CN114239655A - Seismic facies recognition model training method and device and seismic facies prediction method and device - Google Patents

Seismic facies recognition model training method and device and seismic facies prediction method and device Download PDF

Info

Publication number
CN114239655A
CN114239655A CN202111549826.XA CN202111549826A CN114239655A CN 114239655 A CN114239655 A CN 114239655A CN 202111549826 A CN202111549826 A CN 202111549826A CN 114239655 A CN114239655 A CN 114239655A
Authority
CN
China
Prior art keywords
seismic
seismic data
facies
data set
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
CN202111549826.XA
Other languages
Chinese (zh)
Inventor
袁三一
贺粟梅
陈�胜
宋朝辉
王尚旭
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 University of Petroleum Beijing
Original Assignee
China University of Petroleum Beijing
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 University of Petroleum Beijing filed Critical China University of Petroleum Beijing
Priority to CN202111549826.XA priority Critical patent/CN114239655A/en
Publication of CN114239655A publication Critical patent/CN114239655A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/22—Source localisation; Inverse modelling

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Signal Processing (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

The invention relates to the field of oil and gas exploration, in particular to a seismic facies recognition model training method and device and a seismic facies prediction method and device, wherein the training method comprises the following steps of 1, acquiring seismic data; step 2, determining a similar data set of each seismic data in the first seismic data set according to the correlation of the seismic data in the first seismic data set and the seismic data in the second seismic data set; step 3, assigning the seismic facies labels of the seismic data in the first seismic data set to the related similar data sets; step 4, training a seismic facies recognition model; step 5, judging whether the second seismic data set is empty, if so, ending the training; if not, executing the step 6; step 6, recognizing boundary seismic data in the similar data set by using the seismic facies recognition model obtained by training; step 7, updating the first seismic data set according to the boundary seismic data; returning to the step 2 to the step 5, the method can reduce the dependence degree on the seismic facies label and improve the seismic facies prediction precision.

