Class adaptive seismic data clustering method and system

Abstract

The invention provides a category adaptive seismic data clustering method and system, and belongs to the technical field of seismic exploration. The method comprises the steps of obtaining a to-be-processed data sample set based on pre-processed seismic data; obtaining seismic attributes based on the clustering task demand parameters and the to-be-processed data sample set; establishing a joint metric function based on seismic waveform data and corresponding seismic attributes in the preprocessed seismic data; and based on the joint metric function and a preset determination category threshold, performing clustering classification on each piece of to-be-processed sample data in the to-be-processed data sample set to obtain a clustering result. According to the method and the system, correlation coefficients are used as judgment standards of clustering categories, more attention is paid to morphological differences among different data, and the method and the system have remarkable category interpretability. And when the adaptive category number is determined, the algorithm is used for self-determination according to a measurement standard, so that the uncertainty of a clustering result is effectively reduced, and the stability between data of the same category is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of seismic exploration, and particularly relates to a category adaptive seismic data clustering method, a category adaptive seismic data clustering system, a machine-readable storage medium, and an electronic device. Background Art

[0002] Seismic exploration is one of the most effective means to solve oil and gas exploration problems. The seismic signals obtained through seismic exploration provide a necessary basis for the identification of oil and gas reservoirs. Seismic facies analysis technology is one of the important technologies for seismic signal analysis in oil and gas exploration. Seismic facies analysis is an effective technology for interpreting reservoir distribution, sedimentary facies, etc. based on seismic reflection data. It mainly identifies and maps seismic facies units according to a series of characteristic parameters of seismic wave reflections, and interprets the reservoir types, sedimentary facies, sedimentary systems, etc. represented by these seismic facies. With the rapid development of computer technology and pattern recognition theory, a series of supervised methods or unsupervised methods have been widely applied to seismic facies analysis. A large number of practical applications have shown that the existing method technologies have significantly improved the efficiency and accuracy of seismic facies analysis.

[0003] Due to the particularity of seismic data itself, it is difficult to obtain labeled seismic data, the annotation cost is high, and the uncertainty is strong. There are significant limitations in the analysis of seismic data by supervised methods. Therefore, for seismic facies analysis, unsupervised methods based on the characteristics of the data itself for measurement and clustering have better adaptability and reliability. Currently, in related algorithms, K-means clustering and self-organizing mapping algorithms, as classic unsupervised clustering algorithms, have been widely applied in seismic facies analysis and achieved good results. Among them, the K-means clustering algorithm is simple and effective and has good convergence. However, this algorithm must pre-specify the number of clusters, and the selection of the initial clustering points has a significant impact on the final clustering result, with poor stability. And K-means clustering generally selects multiple distances as measurement conditions, and the method lacks flexibility and is difficult to meet the actual and variable clustering requirements; although the SOM method can achieve an adaptive number of clusters, this method also has problems that the random initial value of the network has too much influence on the result and the actual clustering result lacks specific measurement indicators.

[0004] However, the above commonly used unsupervised seismic data clustering algorithms mostly have problems such as difficulty in selecting measurement methods and lack of physical theory support for clustering results, and there are significant deficiencies in the actual seismic facies interpretability and reliability; in addition, there is also uncertainty in determining the number of clustering categories in the above traditional methods, and it is often difficult to find significant and easy-to-understand measurement criteria in seismic data of the same category. Summary of the Invention

[0005] The objective of the embodiments of the present invention is to provide a class - adaptive seismic data clustering method and system, so as to at least solve the problem that the random initial value of the network has too great an impact on the result, resulting in the lack of specific measurement indicators for the actual clustering result.

[0006] To achieve the above objective, a first aspect of the present invention provides a class - adaptive seismic data clustering method, including:

[0007] Based on the pre - processed seismic data, obtain a set of data samples to be processed;

[0008] Based on the clustering task requirement parameters and the set of data samples to be processed, perform seismic attribute acquisition;

[0009] Based on the seismic waveform data and the corresponding seismic attributes in the pre - processed seismic data, establish a joint metric function;

[0010] Based on the joint metric function and a preset decision class threshold, perform clustering classification on each data sample to be processed in the set of data samples to be processed, and obtain a clustering result.

