Method, device and equipment for acquiring low-frequency attribute model and readable storage medium

Abstract

The invention discloses a low-frequency attribute model acquisition method, device and equipment and a readable storage medium, and belongs to the technical field of seismic inversion. The method comprises the following steps: acquiring a three-dimensional geological structure model of a target area, a first attribute value of each sampling point included in an observation well located in the target area and a second attribute value of each seismic node included in a seismic body corresponding to the target area; generating a first sequence small layer model according to the three-dimensional geological structure model, each sampling point and each seismic node; domain transformation is carried out on the first sequence small-layer model to obtain a second sequence small-layer model under the sedimentary domain; according to a first attribute value and a second attribute value of a target node in nodes included in each layer in the second sequence small-layer model, obtaining a layer model corresponding to each layer in the second sequence small-layer model; and generating a low-frequency attribute model according to the second attribute value of each node included in the second layer sequence small-layer model and the layer model corresponding to the layer where each node is located. And the precision of the low-frequency attribute model is improved.

Description

Technical Field

[0001] This application relates to the field of seismic inversion technology, and in particular to a method, apparatus, device, and readable storage medium for obtaining a low-frequency attribute model. Background Technology

[0002] Seismic inversion, as an important means of reservoir prediction, can effectively predict the parameters of subsurface models that correspond to seismic responses. The subsurface model parameters obtained from the prediction can be used to characterize subsurface reservoirs. Therefore, the accuracy of seismic inversion directly restricts the economic efficiency and effectiveness of oil and gas development.

[0003] Since low-frequency attribute models can be used for seismic inversion, their accuracy directly affects the accuracy of seismic inversion. Therefore, a method for obtaining high-accuracy low-frequency attribute models is needed to improve the accuracy of seismic inversion and reservoir prediction. Summary of the Invention

[0004] This application provides a method, apparatus, device, and readable storage medium for acquiring low-frequency attribute models, which can be used to improve the accuracy of the acquired low-frequency attribute models, thereby improving the accuracy of seismic inversion and reservoir prediction. The technical solution is as follows:

[0005] On one hand, embodiments of this application provide a method for obtaining a low-frequency attribute model, the method comprising:

[0006] A three-dimensional geological structure model of the target area is obtained, along with the first attribute values ​​of each sampling point in the observation wells located in the target area and the second attribute values ​​of each seismic node in the seismic body corresponding to the target area. The first and second attributes are correlated.

[0007] Based on the three-dimensional geological structure model, the sampling points, and the seismic nodes, a first sequence sub-layer model is generated, which includes the sampling points and the seismic nodes.

[0008] A domain transformation is performed on the first sequence sublayer model to obtain a second sequence sublayer model under the sedimentation domain. The layers in the second sequence sublayer model have uniform and identical thicknesses.

[0009] Based on the first attribute value and the second attribute value of the target node in each layer of the second hierarchical sub-layer model, obtain the layer model corresponding to each layer in the second hierarchical sub-layer model. The layer model corresponding to any layer is used to reflect the relationship between the first attribute and the second attribute of each node included in the layer. The target node is a node that has a first attribute value and a second attribute value.

[0010] A low-frequency attribute model is generated based on the second attribute values ​​of each node included in the second sequence sub-layer model and the layer model corresponding to the layer in which each node is located. The low-frequency attribute model includes the first attribute values ​​of each seismic node.

[0011] In one possible implementation, generating the first sequence sub-layer model based on the three-dimensional geological structure model, the various sampling points, and the various seismic nodes includes:

[0012] Based on the three-dimensional geological structure model, a reference sequence sub-layer model corresponding to the target area is generated;

[0013] Each sampling point is added to the reference sequence sub-layer model to obtain a reference sequence sub-layer model including each sampling point.

[0014] Each seismic node is added to a reference sequence sub-layer model that includes each sampling point to obtain the first sequence sub-layer model.

[0015] In one possible implementation, generating a reference sequence sub-layer model corresponding to the target region based on the three-dimensional geological structure model includes:

[0016] Spatial parameterization is performed on the three-dimensional geological structure model to obtain the positions of each node in the sequence sub-layer model.

[0017] Based on the well trajectories of each observation well and the top and bottom surfaces of each stratum, determine the number of sequence sublayers included in each stratum;

[0018] Based on the positions of each node in the sequence sub-layer model included in the three-dimensional geological structure model and the number of sequence sub-layers included in each stratum, the reference sequence sub-layer model is generated according to the sedimentary pattern.

[0019] In one possible implementation, determining the number of sequence sublayers included in each formation based on the well trajectories of each observation well and the top and bottom surfaces of each formation includes:

[0020] For any stratum in each of the strata, the thickness of the stratum corresponding to each observation well is determined based on the first intersection point and the second intersection point of each observation well. The first intersection point of any observation well is the intersection point between the well trajectory of any observation well and the top surface of any stratum, and the second intersection point of any observation well is the intersection point between the well trajectory of any observation well and the bottom surface of any stratum.

[0021] Determine the maximum thickness among the formation thicknesses corresponding to each observation well;

[0022] The number of sequence sublayers included in any stratum is determined based on the maximum thickness and the sampling interval of the low-frequency attribute model.

[0023] In one possible implementation, adding each sampling point to the reference sequence sub-layer model to obtain a reference sequence sub-layer model including each sampling point includes:

[0024] Each sampling point is resampled to obtain target sampling points, and the sampling interval between the target sampling points is the sampling interval of the low-frequency attribute model.

[0025] The target sampling point is processed into a linear channel to obtain the channel position of the target sampling point in the hierarchical sub-layer model;

[0026] The target sampling point is parameterized to obtain its layer position in the hierarchical sub-layer model;

[0027] Based on the line position and layer position of the target sampling point in the hierarchical sub-layer model, the target sampling point is added to the reference hierarchical sub-layer model to obtain a reference hierarchical sub-layer model including the target sampling point.

[0028] In one possible implementation, adding each seismic node to a reference sequence sub-layer model including each sampling point to obtain the first sequence sub-layer model includes:

[0029] Each of the earthquake nodes is resampled to obtain the target earthquake nodes, and the sampling interval between the target earthquake nodes is the sampling interval of the low-frequency attribute model.

[0030] The target seismic node is parameterized to obtain its layer position in the sequence sub-layer model.

[0031] Based on the layer position of the target seismic node in the sequence sub-layer model, the target seismic node is added to the reference sequence sub-layer model including the sampling points to obtain the first sequence sub-layer model.

[0032] In one possible implementation, generating a low-frequency attribute model based on the second attribute values ​​of each node included in the second hierarchical sub-layer model and the layer model corresponding to the layer in which each node is located includes:

[0033] The values ​​of each node are determined based on the second attribute values ​​of each node included in the second hierarchical sub-layer model and the layer model corresponding to the layer in which each node is located.

[0034] Map the values ​​of each node to the construction domain to obtain the values ​​of each node under the construction domain;

[0035] Based on the values ​​of each node in the tectonic domain, each node is resampled to obtain the values ​​of each seismic node.

[0036] Based on the values ​​of each seismic node, determine the first attribute value of each seismic node;

[0037] The low-frequency attribute model is generated based on the first attribute values ​​of the seismic body and each seismic node.

[0038] In one possible implementation, the method further includes:

[0039] For a reference node in the second hierarchical sub-layer model, an error field is determined based on the first attribute value and the numerical value of the reference node. The reference node is a node in the second hierarchical sub-layer model that has both a first attribute value and a second attribute value.

[0040] The step of determining the first attribute value of each seismic node based on the values ​​of each seismic node includes:

[0041] Based on the values ​​of each seismic node and the error field, the first attribute value of each seismic node is determined.

[0042] In one possible implementation, determining the error field based on the first attribute value of the reference node and the numerical value of the reference node includes:

[0043] Determine the error between the first attribute value of the reference node and the numerical value of the reference node;

[0044] The error field is determined based on the error between the first attribute value of the reference node and the numerical value of the reference node.

