic data body as training samples and acquire seismic phase labels of the training samples, wherein selecting part of the seismic data from the seismic data body comprises: numbering the seismic data in the seismic data body; screening out seismic data whose numbers satisfy an arithmetic progression from the seismic data body; A seismic phase identification model training unit, used to train a seismic phase identification model using the training samples and their seismic phase labels; an initial seismic phase identification result determination unit, configured to predict the seismic data in the seismic data volume using the seismic phase identification model to determine an initial seismic phase identification result; The final seismic phase identification result determination unit is used to adjust the initial seismic phase identification result by using geological space feature constraints to determine the final seismic phase identification result, including performing the following constraint processing on the initial seismic phase identification result of each pixel point in the seismic data: calculating the distribution of the initial seismic phase identification results within a predetermined geological space range centered on each pixel point; and taking the seismic phase identification result with the largest distribution as the final seismic phase identification result of the pixel point.

7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

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 method according to any one of claims 1 to 5 is implemented.

Images