A method and device for obtaining a well logging curve, and an electronic device

Abstract

Embodiments of the present application provide a method and device for obtaining well logging curves and an electronic device. The method for obtaining well logging curves comprises: obtaining seismic response data of each well logging in a target work area; processing the seismic response data of each well logging using a trained well logging facies recognition model to obtain predicted well logging facies of each well logging in the target work area; obtaining a plurality of target well loggings according to the predicted well logging facies of each well logging in the target work area, wherein the target well logging is a well logging within a dominant facies range; and performing the following processing on each target well logging: obtaining a plurality of original well logging curves of the target well logging using different curve logging methods; determining a sensitive curve from the plurality of original well logging curves using a preset sensitive curve screening algorithm; and obtaining a final well logging curve of the target well logging according to the sensitive curve using a preset curve optimization algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geophysical exploration, and in particular to a method and device for obtaining well logging curves and an electronic device. BACKGROUND

[0002] In the process of seismic data interpretation, in complex reservoir conditions such as low permeability, tightness, and multiple lithological components, improving the identification ability of electrical properties to reservoirs through curve optimization is a common and important work.

[0003] In related technologies, different reservoir electrical properties are not distinguished, which affects the inversion accuracy. There are various solutions to this problem, for example, using curve sensitivity intersection analysis and other means to refit a curve so that it can distinguish different reservoirs and improve the identification ability of electrical properties to reservoirs. However, the above method does not fully consider the spatial distribution of wells participating in curve reconstruction, resulting in wells that lack indicative characteristics also participating in curve reconstruction, which leads to poor curve reconstruction results.

[0004] Therefore, how to optimize the well logging curve and improve the identification ability of electrical properties to reservoirs is a technical problem to be solved. SUMMARY

[0005] The embodiments of the present application provide a method and device for obtaining well logging curves, an electronic device and a storage medium, which are used to optimize the well logging curve and improve the identification ability of electrical properties to reservoirs.

[0006] One of the embodiments of the present application provides a method for obtaining well logging curves, the method comprising: obtaining seismic response data of each well logging in a target work area; using a trained well logging facies recognition model to process the seismic response data of each well logging to obtain predicted well logging facies of each well logging in the target work area; wherein the well logging facies recognition model is a trained machine learning model; obtaining a plurality of target wells according to the predicted well logging facies of each well logging in the target work area; wherein the target well is a well within the dominant facies range; for each target well, the following processing is performed: using different curve logging methods to obtain a plurality of original well logging curves of the target well; using a preset sensitive curve screening algorithm to determine a sensitive curve from the plurality of original well logging curves; using a preset curve optimization algorithm to obtain a final well logging curve of the target well according to the sensitive curve.

[0007] In some embodiments, the well logging facies identification model is obtained by: obtaining a training sample set; wherein the training sample set comprises a plurality of groups of training samples, each group of training samples comprising seismic response data as sample data and well logging facies as labels; processing the seismic response data in each group of training samples using an initial well logging facies identification model to obtain predicted well logging facies corresponding to the seismic response data in each group of training samples; wherein the initial well logging facies identification model is a clustering model; adjusting parameters of the initial well logging facies identification model according to differences between the predicted well logging facies corresponding to the seismic response data in each group of training samples and the labels of the seismic response data until a convergence condition of the model is met to obtain a trained well logging facies identification model.

[0008] In some embodiments, the obtaining of the training sample set comprises: processing well logging facies and seismic response data of a plurality of wells using well-seismic joint technology to obtain a correspondence between the well logging facies and the seismic response data; and according to the correspondence between the well logging facies and the seismic response data, taking each type of seismic response data and the well logging facies corresponding thereto as a group of training samples in the training sample set.

[0009] In some embodiments, the well logging facies identification model is a clustering model, and the processing of the seismic response data of each well using the trained well logging facies identification model to obtain predicted well logging facies of each well in the target work area comprises: clustering the seismic response data of each well using the well logging facies identification model to obtain a plurality of sets of seismic response data and predicted well logging facies corresponding thereto; and taking the predicted well logging facies corresponding to each set of seismic response data as predicted well logging facies of each well corresponding to the seismic response data in the set of seismic response data.