[0045] In one possible implementation, determining the first attribute value of each seismic node based on the values ​​of each seismic node and the error field includes:

[0046] For any one of the earthquake nodes, the sum of the numerical value of the earthquake node and the error field is determined to be the first attribute value of the earthquake node.

[0047] On the other hand, embodiments of this application provide a device for obtaining a low-frequency attribute model, the device comprising:

[0048] The acquisition module is used to acquire a three-dimensional geological structure model of the target area, the first attribute values ​​of each sampling point of the observation well located in the target area, and the second attribute values ​​of each seismic node of the seismic body corresponding to the target area. The first attribute and the second attribute are related.

[0049] The generation module is used to generate a first sequence sub-layer model based on the three-dimensional geological structure model, the sampling points and the seismic nodes. The first sequence sub-layer model includes the sampling points and the seismic nodes.

[0050] The transformation module is used to perform domain transformation on the first sequence sublayer model to obtain a second sequence sublayer model under the deposition domain, wherein the thickness of the layers in the second sequence sublayer model is uniform and the same.

[0051] The acquisition module is further configured to acquire the layer model corresponding to each layer in the second hierarchical sub-layer model based on the first attribute value and the second attribute value of the target node in each layer of the second hierarchical sub-layer model. The layer model corresponding to any layer is used to reflect the relationship between the first attribute and the second attribute of each node included in the layer. The target node is a node that has a first attribute value and a second attribute value.

[0052] The generation module is further configured to generate a low-frequency attribute model based on the second attribute values ​​of each node included in the second hierarchical sub-layer model and the layer model corresponding to the layer in which each node is located, wherein the low-frequency attribute model includes the first attribute values ​​of each seismic node.

[0053] In one possible implementation, the generation module is used to generate a reference sequence sub-model corresponding to the target area based on the three-dimensional geological structure model; add each sampling point to the reference sequence sub-model to obtain a reference sequence sub-model including each sampling point; and add each seismic node to the reference sequence sub-model including each sampling point to obtain the first sequence sub-model.

[0054] In one possible implementation, the generation module is used to perform spatial parameterization processing on the three-dimensional geological structure model to obtain the positions of each node included in the three-dimensional geological structure model in the sequence sub-layer model; determine the number of sequence sub-layers included in each stratum based on the well trajectories of each observation well, the top surface and bottom surface of each stratum; and generate the reference sequence sub-layer model according to the sedimentary mode based on the positions of each node included in the three-dimensional geological structure model in the sequence sub-layer model and the number of sequence sub-layers included in each stratum.

[0055] In one possible implementation, the generation module is configured to, for any stratum among the various formations, determine the formation thickness corresponding to each observation well based on the first intersection point and the second intersection point of each observation well, wherein the first intersection point of any observation well is the intersection point between the well trajectory of any observation well and the top surface of any formation, and the second intersection point of any observation well is the intersection point between the well trajectory of any observation well and the bottom surface of any formation; determine the maximum thickness among the formation thicknesses corresponding to each observation well; and determine the number of sequence sublayers included in any formation based on the maximum thickness and the sampling interval of the low-frequency attribute model.

[0056] In one possible implementation, the generation module is configured to resample each sampling point to obtain target sampling points, wherein the sampling interval between the target sampling points is the sampling interval of the low-frequency attribute model; perform line-based processing on the target sampling points to obtain the line position of the target sampling points in the hierarchical sub-layer model; perform parameterization processing on the target sampling points to obtain the layer position of the target sampling points in the hierarchical sub-layer model; and add the target sampling points to the reference hierarchical sub-layer model according to the line position and the layer position of the target sampling points in the hierarchical sub-layer model to obtain a reference hierarchical sub-layer model including the target sampling points.

[0057] In one possible implementation, the generation module is configured to resample each of the seismic nodes to obtain target seismic nodes, wherein the sampling interval between the target seismic nodes is the sampling interval of the low-frequency attribute model; to parameterize the target seismic nodes to obtain the layer position of the target seismic nodes in the sequence sublayer model; and to add the target seismic nodes to a reference sequence sublayer model including the sampling points according to the layer position of the target seismic nodes in the sequence sublayer model to obtain the first sequence sublayer model.

[0058] In one possible implementation, the generation module is configured to: determine the value of each node based on the second attribute value of each node included in the second hierarchical sub-layer model and the layer model corresponding to the layer in which each node is located; map the value of each node to a structural domain to obtain the value of each node in the structural domain; resample each node based on the value of each node in the structural domain to obtain the value of each seismic node; determine the first attribute value of each seismic node based on the value of each seismic node; and generate the low-frequency attribute model based on the seismic body and the first attribute value of each seismic node.

[0059] In one possible implementation, the device further includes:

[0060] The determination module is used to determine the error field for a reference node in the second hierarchical sub-layer model based on the first attribute value of the reference node and the numerical value of the reference node, wherein the reference node is a node in the second hierarchical sub-layer model that has a first attribute value and a second attribute value.

[0061] The generation module is used to determine the first attribute value of each seismic node based on the values ​​of each seismic node and the error field.

[0062] In one possible implementation, the determining module is configured to determine the error between the first attribute value of the reference node and the numerical value of the reference node; and to determine the error field based on the error between the first attribute value of the reference node and the numerical value of the reference node.

[0063] In one possible implementation, the generation module is configured to, for any one of the various seismic nodes, determine that the sum of the numerical value of the seismic node and the error field is the first attribute value of the seismic node.

[0064] On the other hand, embodiments of this application provide a computer device, the computer device including a processor and a memory, the memory storing at least one piece of program code, the at least one piece of program code being loaded and executed by the processor, so that the computer device implements any of the above-described methods for obtaining low-frequency attribute models.

[0065] On the other hand, a computer-readable storage medium is also provided, wherein at least one piece of program code is stored in the computer-readable storage medium, the at least one piece of program code being loaded and executed by a processor to enable a computer to implement the method for obtaining the low-frequency attribute model as described above.

[0066] On the other hand, a computer program or computer program product is also provided, wherein the computer program or computer program product stores at least one computer instruction, which is loaded and executed by a processor to enable the computer to implement any of the above-mentioned methods for obtaining low-frequency attribute models.

[0067] The technical solution provided in this application has at least the following beneficial effects:

[0068] The technical solution provided in this application employs a three-dimensional geological structure model of the target area, first attribute values ​​of sampling points included in observation wells located in the target area, and second attribute values ​​of seismic nodes included in seismic bodies of the target area. Under the constraints of the three-dimensional geological structure model, a low-frequency attribute model is established based on the first attribute values ​​of sampling points included in observation wells located in the target area and the second attribute values ​​of seismic nodes included in seismic bodies of the target area. This method comprehensively utilizes the three-dimensional geological structure model, well logging data, and seismic data to establish a high-precision low-frequency attribute model, overcoming the problems of excessive mathematical and model-based approaches in conventional methods, resulting in a higher accuracy of the established low-frequency attribute model. Since the low-frequency attribute model is used for seismic inversion, and seismic inversion is used for reservoir prediction, the higher accuracy of the low-frequency attribute model improves the accuracy of both seismic inversion and reservoir prediction. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0070] Figure 1 This is a schematic diagram of a "bull's-eye" phenomenon provided in an embodiment of this application;

[0071] Figure 2 This is a schematic diagram of the implementation environment for a method for obtaining a low-frequency attribute model provided in an embodiment of this application;

[0072] Figure 3 This is a flowchart of a method for obtaining a low-frequency attribute model provided in an embodiment of this application;

[0073] Figure 4 This is a schematic diagram of a spatial parameterization technique provided in an embodiment of this application;

[0074] Figure 5 This is a schematic diagram illustrating a process for determining the formation thickness corresponding to any observation well, as provided in an embodiment of this application.

