CN114429161A - Seismic waveform clustering method, electronic device, and medium - Google Patents

Seismic waveform clustering method, electronic device, and medium Download PDF

Info

Publication number
CN114429161A
CN114429161A CN202010909915.XA CN202010909915A CN114429161A CN 114429161 A CN114429161 A CN 114429161A CN 202010909915 A CN202010909915 A CN 202010909915A CN 114429161 A CN114429161 A CN 114429161A
Authority
CN
China
Prior art keywords
data
centroid
distance
seismic
seismic data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010909915.XA
Other languages
Chinese (zh)
Inventor
夏红敏
刘兰锋
刘俊州
史云清
王箭波
温立峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Petroleum and Chemical Corp
Sinopec Exploration and Production Research Institute
Original Assignee
China Petroleum and Chemical Corp
Sinopec Exploration and Production Research Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Petroleum and Chemical Corp, Sinopec Exploration and Production Research Institute filed Critical China Petroleum and Chemical Corp
Priority to CN202010909915.XA priority Critical patent/CN114429161A/en
Publication of CN114429161A publication Critical patent/CN114429161A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
    • G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V2210/00—Details of seismic processing or analysis
    • G01V2210/60—Analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Acoustics & Sound (AREA)
  • Environmental & Geological Engineering (AREA)
  • Geology (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geophysics (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

The application discloses a seismic waveform clustering method, electronic equipment and a medium. The method can comprise the following steps: step 1: inputting seismic data and normalizing; step 2: randomly selecting K channels of seismic data from the seismic data as a centroid; and step 3: calculating the centroid distances between the residual seismic data and the K centroids, and dividing the seismic data into K clusters of data; and 4, step 4: calculating the mass center of the next iteration for each cluster of data; and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result. According to the method, the K-Means improved KMedoids-DTW algorithm is adopted, so that geological features such as a heterogeneous reservoir stratum with rapid transverse change, a small crack group and the like can be identified; the stratigraphic structure edge can be clearly depicted.