[0010] One of the embodiments of the present application provides a device for obtaining well logging curves, the device comprising: a first obtaining module configured to obtain seismic response data of each well in a target work area; a second obtaining module configured to process the seismic response data of each well using a trained well logging facies identification model to obtain predicted well logging facies of each well in the target work area; wherein the well logging facies identification model is a trained machine learning model; a third obtaining module configured to obtain a plurality of target wells according to the predicted well logging facies of each well in the target work area; wherein the target well is a well within a dominant facies range; and a fourth obtaining module configured to perform the following processing on each target well: obtaining a plurality of original well logging curves of the target well using different curve logging methods; determining a sensitive curve from the plurality of original well logging curves using a preset sensitive curve screening algorithm; and obtaining a final well logging curve of the target well according to the sensitive curve using a preset curve optimization algorithm.

[0011] The electronic device provided in the embodiments of the present application comprises a memory and a processor, the memory stores a computer program, and the processor executes the program to perform the method described above.

[0012] The embodiments of the present application provide a storage medium for storing a computer readable program, which, when executed, performs the method described above.

[0013] The above technical solutions provided in the embodiments of the present application have at least the following advantages compared with the prior art.

[0014] In the embodiments provided in the present application, the trained logging facies recognition model is used to process the seismic response data of each well to obtain the predicted logging facies of each well in the target work area; the multiple target wells are obtained according to the predicted logging facies of each well in the target work area; for each target well, the following processing is performed: different curve logging methods are used to obtain multiple original logging curves of the target well; the preset sensitive curve screening algorithm is used to determine the sensitive curve from the multiple original logging curves; and the preset curve optimization algorithm is used to obtain the final logging curve of the target well according to the sensitive curve. Thus, the logging curve can be effectively optimized, and the identification ability of electrical properties to reservoirs can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] The present application will be further illustrated in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same reference numbers represent the same structures, wherein:

[0016] Figure 1 is an exemplary flowchart of a logging curve acquisition method according to some embodiments of the present application;

[0017] Figure 2 is an exemplary flowchart of a logging facies recognition model training method according to some embodiments of the present application;

[0018] Figure 3 is an exemplary schematic diagram of a seismic facies and logging facies correspondence relationship according to some embodiments of the present application;

[0019] Figure 4 is an exemplary schematic diagram of a clustering result according to some embodiments of the present application;

[0020] Figure 5 is an exemplary schematic diagram of a crossplot method preferred sensitive curve according to some embodiments of the present application;

[0021] Figure 6a is an exemplary schematic diagram of an original natural gamma-ray histogram according to some embodiments of the present application;

[0022] Figure 6b is an exemplary schematic diagram of a natural gamma histogram after direct curve optimization according to some embodiments of the present application;

[0023] Figure 6c is an exemplary schematic diagram of a natural gamma histogram after phase control curve optimization according to some embodiments of the present application;

[0024] Figure 7 is an exemplary schematic diagram of a logging curve acquisition device according to some embodiments of the present application;

[0025] Figure 8 is an exemplary structural schematic diagram of an electronic device according to some embodiments of the present application. DETAILED DESCRIPTION

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some examples or embodiments of the present application, and for those skilled in the art, the present application can also be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language context or otherwise indicated, the same reference numbers in the drawings represent the same structures or operations.

[0027] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0028] As shown in the present application and claims, unless the context clearly indicates otherwise, "one", "a", "an" and / or "the" do not refer to the singular, but can also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0029] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or subsequent operations are not necessarily performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can be added to these processes, or one or more steps of the operation can be removed from these processes.

[0030] In order to facilitate understanding, the technical solutions of the present application will be introduced below in combination with the drawings and embodiments.

[0031] Figure 1is an exemplary flowchart of a method for obtaining a well log curve according to some embodiments of the present application. As shown in Figure 1 the method for obtaining a well log curve includes the following steps:

[0032] In step S110, seismic response data of each well log of a target work area is obtained.

[0033] The target work area is a work area for which the geological structure of the reservoir in which it is located needs to be identified through a well log curve. In some embodiments, the target work area can be a region with complex reservoir conditions such as low permeability, tightness, and multiple lithological components.

[0034] The seismic response data can include, but is not limited to, the amplitude, continuity, frequency, and other characteristics of the seismic reflection wave.

[0035] In step S120, the seismic response data of each well log is processed using a trained well facies identification model to obtain the predicted well facies of each well log of the target work area; wherein the well facies identification model is a trained machine learning model.

[0036] Well facies refers to the curve characteristics, lithology, and other characteristics of a well log at different depth sections.