[0011] Optionally, the above - mentioned class - adaptive seismic data clustering method further includes: pre - processing the seismic data;

[0012] The pre - processing rules of the seismic data are as follows:

[0013] According to the data type of the seismic data, perform dimension conversion on the seismic data;

[0014] The data type of the seismic data includes prestack data Data pre and post - stack data Data post ; where

[0015] The dimension conversion formula for prestack data Data pre is:

[0016] SIZE(Data pre ) = n*(t*a) = n*m;

[0017] The dimension conversion formula for post - stack data Data post is:

[0018] SIZE(Data post ) = n*t = n*m;

[0019] SIZE() represents the dimension of the data set, n represents the number of data samples, t represents the length of the single - data time axis, a represents the number of trace gathers of the single - data prestack gather, and m represents the dimension of a single data sample.

[0020] Optionally, the sliding window method is used to obtain seismic attributes;

[0021] Among them, the formula for obtaining seismic attributes is as follows:

[0022]

[0023] Attr RMS represents the seismic attribute, N is the size of the sliding window, and a i represents the amplitude value of the i-th sampling point within the sliding window.

[0024] Optionally, obtaining seismic attributes based on the clustering task requirement parameters and the set of data samples to be processed includes:

[0025] Obtaining different seismic attributes according to different clustering task requirement parameters;

[0026] Among them, the seismic attributes at least include root mean square amplitude, average absolute amplitude, reflection intensity slope, seismic wave geometric characteristics, seismic wave kinematic characteristics, and / or seismic wave statistical characteristics.

[0027] Optionally, the expression of the above joint metric function is as follows:

[0028]

[0029] Among them, K represents the set of clustering centers, θ j is the clustering center of the j-th type of seismic data, is the seismic waveform data center, is the seismic attribute data center, 1 ≤ j ≤ k indicates that there are currently k different categories, represents the data sample to be processed including seismic waveform data and seismic attributes , Data represents the set of data samples to be processed, μ is the weight attribute between the actual seismic data and the seismic attributes, and r() represents the Pearson correlation coefficient between data of the same dimension.

[0030] Optionally, clustering and classifying each data sample to be processed in the set of data samples to be processed based on the joint metric function and a preset decision class threshold includes:

[0031] For each data sample to be processed λ i , calculate the minimum value R i of the correlation coefficients between each data sample to be processed λ t and each clustering center in the current set of clustering centers;

[0032] Based on the preset decision class threshold and each data sample to be processed λ iThe minimum value R of the correlation coefficients between each of the to-be-processed sample data λ and the cluster centers in the current cluster center set t , determine whether each to-be-processed sample data λ i is classified into the current cluster category;

[0033] For the to-be-processed sample data λ that is classified into the current cluster category i , directly assign it to the corresponding cluster category and recalculate the new cluster center of this cluster category;

[0034] In the case where the to-be-processed sample data λ i cannot be classified into the current cluster category, use this to-be-processed sample data λ i as the cluster center of a new category, and at the same time, the number of cluster categories in the cluster center set increases by one.

[0035] Optionally, the above determination of whether each to-be-processed sample data λ i is classified into the current cluster category based on a preset determination category threshold and the minimum value R of the correlation coefficients between each to-be-processed sample data λ and the cluster centers in the current cluster center set t , includes: i For each to-be-processed sample data λ

[0036] , if the minimum value R of the correlation coefficients between the to-be-processed sample data λ and the cluster centers in the current cluster center set i is less than the preset determination category threshold, then determine that this to-be-processed sample data λ i is classified into the current cluster category, otherwise, determine that this to-be-processed sample data λ t cannot be classified into the current cluster category. i i i

[0037] Optionally, the above direct assignment of the to-be-processed sample data λ that is classified into the current cluster category to the corresponding cluster category and recalculation of the new cluster center of this cluster category includes: i

[0038] Recalculate the new cluster center of this cluster category according to the formula ;

[0039] where θ j is the original cluster center of this cluster category, is the new cluster center of this cluster category after update.