Description

Seismic waveform clustering method, electronic device, and medium
Technical Field
The invention relates to the field of oil and gas seismic data interpretation and complex reservoir prediction, in particular to a seismic waveform clustering method, electronic equipment and a medium.
Background
From the characteristics of the waveform, the change of the seismic wave characteristics on the seismic data can reflect the change of the parameters of the underground medium, and the change of the seismic wave waveform is the dynamic characteristics of the seismic reflection wave. The physical properties, internal structure, oil and gas content, distribution range and thickness of the geologic body can be expressed by the waveform characteristics of seismic waves. Seismic waveform characteristics (dynamics) can reflect stratum attributes to a certain extent, and stratum characteristics corresponding to waveforms can be represented by using certain structural characteristic parameters of the waveforms. From the aspect of the waveform classification technology, the corresponding relation between the seismic phase diagram and the actual geologic body can be established through an effective waveform classification technology, underground rock properties and physical property changes are estimated, local stratum interpretation is carried out, a deposition system is determined, an oil favorable region is presumed, and a later-stage mining target is clarified. The waveform classification is an important technology of geophysical exploration technology, especially plays a lot of roles in identifying sedimentary facies structures for oil and gas prediction and exploration and development, and is an important technology in the field of oil exploration.
At present, the seismic waveform classification method mainly comprises the following steps: classifying seismic waveforms by a discriminant factor analysis method; classifying the seismic waveform based on the neural network; unsupervised waveform classification based on wavelet transformation and self-organizing mapping, the method has the advantage of insensitivity to horizon interpretation errors, and seismic facies analysis results are improved; a mixed waveform classification method based on an artificial immune system and self-organizing mapping compresses data and determines the number of seismic facies through the artificial immune system method, and has better robustness on the analysis of seismic data with noise; classifying three-dimensional waveforms based on K-means and SOM, wherein the method has the main advantage that thin layers with vertical thickness smaller than vertical seismic resolution can be classified; sixthly, a semi-supervised-based rapid seismic waveform classification method is adopted, and a linear transformation-based semi-supervised dimension reduction algorithm is adopted to reduce the dimension of the samples to be classified.
The waveform clustering methods have better effects in seismic facies analysis, but the methods have certain defects. In unsupervised clustering, the number of classes is uncertain, and the accuracy of the number of classes determines the classification effect and fineness. The existing K-Means algorithm adopts an Euclidean distance algorithm, and the center of each cluster is obtained by averaging all kinds of cluster data of a sample. Therefore, when the method is applied, if individual outliers exist in a sample, the clustering center is interfered by an abnormal value, so that the position deviation between the mean value center and the actual center is overlarge, and the cluster is distorted; and by applying the Euclidean distance, if a horizon error caused by artificial horizon identification exists, the Euclidean distance serving as a similarity measure forms an error due to waveform misalignment.
Therefore, there is a need to develop a seismic waveform clustering method, an electronic device and a medium based on the KMedges-DTW algorithm.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
The invention provides a seismic waveform clustering method, electronic equipment and a medium, which can identify geological features such as a heterogeneous reservoir stratum, a small crack group and the like with rapid transverse change by adopting a K-Means-based improved KMedoids-DTW algorithm and clearly depict the edge of a stratum structure.
In a first aspect, an embodiment of the present disclosure provides a seismic waveform clustering method, including:
step 1: inputting seismic data and normalizing;
step 2: randomly selecting K channels of seismic data from the seismic data as a centroid;
and step 3: calculating the centroid distances between the residual seismic data and K centroids, and dividing the seismic data into K clusters of data;
and 4, step 4: calculating the mass center of the next iteration for each cluster of data;
and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
Preferably, calculating the centroid distance comprises:
calculating the DTW distance of the two seismic data;
and calculating the centroid distance according to the DTW distance.
Preferably, the DTW distance is calculated by equation (1):
Figure BDA0002662886560000031
wherein, D (Q)m,pn) DTW distance, Q ═ l, representing the two seismic data1,l2,l3,...lM),P＝(g1,g2,g3,...gM)，d(lm,gn) Is amAnd gnThe distance of (a) to (b),
Figure BDA0002662886560000032
is the best path.
Preferably, the centroid distance is calculated by equation (2):
Figure BDA0002662886560000033
therein, IndexM×1Is the centroid distance.
Preferably, the step 3 comprises:
and dividing each channel of data into a centroid cluster with the minimum distance to obtain K cluster data.
Preferably, the step 4 comprises:
for each cluster of data, calculating average data of the cluster of data;
and calculating the centroid distance between each seismic data in the cluster of data and the average data, and selecting the seismic data with the minimum distance as the centroid of the next iteration.
As a specific implementation of the embodiments of the present disclosure,
in a second aspect, an embodiment of the present disclosure further provides an electronic device, including:
a memory storing executable instructions;
a processor executing the executable instructions in the memory to implement the steps of:
step 1: inputting seismic data and normalizing;
step 2: randomly selecting K channels of seismic data from the seismic data as a centroid;
and step 3: calculating the centroid distances between the residual seismic data and K centroids, and dividing the seismic data into K clusters of data;
and 4, step 4: calculating the mass center of the next iteration for each cluster of data;
and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
Preferably, calculating the centroid distance comprises:
calculating the DTW distance of the two seismic data;
and calculating the centroid distance according to the DTW distance.
Preferably, the DTW distance is calculated by equation (1):
Figure BDA0002662886560000041
wherein, D (Q)m,pn) DTW distance, Q ═ l, representing the two seismic data1,l2,l3,...lM),P＝(g1,g2,g3,...gM)，d(lm,gn) Is amAnd gnThe distance of (a) to (b),
Figure BDA0002662886560000042
is the best path.
Preferably, the centroid distance is calculated by equation (2):
Figure BDA0002662886560000051
therein, IndexM×1Is the centroid distance.
Preferably, the step 3 comprises:
and dividing each channel of data into a centroid cluster with the minimum distance to obtain K cluster data.
Preferably, the step 4 comprises:
for each cluster of data, calculating average data of the cluster of data;
and calculating the centroid distance between each seismic data in the cluster of data and the average data, and selecting the seismic data with the minimum distance as the centroid of the next iteration.