[0037] In the specific implementation process, the well facies identification model can be constructed based on various machine learning models, including but not limited to classification models (such as decision trees, random forests, etc.), clustering models (such as K-means clustering algorithm), etc.

[0038] For the training method of the well facies identification model, please refer to the related content in Figure 2 , which will not be described here.

[0039] In some embodiments, the well facies identification model is a clustering model, and the input of the well facies identification model is the seismic response data; the output is a set of seismic response data, which can be represented using various identifiers, for example, Arabic numerals (1, 2, 3, 4, etc.) can be used to represent different sets of seismic response data.

[0040] In some embodiments, the trained well facies identification model can be used to process the seismic response data of each well log to obtain the predicted well facies of each well log of the target work area in the following way:

[0041] The seismic response data of each well log is clustered using the well facies identification model to obtain multiple sets of seismic response data and their corresponding predicted well facies; the predicted well facies corresponding to each set of seismic response data is taken as the predicted well facies of the well corresponding to each seismic response data in the set of seismic response data.

[0042] As an example only, a well logging facies identification model can be used to obtain, for instance, the following results. Figure 4 The clustering results shown ( Figure 4 Within the area marked by the black line, the clustering result represents a planar distribution range with similar river phase reflection characteristics.

[0043] Step S130: Based on the predicted logging facies of each well in the target work area, multiple target wells are obtained; wherein, the target wells are wells within the dominant facies range.

[0044] The dominant facies refers to the facies zone in which the target geological body is most developed.

[0045] In the specific implementation process, based on the predicted logging facies, it can be determined whether the geological body corresponding to each logging in the target work area is the most developed, and the logging of the most developed geological body is taken as the target logging.

[0046] For each target logging obtained in step S130, perform the following steps.

[0047] Step S140: Using different curve logging methods, various original logging curves of the target well are obtained.

[0048] The original logging curve is the curve formed during logging. It can reflect the different lithology and stratigraphic characteristics of different reservoirs, and then the lithology, stratigraphic characteristics, etc. of a specific reservoir can be determined based on the logging curve.

[0049] Commonly used curve determination methods include: natural gamma logging, resistance logging, and sonic transit logging.

[0050] Step S150: Using a preset sensitive curve screening algorithm, a sensitive curve is determined from a variety of original logging curves.

[0051] Sensitive curves refer to one or more logging curves that best reflect the characteristics of the target geological body among various logging curves.

[0052] In some embodiments, the preset sensitive curve screening algorithm can be a cross-plot analysis algorithm, which can be used to select the sensitive curve of the target well from the various original logging curves of the target well obtained in step S140.

[0053] For example only, such as Figure 5 As shown, histogram cross-plot analysis was performed on the natural gamma curve, resistivity curve, sonic curve, and density curve to determine the identification ability of different curves for the target well logging of sandstone and mudstone. The analysis results show that the natural gamma curve has the best identification effect, so the natural gamma curve can be used as the sensitive curve.

[0054] In step S160, according to the sensitive curve, a preset curve optimization algorithm is used to obtain the final logging curve of the target well.

[0055] The curve optimization algorithm is an algorithm for re-fitting the sensitive curve into a new curve according to certain rules, so as to improve the identification ability of the sensitive curve to the reservoir.

[0056] In the specific implementation process, the preset curve optimization algorithm can be implemented by using a neural network according to the processing effect, or can be implemented by using mathematical operation, and is not limited by the description in the specification.

[0057] For example, according to the comparative effect of FIGS. 6a, 6b and 6c, the phase-controlled optimized natural gamma ray curve has the best effect on distinguishing sand and shale, that is, the overlapping area of sand and shale in the histogram is the smallest, and therefore, the curve after the phase optimization processing can be used as the final logging curve of the target well. Figure 6a 6b

[0058] In the embodiment of the present application, the trained logging facies recognition model is used to process the seismic response data of each well to obtain the predicted logging facies of each well in the target work area, and according to the predicted logging facies of each well in the target work area, a plurality of target wells are obtained, which fully develops the spatial constraint advantage of the seismic response data, selects the well with the most developed geological body, and performs curve logging, effectively improves the identification ability of the electrical property to the reservoir, and has important significance for subsequent inversion and reservoir prediction.

[0059] Figure 2 is an example flowchart of a training method of a logging facies recognition model according to some embodiments of the present application. As shown in Figure 2 , the method comprises the following steps.