[0075] Figure 6 This is a schematic diagram of the generated reference sequence sublayer model provided in the embodiments of this application;

[0076] Figure 7 This is a schematic diagram of a sampling point linearization process provided in an embodiment of this application;

[0077] Figure 8 This is a schematic diagram of a reference sequence sublayer model including target sampling points provided in an embodiment of this application;

[0078] Figure 9 This is a schematic diagram of a first-order sub-layer model provided in an embodiment of this application;

[0079] Figure 10 This is a schematic diagram of the structure of an RBF network provided in an embodiment of this application;

[0080] Figure 11 This is a schematic diagram of a second-order sub-layer model provided in an embodiment of this application;

[0081] Figure 12 This is a schematic diagram of a method for determining the sub-layer model corresponding to any layer, provided in an embodiment of this application;

[0082] Figure 13 This is a traditional stratabound inversion result for a certain work area in an oilfield, provided in an embodiment of this application;

[0083] Figure 14 A schematic diagram illustrating an RBF neural network phase-controlled attribute modeling technique based on complex structural constraints, provided in an embodiment of this application;

[0084] Figure 15 This is a seismic inversion profile provided in an embodiment of this application;

[0085] Figure 16 This is a schematic diagram of a low-frequency attribute model established by a conventional method according to an embodiment of this application;

[0086] Figure 17 This application provides a low-frequency attribute model established under construction constraints.

[0087] Figure 18 This application provides a low-frequency attribute model for establishing RBF phase control attributes under construction constraints.

[0088] Figure 19 Seismic inversion results of two different methods provided in the embodiments of this application;

[0089] Figure 20 This is a schematic diagram of the structure of a device for obtaining a low-frequency attribute model provided in an embodiment of this application;

[0090] Figure 21 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application;

[0091] Figure 22 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation

[0092] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0093] It should be noted that the terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0094] Seismic inversion, as a crucial method for reservoir prediction, can effectively predict subsurface model parameters that correspond to seismic responses. These predicted parameters can then be used to characterize subsurface reservoirs. The accuracy of seismic inversion directly impacts the economic viability and effectiveness of oil and gas development.

[0095] The establishment of low-frequency attribute models is a crucial part of seismic inversion, and the accuracy of these models directly impacts the precision of the inversion. Methods for constructing low-frequency attribute models can be broadly categorized into two types: one is theory-driven, which utilizes a series of geophysical, sequence geography, facies control, iterative, and trend surface constraints theories and related techniques to compensate for missing low-frequency information in seismic data. Theory-driven methods include constant low-frequency attribute models, compaction trend low-frequency attribute models, and well interpolation low-frequency attribute models constrained by stratigraphic frameworks. Constant low-frequency attribute models are only suitable for areas with gentle terrain and stable sedimentation; compaction trend low-frequency attribute models are suitable for areas with obvious stratigraphic compaction patterns; well interpolation low-frequency attribute models constrained by stratigraphic frameworks, compared to the former two, can adapt to complex structures and have better applicability in areas with a large number of wells and abundant distribution, but the model relies entirely on well logging curves, making it prone to "bull's-eye" phenomena, and exhibiting a strong tendency towards mathematical manipulation when the number of wells is small. Figure 1 This is a schematic diagram of a "bull's eye" phenomenon provided in an embodiment of this application.

[0096] Another type is the data-driven approach, which uses artificial intelligence algorithms such as deep learning to learn the complex nonlinear relationship between seismic data and well logging curves, aiming to build a more accurate low-frequency attribute model. There is currently limited publicly available research on data-driven approaches. These methods directly train the entire well logging curve and the entire seismic trace as a whole, and then build the attribute model based on the differences in the trained network. Because this approach does not consider spatial tectonic sequence information and lacks geological information, the resulting low-frequency attribute model is overly model-based.

[0097] As exploration deepens, the exploration targets have gradually shifted from shallow to deep, from conventional reservoirs to concealed reservoirs, and from conventional structures to complex structures. This shift in exploration targets also places higher demands on the accuracy of seismic inversion methods. For work areas characterized by rapid lateral changes in stratigraphic sedimentation, complex structures, and few wells, conventional difference methods struggle to obtain accurate low-frequency attribute models, thus limiting the accuracy of seismic inversion.

[0098] In view of the deficiencies of related technologies, this application provides a method for obtaining a low-frequency attribute model. This method uses well logging curves as labels and seismic attributes as training samples. It performs local training of an RBF (Radial Basis Function) neural network within the structural model to establish a low-frequency attribute model. This method comprehensively utilizes the structural model, well logging curves, and seismic attributes to establish a high-precision low-frequency attribute model, overcoming the serious problems of mathematical and model-based approaches in related technologies. This improves the accuracy of seismic inversion and storage prediction.

[0099] Figure 2 This is a schematic diagram illustrating the implementation environment of a method for obtaining a low-frequency attribute model provided in this application embodiment, such as... Figure 2 As shown, the implementation environment includes a computer device 201, which can be a terminal device or a server; this embodiment does not limit the specific type of device. The computer device 201 is used to execute the method for obtaining the low-frequency attribute model provided in this embodiment.

[0100] Optionally, computer device 201 is a terminal device. A terminal device can be any electronic device that can interact with a user through one or more methods such as a keyboard, touchpad, remote control, voice interaction, or handwriting device. Examples include PCs (Personal Computers), mobile phones, smartphones, PDAs (Personal Digital Assistants), wearable devices, PPCs (Pocket PCs), tablets, smart car systems, smart TVs, smart speakers, and smartwatches.

[0101] A terminal device can refer to one of multiple terminal devices; this embodiment uses only one terminal device as an example. Those skilled in the art will understand that the number of terminal devices can be more or less. For example, there may be only one terminal device, or there may be dozens or hundreds, or even more. This application embodiment does not limit the number or type of terminal devices.

[0102] When computer device 201 is a server, the server can be a single server, a server cluster consisting of multiple servers, or any of the following: a cloud computing platform or a virtualization center. This application embodiment does not limit this. The server and terminal devices communicate via a wired or wireless network. The server has data receiving, data processing, and data sending functions. Of course, the server may also have other functions, which this application embodiment does not limit.

[0103] Those skilled in the art should understand that the above-described terminal devices and servers are merely illustrative examples. Other existing or future terminal devices or servers that are applicable to this application should also be included within the scope of protection of this application, and are hereby incorporated by reference.

[0104] This application provides a method for obtaining a low-frequency attribute model, which can be applied to the above-mentioned... Figure 2 The implementation environment shown is as follows: Figure 3 The flowchart shown in this embodiment of the present application illustrates a method for obtaining a low-frequency attribute model. This method can be implemented by... Figure 2 The computer device 201 in the middle executes. For example... Figure 3 As shown, the method includes the following steps 301 to 305.

[0105] In step 301, a three-dimensional geological structure model of the target area is obtained, along with the first attribute values ​​of each sampling point in the observation wells located in the target area and the second attribute values ​​of each seismic node in the seismic body corresponding to the target area. The first and second attributes are correlated.

[0106] In the exemplary embodiments of this application, the three-dimensional geological structure model can reflect the geometric morphology of the geological structure and the macroscopic contact relationships between various geological structural elements, but does not include sequence stratigraphic information within each stratum. The number of observation wells located in the target area is at least one.

[0107] In one possible implementation, the correlation between the first attribute and the second attribute can be positive, negative, or other than that specified in this embodiment. For example, the first attribute is porosity, the second attribute is density, and the correlation between porosity and density is negative.

[0108] In step 302, a first sequence sub-layer model is generated based on the three-dimensional geological structure model, each sampling point, and each seismic node. The first sequence sub-layer model includes each sampling point and each seismic node.