In a third aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the method for clustering seismic waveforms is implemented.
The beneficial effects are that: the method can well reduce the influence of abnormal values, has robustness on horizon interpretation errors caused by manual identification, and can more clearly depict the terrain structures such as small crack groups, faults and the like; clustering distortion caused by abnormal values is reduced, classification deviation caused by horizon errors is weakened, and technical support is provided for underground structure identification.
The method and apparatus of the present invention have other features and advantages which will be apparent from or are set forth in detail in the accompanying drawings and the following detailed description, which are incorporated herein, and which together serve to explain certain principles of the invention.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent by describing in more detail exemplary embodiments thereof with reference to the attached drawings, in which like reference numerals generally represent like parts.
FIG. 1 shows a flow diagram of the steps of a method of clustering seismic waveforms according to one embodiment of the invention.
FIG. 2 shows a schematic diagram of a two-dimensional forward modeling work area, according to one embodiment of the present invention.
Fig. 3 shows a schematic diagram of forward data with added horizon error and noise according to fig. 2.
FIG. 4 shows a schematic diagram of the DTW distance of each trace of seismic data from a standard trace according to FIG. 3.
FIG. 5 shows a schematic representation of Euclidean distance of each trace of seismic data from a standard trace according to FIG. 3.
FIG. 6 shows a schematic diagram of a destination layer slice of a physical model according to one embodiment of the invention.
FIG. 7 shows a schematic diagram of a seismic data profile and horizon display of a physical model according to one embodiment of the invention.
FIG. 8 shows a schematic diagram of the spread of the destination layer of the physical model over a three-dimensional space, according to one embodiment of the invention.
FIG. 9 shows a schematic representation of the root mean square amplitude of each trace of seismic data 60ms below the top of the destination layer of the physical model, according to one embodiment of the invention.
FIG. 10 shows a schematic diagram of the phase classification results according to the K-Means algorithm of FIG. 9.
Fig. 11 shows a schematic diagram of the result of the classification of a facies according to the present algorithm of fig. 9.
Detailed Description
Preferred embodiments of the present invention will be described in more detail below. While the following describes preferred embodiments of the present invention, it should be understood that the present invention may be embodied in various forms and should not be limited by the embodiments set forth herein.
The invention provides a seismic waveform clustering method, which comprises the following steps:
step 1: inputting seismic data and normalizing;
step 2: randomly selecting K channels of seismic data from the seismic data as a centroid;
and step 3: calculating the centroid distances between the residual seismic data and the K centroids, and dividing the seismic data into K clusters of data;
and 4, step 4: calculating the mass center of the next iteration for each cluster of data;
and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
In one example, calculating the centroid distance comprises:
calculating the DTW distance of the two seismic data;
the centroid distance is calculated from the DTW distance.
In one example, the DTW distance is calculated by equation (1):
Figure BDA0002662886560000071
wherein, D (Q)m,pn) DTW distance, Q ═ l, representing the two seismic data1,l2,l3,...lM),P＝(g1,g2,g3,...gM)，d(lm,gn) Is amAnd gnThe distance of (a) to (b),
Figure BDA0002662886560000072
is the best path.
In one example, the centroid distance is calculated by equation (2):
Figure BDA0002662886560000073
therein, IndexM×1Is the centroid distance.
In one example, step 3 comprises:
and dividing each channel of data into a centroid cluster with the minimum distance to obtain K cluster data.
In one example, step 4 comprises:
for each cluster of data, calculating average data of the cluster of data;
and calculating the centroid distance between each seismic data in the cluster of data and the average data, and selecting the seismic data with the minimum distance as the centroid of the next iteration.
Specifically, the seismic waveform clustering method according to the invention comprises the following steps:
step 1: inputting seismic data and normalizing;
step 2: randomly selecting K channels of seismic data from the seismic data as a centroid;
and step 3: calculating the centroid distances between the residual seismic data and the K centroids, dividing each channel of data into a centroid cluster with the minimum distance, and dividing the seismic data into K clusters of data; calculating the centroid distance comprises: calculating the DTW distance of two paths of seismic data through a formula (1); from the DTW distance, the centroid distance is calculated by equation (2).
And 4, step 4: for each cluster of data, calculating average data of the cluster of data; and calculating the centroid distance between each seismic data in the cluster of data and the average data through the formula (1) and the formula (2), and selecting the seismic data with the minimum distance as the centroid of the next iteration.
And 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
The present invention also provides an electronic device, comprising: a memory storing executable instructions; a processor executing executable instructions in the memory to implement the steps of:
step 1: inputting seismic data and normalizing;
step 2: randomly selecting K channels of seismic data from the seismic data as a centroid;
and step 3: calculating the centroid distances between the residual seismic data and the K centroids, and dividing the seismic data into K clusters of data;
and 4, step 4: calculating the mass center of the next iteration for each cluster of data;
and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
In one example, calculating the centroid distance comprises:
calculating the DTW distance of the two seismic data;
the centroid distance is calculated from the DTW distance.
In one example, the DTW distance is calculated by equation (1):
Figure BDA0002662886560000091
wherein, D (Q)m,pn) DTW distance, Q ═ l, representing the two seismic data1,l2,l3,...lM),P＝(g1,g2,g3,...gM)，d(lm,gn) Is amAnd gnThe distance of (a) to (b),
Figure BDA0002662886560000092
is the best path.
In one example, the centroid distance is calculated by equation (2):
Figure BDA0002662886560000093
therein, IndexM×1Is the centroid distance.
In one example, step 3 comprises:
and dividing each channel of data into a centroid cluster with the minimum distance to obtain K cluster data.
In one example, step 4 comprises:
for each cluster of data, calculating average data of the cluster of data;
and calculating the centroid distance between each seismic data in the cluster of data and the average data, and selecting the seismic data with the minimum distance as the centroid of the next iteration.
The present invention also provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the seismic waveform clustering method described above.