[0060] Step S210, obtaining a training sample set; wherein the training sample set comprises a plurality of groups of training samples, and each group of training samples comprises seismic response data as sample data and logging facies as a label.

[0061] For detailed description of the seismic response data and the logging facies, refer to the related content in Figure 1 , which will not be described here.

[0062] In the specific implementation process, the logging facies and the seismic response data of a plurality of wells can be processed by using well-seismic joint technology to obtain the corresponding relationship between the logging facies and the seismic response data; according to the corresponding relationship between the logging facies and the seismic response data, each type of seismic response data and the corresponding logging facies are used as a group of training samples in the training sample set.

[0063] ​​Well-seismic joint refers to using time-depth calibration means to establish a relationship between depth domain logging and time domain seismic data in the time domain. The well-seismic joint can determine the corresponding seismic response characteristics of different logging sections.

[0064] For example, for multiple wells (such as W1-W4) in a target work area, the correspondence between the logging facies of each well and the seismic response can be determined by the well-seismic joint technology, and the correspondence between the lithological characteristics of multiple well points and the wave group reflection characteristics of the seismic can be established as shown in FIG. 1C, to obtain a training sample set. Figure 3 Figure 3

[0065] In the specific implementation process, in order to comprehensively understand the geological information of the target work area, the macro-geological data of the target work area can be collected. The macro-geological data is helpful for geophysicists to understand the sedimentary environment of the area, and has guiding significance for improving the understanding of seismic facies. The macro-geological data can include but is not limited to the paleogeomorphology, stratigraphic characteristics, sedimentary model, and sedimentary facies of an area.

[0066] In step S220, the seismic response data in each training sample is processed using an initial logging facies recognition model to obtain the predicted logging facies corresponding to the seismic response data in each training sample. The initial logging facies recognition model is a clustering model.

[0067] For details of using the initial logging facies recognition model to process the seismic response data in each training sample to obtain the predicted logging facies corresponding to the seismic response data in each training sample, refer to the related content in step S120, which will not be repeated here.

[0068] In step S230, the parameters of the initial logging facies recognition model are adjusted according to the difference between the predicted logging facies corresponding to the seismic response data in each training sample and the label of the seismic response data, until the convergence condition of the model is met, to obtain the trained logging facies recognition model.

[0069] For example, the initial logging facies recognition model is a K-mean clustering model. The difference between the actual result and the recognition result can be determined by using a loss function, and the parameters of the K-mean clustering model can be adjusted by using an optimization algorithm to iteratively train the initial logging facies recognition model.

[0070] The convergence condition of the model can be that the accuracy of the recognition of the initial logging facies recognition model is greater than a preset threshold (for example, 99%), or the number of training reaches a preset value.

[0071] Figure 7 FIG. 1A is an example schematic diagram of an acquisition device of a logging curve according to some embodiments of the present application. ​​

[0072] As Figure 7 shown in FIG. 7, the log curve acquisition device includes a first acquisition module 710, a second acquisition module 720, a third acquisition module 730, and a fourth acquisition module 740.

[0073] The first acquisition module 710 is configured to acquire seismic response data of each log in a target work area.

[0074] The second acquisition module 720 is configured to process the seismic response data of each log using a trained log facies identification model to obtain a predicted log facies of each log in the target work area; wherein the log facies identification model is a trained machine learning model.

[0075] The third acquisition module 730 is configured to obtain a plurality of target logs according to the predicted log facies of each log in the target work area; wherein the target log is a log within a dominant facies range.

[0076] The fourth acquisition module 740 is configured to perform the following processing on each target log: using different curve logging methods to obtain a plurality of original log curves of the target log; using a preset sensitive curve screening algorithm to determine a sensitive curve from the plurality of original log curves; and using a preset curve optimization algorithm to obtain a final log curve of the target log according to the sensitive curve.

[0077] In some embodiments, the log facies identification model is obtained by: obtaining a training sample set; wherein the training sample set includes a plurality of groups of training samples, and each group of training samples includes seismic response data as sample data and a log facies as a label; using an initial log facies identification model to process the seismic response data in each group of training samples to obtain a predicted log facies corresponding to the seismic response data in each group of training samples; wherein the initial log facies identification model is a clustering model; adjusting parameters of the initial log facies identification model according to a difference between the predicted log facies corresponding to the seismic response data in each group of training samples and the label of the seismic response data until a convergence condition of the model is met to obtain a trained log facies identification model.