[0040] Optionally, the setting rule of the determination category threshold is as follows:

[0041] Determine the first threshold ε for judging categories according to the similarity between seismic data;

[0042] Based on the first threshold ε, determine the final decision class threshold

[0043] Wherein,

[0044] The second aspect of the present invention provides a class - adaptive seismic data clustering system, including:

[0045] A sample set obtaining module, configured to obtain a set of data samples to be processed based on the pre - processed seismic data;

[0046] A seismic attribute obtaining module, configured to obtain seismic attributes based on the clustering task requirement parameters and the set of data samples to be processed;

[0047] A joint metric function establishing module, configured to establish a joint metric function based on the seismic waveform data and the corresponding seismic attributes in the pre - processed seismic data;

[0048] A clustering and classification module, configured to perform clustering and classification on each data sample to be processed in the set of data samples to be processed based on the joint metric function and a preset decision class threshold, and obtain a clustering result.

[0049] The third aspect of the present invention provides a machine - readable storage medium, on which instructions are stored. When the instructions are executed by a processor, the processor is configured to execute the above - mentioned class - adaptive seismic data clustering method.

[0050] The fourth aspect of the present invention provides an electronic device. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above - mentioned class - adaptive seismic data clustering method is implemented.

[0051] Through the above technical solutions, a method and system for class - adaptive seismic data clustering are provided. The seismic data is pre - processed according to the data type, and a set of data samples to be processed is obtained by using the pre - processed seismic data. Based on the set of data samples to be processed, corresponding seismic attributes are obtained according to the parameter requirements of the clustering task. By using the correlation coefficient between the seismic waveform data in the pre - processed seismic data and the corresponding seismic attributes as a metric function to establish a joint metric function, the adaptability and flexibility to different tasks can be improved. Based on a preset decision - class threshold, a class decision condition and the joint metric function are used for clustering and classification, and a sample data to be processed in the set of data samples to be processed is randomly selected for clustering and classification and the initialization of the clustering center. Finally, by traversing all the sample data to be processed in the set of data samples to be processed, the joint metric function is calculated to continuously update the total number of classes and the clustering center, so as to determine the clustering class of each seismic data. Eventually, a clustering result with an adaptive number of classes based on the correlation - coefficient metric can be obtained. The results obtained by this method and system meet the expected task requirements, have a relatively complete physical meaning, and have stronger robustness and flexibility. This method and system use the correlation coefficient as the criterion for determining the clustering class. This criterion is more in line with the discrimination criteria of earth and geological experts, pays more attention to the morphological differences between different data, and has significant class interpretability. Moreover, when this method and system determine the adaptive number of classes, it is not necessary to pre - specify the number of clustering classes in advance, but the algorithm itself determines it according to the metric standard, which can effectively reduce the uncertainty of the clustering results and improve the stability between data of the same class. This method and system can obtain more reliable seismic data clustering results that meet the needs of geological experts, overcoming the defects of traditional clustering methods in terms of interpretability, stability, and flexibility.

[0052] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent specific embodiments section. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0054] Figure 1 is a flowchart of a method for class - adaptive seismic data clustering provided by an embodiment of the present invention;

[0055] Figure 2 is a flowchart of another method for class - adaptive seismic data clustering provided by an embodiment of the present invention;

[0056] Figure 3 is a schematic diagram of the velocity model of synthetic data provided by an embodiment of the present invention;

[0057] Figure 4 It is a schematic diagram of synthetic seismic data provided by an embodiment of the present invention;

[0058] Figure 5 It is a schematic diagram of a clustering result provided by an embodiment of the present invention;

[0059] Figure 6 It is a block diagram of a class - adaptive seismic data clustering system provided by an embodiment of the present invention;

[0060] Figure 7 It is a schematic diagram of the structure of an electronic device provided by a preferred embodiment of the present invention.

[0061] Description of reference numerals

[0062] 10 - Electronic device, 100 - Processor, 101 - Memory, 102 - Computer program. Detailed implementation manners

[0063] The following will describe the detailed implementation manners of the present invention with reference to the accompanying drawings. It should be understood that the detailed implementation manners described herein are only for explaining and understanding the present invention, and are not used to limit the present invention.