Description

Seismic facies recognition model training method and device and seismic facies prediction method and device
Technical Field
The invention relates to the field of oil and gas exploration, in particular to a seismic facies recognition model training method and device and a seismic facies prediction method and device.
Background
With the continuous deepening of oil and gas exploration, the accurate division and identification of seismic facies categories play an important role in the exploration of underground geological environment and geological structure, high-resolution processing, seismic inversion, reservoir prediction, structural interpretation and other works.
The traditional seismic facies identification needs interpreters to visually observe seismic section information, and reasonable seismic facies division results are given by combining professional knowledge and experience of the interpreters. The process has great randomness, is easy to lose detail information, and is difficult to construct a complex nonlinear intrinsic relation between seismic facies categories and seismic data. In the prior art, a method for identifying seismic facies by using a deep learning network is available, but a large amount of manually explained seismic data is required to be used for training, and then the seismic data which is not manually explained is predicted. The manual interpretation cost is high, and the work difficulty of acquiring the strong supervision information of the seismic phase truth value label is large.
Aiming at the problems that in the prior art, detailed information is easy to lose, the manual operation cost is high, the difficulty in obtaining tagged data is high and the like, an earthquake phase automatic identification method is urgently needed to be researched.
Disclosure of Invention
To solve the above problems of the prior art, embodiments herein provide a training method for a seismic facies recognition model. The problem of weak dependence automatic identification of the seismic facies label under the conditions of less seismic sample data volume for training and single seismic data distribution is solved, the dependence degree on the seismic facies label is reduced, the interpretation efficiency is improved, meanwhile, the seismic facies automatic identification precision is greatly improved, and the prediction precision of the whole seismic data set is improved.
The embodiment of the invention provides a training method of a seismic facies recognition model, which comprises the following steps: step 1, acquiring seismic data, wherein the seismic data comprises a first seismic data set and a second seismic data set, the first seismic data set comprises a plurality of seismic data with seismic facies labels, and the second seismic data set comprises a plurality of seismic data without seismic facies labels; step 2, determining a similar data set of each seismic data in the first seismic data set according to the correlation between the seismic data in the first seismic data set and the seismic data in the second seismic data set; step 3, assigning the seismic facies labels of the seismic data in the first seismic data set to related similar data sets, and updating the second seismic data set according to the similar data sets; step 4, training a seismic facies recognition model by utilizing the first seismic data set and the similar data set; step 5, judging whether the second seismic data set is empty, if so, ending the training; if not, executing the step 6; step 6, recognizing boundary seismic data in the similar data set by using a seismic facies recognition model obtained by training so as to update a seismic facies label of the boundary seismic data; step 7, updating the first seismic data set according to the boundary seismic data; and returning to execute the step 2 to the step 5.
According to one aspect of embodiments herein, determining a similar dataset for each seismic data in the first seismic dataset based on the correlation of the first seismic dataset with the seismic data in the second seismic dataset comprises: calculating a correlation of each seismic data in the first set of seismic data with each seismic data in the second set of seismic data; and for each seismic data in the first seismic data set, screening out seismic data of which the correlation exceeds a preset threshold value from the second seismic data set, and taking the screened-out seismic data as a similar data set of the seismic data in the first seismic data set.
According to one aspect of embodiments herein, calculating the correlation of each seismic data in the first set of seismic data with each seismic data in the second set of seismic data comprises calculating the correlation using the formula:
Figure BDA0003416848530000021
wherein A is a space vector in one seismic data in the first seismic data set; b is a space vector of seismic data of the seismic data in the second set of seismic data; m is the total number of the transverse vector dimensions of the seismic data, and n is the total number of the longitudinal vector dimensions of the seismic data; i represents the ith transverse vector of the seismic data; j represents the jth longitudinal vector of the seismic data.
According to one aspect of embodiments herein, acquiring the seismic data comprises: segmenting the initial seismic data volume to obtain a plurality of initial seismic data; sending the initial seismic data to a user terminal; receiving a seismic facies label of part of initial seismic data sent by a user terminal; and carrying out data cleaning and normalization processing on the initial seismic data to obtain the seismic data.
According to one aspect of embodiments herein, updating the second set of seismic data from the similar set of data comprises: deleting the similar dataset from the second seismic dataset; updating the first seismic dataset from the boundary seismic data comprises: replacing seismic data in the first seismic dataset with the boundary seismic data.
According to an aspect of an embodiment herein, the seismic facies recognition model is trained by using any one of the methods described above, the method including: acquiring seismic data; and inputting the seismic data into a seismic facies recognition model, and predicting to obtain the seismic facies category of the seismic data.
Embodiments herein also provide a training apparatus for a seismic facies recognition model, including: the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring seismic data, the seismic data comprises a first seismic data set and a second seismic data set, the first seismic data set comprises a plurality of seismic data with seismic facies labels, and the second seismic data set comprises a plurality of seismic data without seismic facies labels; a determining unit, configured to determine a similar data set of each seismic data in the first seismic data set according to a correlation between the seismic data in the first seismic data set and the seismic data in the second seismic data set; the first updating unit is used for assigning the seismic facies labels of the seismic data in the first seismic data set to related similar data sets and updating the second seismic data set according to the similar data sets; the training unit is used for training a seismic facies recognition model by utilizing the first seismic data set and the similar data set; the control unit is used for judging whether the second seismic data set is empty or not, and if yes, finishing training; if not, starting the identification unit; the first updating unit is used for updating the first seismic data set according to the first seismic data set; the recognition unit is used for recognizing boundary seismic data in the similar data set by using a seismic facies recognition model obtained through training so as to update a seismic facies label of the boundary seismic data; a second updating unit for updating the first seismic dataset according to the boundary seismic data.
Embodiments herein also provide a seismic facies prediction apparatus, comprising: an acquisition unit for acquiring seismic data; and the prediction unit is used for inputting the seismic data into the seismic facies recognition model and predicting the seismic facies category.
Embodiments herein also provide a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the above-mentioned method when executing the computer program.
Embodiments herein also provide a computer-readable storage medium having stored thereon computer instructions, which when executed by a processor, implement the above-described method.
In the embodiment of the invention, the explained seismic data (first seismic data) is used as the priori knowledge of the unexplained seismic data (second seismic data), and the difference of the characteristic domains of the similar seismic data is used as the increment constraint, so that the problems of low seismic sample data volume, single seismic data distribution, low dependence degree on seismic facies labels, high seismic facies automatic prediction precision, large manual workload, difficult cross-domain identification and the like can be solved.