[0109] Sequence layers are crucial for spatial property calculations. The key to extracting sequence layer models lies in spatial parameterization techniques. Based on spatial parameterization, sequence layer models can be generated according to the number of sequence layers and their depositional patterns. For example... Figure 4 This is a schematic diagram of a spatial parameterization technique provided in an embodiment of this application.

[0110] In one possible implementation, the process of generating the first sequence sub-layer model based on the three-dimensional geological structure model, each sampling point, and each seismic node includes: generating a reference sequence sub-layer model corresponding to the target area based on the three-dimensional geological structure model; adding each sampling point to the reference sequence sub-layer model to obtain a reference sequence sub-layer model including each sampling point; and adding each seismic node to the reference sequence sub-layer model including each sampling point to obtain the first sequence sub-layer model.

[0111] The process of generating a reference sequence sub-model for the target area based on the three-dimensional geological structure model includes: spatial parameterization of the three-dimensional geological structure model to obtain the positions of each node in the sequence sub-model; determining the number of sequence sub-models in each stratum based on the well trajectories of each observation well and the top and bottom surfaces of each stratum; and generating a reference sequence sub-model according to the sedimentary pattern based on the positions of each node in the sequence sub-model and the number of sequence sub-models in each stratum.

[0112] Spatial parameterization utilizes bedding and cross-sectional information to establish sequence sub-layer models within each stratigraphic unit under a specified sedimentary model. The parameterization results are recorded on the geological interface, allowing for rapid calculation of the parameterized value of any sampling point within the stratigraphic unit. This parameterized value represents the location of the sampling point at a specific bedding layer within the underlying unit. When interpolating this sampling point, it ensures that suitable attribute control points are found from the well logging curves, and that the attribute difference distribution within the stratigraphic unit is consistent with the sub-layer distribution trend, thus satisfying the constraints of the specified sedimentary model on attribute interpolation.

[0113] Sequence layers are the carriers for low-frequency attribute model calculations, and their number affects the accuracy and computational efficiency of the low-frequency attribute model. Too few sequence layers will lead to low accuracy, while too many will affect computational efficiency. In this embodiment, the number of sequence layers is primarily determined based on the vertical sampling interval of the low-frequency attribute model. In one possible implementation, the process of determining the number of sequence layers in each formation based on the well trajectories of each observation well and the top and bottom surfaces of each formation includes: for any formation within each formation, determining the formation thickness corresponding to each observation well based on the first and second intersection points of each observation well, where the first intersection point is the intersection between the well trajectory of any observation well and the top surface of any formation, and the second intersection point is the intersection between the well trajectory of any observation well and the bottom surface of any formation; determining the maximum thickness among the formation thicknesses corresponding to each observation well; and determining the number of sequence layers in any formation based on the maximum thickness and the sampling interval of the low-frequency attribute model.

[0114] The process of determining the formation thickness corresponding to each observation well based on the first intersection point and the second intersection point of each observation well includes: taking the vertical distance between the first intersection point and the second intersection point of any observation well as the formation thickness corresponding to any observation well.

[0115] like Figure 5 This is a schematic diagram illustrating the process of determining the formation thickness corresponding to any observation well, as provided in an embodiment of this application. Wherein, A is the intersection point between the well trajectory of observation well 1 and the top surface of any formation; B is the intersection point between the well trajectory of observation well 1 and the bottom surface of any formation; H1 is the formation thickness corresponding to observation well 1; C is the intersection point between the well trajectory of observation well 2 and the top surface of any formation; D is the intersection point between the well trajectory of observation well 2 and the bottom surface of any formation; H2 is the formation thickness corresponding to observation well 2; E is the intersection point between the well trajectory of observation well 3 and the top surface of any formation; F is the intersection point between the well trajectory of observation well 3 and the bottom surface of any formation; H3 is the formation thickness corresponding to observation well 3; G is the intersection point between the well trajectory of observation well 4 and the top surface of any formation; H is the intersection point between the well trajectory of observation well 4 and the bottom surface of any formation; H4 is the formation thickness corresponding to observation well 4.

[0116] In one possible implementation, the process of determining the maximum thickness among the formation thicknesses corresponding to each observation well includes: taking the maximum value among the formation thicknesses corresponding to each observation well as the maximum thickness. For example... Figure 5 The formation thickness H4 corresponding to observation well 4 is the maximum value among all observation wells. Therefore, H4 is taken as the maximum thickness.

[0117] The process of determining the number of sequence sublayers included in any stratum based on the maximum thickness and the sampling interval of the low-frequency attribute model includes: determining the depth corresponding to the sampling interval of the low-frequency attribute model, and determining the quotient between the maximum thickness and the depth corresponding to the sampling interval of the low-frequency attribute model as the number of sequence sublayers included in any stratum.

[0118] The sampling interval of the low-frequency attribute model is set based on experience or adjusted according to the implementation environment, and this application embodiment does not limit this. For example, the sampling interval of the low-frequency attribute model is 1 millisecond.

[0119] In one possible implementation, the sedimentary models include a scale model, a parallel top model, and a parallel bottom model. The process of generating a reference sequence layer model according to the sedimentary models, based on the positions of the nodes in the three-dimensional geological structure model within the sequence layer model and the number of sequence layers in each stratum, includes: generating a reference sequence layer model based on the positions of the nodes in the three-dimensional geological structure model within the sequence layer model, according to the sedimentary models and the number of sequence layers.

[0120] like Figure 6 This is a schematic diagram of the generated reference sequence sublayer model provided in the embodiments of this application. Figure 6 In the model, (1) is the reference sequence sub-layer model generated according to the proportional model, (2) is the reference sequence sub-layer model generated according to the parallel bottom model, and (3) is the reference sequence sub-layer model generated according to the parallel top model.

[0121] In one possible implementation, the process of adding each sampling point to a reference sequence sub-layer model to obtain a reference sequence sub-layer model including each sampling point includes: resampling each sampling point to obtain target sampling points, wherein the sampling interval between target sampling points is the sampling interval of the low-frequency attribute model; performing line-based processing on the target sampling points to obtain the line positions of the target sampling points in the sequence sub-layer model; performing parameterization processing on the target sampling points to obtain the layer positions of the target sampling points in the sequence sub-layer model; and adding the target sampling points to the reference sequence sub-layer model based on the line positions and layer positions of the target sampling points in the sequence sub-layer model to obtain a reference sequence sub-layer model including the target sampling points.

[0122] Sampling points are generally uniformly sampled at a depth of 0.125 meters. The sampling interval is much lower than that of seismic nodes, so resampling is required. The resampling process includes converting the sampling points into time-domain sampling points according to the time-depth relationship, and then using a wavenumber filtering algorithm to resample the sampling points to obtain the target sampling points.

[0123] The sampling points are represented in planar geodetic coordinates, while the sequence sub-layer model uses line coordinates. Therefore, it is necessary to convert the planar geodetic coordinates of the sampling points to line coordinates. Sampling points between lines are assigned to the nearest line. For example... Figure 7 This is a schematic diagram of a sampling point linearization process provided in an embodiment of this application. Figure 7 (1) is a schematic diagram before the line-tracking process, and (2) is a schematic diagram after the line-tracking process.

[0124] Based on the parameterized information of the hierarchical sub-layer model, time-domain data can be transformed into hierarchical information. The hierarchical sub-layer model uses a 0-1 representation between layers. Based on the parameterized information, the target sampling point is also parameterized and transformed into a 0-1 representation, thus obtaining the layer position of the target sampling point in the hierarchical sub-layer model. Vertically, the target sampling point will be located between the sub-layers of the hierarchical sub-layer model. Therefore, using the target sampling point as the center, a distance-weighted algorithm is used to determine the layer position of the target sampling point in the hierarchical sub-layer model.

[0125] After determining the trace position and layer position of the target sampling point in the hierarchical sub-layer model, the target sampling point is assigned to the reference hierarchical sub-layer model, thus obtaining the reference hierarchical sub-layer model including the target sampling point. For example... Figure 8 This is a schematic diagram of a reference sequence sublayer model including target sampling points provided in an embodiment of this application.