[0064] Figure 1 It is a flowchart of a class - adaptive seismic data clustering method provided by an embodiment of the present invention, Figure 2 It is a flowchart of another class - adaptive seismic data clustering method provided by an embodiment of the present invention. As Figure 1 and Figure 2 shown, the embodiment of the present invention provides a class - adaptive seismic data clustering method, including:

[0065] S110: Based on the pre - processed seismic data, obtain a set of data samples to be processed;

[0066] In some embodiments of this embodiment, the above - mentioned class - adaptive seismic data clustering method further includes: pre - processing the seismic data; the pre - processing rules of the seismic data are as follows: according to the data type of the seismic data, perform dimensionality conversion on the seismic data; the data types of the seismic data include prestack data Data pre and post - stack data Data post ; among them, the dimensionality conversion formula for prestack data Data pre is: SIZE(Data pre ) = n*(t*a) = n*m; the dimensionality conversion formula for post - stack data Data post is: SIZE(Data post) = n * t = n * m; SIZE() represents the dimension of the data set, n represents the number of data samples, t represents the length of the time axis of a single data, a represents the number of gathers in the prestack gather of a single data, and m represents the dimension of a single data sample.

[0067] Specifically, first obtain the seismic data of the target formation and perform data preprocessing according to the data type of the seismic data: For prestack data SIZE(Data pre ) = n * t * a, perform dimensionality conversion on the data, that is, expand each prestack data according to the prestack gather, and there is SIZE(Data) = n * (t * a) = n * m; for post-stack data, there is SIZE(Data post ) = n * t = n * m = SIZE(Data). By performing dimensionality conversion on the seismic data, a set of data samples to be processed is obtained where λ i represents the i-th data sample to be processed in the set of data samples to be processed, 1 ≤ i ≤ n, and the data dimension of each sample is m * 1.

[0068] S120: Based on the clustering task requirement parameters and the set of data samples to be processed, obtain seismic attributes;

[0069] In some embodiments of this embodiment, the sliding window method is used to obtain seismic attributes; among them, the formula for obtaining seismic attributes is: Attr RMS represents the seismic attribute, N is the size of the sliding window, N can be selected according to the actual situation, and a i represents the amplitude value of the i-th sampling point within the sliding window.

[0070] Specifically, according to the specific problems of the actual work area and in order to conform to the actual clustering calculation process, the sliding window method is used to obtain the corresponding seismic attributes. Exemplarily, the root mean square amplitude (RMSAmplitude) can be obtained by using a sliding window as the seismic attribute, and this seismic attribute can reflect the energy characteristics of the seismic data, as a good measure factor supplement for the seismic waveform correlation coefficient, and is very helpful for the identification of deltas, channels, and gas-bearing sandstones.

[0071] In some embodiments of this embodiment, the above-mentioned obtaining seismic attributes based on the clustering task requirement parameters and the set of data samples to be processed includes: obtaining different seismic attributes according to different clustering task requirement parameters; among them, the seismic attributes at least include root mean square amplitude, average absolute amplitude, reflection intensity slope, seismic wave geometric characteristics, seismic wave kinematic characteristics, and / or seismic wave statistical characteristics.

[0072] Exemplarily, the clustering task requirement parameters may include various clustering tasks such as lithology identification or gas-bearing sandstone identification, and vertical stratigraphic sequence trend discrimination. Among them, for lithology identification or gas-bearing sandstone identification, the average absolute amplitude can be selected as the seismic attribute; for vertical stratigraphic sequence trend discrimination, the slope of reflection strength can be selected as the seismic attribute. For different clustering tasks, one or more other seismic attributes representing the geometry, kinematics, or statistics of seismic data, such as amplitude, frequency, phase, and energy, can also be selected as metric supplements.

[0073] Among them, assuming that the corresponding attributes of the pre-stack data and post-stack data after processing are Attr pre and Attr post , then SIZE(Attr pre ) = n * t * a = n * m and SIZE(Attr post ) = n * t = n * m, that is, the obtained seismic attribute has the same data dimension as the seismic waveform data.

[0074] S130: Based on the seismic waveform data and the corresponding seismic attributes in the preprocessed seismic data, establish a joint metric function;

[0075] Specifically, establish corresponding metric functions based on the correlation coefficient for the seismic waveform data and the obtained seismic attributes, that is, the joint metric function. The existing clustering center set θ j is the clustering center of the j-th class of data, where is the waveform data center, is the attribute data center, whose dimension is the same as that of the sample data, and 1 ≤ j ≤ k indicates that there are currently k different classes. Then for the data containing the seismic waveform data and the attribute data , the joint metric function of its corresponding seismic attribute is: (1 ≤ i ≤ n)(1 ≤ j ≤ k); where, K represents the clustering center set, θ j is the clustering center of the j-th class of seismic data, is the seismic waveform data center, is the seismic attribute data center, 1 ≤ j ≤ k indicates that there are currently k different classes, represents the sample data to be processed containing the seismic waveform data and the seismic attribute , Data represents a set of data samples to be processed. μ is the weight attribute between the actual seismic data and seismic attributes, which is specified according to the specific task. r() represents the Pearson correlation coefficient between data of the same dimension.