Drawings
In order to more clearly illustrate the embodiments or technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram illustrating a seismic phase identification system according to an embodiment of the disclosure;
FIG. 2 is a flow chart illustrating a method for training a seismic facies recognition model according to an embodiment of the present disclosure;
FIG. 3 is a flow chart illustrating a method of determining similar data sets according to an embodiment of the present disclosure;
FIG. 4 is a flow chart illustrating a method of determining seismic data according to an embodiment of the disclosure;
FIG. 5 is a flow diagram illustrating a method for predicting seismic facies labels for seismic data according to an embodiment herein;
FIG. 6 is a schematic structural diagram illustrating a training apparatus for a seismic facies recognition model according to an embodiment of the present disclosure;
FIG. 7 is a schematic structural diagram of a seismic facies recognition model training apparatus according to an embodiment of the present disclosure;
FIG. 8 is a schematic diagram illustrating an initial seismic data volume according to embodiments herein;
FIGS. 9A and 9B are schematic diagrams illustrating seismic data and corresponding seismic facies tags according to embodiments herein;
fig. 10 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure.
Description of the symbols of the drawings:
101. a server;
102. a user terminal;
601. an acquisition unit;
6011. a segmentation module;
6012. a sending module;
6013. a receiving module;
6014. a data processing module;
602. a determination unit;
6021. a correlation calculation module;
603. a first update unit;
604. a training unit;
605. a control unit;
606. an identification unit;
6061. a prediction module;
607. a second updating unit;
1002. a computer device;
1004. a processor;
1006. a memory;
1008. a drive mechanism;
1010. an input/output module;
1012. an input device;
1014. an output device;
1016. a presentation device;
1018. a graphical user interface;
1020. a network interface;
1022. a communication link;
1024. a communication bus.
Detailed Description
In order to make the technical solutions in the present specification better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments, but not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments herein without making any creative effort, shall fall within the scope of protection.
It should be noted that the terms "first," "second," and the like in the description and claims herein and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments herein described are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, apparatus, article, or device that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or device.
The present specification provides method steps as described in the examples or flowcharts, but may include more or fewer steps based on routine or non-inventive labor. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. When an actual system or apparatus product executes, it can execute sequentially or in parallel according to the method shown in the embodiment or the figures.
It should be noted that the method and apparatus for training the seismic facies recognition model herein can be used in the field of oil and gas exploration, and the application field of the method and apparatus for training the seismic facies recognition model herein is not limited.
Fig. 1 is a schematic structural diagram of a seismic facies recognition system according to an embodiment of the present disclosure, in which a seismic facies recognition model training method combining a server 101 and a user terminal 102 is described. Data interaction can be performed between the server 101 and the user terminal 102.
In some embodiments of the present specification, the server 101 may be an electronic device having a network interaction function, or may be software running in the electronic device and providing service logic for data processing and network interaction. The server 101 may obtain a seismic facies tag of a part of initial seismic data sent by the user terminal 102; the server 101 may be used to obtain seismic data, calculate correlations of seismic data in the first set of seismic data and the second set of seismic data, train seismic facies recognition models, and the like.
In some embodiments of the present description, 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, and the like. The user terminal 102 is not limited to the electronic device with a certain entity, and may also be software running in the electronic device. The user terminal 102 is configured to receive initial seismic data sent by the server 101, and the user manually interprets a part of the initial seismic data and labels a seismic facies tag on the part of the initial seismic data. The user terminal 102 is further configured to send seismic facies tags of a portion of the initial seismic data to the server 101 for subsequent data cleansing, normalization, and the like. The user terminal 102 further includes displaying the portion of the initial seismic data and the labeled seismic facies label on an interactive display interface.
Fig. 2 is a flowchart illustrating a method for training a seismic facies recognition model according to an embodiment of the present disclosure.
In the field of seismic exploration, the nature and the form of a subterranean rock stratum can be inferred by observing and analyzing the response of the stratum to seismic waves by utilizing the difference of the elasticity and the density of the subterranean medium. Specifically, a three-dimensional seismic data volume is obtained by collecting seismic waves, and the three-dimensional seismic data volume is subjected to stacking processing to form a stacked three-dimensional seismic data volume. The three-dimensional seismic data volume is combined with the interpretation label, can describe seismic facies characteristics, interpret stratum attributes and analyze reservoir characteristics of different types of terrains, and provides a basis for oil field exploration. In some embodiments of the present description, a seismic facies recognition model is trained by using a priori knowledge of some of the seismic data in the stacked three-dimensional seismic data volume as unexplained seismic data, the number of seismic sample data used for training is gradually increased by using the difference of the characteristic domains of the seismic data as an incremental constraint, a label propagation process is implemented by using the similarity of similar seismic data, and the seismic facies category of the unexplained seismic data is predicted.
Step 201, acquiring seismic data, where the seismic data includes a first seismic data set and a second seismic data set, where the first seismic data set includes a plurality of seismic data with seismic facies labels, and the second seismic data set includes a plurality of seismic data without seismic facies labels.
In this step, the first seismic data set and the second seismic data set in the seismic data acquired in this step may be three-dimensional seismic images, two-dimensional seismic images, or data obtained by performing vectorization processing on seismic images.
In some embodiments of the present description, seismic data is described by way of example as a two-dimensional seismic image. The first seismic dataset and the second seismic dataset in the seismic data are obtained by processing an initial seismic dataset, and a detailed description about the initial seismic dataset can be seen in fig. 4. The first seismic data set is a small amount of manually interpreted seismic data, and each seismic data in the first seismic data set has a corresponding seismic facies label. The seismic data in the second set of seismic data is seismic data that has not been manually interpreted, and thus each seismic data set has no corresponding seismic facies label. Furthermore, the plurality of seismic facies tags in the first seismic dataset correspond to subsurface terrain and reservoir resource distribution. For example, seismic facies tags may include, but are not limited to, one or any combination of basement, sloped mudstone a, sloped mudstone B, bulk handling sedimentary rock, valley, sea canyon, and the like. In some embodiments of the present description, the artificially interpreted seismic facies tags may be adjusted based on the actual conditions of the survey area, terrain. The application is not limited to the type or form of seismic facies tags.
Step 202, determining a similar data set of the seismic data in the first seismic data set according to the correlation of the seismic data in the first seismic data set and the seismic data in the second seismic data set.