[0078] In some embodiments, the training sample set is obtained by: using well-seismic joint technology to process log facies and seismic response data of a plurality of logs to obtain a correspondence between the log facies and the seismic response data; and according to the correspondence between the log facies and the seismic response data, taking each type of seismic response data and the log facies corresponding thereto as a group of training samples in the training sample set.

[0079] In some embodiments, the well facies identification model is a clustering model, and the processing of the seismic response data of each well by using the trained well facies identification model to obtain the predicted well facies of each well in the target work area comprises: clustering the seismic response data of each well by using the well facies identification model to obtain a plurality of sets of seismic response data and corresponding predicted well facies; and taking the predicted well facies corresponding to each set of seismic response data as the predicted well facies of the well corresponding to each piece of seismic response data in the set of seismic response data.

[0080] In the embodiments of the well curve acquisition device, the specific processing of each module and the technical effects brought by the specific processing can be referred to the related descriptions in the corresponding method embodiments, which will not be repeated here.

[0081] Figure 8 is an exemplary structural schematic diagram of an electronic device according to some embodiments of the present application.

[0082] As Figure 8 shown, the electronic device comprises at least one processor 801, at least one communication interface 802, at least one memory 803, and at least one communication bus 804. Optionally, the communication interface 802 can be an interface of a communication module, such as an interface of a GSM module. The processor 801 can be a processor CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present application. The memory 803 can include a high-speed RAM memory, and can also include a non-volatile memory, such as at least one disk memory. The memory 803 stores a program, and the processor 801 invokes the program stored in the memory 803 to execute part or all of the method embodiments described above.

[0083] The present application relates to a storage medium for storing a computer readable program, which, when executed, performs part or all of the method embodiments described above.

[0084] Optionally, the storage medium can be a non-transitory computer readable storage medium, for example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0085] Based on the same inventive concept, the embodiments of the present application also provide a computer program product comprising a computer program, which, when executed by a processor, implements part or all of the method embodiments described above.

[0086] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this application. Such modifications, improvements, and corrections are suggested in this application, and therefore remain within the spirit and scope of the exemplary embodiments of this application.

[0087] Furthermore, this application uses specific terms to describe its embodiments. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic related to at least one embodiment of this application. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this application do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of this application can be appropriately combined.

[0088] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this application are not intended to limit the order of the processes and methods of this application. Although the foregoing disclosure has discussed some currently considered useful embodiments of the invention through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments of this application. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely through software solutions, such as installing the described system on existing servers or mobile devices.

[0089] Similarly, it should be noted that, in order to simplify the description of the present application and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of the embodiments of the present application sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of the application requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0090] In some embodiments, numbers that describe amounts, dimensions, and so forth, are used in the description of the embodiments. It should be understood that such numbers are used only to illustrate certain embodiments and that the application is not limited to the numbers. In some examples, such numbers are modified by the modifier "about" or "approximately." Unless otherwise indicated, "about" or "approximately" shall be understood to refer to a variation of ±20% of the value. Accordingly, in some embodiments, the numerical parameters in the description and claims are approximations that can vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, numerical parameters are determined by the use of standard techniques. Although the numerical ranges and parameters setting forth the broad scope of the application are approximations, the numerical values set forth in the specific examples are reported as precisely as practicable. The numerical values set forth in the specific examples are provided to be as precise as reasonably possible. However, some variations may occur depending on the choice of the desired end result.

[0091] Each patent, patent application, publication, and other material cited in this application is hereby incorporated by reference in its entirety. In the event of inconsistencies between the disclosure of this application and the materials incorporated by reference, the disclosure of this application shall prevail. To the extent the documents incorporated by reference contradict the disclosure of this application, the disclosure of this application shall control. It is specifically intended that the description of the present application, as set forth in the specification and the claims, be considered as illustrative only and not restrictive. It is further intended that the scope of the application should not be limited to the embodiments set forth herein but should be given the broadest scope consistent with the specification and the appended claims.

[0092] Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of this application. Numerous modifications may be made by one skilled in the art that will apply to the principles described herein. Accordingly, the application is not limited to the embodiments described herein but rather is intended to embrace all modifications that fall within the scope of the claims.