To facilitate understanding of the scheme of the embodiments of the present invention and the effects thereof, four specific application examples are given below. It will be understood by those skilled in the art that this example is merely for the purpose of facilitating an understanding of the present invention and that any specific details thereof are not intended to limit the invention in any way.
Example 1
FIG. 1 shows a flow diagram of the steps of a method of clustering seismic waveforms according to one embodiment of the invention.
As shown in fig. 1, the seismic waveform clustering method includes: step 1: inputting seismic data and normalizing; step 2: randomly selecting K channels of seismic data from the seismic data as a centroid; and step 3: calculating the centroid distances between the residual seismic data and the K centroids, and dividing the seismic data into K clusters of data; and 4, step 4: calculating the mass center of the next iteration for each cluster of data; and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
FIG. 2 shows a schematic diagram of a two-dimensional forward modeling work area, according to one embodiment of the present invention.
Fig. 3 shows a schematic diagram of forward data with added horizon error and noise according to fig. 2.
FIG. 4 shows a schematic diagram of the DTW distance of each trace of seismic data from a standard trace according to FIG. 3.
FIG. 5 shows a schematic representation of Euclidean distance of each trace of seismic data from a standard trace according to FIG. 3.
By utilizing software forward modeling, three layers of stratum models are set, forward wavelets are Rake wavelets, the frequency is 30HZ, the forward algorithm is a finite difference algorithm, the wave field propagation mode is vertical wave field propagation, and as shown in FIG. 2, the research area is 40ms above and below the first layer of reflecting layer. FIG. 3 is seismic data with a random horizon error of 1-3ms, variance 0.01 added. And selecting the first path of seismic data as a standard trace, and calculating the DTW distance and Euclidean distance between the residual trace seismic data and the standard trace. As shown in fig. 4 and 5, comparing the DTW distance and the euclidean distance between each seismic data and the standard trace, it can be seen that the DTW distance can more clearly divide the seismic data into three categories, which are consistent with the forward model, while the euclidean distance is extremely chaotic and inaccurate, and the corresponding stratigraphic model cannot be identified.
FIG. 6 shows a schematic diagram of a destination layer slice of a physical model according to one embodiment of the invention.
FIG. 7 shows a schematic diagram of a seismic data profile and horizon display of a physical model according to one embodiment of the invention.
FIG. 8 shows a schematic diagram of the spread of the destination layer of the physical model over a three-dimensional space, according to one embodiment of the invention.
FIG. 9 shows a schematic representation of the root mean square amplitude of each trace of seismic data 60ms below the top of the destination layer of the physical model, according to one embodiment of the invention.
And carrying out physical model experiments, wherein the physical model is established according to a certain region. The post-stack seismic data is obtained through the physical model, and the target layer is taken and applied to the method. In which FIG. 6 shows a slice of the target layer along the layer, the upper half of the slice is mostly a small crack group, the lower half is mostly a larger fault, and there is a small crack group. FIG. 7 is a section of seismic data in the corresponding modeled three-dimensional seismic data volume of FIG. 6. Fig. 8 shows the spread of the destination layer in three dimensions. FIG. 9 shows the root mean square amplitude of each trace of seismic data 60ms below the top of the target layer, and it can be seen that there is some outlier in the seismic data.
FIG. 10 shows a schematic diagram of the phase classification results according to the K-Means algorithm of FIG. 9.
Fig. 11 shows a schematic diagram of the result of the classification of a facies according to the present algorithm of fig. 9.
And taking 60ms below the top of the horizon as a research area, and respectively applying a K-Means algorithm and the KMedoids-DTW algorithm of the invention to obtain the images 10 and 11. The comparison shows that the KMedoids-DTW algorithm provided by the invention is more refined than the K-Means method in the aspect of displaying the small crack groups in the 300ms-500ms part in the CrossLine direction, and the detailed parts are more abundantly depicted. Whereas in the portion of CrossLine orientation 100ms-300ms, the KMedoids-DTW algorithm can clearly describe the fault edge in fault identification, the small crack groups 1, 2, 3 in FIG. 6 can clearly be shown in the circle of FIG. 11 but cannot be seen in FIG. 10. Therefore, the method provided by the invention can better identify small cracks, faults and the like, clearly depict fault edges, and has the advantages of accurately identifying topographic features and the like. Compared with the K-Means algorithm, the centroids obtained by the KMedoids-DTW algorithm exist in the seismic data set, and the rock physical significance of each centroid waveform can be given when logging information exists; the centroid waveform obtained by the K-Means algorithm is the average value of each cluster of waveforms, is not really existed in the seismic data set, and cannot give rock physical significance.
Example 2
The present disclosure provides an electronic device including: a memory storing executable instructions; and the processor runs the executable instructions in the memory to realize the seismic waveform clustering method.
An electronic device according to an embodiment of the present disclosure includes a memory and a processor.
The memory is to store non-transitory computer readable instructions. In particular, the memory may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc.
The processor may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of the disclosure, the processor is configured to execute the computer readable instructions stored in the memory.
Those skilled in the art should understand that, in order to solve the technical problem of how to obtain a good user experience, the present embodiment may also include well-known structures such as a communication bus, an interface, and the like, and these well-known structures should also be included in the protection scope of the present disclosure.
For the detailed description of the present embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, which are not repeated herein.
Example 3
The disclosed embodiments provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the seismic waveform clustering method.
A computer-readable storage medium according to an embodiment of the present disclosure has non-transitory computer-readable instructions stored thereon. The non-transitory computer readable instructions, when executed by a processor, perform all or a portion of the steps of the methods of the embodiments of the disclosure previously described.
The computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROMs and DVDs), magneto-optical storage media (e.g., MOs), magnetic storage media (e.g., magnetic tapes or removable disks), media with built-in rewritable non-volatile memory (e.g., memory cards), and media with built-in ROMs (e.g., ROM cartridges).
It will be appreciated by persons skilled in the art that the above description of embodiments of the invention is intended only to illustrate the benefits of embodiments of the invention and is not intended to limit embodiments of the invention to any examples given.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments.