Corresponding to the continuity of the underground structure, there is a strong similarity between adjacent post-stack sections of the underground three-dimensional seismic data volume, that is, there is also a certain continuity in space in the seismic data acquired in step 201. Characterized from seismic data as: there is a correlation between spatially adjacent seismic data. Due to this characteristic, the result of the classification of the seismic facies classes is substantially consistent between spatially adjacent seismic data. And based on the geological rule, performing seismic facies identification by taking the difference of the characteristic domains of the similar seismic data as a constraint. Thus, a similar dataset to each seismic data in the first seismic dataset needs to be determined. The specific steps for determining similar datasets are shown in fig. 3.
Step 203, assigning the seismic facies labels of the seismic data in the first seismic data set to the related similar data sets, and updating the second seismic data set according to the similar data sets. And calculating a similar data set of certain seismic data in the screened first seismic data set according to the correlation in the step 202, and assigning a seismic facies label of the seismic data to the similar data set, so that each seismic data in the similar data set has the same seismic facies label as the seismic data. For example, the seismic data acquired in step 201 is spatially divided into five hundred consecutive two-dimensional seismic maps. The first seismic data set is the 50 th, 100 th, 270 th, 380 th and 440 th images and has corresponding seismic facies labels. Specifically, the two-dimensional seismic map in the first seismic dataset is divided according to pixels, and each pixel is manually explained in advance and has a corresponding seismic facies label. The seismic facies labels of each pixel can be classified into a plurality of categories, and the categories respectively correspond to underground structures such as basement or other structures, slope mudstone, block-shaped transportation sedimentary deposit, slope valleys, submarine canyons and the like. For example, the 50 th image is a picture of 10 by 10 pixels, the 50 th image is manually interpreted according to the pixels, and the seismic facies label of the 1 st to 3 rd by 10 th pixel area along the longitudinal direction is a label 0; the seismic facies label of the 4 th to 8 th by 10 pixel area is label 2; the seismic facies label for the 9 th to 10 th by 10 th pixel regions is label 3. By analogy, the 100 th, 270 th, 380 th, 440 th plots in the first seismic dataset also have a seismic facies label corresponding to each pixel.
The seismic facies labels of the 5 seismic data in the first seismic data set are assigned to their similar data sets. In some embodiments of the present description, labels corresponding to respective pixels of the 50 th image may be assigned to their corresponding similar data sets as a class of large labels. Assigning the seismic facies labels of the 50 th graph to the 40 th through 49 th graphs of the similar data set associated with the graph; assigning the seismic facies labels of the 100 th graph to the 80 th through 99 th graphs of the similar data set associated with the graph; assigning the seismic facies labels of the 270 th graph to the 271 th graph to the 290 th graph of the similar dataset associated with the graph; assigning the seismic facies labels of the 380 th graph to the 360 th through 379 th graphs of the similar data set associated with the graph; the seismic facies labels of the 440 th graph are assigned to the 420 th through 439 th graphs of the similar dataset associated with the graph. Thus, the seismic data associated with each seismic data in the first seismic data set has a corresponding seismic facies label. In some embodiments of the present description, the similar dataset for each seismic data in the first seismic dataset may be other seismic data or any variant thereof, and the present application does not limit the similar dataset for each seismic data in the first seismic dataset.
Further, this step includes updating the second set of seismic data with the similar set of data. In accordance with the above description, the determined similar dataset associated with the first seismic dataset originally belongs to the second seismic dataset, and currently, each seismic data in the similar dataset has an assigned seismic facies label, so that both the similar dataset needs to be deleted from the second seismic dataset to update the second seismic dataset.
And step 204, training a seismic facies recognition model by using the first seismic data set and the similar data set. And training a data driving model by utilizing a plurality of seismic data with seismic facies labels in the first seismic data set and a plurality of seismic data in the similar data set, and acquiring a seismic facies identification initial model. And inputting the first seismic data set with the seismic facies labels and the similar data set with the assigned seismic facies labels into a data driving model for supervised training to obtain an initial seismic facies recognition model.
The specific process of training the seismic facies recognition initial model can be formulated as:
Predicti+1＝MODELi[SIMILARITY(Datai),Predicti]
wherein i represents the number of model iterations; MODEL is a trained seismic facies recognition MODEL; presectiA prediction tag representing the seismic data after the ith model iteration; data representing existing earthquakeSeismic data of facies tags. Predict when i is 0, i.e. model training is not started0＝Label，Predict0As Data0A corresponding real tag. SIMILARITY is similarity discriminator, which uses seismic Data with existing seismic facies labelsiAnd searching a similar data set meeting a similarity threshold value for reference, and forming a training data set by the similar data set and the corresponding seismic facies label.
Step 205, judging whether the second seismic data set is empty, if so, ending the training; if not, go to step 206. This step is used to judge the end condition of the model training. And when all the seismic data in the seismic data are endowed with corresponding labels, namely, the second seismic data set is updated, and no data exist in the second seismic data set, the training of the seismic facies label identification model is determined to be finished. And if a part of seismic data still has seismic facies labels which do not correspond to the seismic data, namely the second seismic data set is not empty, continuing to train the seismic facies label identification model.
And step 206, identifying the boundary seismic data in the similar data set by using the seismic facies identification model obtained by training so as to update the seismic facies labels of the boundary seismic data. And inputting the boundary seismic data in the similar data set and the seismic facies labels corresponding to the boundary seismic data into a seismic facies identification model, and predicting by the seismic facies identification model to obtain a seismic facies prediction result of the boundary seismic data. In this step, the boundary seismic data in the similar data set are seismic data at both ends in the similar data set. For example, the similar dataset corresponds to the 20 th to 30 th seismic image in the seismic data, and the boundary seismic data in the similar dataset is the 20 th and 30 th seismic image in the seismic data. And predicting the seismic facies of the 20 th and 30 th seismic images by using the seismic facies recognition model, and regarding the seismic facies prediction results as real labels. By analogy, for other boundary seismic data in other similar data sets, the corresponding seismic facies prediction result can be identified by using the seismic facies identification model.
Step 207, updating the first seismic dataset according to the boundary seismic data; and returning to execute the step 202 to the step 205.
In this step, the boundary seismic data and 2 seismic facies labels obtained by prediction thereof are substituted for the seismic data in the first seismic data set. Specifically, the original seismic data in the first seismic data set is deleted, and 2 seismic facies labels obtained by predicting the boundary seismic data set are used as new seismic data of the first seismic data. For example, 5 similar datasets of 5 seismic data in the first seismic dataset are determined in step 202, which are respectively the 40 th image-49 th image, the 80 th image-99 th image, the 271 th image-290 th image, the 360 th image-379 th image and the 420 th image-439 th image of the seismic data. Step 203 assigns 5 seismic facies labels corresponding to the seismic data to the 5 similar data sets. The boundary seismic data in the 5 similar data sets are the 40 th image, the 49 th image, the 80 th image, the 99 th image, the 271 th image, the 290 th image, the 360 th image, the 379 th image, the 420 th image and the 439 th image. In this step, the boundary data and the seismic facies labels thereof are substituted for seismic data in the first seismic data set. Then the current first seismic dataset includes 10 seismic data, respectively 10 boundary seismic data in 5 similar datasets.