Claims

1. A method of obtaining a well log, characterized in that, The method comprises: obtaining seismic response data of each well log in a target work area; processing the seismic response data of each well log using a trained well log facies identification model to obtain predicted well log facies of each well log in the target work area; wherein the well log facies identification model is a trained machine learning model; obtaining a plurality of target well logs according to the predicted well log facies of each well log in the target work area; wherein the target well log is a well log within a dominant facies range; for each target well log, the following processing is performed: obtaining a plurality of original well log curves of the target well log using different curve logging methods; determining a sensitive curve from the plurality of original well log curves using a preset sensitive curve screening algorithm; obtaining a final well log curve of the target well log according to the sensitive curve using a preset curve optimization algorithm; The well log facies identification model is obtained by: obtaining a training sample set; wherein the training sample set comprises a plurality of training samples, and each training sample comprises seismic response data as sample data and well log facies as a label; processing the seismic response data in each training sample using an initial well log facies identification model to obtain predicted well log facies corresponding to the seismic response data in each training sample; wherein the initial well log facies identification model is a clustering model; adjusting parameters of the initial well log facies identification model according to differences between the predicted well log facies corresponding to the seismic response data in each training sample and the labels of the seismic response data until a convergence condition of the model is met to obtain a trained well log facies identification model.

2. The method of claim 1, wherein, The obtaining of the training sample set comprises: processing well log facies and seismic response data of a plurality of well logs using well-seismic joint technology to obtain a correspondence between the well log facies and the seismic response data; according to the correspondence between the well log facies and the seismic response data, each type of seismic response data and the well log facies corresponding thereto are taken as a training sample in the training sample set.

3. The method of claim 1, wherein, The well log facies identification model is a clustering model, and the processing of the seismic response data of each well log using the trained well log facies identification model to obtain predicted well log facies of each well log in the target work area comprises: clustering the seismic response data of each well log using the well log facies identification model to obtain a plurality of seismic response data sets and predicted well log facies corresponding thereto; taking the predicted well log facies corresponding to each seismic response data set as predicted well log facies of each well log corresponding to the seismic response data in the seismic response data set.

4. An apparatus for obtaining a well log, characterized in that The device comprises: a first obtaining module configured to obtain seismic response data of each well log in a target work area; a second obtaining module configured to process the seismic response data of each well log using a trained well log facies identification model to obtain predicted well log facies of each well log in the target work area; wherein the well log facies identification model is a trained machine learning model; a third obtaining module configured to obtain a plurality of target well logs according to the predicted well log facies of each well log in the target work area; wherein the target well log is a well log within a dominant facies range. A fourth obtaining module is configured to perform the following processing on each target well log: Different curve logging methods are used to obtain multiple original well logs of the target well log; A preset sensitive curve screening algorithm is used to determine a sensitive curve from the multiple original well logs; A preset curve optimization algorithm is used to obtain a final well log of the target well log according to the sensitive curve; The well facies recognition model is obtained by the following method: A training sample set is obtained; the training sample set includes multiple groups of training samples, and each group of training samples includes seismic response data as sample data and well facies as a label; An initial well facies recognition model is used to process the seismic response data in each group of training samples to obtain a predicted well facies corresponding to the seismic response data in each group of training samples; the initial well facies recognition model is a clustering model; Parameters of the initial well facies recognition model are adjusted according to a difference between the predicted well facies corresponding to the seismic response data in each group of training samples and the label of the seismic response data until a convergence condition of the model is met, to obtain a trained well facies recognition model.

5. The apparatus of claim 4, wherein, The training sample set is obtained by: Using well-seismic joint technology, well facies and seismic response data of multiple wells are processed to obtain a corresponding relationship between the well facies and the seismic response data; According to the corresponding relationship between the well facies and the seismic response data, each type of seismic response data and the well facies corresponding thereto are taken as a group of training samples in the training sample set.

6. The apparatus of claim 4, wherein, The well facies recognition model is a clustering model, and the trained well facies recognition model is used to process the seismic response data of each well to obtain a predicted well facies of each well in the target work area, including: The well facies recognition model is used to cluster the seismic response data of each well to obtain multiple seismic response data sets and predicted well facies corresponding thereto; The predicted well facies corresponding to each seismic response data set is taken as a predicted well facies of a well corresponding to each seismic response data in the seismic response data set.

7. An electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor executing the program to perform the method of any one of claims 1 to 3.

8. A storage medium for storing a computer readable program, the computer readable program being executed to perform the method of any one of claims 1 to 3.

Images