Claims (10)

1. A method for clustering seismic waveforms, comprising:
step 1: inputting seismic data and normalizing;
step 2: randomly selecting K channels of seismic data from the seismic data as a centroid;
and step 3: calculating the centroid distances between the residual seismic data and K centroids, and dividing the seismic data into K clusters of data;
and 4, step 4: calculating the mass center of the next iteration for each cluster of data;
and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
2. The seismic waveform clustering method of claim 1, wherein calculating a centroid distance comprises:
calculating the DTW distance of the two seismic data;
and calculating the centroid distance according to the DTW distance.
3. The seismic waveform clustering method according to claim 2, wherein the DTW distance is calculated by formula (1):
Figure FDA0002662886550000011
wherein, D (Q)m,pn) DTW distance, Q ═ l, representing the two seismic data1,l2,l3,...lM),P＝(g1,g2,g3,...gM)，d(lm,gn) Is amAnd gnThe distance of (a) to (b),
Figure FDA0002662886550000012
is the best path.
4. The seismic waveform clustering method according to claim 3, wherein the centroid distance is calculated by formula (2):
Figure FDA0002662886550000021
therein, IndexM×1Is the centroid distance.
5. The seismic waveform clustering method according to claim 1, wherein the step 3 comprises:
and dividing each channel of data into a centroid cluster with the minimum distance to obtain K cluster data.
6. The seismic waveform clustering method according to claim 1, wherein the step 4 comprises:
for each cluster of data, calculating average data of the cluster of data;
and calculating the centroid distance between each seismic data in the cluster of data and the average data, and selecting the seismic data with the minimum distance as the centroid of the next iteration.
7. An electronic device, characterized in that the electronic device comprises:
a memory storing executable instructions;
a processor executing the executable instructions in the memory to perform the steps of:
step 1: inputting seismic data and normalizing;
step 2: randomly selecting K channels of seismic data from the seismic data as a centroid;
and step 3: calculating the centroid distances between the residual seismic data and K centroids, and dividing the seismic data into K clusters of data;
and 4, step 4: calculating the mass center of the next iteration for each cluster of data;
and 5: and (5) comparing whether the centroid after the iteration is the same as the previous centroid, if so, repeating the steps 3-5, and if so, outputting a classification result.
8. The electronic device of claim 7, wherein calculating a centroid distance comprises:
calculating the DTW distance of the two seismic data;
and calculating the centroid distance according to the DTW distance.
9. The electronic device of claim 8, wherein the DTW distance is calculated by equation (1):
Figure FDA0002662886550000031
wherein, D (Q)m,pn) DTW distance, Q ═ l, representing the two seismic data1,l2,l3,...lM),P＝(g1,g2,g3,...gM)，d(lm,gn) Is amAnd gnThe distance of (a) to (b),
Figure FDA0002662886550000032
is the best path;
the centroid distance is calculated by equation (2):
Figure FDA0002662886550000033
therein, IndexM×1Is the centroid distance.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a processor, implements the seismic waveform clustering method according to any one of claims 1 to 6.
CN202010909915.XA 2020-09-02 2020-09-02 Seismic waveform clustering method, electronic device, and medium Pending CN114429161A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010909915.XA CN114429161A (en) 2020-09-02 2020-09-02 Seismic waveform clustering method, electronic device, and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010909915.XA CN114429161A (en) 2020-09-02 2020-09-02 Seismic waveform clustering method, electronic device, and medium