And then returning to the step 2 to the step 5, determining similar datasets related to 10 boundary seismic data in the first seismic dataset by using the seismic data in the current first seismic dataset, assigning seismic facies labels of the 10 boundary seismic data to the similar datasets, and updating the second seismic dataset. And judging whether the second seismic data set is empty, identifying the boundary seismic data in the similar data set by using the seismic facies identification model again under the condition that the second seismic data set is not empty, updating the seismic facies label of the boundary seismic data, and updating the first seismic data set again. And a cycle is formed, so that the process that the seismic facies labels are transmitted from the explained seismic data to the unexplained seismic data is realized, and the generalization capability of the data driving model is improved by fully utilizing the unexplained data.
Fig. 3 is a flow chart of a method of determining a similar data set according to an embodiment herein.
Step 301, calculating a correlation of each seismic data in the first seismic data set with each seismic data in the second seismic data set. Wherein each seismic data in the seismic dataset may reflect seismic waveform characteristics, such as: amplitude, phase, continuity, etc. Similarity calculation is carried out according to each seismic data reflecting the seismic wave amplitude in the first seismic data set and each seismic data reflecting the seismic wave amplitude in the second seismic data set. In particular, a similar dataset is determined that is correlated to each seismic data in the first seismic dataset. And calculating the correlation of each seismic data in the first seismic data set and each seismic data in the second seismic data set by adopting a similarity calculation method. The method for calculating the similarity includes, but is not limited to: cosine similarity formula, pearson correlation coefficient, Jaccard similarity coefficient, Tanimoto coefficient, etc. The method of calculating the correlation between seismic data is not limited in this application.
In the step, a cosine similarity calculation formula is taken as an example, and the correlation between each seismic data in the first seismic data set and each seismic data in the second seismic data set is calculated. The calculation formula is as follows:
Figure BDA0003416848530000101
Figure BDA0003416848530000102
wherein A is a space vector in one seismic data in the first seismic data set; b is a space vector of seismic data in the second seismic data set; m is the total number of the transverse vector dimensions of the seismic data, and n is the total number of the longitudinal vector dimensions of the seismic data; i represents the ith transverse vector of the seismic data; j represents the jth longitudinal vector of the seismic data. Specifically, by performing vectorization processing on the seismic data (two-dimensional seismic image) in the first seismic data and the second seismic data, the seismic data in the first seismic data and the second seismic data can be respectively represented in a vector form. For example, the seismic image in the first seismic data is a 10 by 10 pixel image, and may be represented by a space vector a, where m and n are 10, respectively. Similarly, B may represent individual pixels of the seismic image in the second seismic data.
The method comprises the steps that the seismic data in the first seismic data set and the second seismic data set contain more features, the feature vectorization processing of each two-dimensional seismic image in the first seismic data set and the second seismic data set is carried out and is mapped to corresponding space vectors, and the cosine representation of the included angle between each space vector of each two-dimensional seismic image in the first seismic data set and each space vector of each two-dimensional seismic image in the second seismic data set is calculated. The similarity between two seismic data is determined by measuring the cosine value of the inner product space of the two vectors, and the difference between the two-dimensional images is further measured.
In some embodiments of the present description, the correlation of each seismic data in the first set of seismic data with each seismic data in the second set of seismic data may also be calculated separately using a plurality of similarity calculation methods. The correlation results obtained by the calculation of the multiple similarity calculation methods are averaged, so that the similarity between two seismic data can be determined, and the difference between two-dimensional images can be further measured.
Step 302, for each seismic data in the first seismic data set, selecting seismic data with a correlation exceeding a preset threshold from the second seismic data set, and using the selected seismic data as a similar data set of the seismic data in the first seismic data set. Seismic data having a correlation exceeding a preset threshold may be understood as seismic data having a greater correlation with seismic data in the first seismic data. The preset threshold value can be preset by the server, and can also be adjusted according to the actual similarity calculation condition.
According to the similarity, a similarity data set consisting of the screened seismic data is calculated, and the similarity exists between the space and the corresponding first seismic data, so that the corresponding first seismic data can give a seismic facies label to the seismic facies label.
FIG. 4 is a flow chart of a method of determining seismic data according to an embodiment herein.
Step 401, segmenting the initial seismic data volume to obtain a plurality of initial seismic data. In this step, seismic waves are excited to collect field data by using explosive sources according to a certain sampling time, and an initial seismic data volume is obtained (see fig. 8). As shown in FIG. 8, the initial seismic data volume forms a subsurface three-dimensional seismic data volume along the longitudinal direction of the formation, the horizontal direction of the surface, which may characterize the nature and morphology of the subsurface geological formation. The initial seismic data volume is segmented along the ground horizontal direction, so that a plurality of initial seismic data can be obtained, namely a plurality of two-dimensional section seismic images continuously distributed in the horizontal direction (as shown in fig. 9A). As shown in fig. 9A, one of the two-dimensional sectional seismic images in which the initial seismic data are continuously distributed in the horizontal direction is one of the initial seismic data. The abscissa in fig. 9A is a common depth point gather, which represents a point gather collected along the ground horizontal direction, and the ordinate in fig. 9A corresponds to the time of receiving seismic waves, further reflecting the subsurface longitudinal depth. FIG. 9A may provide feedback of various signatures of seismic waves, and further feedback of subsurface lithology and formation morphology.
Step 402, sending the initial seismic data to a user terminal.
And receiving the seismic facies label of the part of the initial seismic data sent by the user terminal. And the user manually interprets the received initial seismic data at the user terminal, namely, the seismic facies label is marked on the initial seismic data. Specifically, the initial seismic data is marked as a base or other seismic facies labels through the characteristic of low signal-to-noise ratio of the initial seismic data; marking initial seismic data with strong amplitude of upper and lower boundaries and mostly low amplitude or semi-continuous internal reflection as a seismic facies label of a slope mudstone A; marking initial seismic data mixed with chaotic phase and low-amplitude parallel reflection as seismic phase labels for carrying sedimentary rocks in block shapes; marking the initial seismic data of the strong-amplitude parallel reflection as a seismic facies label of the sloping mudstone B; marking the initial seismic data as a seismic facies label of a slope valley according to a waveform which presents a high-amplitude lower cut valley and relatively low fluctuation in the initial seismic data; and marking the initial seismic data as a seismic facies label of the strait valley according to the waveform that the erosion base in the initial seismic data is U-shaped and has larger local fluctuation. As shown in fig. 9B, a schematic diagram of a seismic facies label corresponding to the initial seismic data of fig. 9A is shown, the initial seismic data is manually interpreted along the ground horizontal direction and the subsurface longitudinal depth, and the seismic facies label is marked.
And step 403, receiving the seismic facies label of the part of the initial seismic data sent by the user terminal. And the user manually explains part of the initial seismic data and marks the corresponding seismic facies label on the part of the seismic data. This step may be performed by receiving the seismic facies tag of a portion of the initial seismic data stored at the user terminal.
And step 404, performing data cleaning and normalization processing on the initial seismic data to obtain the seismic data. When the initial seismic data is manually interpreted, there may be situations where the tag label is omitted, the interpretation is wrong, and the data label of the image edge of the initial seismic data does not correspond. Therefore, data cleaning and normalization processing need to be performed on the initial seismic data to acquire the seismic data. The seismic data acquired in this step is the seismic data for which the seismic facies recognition model training is performed in fig. 2.
FIG. 5 is a flow chart of a method for predicting seismic facies labels for seismic data in accordance with an embodiment of the present application. The method comprises the following steps:
step 501, acquiring seismic data;
step 502, inputting the seismic data into a seismic facies recognition model, and predicting to obtain the seismic facies category of the seismic data.
FIG. 5 illustrates a method for predicting seismic facies categories for seismic data using a seismic facies recognition model. The seismic data includes seismic data without seismic facies tags. And obtaining the seismic facies label of the input data by using the trained seismic facies identification model.
Fig. 6 is a schematic structural diagram of a seismic facies recognition model training apparatus according to an embodiment of the present disclosure, in which a basic structure of the seismic facies recognition model training apparatus is described, where functional units and modules may be implemented in a software manner, or may also be implemented in a general chip or a specific chip, and a part or all of the functional units and modules may be on a server, or a part of the functional units and modules may also be on a user terminal, and the seismic facies recognition model training apparatus is implemented through cooperation with the server, and the apparatus specifically includes:
an obtaining unit 601, configured to obtain seismic data, where the seismic data includes a first seismic dataset and a second seismic dataset, where the first seismic dataset includes a plurality of seismic data with seismic facies labels, and the second seismic dataset includes a plurality of seismic data without seismic facies labels;
a determining unit 602, configured to determine a similar dataset of each seismic data in the first seismic dataset according to a correlation between the seismic data in the first seismic dataset and the seismic data in the second seismic dataset;
a first updating unit 603, configured to assign a seismic facies label of each seismic data in the first seismic data set to a related similar data set, and update the second seismic data set according to the similar data set;
a training unit 604, configured to train a seismic facies recognition model using the first seismic dataset and the similar dataset;
a control unit 605, configured to determine whether the second seismic data set is empty, and if so, end the training; if not, the identification unit 606 is started; the second updating unit 607 is also used for restarting the determining unit 602, the first updating unit 603 and the training unit 604 after the first seismic data set is updated;
the recognition unit 606 is configured to recognize boundary seismic data in the similar data set by using the trained seismic facies recognition model, so as to update a seismic facies tag of the boundary seismic data;
a second updating unit 607 for updating said first seismic dataset in dependence of said boundary seismic data.
In the embodiment of the invention, the explained seismic data (first seismic data) is used as the prior knowledge of the unexplained seismic data (second seismic data), and the difference of the characteristic domains of the similar seismic data is used as the increment constraint, so that the problems of low seismic sample data volume, single seismic data distribution, reduced dependence degree on seismic facies labels, high seismic facies automatic prediction precision and seismic data set prediction precision, large manual workload, difficult cross-domain identification and the like are solved.
As an embodiment of this document, referring to a specific structural schematic diagram of the training apparatus for seismic facies recognition model shown in fig. 7, the obtaining unit 601 is further configured to process an initial seismic data volume before obtaining seismic data;
as an embodiment herein, the obtaining unit 601 further includes:
a segmentation module 6011, configured to segment the initial seismic data volume to obtain a plurality of initial seismic data;
a sending module 6012, configured to send the initial seismic data to a user terminal;
a receiving module 6013, configured to receive a seismic facies tag of part of initial seismic data sent by a user side;
and a data processing module 6014, configured to perform data cleaning and normalization on the initial seismic data.
As an embodiment herein, the determining unit 602 is configured to calculate a correlation between the seismic data in the first seismic data set and the seismic data in the second seismic data set, and further includes:
a correlation calculation module 6021 for calculating a correlation of the seismic data in the first seismic data set with the seismic data in the second seismic data set;
as an embodiment herein, the identifying unit 606 further comprises:
a prediction module 6061 for predicting seismic facies labels for the boundary seismic data in the similar dataset using the seismic facies recognition model.
As shown in fig. 10, for a computer device provided for embodiments herein, the computer device 1002 may include one or more processors 1004, such as one or more Central Processing Units (CPUs), each of which may implement one or more hardware threads. The computer device 1002 may also include any memory 1006 for storing any kind of information, such as code, settings, data, etc. For example, and without limitation, the memory 1006 may include any one or more of the following in combination: any type of RAM, any type of ROM, flash memory devices, hard disks, optical disks, 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 fixed or removable components of computer device 1002. In one case, when the processor 1004 executes the associated instructions, which are stored in any memory or combination of memories, the computer device 1002 can perform any of the operations of the associated instructions. The computer device 1002 also includes one or more drive mechanisms 1008, such as a hard disk drive mechanism, an optical disk drive mechanism, or the like, for interacting with any memory.
Computer device 1002 may also include an input/output module 1010(I/O) for receiving various inputs (via input device 1012) and for providing various outputs (via output device 1014). One particular output mechanism may include a presentation device 1016 and an associated Graphical User Interface (GUI) 1018. In other embodiments, input/output module 1010(I/O), input device 1012, and output device 1014 may also be excluded, as only one computer device in a network. Computer device 1002 can 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 above-described components together.
Communication link 1022 may be implemented in any manner, such as over a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. Communications 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.
Corresponding to the methods in fig. 1-5, the embodiments herein also provide a computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, performs the steps of the above-described method.
Embodiments herein also provide computer readable instructions, wherein when executed by a processor, a program thereof causes the processor to perform the method as shown in fig. 1-5.
It should be understood that, in various embodiments herein, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments herein.
It should also be understood that, in the embodiments herein, the term "and/or" is only one kind of association relation describing an associated object, meaning that three kinds of relations may exist. For example, a and/or B, may represent: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described in a functional general in the foregoing description for the purpose of illustrating clearly the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided herein, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may also be an electric, mechanical or other form of connection.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments herein.
In addition, functional units in the embodiments herein may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present invention may be implemented in a form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The principles and embodiments of this document are explained herein using specific examples, which are presented only to aid in understanding the methods and their core concepts; meanwhile, for the general technical personnel in the field, according to the idea of this document, there may be changes in the concrete implementation and the application scope, in summary, this description should not be understood as the limitation of this document.

Claims (10)

1. A method for training a seismic facies recognition model, the method comprising:
step 1, acquiring seismic data, wherein the seismic data comprises a first seismic data set and a second seismic data set, the first seismic data set comprises a plurality of seismic data with seismic facies labels, and the second seismic data set comprises a plurality of seismic data without seismic facies labels;
step 2, determining a similar data set of each seismic data in the first seismic data set according to the correlation between the seismic data in the first seismic data set and the seismic data in the second seismic data set;
step 3, assigning the seismic facies labels of the seismic data in the first seismic data set to related similar data sets, and updating the second seismic data set according to the similar data sets;
step 4, training a seismic facies recognition model by utilizing the first seismic data set and the similar data set;
step 5, judging whether the second seismic data set is empty, if so, ending the training; if not, executing the step 6;
step 6, recognizing boundary seismic data in the similar data set by using a seismic facies recognition model obtained by training so as to update a seismic facies label of the boundary seismic data;
and 7, updating the first seismic data set according to the boundary seismic data, and returning to execute the steps 2 to 5.
2. The method of seismic facies recognition model training as claimed in claim 1, wherein determining a similar dataset for each seismic data in the first seismic dataset based on the correlation of the seismic data in the first seismic dataset and the second seismic dataset comprises:
calculating a correlation of each seismic data in the first set of seismic data with each seismic data in the second set of seismic data;
and for each seismic data in the first seismic data set, screening out seismic data of which the correlation exceeds a preset threshold value from the second seismic data set, and taking the screened-out seismic data as a similar data set of the seismic data in the first seismic data set.
3. The method of seismic facies recognition model training as claimed in claim 1, wherein calculating the correlation of each seismic data in the first set of seismic data with each seismic data in the second set of seismic data comprises calculating the correlation using the formula:
Figure FDA0003416848520000021
wherein A is a space vector in one seismic data in the first seismic data set; b is a space vector of seismic data of the seismic data in the second set of seismic data; m is the total number of the transverse vector dimensions of the seismic data, and n is the total number of the longitudinal vector dimensions of the seismic data; i represents the ith transverse vector of the seismic data; j represents the jth longitudinal vector of the seismic data.
4. The seismic facies recognition model training method of claim 1, wherein said acquiring seismic data comprises:
segmenting the initial seismic data volume to obtain a plurality of initial seismic data;
sending the initial seismic data to a user terminal;
receiving a seismic facies label of part of initial seismic data sent by a user terminal;
and carrying out data cleaning and normalization processing on the initial seismic data to obtain the seismic data.
5. The seismic facies recognition model training method of claim 1, wherein updating the second set of seismic data based on the similar set of data comprises: deleting the similar dataset from the second seismic dataset;
updating the first seismic dataset from the boundary seismic data comprises: replacing seismic data in the first seismic dataset with the boundary seismic data.
6. A method for seismic facies prediction, characterized in that a seismic facies recognition model is trained using the method of any one of claims 1 to 5, the method comprising:
acquiring seismic data;
and inputting the seismic data into a seismic facies recognition model, and predicting to obtain the seismic facies category of the seismic data.
7. A seismic facies recognition model training apparatus, said apparatus comprising:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring seismic data, the seismic data comprises a first seismic data set and a second seismic data set, the first seismic data set comprises a plurality of seismic data with seismic facies labels, and the second seismic data set comprises a plurality of seismic data without seismic facies labels;
a determining unit, configured to determine a similar data set of each seismic data in the first seismic data set according to a correlation between the seismic data in the first seismic data set and the seismic data in the second seismic data set;
the first updating unit is used for assigning the seismic facies labels of the seismic data in the first seismic data set to related similar data sets and updating the second seismic data set according to the similar data sets;
the training unit is used for training a seismic facies recognition model by utilizing the first seismic data set and the similar data set;
the control unit is used for judging whether the second seismic data set is empty or not, and if yes, finishing training; if not, the identification unit is started, and the identification unit is also used for restarting the determining unit, the first updating unit and the training unit after the second updating unit updates the first seismic data set;
the recognition unit is used for recognizing boundary seismic data in the similar data set by using a seismic facies recognition model obtained through training so as to update a seismic facies label of the boundary seismic data;
a second updating unit for updating the first seismic dataset according to the boundary seismic data.
8. A seismic phase prediction apparatus, wherein a seismic phase recognition model is trained using the apparatus of claim 7, the apparatus comprising:
an acquisition unit for acquiring seismic data;
and the prediction unit is used for inputting the seismic data into the seismic facies recognition model and predicting the seismic facies category.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1-6 when executing the computer program.
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 method of any one of claims 1-6.
CN202111549826.XA 2021-12-17 2021-12-17 Seismic facies recognition model training method and device and seismic facies prediction method and device Pending CN114239655A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111549826.XA CN114239655A (en) 2021-12-17 2021-12-17 Seismic facies recognition model training method and device and seismic facies prediction method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111549826.XA CN114239655A (en) 2021-12-17 2021-12-17 Seismic facies recognition model training method and device and seismic facies prediction method and device

Publications (1)

Publication Number Publication Date
CN114239655A true CN114239655A (en) 2022-03-25

Family

ID=80758123

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111549826.XA Pending CN114239655A (en) 2021-12-17 2021-12-17 Seismic facies recognition model training method and device and seismic facies prediction method and device

Country Status (1)

Country Link
CN (1) CN114239655A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116563298A (en) * 2023-07-12 2023-08-08 南京茂莱光学科技股份有限公司 Cross line center sub-pixel detection method based on Gaussian fitting

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116563298A (en) * 2023-07-12 2023-08-08 南京茂莱光学科技股份有限公司 Cross line center sub-pixel detection method based on Gaussian fitting
CN116563298B (en) * 2023-07-12 2023-09-08 南京茂莱光学科技股份有限公司 Cross line center sub-pixel detection method based on Gaussian fitting

Similar Documents

Publication Publication Date Title
Shi et al. Development of subsurface geological cross-section from limited site-specific boreholes and prior geological knowledge using iterative convolution XGBoost
Arnold et al. Optimisation of decision making under uncertainty throughout field lifetime: A fractured reservoir example
CN111596978A (en) Web page display method, module and system for lithofacies classification by artificial intelligence
Bruna et al. A new methodology to train fracture network simulation using multiple-point statistics
CN104335074A (en) System and method for facies classification
US20230161061A1 (en) Structured representations of subsurface features for hydrocarbon system and geological reasoning
Tokpanov et al. Deep-learning-based automated stratigraphic correlation
CN117576335B (en) Three-dimensional space model data processing method and system for mineral area investigation
CN114548299A (en) Earthquake phase identification method and device, computer equipment and storage medium
WO2022197846A1 (en) Method to automatically pick formation tops using optimization algorithm
Reolon et al. Unlocking data analytics for the automatic evaluation of cement bond scenarios
CN114239655A (en) Seismic facies recognition model training method and device and seismic facies prediction method and device
US10605940B2 (en) Method for selecting horizon surfaces
WO2022026827A1 (en) Systems and methods for identifying subsurface features as a function of position in a subsurface volume of interest
Udegbe et al. Big Data Analytics for Seismic Fracture Identification, Using Amplitude-Based Statistics
CN114402233A (en) Automatic calibration of forward deposition model
Li et al. Reservoir ranking map sketching for selection of infill and replacement drilling locations using machine learning technique
EP3579026B1 (en) Method of characterising a subsurface region using multiple point statistics
CA3108964C (en) Petroleum play analysis and display
CN114063169B (en) Wave impedance inversion method, system, equipment and storage medium
Mohammadpour et al. Effect of spatial variability of downhole geophysical logs on machine learning exercises
Yousefzadeh et al. Geological Uncertainty Quantification
Ramadhan et al. Unravel Carbonate Reservoir Heterogeneities Through 3D Seismic Neural Network Application: A Machine Learning Based Approach and Workflow
US20230235660A1 (en) System and method for uncertainty calculation in unconventional hydrocarbon reservoirs
CN116068640A (en) Salt dome identification method and device and computer equipment

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