[0076] Among them, according to the definition of the Pearson correlation coefficient, the Pearson correlation coefficient r between two different data X and Y with dimension n*1 is defined as:

[0077] Among them, and are the sample averages:

[0078] S140: Based on the joint metric function and a preset decision class threshold, perform clustering classification on each data sample to be processed in the set of data samples to be processed, and obtain a clustering result.

[0079] In the above implementation process, the method preprocesses according to the data type of the seismic data, and uses the preprocessed seismic data to obtain a set of data samples to be processed. Based on the set of data samples to be processed, corresponding seismic attributes are obtained according to the clustering task requirement parameters. By using the correlation coefficient between the seismic waveform data in the preprocessed seismic data and the corresponding seismic attributes as a metric function to establish a joint metric function, the adaptability and flexibility to different tasks can be improved. Establish a class decision condition and a joint metric function based on a preset decision class threshold for clustering classification, and randomly select the data samples to be processed in the set of data samples to be processed for clustering classification and initialization of the clustering center. Finally, by traversing all the data samples to be processed in the set of data samples to be processed, calculate the joint metric function to continuously update the total number of classes and the clustering center, and realize the determination of the clustering class of each seismic data. Then, finally, a clustering result with an adaptive number of classes based on the correlation coefficient metric can be obtained. The result obtained by this method meets the expected task requirements, has a relatively complete physical meaning, and has stronger robustness and flexibility. This method uses the correlation coefficient as the decision criterion for clustering classes. This criterion is more in line with the discrimination criteria of earth and geological experts, pays more attention to the morphological differences between different data, and has significant class interpretability. And when this method determines the adaptive number of classes, it does not need to pre-give the number of classes for clustering, but is determined by the algorithm itself according to the metric standard, which can effectively reduce the uncertainty of the clustering result and improve the stability between data of the same class. This method can obtain more reliable seismic data clustering results that meet the needs of geological experts, and overcomes the defects of traditional clustering methods in terms of interpretability, stability and flexibility.

[0080] In addition, in order to verify the implementation effect of this method, this study uses artificially synthesized post-stack seismic data for experimental verification. Figure 3It is a schematic diagram of the velocity model of synthetic data provided by an embodiment of the present invention. From left to right, the model has three different stratigraphic layer models, including high-velocity sand bodies with different shapes. A Ricker wavelet with a main frequency of 35 Hz is selected for convolution to obtain the synthetic seismic data as shown in Figure 4 shown. Figure 4 It is a schematic diagram of synthetic seismic data provided by an embodiment of the present invention. It can be seen that the synthetic seismic data conforms to the velocity model, and the obtained seismic data can be divided into three categories. Using this method to cluster the synthetic data, Figure 5 the clustering results with the threshold ε set to 0.7, 0.8, and 0.9 are respectively shown, and the clustering results are consistent with the actual prediction.

[0081] In some embodiments of this embodiment, clustering and classifying each to-be-processed sample data in the to-be-processed data sample set based on the combined metric function and the preset determination category threshold includes: for each to-be-processed sample data λ i , calculate the minimum value R i of the correlation coefficients between each to-be-processed sample data λ t and each cluster center in the current cluster center set; based on the preset determination category threshold and the minimum value R i of the correlation coefficients between each to-be-processed sample data λ t and each cluster center in the current cluster center set, determine whether each to-be-processed sample data λ i belongs to the current cluster category; directly assign the to-be-processed sample data λ i that belongs to the current cluster category to the corresponding cluster category, and recalculate the new cluster center of this cluster category; in the case where the to-be-processed sample data λ i cannot be classified into the current cluster category, use this to-be-processed sample data λ i as the cluster center of a new category, and at the same time, the number of cluster categories in the cluster center set increases by one.

[0082] In some embodiments of this embodiment, based on the preset determination category threshold and the minimum value R i of the correlation coefficients between each to-be-processed sample data λ t and each cluster center in the current cluster center set, determining whether each to-be-processed sample data λ i belongs to the current cluster category includes: for each to-be-processed sample data λ i , if the minimum value R i of the correlation coefficients between the to-be-processed sample data λ t and each cluster center in the current cluster center set is less than the preset determination category threshold, then determine that this to-be-processed sample data λ iIf it is classified into the current clustering category; otherwise, it is determined that the sample data λ to be processed i cannot be classified into the current clustering category.

[0083] In some embodiments of the present embodiment, the sample data λ to be processed that is classified into the current clustering category i is directly assigned to the corresponding clustering category, and the new clustering center of this clustering category is recalculated, including: according to the formula recalculate the new clustering center of this clustering category; where θ j is the original clustering center of this clustering category, is the new clustering center of this clustering category after update.

[0084] Specifically, for each sample data λ to be processed i , assume R t = min(R K (i, 1), R K (i, 2),..., R K (i, k)), that is, R t is the minimum correlation coefficient between λ i and any clustering center in the existing clustering center set K. Then there are the following discrimination conditions:

[0085]

[0086] That is, if the minimum correlation coefficient between λ i and all existing clustering centers is less than the preset determination category threshold then λ i is classified into the current clustering category and the new clustering center of this clustering category is recalculated using the expression . Otherwise, this λ i is used as the clustering center of a new category, and the total number of categories in the clustering center set is increased by 1.

[0087] In some embodiments of the present embodiment, the setting rule of the determination category threshold is as follows: According to the mutual similarity between seismic data, determine the first threshold ε for judging categories; based on the first threshold ε, determine the final determination category threshold where

[0088] Specifically, according to the mutual similarity between seismic data (such as the similarity of seismic waveforms of the same category, the similarity between seismic attributes, etc.), give the threshold ε for judging categories. In order to improve the sensitivity of the metric and the convenience of parameter adjustment, take Take as the actual discrimination condition.

[0089] It should be noted that the first threshold ε for judging data of the same category can be set artificially, thus improving the flexibility and adaptability of this method when performing clustering classification.

[0090] Figure 6 It is a block diagram of a category-adaptive seismic data clustering system provided by an embodiment of the present invention. As Figure 6 shown, the embodiment of the present invention provides a category-adaptive seismic data clustering system, including:

[0091] A sample set obtaining module, configured to obtain a set of data samples to be processed based on the preprocessed seismic data;

[0092] A seismic attribute obtaining module, configured to obtain seismic attributes based on the clustering task requirement parameters and the set of data samples to be processed;

[0093] A joint metric function establishing module, configured to establish a joint metric function based on the seismic waveform data and the corresponding seismic attributes in the preprocessed seismic data;

[0094] A clustering and classification module, configured to perform clustering and classification on each sample data to be processed in the set of data samples to be processed based on the joint metric function and a preset determination category threshold, so as to obtain a clustering result.

[0095] Specifically, the system preprocesses according to the data type of seismic data, and uses the preprocessed seismic data to obtain a set of data samples to be processed. Based on the set of data samples to be processed and according to the parameter requirements of the clustering task, the corresponding seismic attributes are obtained. By using the correlation coefficient between the seismic waveform data in the preprocessed seismic data and the corresponding seismic attributes as a metric function to establish a joint metric function, the adaptability and flexibility for different tasks can be improved. Based on a preset decision category threshold, a category decision condition and the joint metric function are established for clustering classification, and the data samples to be processed in the set of data samples to be processed are randomly selected for clustering classification and the initialization of the clustering center. Finally, by traversing all the data samples to be processed in the set of data samples to be processed, the joint metric function is calculated to continuously update the total number of categories and the clustering center, so as to determine the clustering category of each seismic data. Eventually, a clustering result with an adaptive number of categories based on the correlation coefficient metric can be obtained. The result obtained by the system meets the expected task requirements, has a relatively complete physical meaning, and has stronger robustness and flexibility. The system uses the correlation coefficient as the criterion for determining the clustering category. This criterion is more in line with the discrimination criteria of geologists and pays more attention to the morphological differences between different data, and has significant category interpretability. Moreover, when the system determines the adaptive number of categories, it does not need to pre-give the number of clustering categories, but is determined by the algorithm itself according to the metric standard, which can effectively reduce the uncertainty of the clustering result and improve the stability between data of the same category. The system can obtain a more reliable seismic data clustering result that meets the requirements of geologists, overcoming the defects of traditional clustering methods in terms of interpretability, stability and flexibility.

[0096] An embodiment of the present invention further provides a machine-readable storage medium, on which instructions are stored. When the instructions are executed by the processor 100, the processor 100 is configured to execute the above-mentioned category-adaptive seismic data clustering method.

[0097] A machine-readable storage medium includes permanent and non-permanent, removable and non-removable media, and can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media do not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0098] An embodiment of the present invention further provides an electronic device 10, which includes a memory 101, a processor 100, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the above-mentioned category adaptive seismic data clustering method is implemented.

[0099] As Figure 7 shown is a schematic diagram of an electronic device provided by an embodiment of the present invention. As Figure 7 shown, the electronic device 10 of this embodiment includes: a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the steps in the above method embodiment are implemented. Alternatively, when the processor 100 executes the computer program 102, the functions of each module / unit in the above device embodiment are implemented.

[0100] Exemplarily, the computer program 102 can be divided into one or more modules / units. One or more modules / units are stored in the memory 101 and executed by the processor 100 to complete the present invention. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program 102 in the electronic device 10. For example, the computer program 102 can be divided into a module for obtaining a sample set, a module for obtaining seismic attributes, a module for establishing a joint metric function, and a clustering and classification module.

[0101] The electronic device 10 can be a computing device such as a desktop computer, a notebook, a handheld computer, and a cloud server. The electronic device 10 may include, but is not limited to, a processor 100 and a memory 101. Those skilled in the art can understand that Figure 7 merely examples of the electronic device 10, which do not constitute a limitation on the electronic device 10, may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, a bus, etc.

[0102] The processor 100 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0103] The memory 101 may be an internal storage unit of the electronic device 10, such as the hard disk or memory of the electronic device 10. The memory 101 may also be an external storage device of the electronic device 10, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 10. Further, the memory 101 may also include both an internal storage unit and an external storage device of the electronic device 10. The memory 101 is used to store computer programs and other programs and data required by the electronic device 10. The memory 101 may also be used to temporarily store data that has been output or will be output.

[0104] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0105] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program 102 product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program 102 product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0106] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program 102 products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program 102 instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program 102 instructions. These computer program 102 instructions can be provided to the processor 100 of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor 100 of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0107] These computer program 102 instructions can also be stored in a computer-readable memory 101 that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory 101 generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0108] These computer program 102 instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide for implementing the steps in a process Figure 1 one process or multiple processes and / or blocks Figure 1 steps of the functions specified in one block or multiple blocks.

[0109] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of another identical element in the process, method, commodity or device comprising the element.

[0110] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A class - adaptive seismic data clustering method, characterized in that, it includes: Based on the pre - processed seismic data, obtain a set of data samples to be processed; Based on the clustering task requirement parameters and the set of data samples to be processed, perform seismic attribute acquisition; Based on the seismic waveform data and the corresponding seismic attributes in the pre - processed seismic data, establish a joint metric function; Based on the joint metric function and a preset decision class threshold, perform clustering classification on each data sample to be processed in the set of data samples to be processed, and obtain a clustering result.

2. The class - adaptive seismic data clustering method according to claim 1, characterized in that, the method further includes: pre - processing the seismic data; The pre - processing rules of the seismic data are as follows: According to the data type of the seismic data, perform dimensional conversion on the seismic data; The data types of the seismic data include prestack data Data pre and poststack data Data post ; among which, The pre-stack data Data pre has a dimensional conversion formula as follows: SIZE(Data pre ) = n*(t*a) = n*m; The post-stack data Data post has a dimensional conversion formula as follows: SIZE(Data post ) = n * t = n * m; SIZE() represents the dimension of the data set, n represents the number of data samples, t represents the length of the time axis of a single data, a represents the number of gathers in the prestack gather of a single data, and m represents the dimension of a single data sample.

3. The class - adaptive seismic data clustering method according to claim 1, characterized in that, Adopt a sliding window method to calculate seismic attributes; Among them, the formula for calculating seismic attributes is: Attr RMS Indicates the seismic attribute, where N is the size of the sliding window, and a i represents the amplitude value of the i-th sampling point within the sliding window.

4. The class - adaptive seismic data clustering method according to claim 1, characterized in that, The performing seismic attribute acquisition based on the clustering task requirement parameters and the set of data samples to be processed includes: According to different clustering task requirement parameters, calculate different seismic attributes; Among them, the seismic attributes at least include root - mean - square amplitude, average absolute amplitude, reflection intensity slope, seismic wave geometric characteristics, seismic wave kinematic characteristics, and / or seismic wave statistical characteristics.

5. The class - adaptive seismic data clustering method according to claim 1, characterized in that, The expression of the joint metric function is as follows: Among them, K represents the set of clustering centers, θ j is the clustering center of the j-th type of seismic data, is the seismic waveform data center, is the seismic attribute data center, 1 ≤ j ≤ k indicates that there are currently k different categories, represents the seismic waveform data λ i d and the seismic attribute λ i a of the sample data to be processed, Data represents the set of data samples to be processed, μ is the weight attribute between the actual seismic data and the seismic attributes, and r() represents the Pearson correlation coefficient between data of the same dimension.

6. The class - adaptive seismic data clustering method according to claim 5, characterized in that, The performing clustering classification on each data sample to be processed in the set of data samples to be processed based on the joint metric function and a preset decision class threshold includes: For each sample data λ to be processed i , calculate each sample data λ to be processed i and the minimum value R of the correlation coefficients between each sample data λ to be processed t ; Based on a preset determination category threshold and each sample data λ to be processed i The minimum value R of the correlation coefficients between each sample data λ to be processed and each cluster center in the current cluster center set t , determine whether each sample data λ to be processed i should be classified into the current cluster category; The to-be-processed sample data λ classified into the current clustering category i is directly assigned to the corresponding clustering category, and the new clustering center of this clustering category is recalculated; In the case where the sample data λ to be processed i cannot be classified into the current clustering category, the sample data λ to be processed i is used as the clustering center of a new category, and at the same time, the number of clustering categories in the clustering center set increases by one.

7. The class - adaptive seismic data clustering method according to claim 6, characterized in that, Based on the preset determination category threshold and each sample data to be processed λ i and the minimum value R of the correlation coefficients between each sample data to be processed λ and each cluster center in the current cluster center set t , it is determined whether each sample data to be processed λ i belongs to the current cluster category, including: For each sample data λ to be processed i , if the sample data λ to be processed i and the minimum value R of the correlation coefficients between the sample data λ to be processed and each cluster center in the current cluster center set t is less than the preset determination category threshold, then it is determined that the sample data λ to be processed i belongs to the current cluster category; otherwise, it is determined that the sample data λ to be processed i does not belong to the current cluster category.

8. The class - adaptive seismic data clustering method according to claim 6, characterized in that, The to-be-processed sample data λ classified into the current clustering category i is directly assigned to the corresponding clustering category, and the new clustering center of this clustering category is recalculated, including: According to the formula Recalculate the new cluster center of this cluster category; Among them, θ j is the original clustering center of this clustering category, and is the new clustering center of this clustering category after update.

9. The class - adaptive seismic data clustering method according to claim 1, characterized in that, The setting rules of the decision class threshold are as follows: According to the mutual similarity between seismic data, determine the first threshold ε for judging the class; Determine the final decision class threshold based on the first threshold ε Among them, 1. A class - adaptive seismic data clustering system, characterized in that, it includes: A sample set obtaining module, configured to obtain a set of data samples to be processed based on the pre - processed seismic data; A seismic attribute acquisition module, configured to perform seismic attribute acquisition based on the clustering task requirement parameters and the set of data samples to be processed; A joint metric function establishing module, configured to establish a joint metric function based on the seismic waveform data and the corresponding seismic attributes in the pre - processed seismic data; A clustering and classification module, configured to perform clustering and classification on each sample data to be processed in the set of sample data to be processed based on a joint metric function and a preset determination category threshold, so as to obtain a clustering result.

11. A machine-readable storage medium, on which instructions are stored, wherein, when the instructions are executed by a processor, the processor is configured to execute the category-adaptive seismic data clustering method according to any one of claims 1 to 9.

12. An electronic device, the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, when the processor executes the computer program, the category-adaptive seismic data clustering method according to any one of claims 1 to 9 is implemented.

Images