[0126] In one possible implementation, the process of adding each seismic node to a reference sequence sub-model including each sampling point to obtain the first sequence sub-model includes: resampling each seismic node to obtain the target seismic node, where the sampling interval between the target seismic nodes is the sampling interval of the low-frequency attribute model; parameterizing the target seismic node to obtain its layer position in the sequence sub-model; and adding the target seismic node to the reference sequence sub-model including each sampling point based on its layer position in the sequence sub-model to obtain the first sequence sub-model. Figure 9 This is a schematic diagram of a first-order sub-layer model provided in an embodiment of this application. Figure 9 (1) contains earthquake nodes, and (2) is the first sequence sub-layer model.

[0127] In step 303, the first sequence sublayer model is transformed into a domain to obtain the second sequence sublayer model under the sedimentation domain. In the second sequence sublayer model, the thickness between layers is uniform and the same.

[0128] By training the first-level sequence sub-layer model with Radial Basis Functions (RBFs), a low-frequency attribute model is obtained. RBFs are locally distributed, centrosymmetric decaying non-negative nonlinear functions. Local distribution means that the radial basis functions of the hidden units produce a meaningful non-zero response only when the input falls within a very small, specified region of the input space. Centrosymmetric decay means that for inputs radially distanced from the center vector C of the basis functions, the radial basis functions in the hidden nodes produce the same output, and the closer the input is to the center of the RBF, the larger the response of the hidden node. Therefore, the width of the basis function curve can be adjusted by changing the radius.

[0129] Applying Restricted Baseline Flow (RBF) to the hidden layers of a neural network creates an RBF network, which consists of an input layer, hidden layers, and an output layer. For example... Figure 10 This is a schematic diagram of the structure of an RBF network provided in an embodiment of this application.

[0130] The learning algorithm for RBF networks involves constructing and training an RBF network so that the mapping function learns to determine the center and width of the basis functions of each hidden unit neuron, as well as the weights from the hidden unit layer to the output layer, thereby completing the required input-to-output mapping. The center and width of the hidden layer represent the sample space pattern and the relative positions of each center, completing a non-linear mapping from the input space to the hidden layer space. The weights of the output layer achieve a linear mapping from the hidden layer space to the output layer space. It is important to understand that the core of an RBF network is the design of the hidden layers; the appropriateness of the center selection fundamentally affects the performance of the RBF network.

[0131] RBF training is based on regular data, therefore it is necessary to transform the first sequence sublayer model from the tectonic domain to the sedimentary domain, that is, to map the sequence sublayer model from a series of curved surfaces to equally spaced planes, transforming the vertically irregular data into regular data. For example... Figure 11 This is a schematic diagram of a second-order sub-layer model provided in an embodiment of this application. Figure 11 (1) is the first sequence sublayer model, in which the thickness between layers is not uniform, and (2) is the second sequence sublayer model, in which the thickness between layers is uniform and the same.

[0132] In step 304, the layer model corresponding to each layer in the second hierarchical sub-layer model is obtained based on the first attribute value and the second attribute value of the target node in each layer of the second hierarchical sub-layer model.

[0133] The layer model corresponding to any layer is used to reflect the relationship between the first and second attributes of each node included in any layer, and the target node is the node that has both first and second attribute values.

[0134] In one possible implementation, for any layer in the second-order sub-layer model, the layer model corresponding to any layer is obtained based on the first attribute value and the second attribute value of the target node in the nodes included in any layer.

[0135] In another possible implementation, for any layer in the second-order sub-layer model, the layer model corresponding to any layer is obtained based on the first and second attribute values ​​of the target nodes in the nodes included in that layer, and the first and second attribute values ​​of the target nodes in the nodes included in the target layer. The target layer includes at least one of the layers above and below any given layer. Figure 12 This is a schematic diagram illustrating how to determine the layer model corresponding to any given layer, as provided in an embodiment of this application. Figure 12 When obtaining the layer model corresponding to the L2 layer, the layer model corresponding to the L2 layer is generated based on the first and second attribute values ​​of the target nodes in the nodes included in the L1 layer, the first and second attribute values ​​of the target nodes in the nodes included in the L2 layer, and the first and second attribute values ​​of the target nodes in the nodes included in the L3 layer.

[0136] The method provided in this application provides a method for obtaining layer models corresponding to each layer. This method makes the obtained layer models corresponding to each layer more targeted and can improve the accuracy of low-frequency attribute models.

[0137] In one possible implementation, after obtaining the layer model corresponding to each layer, the layer model corresponding to each layer is saved so that it can be used in subsequent steps.

[0138] In step 305, a low-frequency attribute model is generated based on the second attribute values ​​of each node included in the second-sequence sub-layer model and the layer model corresponding to the layer where the node is located. The low-frequency attribute model includes the first attribute values ​​of each seismic node.

[0139] In one possible implementation, the process of generating a low-frequency attribute model based on the second attribute values ​​of each node included in the second-sequence sub-layer model and the layer model corresponding to the layer where the node is located includes: determining the value of each node based on the second attribute values ​​of each node included in the second-sequence sub-layer model and the layer model corresponding to the layer where the node is located; mapping the value of each node to the tectonic domain to obtain the value of each node in the tectonic domain; resampling each node based on the value of each node in the tectonic domain to obtain the value of each seismic node; determining the first attribute value of each seismic node based on the value of each seismic node; and generating a low-frequency attribute model based on the seismic body and the first attribute values ​​of each seismic node, wherein the low-frequency attribute model includes the first attribute values ​​of each seismic node.

[0140] The process of determining the value corresponding to each node is similar. This embodiment of the application only takes the process of determining the value corresponding to any node included in the second-order sub-layer model as an example. Optionally, the process of determining the value corresponding to any node included in the second-order sub-layer model includes: inputting the second attribute value of any node into the layer model corresponding to the layer where the node is located, to obtain the value corresponding to any node.

[0141] In one possible implementation, after obtaining the values ​​of each seismic node, the values ​​of each seismic node can also be stored. The storage process includes: to ensure storage efficiency, the values ​​of the seismic nodes are obtained and stored simultaneously.

[0142] In one possible implementation, the process of determining the first attribute value of each seismic node based on the values ​​of each seismic node includes: determining the value of each seismic node as the first attribute value of each seismic node.

[0143] In another possible implementation, for reference nodes in the second-sequence sub-layer model, the error field is determined based on the first attribute value and the numerical value of the reference node. The reference node is a node in the second-sequence sub-layer model that has both the first attribute value and the second attribute value. The process of determining the first attribute value of each seismic node based on the numerical value of each seismic node includes: determining the first attribute value of each seismic node based on the numerical value of each seismic node and the error field.

[0144] Optionally, the process of determining the error field based on the first attribute value of the reference node and the numerical value of the reference node includes: determining the error between the first attribute value of the reference node and the numerical value of the reference node; and determining the error field based on the error between the first attribute value of the reference node and the numerical value of the reference node. For example, the error field is determined using a Kriging interpolation algorithm based on the error between the first attribute value of the reference node and the numerical value of the reference node.

[0145] The process of determining the error between the first attribute value and the reference node's value includes: using the difference between the first attribute value and the reference node's value as the error between them. Kriging interpolation is a widely used spatial interpolation method that estimates an optimal, unbiased difference with minimal variance for unknown points by minimizing the sum of squared prediction errors. Kriging interpolation considers the spatial correlation between sample points, thus yielding more accurate results.

[0146] In one possible implementation, the process of determining the first attribute value of each seismic node based on the numerical value and error field of each seismic node includes: for any seismic node among the seismic nodes, determining the sum of the numerical value and error field of that seismic node as the first attribute value of that seismic node.

[0147] For example, the first attribute is porosity and the second attribute is density. If the three-dimensional geological structure model of the target area, the porosity of the sampling points of the observation wells located in the target area, and the density of each seismic node of the seismic body in the target area are obtained, the porosity of each seismic node of the seismic body can be obtained through the above steps 301 to 305.

[0148] like Figure 13 This application provides a traditional stratabound inversion result for a specific work area in an oilfield. The cross-section shows discontinuous sand bodies near the fault. The overall sedimentary pattern exhibits a "carpet-like" multi-layered superposition, which contradicts the latest understanding of "land reclamation" type infill-driven sedimentary systems. Therefore, using RBF neural network facies-controlled attribute modeling technology based on complex tectonic constraints, a complex tectonic model (26 layers in total) is first established. Then, an RBF attribute model is established under the constraints of a three-dimensional geological tectonic model. Finally, this attribute model is used as a low-frequency attribute model to conduct seismic inversion, obtaining a seismic inversion cross-section. (See attached image.) Figure 14 A schematic diagram illustrating an RBF neural network phase control attribute modeling technique based on complex structural constraints provided in this application embodiment is shown below. Figure 15 This is a seismic inversion profile provided in an embodiment of this application. It can be seen from the seismic inversion profile that the sand body and the well logging height are in good agreement, the continuity of the sand body is improved, and the depositional pattern is more in line with geological laws.

[0149] In a certain region, the quality of Carboniferous data is poor, the volcanic structure and morphology are not fully preserved, the characterization of volcanic structures and facies zones is difficult, the accumulation model of volcanic rocks is unclear, and the characteristics of sensitive elastic parameters are unknown, making it difficult to predict high-quality volcanic reservoirs. Figure 16 , 17 1 and 18 are low-frequency attribute models established by three different methods, where... Figure 16 A schematic diagram of a low-frequency attribute model established using conventional methods. Figure 17 To construct a low-frequency attribute model under constraints, Figure 18 A low-frequency attribute model was established for RBF phasing-controlled attributes under structural constraints. Comparison shows that the structurally constrained phasing-controlled attribute model is more consistent with the geological patterns of volcanic rocks than both conventional well interpolation attribute models and structurally constrained attribute models. Figure 19 The seismic inversion results of two different methods provided in the embodiments of this application are as follows: Figure 19(1) is the seismic inversion result of the first method. The well-constrained area has a good relationship with the well, while the inversion effect of the unconstrained drilling area is poor. (2) is the seismic inversion result of the second method. The well-constrained area has a good relationship with the well, and the inversion effect of the unconstrained drilling area is also very good.

[0150] The aforementioned method employs a three-dimensional geological structural model of the target area, the first attribute values ​​of sampling points included in observation wells located in the target area, and the second attribute values ​​of seismic nodes included in seismic bodies within the target area. Under the constraints of the three-dimensional geological structural model, a low-frequency attribute model is established based on the first attribute values ​​of sampling points included in observation wells located in the target area and the second attribute values ​​of seismic nodes included in seismic bodies within the target area. This method comprehensively utilizes the three-dimensional geological structural model, well logging data, and seismic data to establish a high-precision low-frequency attribute model, overcoming the problems of excessive mathematical and model-based approaches in conventional methods, resulting in a higher accuracy of the established low-frequency attribute model. Since the low-frequency attribute model is used for seismic inversion, and seismic inversion is used for reservoir prediction, the higher accuracy of the low-frequency attribute model improves both the accuracy of seismic inversion and the accuracy of reservoir prediction.

[0151] Figure 20 The diagram shown is a structural schematic of a device for acquiring a low-frequency attribute model according to an embodiment of this application. Figure 20 As shown, the device includes:

[0152] The acquisition module 2001 is used to acquire the three-dimensional geological structure model of the target area, the first attribute values ​​of each sampling point of the observation well located in the target area, and the second attribute values ​​of each seismic node of the seismic body corresponding to the target area. The first attribute and the second attribute are related.

[0153] The generation module 2002 is used to generate a first sequence sub-layer model based on the three-dimensional geological structure model, each sampling point and each seismic node. The first sequence sub-layer model includes each sampling point and each seismic node.

[0154] Transformation module 2003 is used to perform domain transformation on the first sequence sublayer model to obtain the second sequence sublayer model under the sedimentation domain. The thickness of the layers in the second sequence sublayer model is uniform and the same.

[0155] The acquisition module 2001 is also used to acquire the layer model corresponding to each layer in the second-level order sub-layer model based on the first attribute value and the second attribute value of the target node in each layer of the second-level order sub-layer model. The layer model corresponding to any layer is used to reflect the relationship between the first attribute and the second attribute of each node included in any layer. The target node is a node that has a first attribute value and a second attribute value.

[0156] The generation module 2002 is also used to generate a low-frequency attribute model based on the second attribute values ​​of each node included in the second-sequence sub-layer model and the layer model corresponding to the layer where each node is located. The low-frequency attribute model includes the first attribute values ​​of each seismic node.

[0157] In one possible implementation, the generation module 2002 is used to generate a reference sequence sub-model corresponding to the target area based on the three-dimensional geological structure model; add each sampling point to the reference sequence sub-model to obtain a reference sequence sub-model including each sampling point; add each seismic node to the reference sequence sub-model including each sampling point to obtain a first sequence sub-model.

[0158] In one possible implementation, the generation module 2002 is used to perform spatial parameterization processing on the three-dimensional geological structure model to obtain the position of each node included in the three-dimensional geological structure model in the sequence sub-layer model; determine the number of sequence sub-layers included in each stratum based on the well trajectory of each observation well, the top surface and bottom surface of each stratum; and generate a reference sequence sub-layer model according to the sedimentary mode based on the position of each node included in the three-dimensional geological structure model in the sequence sub-layer model and the number of sequence sub-layers included in each stratum.

[0159] In one possible implementation, the generation module 2002 is used to determine the formation thickness corresponding to each observation well for any formation in each formation based on the first intersection point and the second intersection point of each observation well. The first intersection point of any observation well is the intersection point between the well trajectory of any observation well and the top surface of any formation, and the second intersection point of any observation well is the intersection point between the well trajectory of any observation well and the bottom surface of any formation. The module also determines the maximum thickness among the formation thicknesses corresponding to each observation well and determines the number of sequence sublayers included in any formation based on the maximum thickness and the sampling interval of the low-frequency attribute model.

[0160] In one possible implementation, the generation module 2002 is used to resample each sampling point to obtain target sampling points, wherein the sampling interval between target sampling points is the sampling interval of the low-frequency attribute model; to perform line-based processing on the target sampling points to obtain the line positions of the target sampling points in the hierarchical sub-layer model; to perform parameterization processing on the target sampling points to obtain the layer positions of the target sampling points in the hierarchical sub-layer model; and to add the target sampling points to the reference hierarchical sub-layer model based on the line positions and layer positions of the target sampling points in the hierarchical sub-layer model to obtain a reference hierarchical sub-layer model including the target sampling points.

[0161] In one possible implementation, the generation module 2002 is used to resample each seismic node to obtain a target seismic node, wherein the sampling interval between the target seismic nodes is the sampling interval of the low-frequency attribute model; the target seismic node is parameterized to obtain the layer position of the target seismic node in the sequence sub-layer model; based on the layer position of the target seismic node in the sequence sub-layer model, the target seismic node is added to the reference sequence sub-layer model including each sampling point to obtain the first sequence sub-layer model.

[0162] In one possible implementation, the generation module 2002 is used to determine the value of each node based on the second attribute value of each node included in the second-sequence sub-layer model and the layer model corresponding to the layer where each node is located; map the value of each node to the structural domain to obtain the value of each node in the structural domain; resample each node based on the value of each node in the structural domain to obtain the value of each seismic node; determine the first attribute value of each seismic node based on the value of each seismic node; and generate a low-frequency attribute model based on the seismic body and the first attribute value of each seismic node.

[0163] In one possible implementation, the device further includes:

[0164] The determination module is used to determine the error field for reference nodes in the second-order sub-layer model based on the first attribute value and the numerical value of the reference node. The reference node is a node in the second-order sub-layer model that has both the first attribute value and the second attribute value.

[0165] The generation module 2002 is used to determine the first attribute value of each seismic node based on the numerical values ​​and error field of each seismic node.

[0166] In one possible implementation, a determining module is used to determine the error between the first attribute value of the reference node and the numerical value of the reference node; and to determine the error field based on the error between the first attribute value of the reference node and the numerical value of the reference node.

[0167] In one possible implementation, the generation module 2002 is used to determine, for any seismic node among the various seismic nodes, the sum of the numerical value and the error field of any seismic node is the first attribute value of any seismic node.

[0168] The aforementioned device employs a three-dimensional geological structure model of the target area, the first attribute values ​​of sampling points from observation wells located in the target area, and the second attribute values ​​of seismic nodes from seismic bodies within the target area. Under the constraints of the three-dimensional geological structure model, a low-frequency attribute model is established based on the first attribute values ​​of sampling points from observation wells located in the target area and the second attribute values ​​of seismic nodes from seismic bodies within the target area. This method comprehensively utilizes the three-dimensional geological structure model, well logging data, and seismic data to establish a high-precision low-frequency attribute model, overcoming the problems of excessive mathematical and model-based approaches in conventional methods, resulting in a higher accuracy of the established low-frequency attribute model. Since the low-frequency attribute model is used for seismic inversion, and seismic inversion is used for reservoir prediction, the higher accuracy of the low-frequency attribute model improves both the accuracy of seismic inversion and the accuracy of reservoir prediction.

[0169] It should be understood that the above-described apparatus is only illustrated by the division of the functional modules described above when implementing its functions. In practical applications, the functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0170] Figure 21 This illustration shows a structural block diagram of a terminal device 2100 provided in an exemplary embodiment of this application. The terminal device 2100 can be any electronic device product capable of human-computer interaction with a user through one or more methods such as a keyboard, touchpad, remote control, voice interaction, or handwriting device. Examples include PCs (Personal Computers), mobile phones, smartphones, PDAs (Personal Digital Assistants), wearable devices, PPCs (Pocket PCs), tablet computers, smart car systems, smart TVs, smart speakers, and smartwatches.

[0171] Typically, terminal device 2100 includes a processor 2101 and a memory 2102.

[0172] Processor 2101 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 2101 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 2101 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 2101 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the screen. In some embodiments, processor 2101 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0173] The memory 2102 may include one or more computer-readable storage media, which may be non-transitory. The memory 2102 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 2102 is used to store at least one instruction, which is executed by the processor 2101 to implement the method for obtaining a low-frequency attribute model provided in the method embodiments of this application.

[0174] In some embodiments, the terminal device 2100 may also optionally include a peripheral device interface 2103 and at least one peripheral device. The processor 2101, memory 2102, and peripheral device interface 2103 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 2103 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: radio frequency circuitry 2104, display screen 2105, camera assembly 2106, audio circuitry 2107, and power supply 2108.

[0175] Peripheral device interface 2103 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 2101 and memory 2102. In some embodiments, processor 2101, memory 2102 and peripheral device interface 2103 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 2101, memory 2102 and peripheral device interface 2103 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0176] The radio frequency (RF) circuit 2104 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 2104 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 2104 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 2104 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 2104 can communicate with other terminal devices through at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit 2104 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0177] Display screen 2105 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 2105 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 2101 for processing. In this case, display screen 2105 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 2105, disposed on the front panel of terminal device 2100; in other embodiments, there may be at least two display screens, disposed on different surfaces of terminal device 2100 or in a folded design; in still other embodiments, display screen 2105 may be a flexible display screen, disposed on a curved or folded surface of terminal device 2100. Furthermore, display screen 2105 may be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. The display screen 2105 can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0178] The camera assembly 2106 is used to acquire images or videos. Optionally, the camera assembly 2106 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the terminal device 2100, and the rear-facing camera is located on the back of the terminal device 2100. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 2106 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cool light flash, which can be used for light compensation at different color temperatures.

[0179] The audio circuit 2107 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting the sound waves into electrical signals that are input to the processor 2101 for processing, or input to the radio frequency circuit 2104 for voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each located at a different part of the terminal device 2100. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert electrical signals from the processor 2101 or the radio frequency circuit 2104 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 2107 may also include a headphone jack.

[0180] Power supply 2108 is used to supply power to the various components in terminal device 2100. Power supply 2108 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 2108 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0181] In some embodiments, the terminal device 2100 further includes one or more sensors 2109. The one or more sensors 2109 include, but are not limited to: an acceleration sensor 2110, a gyroscope sensor 2111, a pressure sensor 2112, an optical sensor 2113, and a proximity sensor 2114.

[0182] Accelerometer 2110 can detect the magnitude of acceleration along the three coordinate axes of a coordinate system established by terminal device 2100. For example, accelerometer 2110 can be used to detect the components of gravitational acceleration along the three coordinate axes. Processor 2101 can control display screen 2105 to display the user interface in either a landscape or portrait view based on the gravitational acceleration signal acquired by accelerometer 2110. Accelerometer 2110 can also be used for games or for acquiring user motion data.

[0183] The gyroscope sensor 2111 can detect the orientation and rotation angle of the terminal device 2100. The gyroscope sensor 2111 can work in conjunction with the accelerometer sensor 2110 to collect the user's 3D movements on the terminal device 2100. Based on the data collected by the gyroscope sensor 2111, the processor 2101 can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0184] The pressure sensor 2112 can be disposed on the side bezel of the terminal device 2100 and / or on the lower layer of the display screen 2105. When the pressure sensor 2112 is disposed on the side bezel of the terminal device 2100, it can detect the user's grip signal on the terminal device 2100, and the processor 2101 can perform left / right hand recognition or quick operation based on the grip signal collected by the pressure sensor 2112. When the pressure sensor 2112 is disposed on the lower layer of the display screen 2105, the processor 2101 can control the operable controls on the UI interface based on the user's pressure operation on the display screen 2105. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0185] An optical sensor 2113 is used to collect ambient light intensity. In one embodiment, the processor 2101 can control the display brightness of the display screen 2105 based on the ambient light intensity collected by the optical sensor 2113. Specifically, when the ambient light intensity is high, the display brightness of the display screen 2105 is increased; when the ambient light intensity is low, the display brightness of the display screen 2105 is decreased. In another embodiment, the processor 2101 can also dynamically adjust the shooting parameters of the camera assembly 2106 based on the ambient light intensity collected by the optical sensor 2113.

[0186] The proximity sensor 2114, also known as a distance sensor, is typically located on the front panel of the terminal device 2100. The proximity sensor 2114 is used to detect the distance between the user and the front of the terminal device 2100. In one embodiment, when the proximity sensor 2114 detects that the distance between the user and the front of the terminal device 2100 is gradually decreasing, the processor 2101 controls the display screen 2105 to switch from a screen-on state to a screen-off state; when the proximity sensor 2114 detects that the distance between the user and the front of the terminal device 2100 is gradually increasing, the processor 2101 controls the display screen 2105 to switch from a screen-off state to a screen-on state.

[0187] Those skilled in the art will understand that Figure 21 The structure shown does not constitute a limitation on the terminal device 2100, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0188] Figure 22This is a schematic diagram of the server structure provided in the embodiments of this application. The server 2200 can vary considerably due to different configurations or performance. It may include one or more Central Processing Units (CPUs) 2201 and one or more memories 2202. The one or more memories 2202 store at least one line of program code, which is loaded and executed by the one or more processors 2201 to implement the low-frequency attribute model acquisition method provided in the various method embodiments described above. Of course, the server 2200 may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server 2200 may also include other components for implementing device functions, which will not be elaborated here.

[0189] In an exemplary embodiment, a computer-readable storage medium is also provided, which stores at least one piece of program code that is loaded and executed by a processor to enable a computer to implement any of the above-described methods for obtaining low-frequency attribute models.

[0190] Optionally, the aforementioned computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0191] In an exemplary embodiment, a computer program or computer program product is also provided, which stores at least one computer instruction, which is loaded and executed by a processor to enable the computer to implement any of the above-described methods for obtaining low-frequency attribute models.

[0192] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0193] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0194] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0195] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for obtaining a low-frequency attribute model, characterized in that, The method comprises: obtaining a three-dimensional geological structure model of a target region, first attribute values of each sampling point included in an observation well located in the target region, and second attribute values of each seismic node included in a seismic body corresponding to the target region, the first attribute and the second attribute having a correlation relationship; generating a first sequence sublayer model according to the three-dimensional geological structure model, the each sampling point and the each seismic node, the first sequence sublayer model comprising the each sampling point and the each seismic node; performing domain transformation on the first sequence sublayer model to obtain a second sequence sublayer model under a sedimentary domain, the thickness between layers in the second sequence sublayer model being uniform and the same; obtaining a layer model corresponding to each layer in the second sequence sublayer model according to first attribute values and second attribute values of a target node in the nodes included in each layer in the second sequence sublayer model, the layer model of any layer being used to reflect the relationship between the first attribute and the second attribute of each node included in the any layer, the target node being a node having the first attribute value and the second attribute value; generating a low-frequency attribute model according to second attribute values of each node included in the second sequence sublayer model and the layer model corresponding to the layer in which the each node is located, the low-frequency attribute model comprising the first attribute values of the each seismic node.

2. The method of claim 1, wherein, The method comprises: generating a reference sequence sublayer model corresponding to the target region according to the three-dimensional geological structure model; adding the each sampling point to the reference sequence sublayer model to obtain a reference sequence sublayer model comprising the each sampling point; adding the each seismic node to the reference sequence sublayer model comprising the each sampling point to obtain the first sequence sublayer model.

3. The method of claim 2, wherein, The method comprises: performing spatial parameterization processing on the three-dimensional geological structure model to obtain the positions of each node included in the three-dimensional geological structure model in a sequence sublayer model; determining the number of sequence sublayers included in each stratum according to the well trajectories of each observation well, the top layer surface and the bottom layer surface of each stratum; generating the reference sequence sublayer model according to the positions of each node included in the three-dimensional geological structure model in the sequence sublayer model and the number of sequence sublayers included in each stratum according to a sedimentary mode.

4. The method of claim 3, wherein, The method comprises: for any stratum in the each stratum, determining the stratum thickness corresponding to the each observation well according to the first intersection points of the each observation well and the second intersection points of the each observation well, the first intersection point of any observation well being the intersection point between the well trajectory of the any observation well and the top layer surface of the any stratum, and the second intersection point of the any observation well being the intersection point between the well trajectory of the any observation well and the bottom layer surface of the any stratum; determining a maximum thickness in the formation thickness corresponding to each of the observation wells; determining a number of sequence sublayers included in the any formation according to the maximum thickness and a sampling interval of the low-frequency attribute model.

5. The method of claim 2, wherein, The adding the each sampling point into the reference sequence sublayer model to obtain a reference sequence sublayer model including the each sampling point comprises: resampling the each sampling point to obtain a target sampling point, and a sampling interval between the target sampling points is the sampling interval of the low-frequency attribute model; performing a track processing on the target sampling point to obtain a track position of the target sampling point in the sequence sublayer model; performing a parameterization processing on the target sampling point to obtain a layer position of the target sampling point in the sequence sublayer model; adding the target sampling point into the reference sequence sublayer model according to the track position of the target sampling point in the sequence sublayer model and the layer position of the target sampling point in the sequence sublayer model to obtain a reference sequence sublayer model including the target sampling point.

6. The method of claim 2, wherein, The adding the each seismic node into the reference sequence sublayer model including the each sampling point to obtain the first sequence sublayer model comprises: resampling the each seismic node to obtain a target seismic node, and a sampling interval between the target seismic nodes is the sampling interval of the low-frequency attribute model; performing a parameterization processing on the target seismic node to obtain a layer position of the target seismic node in the sequence sublayer model; adding the target seismic node into the reference sequence sublayer model including the each sampling point according to the layer position of the target seismic node in the sequence sublayer model to obtain the first sequence sublayer model.

7. The method according to any one of claims 1 to 6, characterized in that, The generating the low-frequency attribute model according to the second attribute value of each node included in the second sequence sublayer model and a layer model corresponding to the layer in which the each node is located comprises: determining a value of the each node according to the second attribute value of the each node included in the second sequence sublayer model and the layer model corresponding to the layer in which the each node is located; mapping the value of the each node to a construction domain to obtain a value of the each node under the construction domain; resampling the each node according to the value of the each node under the construction domain to obtain a value of the each seismic node; determining a first attribute value of the each seismic node according to the value of the each seismic node; generating the low-frequency attribute model according to the seismic body and the first attribute value of the each seismic node.

8. The method of claim 7, wherein, The method further comprises: for a reference node in the second sequence sublayer model, determining an error field according to a first attribute value of the reference node and a value of the reference node, the reference node being a node in the second sequence sublayer model having the first attribute value and the second attribute value; The determining the first attribute value of the each seismic node according to the value of the each seismic node comprises: determining the first attribute value of the each seismic node according to the value of the each seismic node and the error field.

9. The method of claim 8, wherein, The determining the error field according to the first attribute value of the reference node and the value of the reference node comprises: determining an error between the first attribute value of the reference node and the numerical value of the reference node; determining the error field according to the error between the first attribute value of the reference node and the numerical value of the reference node.

10. The method of claim 8, wherein, The determining the first attribute value of each seismic node according to the numerical value of each seismic node and the error field comprises: For any seismic node in the seismic nodes, determining a sum value of the numerical value of the any seismic node and the error field as the first attribute value of the any seismic node.

11. A device for acquiring a low-frequency attribute model, characterized in that, The device comprises: An acquisition module is configured to acquire a three-dimensional geological structure model of a target region, first attribute values of each sampling point included in an observation well located in the target region, and second attribute values of each seismic node included in a seismic body corresponding to the target region, the first attribute and the second attribute being in a correlation relationship; A generation module is configured to generate a first sequence sublayer model according to the three-dimensional geological structure model, the each sampling point, and the each seismic node, the first sequence sublayer model including the each sampling point and the each seismic node; A transformation module is configured to perform domain transformation on the first sequence sublayer model to obtain a second sequence sublayer model under a deposition domain, a thickness between layers in the second sequence sublayer model being uniform and identical; The acquisition module is further configured to acquire a layer model corresponding to each layer in the second sequence sublayer model according to first attribute values and second attribute values of target nodes in nodes included in each layer in the second sequence sublayer model, any layer corresponding layer model being used to reflect a relationship between the first attribute and the second attribute of each node included in the any layer, the target nodes being nodes having the first attribute values and the second attribute values; The generation module is further configured to generate a low-frequency attribute model according to second attribute values of each node included in the second sequence sublayer model and the layer model corresponding to a layer in which the each node is located, the low-frequency attribute model including the first attribute values of the each seismic node.

12. A computer device, comprising: The computer device comprises a processor and a memory, the memory storing at least one program code, the at least one program code being loaded and executed by the processor, so that the computer device implements the low-frequency attribute model acquisition method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer readable storage medium stores at least one program code, the at least one program code being loaded and executed by the processor, so that the computer implements the low-frequency attribute model acquisition method according to any one of claims 1 to 10.

14. A computer program product, characterised in that, The computer program product stores at least one computer instruction, the at least one computer instruction being loaded and executed by the processor, so that the computer implements the low-frequency attribute model acquisition method according to any one of claims 1 to 10.

Images