Publications (1)

Publication Number Publication Date
CN114429161A true CN114429161A (en) 2022-05-03

Family

ID=81309247

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010909915.XA Pending CN114429161A (en) 2020-09-02 2020-09-02 Seismic waveform clustering method, electronic device, and medium

Country Status (1)

Country Link
CN (1) CN114429161A (en)

Similar Documents

Publication Publication Date Title
US7248539B2 (en) Extrema classification
CN107688201B (en) RBM-based seismic prestack signal clustering method
AU2017343749A1 (en) System and method for seismic facies identification using machine learning
US9081918B2 (en) Methods and systems regarding models of underground formations
US20130325349A1 (en) Methods for Generating Depofacies Classifications for Subsurface Oil or Gas Reservoirs or Fields
Song et al. Multi-waveform classification for seismic facies analysis
Scheidt et al. Probabilistic falsification of prior geologic uncertainty with seismic amplitude data: Application to a turbidite reservoir case
US11181653B2 (en) Reservoir characterization utilizing ReSampled seismic data
CN110927791B (en) Method and device for predicting fluid by utilizing seismic data based on deep learning
US10884149B2 (en) System and method for assessing the presence of hydrocarbons in a subterranean reservoir based on seismic data
CN106842317A (en) A kind of method and device for predicting oil sand body distribution
US11733424B2 (en) Systems and methods for identifying subsurface features as functions of feature positions in a subsurface volume of interest
CN108226997A (en) A kind of seismic facies analysis method based on earthquake data before superposition
US20150081259A1 (en) Method of calibrating a geologic model
US10677948B2 (en) Context based bounded hydrocarbon formation identification
Yu Using cylindrical surface-based curvature change rate to detect faults and fractures
US20200041677A1 (en) System and method for seismic amplitude analysis
CN114429161A (en) Seismic waveform clustering method, electronic device, and medium
Liu et al. Automatic estimation of traveltime parameters in VTI media using similarity-weighted clustering
Liu et al. Robust prestack seismic facies analysis using shearlet transform-based deep learning
CN112987091A (en) Reservoir detection method and device, electronic equipment and storage medium
Shen et al. Maximum correntropy self-organizing map for waveform classification
US11852768B2 (en) Multimodal approach to target stratigraphic plays through seismic sequence stratigraphy, rock physics, seismic inversion and machine learning
US11768303B2 (en) Automatic data enhancement for full waveform inversion in the midpoint-offset domain
CN114428306A (en) Variable-coefficient strong reflection separation method, device, electronic device and medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination