Relative geological time prediction method and device, computer equipment and storage medium
Through seismic data segmentation and semi-supervised training of the target plot, the problems of relative geological time accuracy and efficiency in traditional methods are solved, and more efficient prediction results and geological structure recognition are achieved.
Patent Information
- Application Number
- CN202310644923.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-01
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2043-06-01
AI Technical Summary
When traditional methods generate relative geological time (RGT), the accuracy is low due to the reliance on subjective experience, and the neural network model lacks constraints in the prediction process, resulting in low efficiency.
By segmenting the seismic data of the target plot, multiple plot sub-data are generated, and semi-supervised training is performed based on the marked first-layer data, and the model is optimized using training losses and structural losses to improve prediction accuracy to improve the prediction accuracy and efficiency.
It improves the accuracy and efficiency of relative geological time prediction, reduces the workload of manual labeling, and better identify and track geological structures.
Smart Images

Figure CN119064993A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of oil and gas seismic exploration and seismic interpretation, and in particular to a relative geological time prediction method, device, computer equipment and storage medium. Background Art
[0002] In seismic exploration, seismic horizon interpretation is the basic work of seismic interpretation, and horizon identification and tracking is an important part of seismic horizon interpretation. The relative geological time (RGT) of seismic data is usually analyzed to identify and track the horizons in the plot corresponding to the seismic data. The traditional method of generating RGT is mainly to manually select as many horizons as possible, and then interpolate the identified horizons to obtain RGT. Although this method is direct, the accuracy of the generated RGT is low due to the heavy reliance on subjective experience when selecting horizons. Therefore, how to obtain a more accurate RGT is a technical problem that needs to be solved.
[0003] With the development of big data artificial intelligence technology, geophysicists have proposed using an end-to-end neural network model to predict the RGT of seismic data. In related technologies, supervised learning methods are usually used to train the neural network model based on multiple seismic data and the RGT of multiple seismic data, so that a more accurate RGT can be predicted based on the trained neural network model.
[0004] However, in the above scheme, since the neural network model has no constraints in the RGT prediction process, the predicted RGT needs to be finely processed, resulting in low RGT prediction efficiency of the neural network model. Summary of the invention
[0005] The embodiment of the present application provides a relative geological time prediction method, device, computer equipment and storage medium, which can improve the prediction accuracy of the relative geological time prediction model while improving the prediction efficiency of the relative geological time. The technical solution is as follows:
[0006] In one aspect, a relative geological time prediction method is provided, the method comprising:
[0007] Segmenting the block seismic data of the target block to be predicted to obtain a plurality of block sub-data of the target block, wherein the block seismic data includes seismic data of a plurality of first layers in the target block with marked relative geological time and seismic data of a plurality of second layers without marked relative geological time, wherein the relative geological time is used to indicate the relative formation time of the layers in the target block, and the block sub-data includes partial seismic data of at least one layer in the target block;
[0008] Based on a plurality of target plot sub-data, a relative geological time prediction model is trained, wherein the target plot sub-data is plot sub-data including at least one portion of seismic data of a first horizon among the plurality of plot sub-data;
[0009] Based on the relative geological time prediction model obtained through training, the plurality of land parcel sub-data are predicted respectively to obtain the predicted relative geological time of each layer in the plurality of land parcel sub-data;
[0010] The predicted relative geological time of each layer in the plurality of land parcel sub-data is merged to obtain the predicted relative geological time of the plurality of second layers.
[0011] On the other hand, a relative geological time prediction device is provided, the device comprising:
[0012] A segmentation module, used for segmenting the block seismic data of the target block to be predicted to obtain a plurality of block sub-data of the target block, wherein the block seismic data includes seismic data of a plurality of first layers in the target block with marked relative geological time and seismic data of a plurality of second layers without marked relative geological time, wherein the relative geological time is used to indicate the relative formation time of the layers in the target block, and the block sub-data includes partial seismic data of at least one layer in the target block;
[0013] A training module, for training a relative geological time prediction model based on a plurality of target plot sub-data, wherein the target plot sub-data is plot sub-data including at least one partial seismic data of a first horizon among the plurality of plot sub-data;
[0014] A prediction module, used to predict the plurality of land parcel sub-data respectively based on the relative geological time prediction model obtained through training, so as to obtain the predicted relative geological time of each layer in the plurality of land parcel sub-data;
[0015] The merging module is used to merge the predicted relative geological time of each layer in the multiple plot sub-data to obtain the predicted relative geological time of the multiple second layers.
[0016] In some embodiments, the training module includes:
[0017] A prediction unit, configured to predict any target land parcel sub-data among the plurality of target land parcel sub-data based on the relative geological time prediction model, and obtain prediction information of the target land parcel sub-data, wherein the prediction information is used to indicate the predicted relative geological time of each horizon in the target land parcel sub-data;
[0018] A determination unit, configured to determine a training loss of the relative geological time prediction model based on the target plot sub-data and the prediction information, wherein the training loss is used to indicate a difference between the prediction information and the relative geological time marked in the target plot sub-data;
[0019] An updating unit is used to update the model parameters of the relative geological time prediction model based on the training loss.
[0020] In some embodiments, the target block sub-data includes partial seismic data of a plurality of horizons, and the plurality of horizons includes at least one first horizon;
[0021] The determining unit comprises:
[0022] an acquisition subunit, configured to acquire a relative geological time of at least one first horizon based on partial seismic data of at least one first horizon in the target land block subdata;
[0023] A first determining subunit is configured to determine a constraint loss of the target plot sub-data based on the predicted relative geological time of the at least one first horizon and the relative geological time of the at least one first horizon, wherein the constraint loss is used to indicate a difference between the predicted relative geological time of the at least one first horizon and the relative geological time of the at least one first horizon;
[0024] A second determination subunit is used to determine the structural loss of the target plot sub-data based on the gradient vector of the target plot sub-data and the gradient vectors of the predicted relative geological time of the multiple horizons, wherein the structural loss is used to indicate the difference between the gradient vector of the predicted relative geological time of the multiple horizons and the gradient vector of the target plot sub-data;
[0025] The third determination subunit is used to perform weighted summation on the structural loss and the constraint loss to obtain the training loss.
[0026] In some embodiments, the second determination subunit is used to determine the tensor matrix of the target plot sub-data based on the gradient vector of the target plot sub-data; perform singular value decomposition on the tensor matrix to obtain the normal vector of the target plot sub-data and a quality parameter of the normal vector, wherein the quality parameter is used to indicate the accuracy of the normal vector; and perform weighted summation of the normal vector, the quality parameter, and the predicted gradient vector relative to geological time to obtain the structural loss.
[0027] In some embodiments, the apparatus further comprises:
[0028] A determination module, configured to determine, based on the well logging position of the target plot, a plurality of layers in the target plot associated with the well logging position;
[0029] A labeling module is used to label the multiple layers based on the well logging interpretation information of the target plot to obtain the seismic data of the plot, wherein the well logging interpretation information is used to indicate the relative geological time of the multiple layers.
[0030] In some embodiments, the apparatus further comprises:
[0031] The determination module is further used to determine, based on the horizon interpretation information of the target land parcel, a plurality of horizons interpreted by the horizon interpretation information, wherein the horizon interpretation information is used to indicate the relative geological time of the plurality of horizons;
[0032] The labeling module is further used to label the multiple target horizons based on the horizon interpretation information to obtain the seismic data of the land block.
[0033] On the other hand, a computer device is provided, which includes a processor and a memory, wherein the memory is used to store at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the relative geological time prediction method in the embodiment of the present application.
[0034] On the other hand, a computer-readable storage medium is provided, in which at least one computer program is stored. The at least one computer program is loaded and executed by a processor to implement a relative geological time prediction method as in an embodiment of the present application.
[0035] On the other hand, a computer program product is provided, including a computer program, wherein the computer program is executed by a processor to implement the relative geological time prediction method provided in the embodiments of the present application.
[0036] The embodiment of the present application provides a relative geological time prediction method, by segmenting the predicted plot seismic data including multiple first layers of seismic data with relative geological time in the target plot and multiple second layers of seismic data without relative geological time, it is possible to obtain multiple plot sub-data including partial seismic data of at least one layer in the target plot. Compared with the traditional supervised training method, the present application is based on the target plot sub-data including partial seismic data of at least one first layer with relative geological time, and can perform semi-supervised training on the relative geological time prediction model, without the need to pre-make a large number of labels, reducing the workload of manual interpretation of seismic data. And in the process of predicting the plot seismic data of the target plot, the first layer of the target plot with relative geological time marked in the training process is introduced to constrain the prediction results of the second layer without relative geological time, while improving the accuracy of the prediction results, and improving the prediction efficiency of the relative geological time prediction model for relative geological time. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0038] Figure 1 It is a schematic diagram of an implementation environment of a relative geological time prediction method provided according to an embodiment of the present application;
[0039] Figure 2 is a flow chart of a relative geological time prediction method provided according to an embodiment of the present application;
[0040] Figure 3 is a flow chart of another relative geological time prediction method provided according to an embodiment of the present application;
[0041] Figure 4 is a schematic diagram of a method for obtaining sub-data of a target plot provided in an embodiment of the present application;
[0042] Figure 5 is a schematic diagram of a plot of seismic data provided according to an embodiment of the present application;
[0043] Figure 6 It is a block diagram of a training process of a relative geological time prediction model provided according to an embodiment of the present application;
[0044] Figure 7 is a schematic diagram of relative geological time provided according to an embodiment of the present application;
[0045] Figure 8 is another schematic diagram of relative geological time provided according to an embodiment of the present application;
[0046] Fig. 9 is a block diagram of a relative geological time prediction device provided according to an embodiment of the present application;
[0047] Fig.10 is a block diagram of another relative geological time prediction device provided according to an embodiment of the present application;
[0048] Fig.11 is a structural block diagram of a terminal provided according to an embodiment of the present application;
[0049] Fig.12 It is a structural diagram of a server provided according to an embodiment of the present application. DETAILED DESCRIPTION
[0050] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.
[0051] In this application, the terms "first", "second", etc. are used to distinguish identical or similar items with basically the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is there any limitation on quantity and execution order.
[0052] In the present application, the term "at least one" means one or more, and the term "plurality" means two or more.
[0053] 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, stored data, displayed data, 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 relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions. For example, the seismic data and relative geological time involved in this application were obtained with full authorization.
[0054] The training method of the relative geological time prediction model provided in the embodiment of the present application can be applied to a computer device. In some embodiments, the computer device is a terminal or a server. The following first takes the computer device as an example to introduce the implementation environment of the training method of the relative geological time prediction model provided in the embodiment of the present application. Figure 1 Schematic diagram of the implementation environment of a training method for a relative geological time prediction model provided in an embodiment of the present application. Figure 1 The implementation environment includes a terminal 101 and a server 102. The terminal 101 and the server 102 can be directly or indirectly connected via wired or wireless communication, which is not limited in this application.
[0055] In some embodiments, the terminal 101 is a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. An application capable of displaying the relative geological time of multiple layers in the target plot is installed and run on the terminal 101. The application is associated with the server 102, and the server 102 provides background services.
[0056] In some embodiments, the server 102 is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. In some embodiments, the server 102 can segment the block seismic data of the target block to be predicted, and based on the relative geological time prediction model obtained by training, predict the multiple block sub-data obtained by segmentation, obtain the relative geological time of each layer in the multiple block sub-data, and merge the prediction results of the multiple block sub-data to obtain the relative geological time of each layer in the target block. The server 102 returns the relative geological time to the terminal 101, which is displayed by the terminal 101 through the application.
[0057] In some embodiments, the server 102 undertakes the main computing work, and the terminal 101 undertakes the secondary computing work; or, the server 102 undertakes the secondary computing work, and the terminal 101 undertakes the main computing work; or, the server 102 and the terminal 101 use a distributed computing architecture to perform collaborative computing. Optionally, the number of the above servers can be more or less, and this embodiment of the application is not limited to this. Of course, the server 102 can also include other functional servers to provide more comprehensive and diversified services.
[0058] Figure 2 is a flow chart of a relative geological time prediction method provided according to an embodiment of the present application, such as Figure 2 As shown, in the embodiment of the present application, the server is used as an example for explanation. The relative geological time prediction method includes the following steps:
[0059] 201. The server divides the block seismic data of the target block to be predicted to obtain multiple block sub-data of the target block, the block seismic data includes seismic data of multiple first layers in the target block with marked relative geological time and seismic data of multiple second layers without marked relative geological time, the relative geological time is used to indicate the relative formation time of the layers in the target block, and the block sub-data includes partial seismic data of at least one layer in the target block.
[0060] In an embodiment of the present application, the target plot is an underground rock formation with a certain geological structure. The target plot includes multiple horizons, and the relative geological time of each horizon is different. The relative geological time of the horizon is used to indicate the relative formation time of the horizon in the target plot. Relative geological time refers to the time sequence of multiple horizons in the target plot during the formation process. This time sequence can only indicate the relative formation time of a certain horizon relative to other horizons, and cannot indicate the true formation time of the horizon. The plot seismic data is two-dimensional seismic data or three-dimensional seismic data used to indicate the geological structure of the target plot, such as a two-dimensional seismic profile or a three-dimensional amplitude image. The plot seismic data includes seismic data of multiple horizons in the target plot, and the multiple horizons include multiple first horizons with marked relative geological time and multiple second horizons without marked relative geological time. The server divides the plot seismic data of the target plot to obtain multiple plot sub-data. Among them, each plot sub-data includes partial seismic data of at least one horizon in the target plot, and the horizon can be a first horizon or a second horizon, which is not limited in the embodiment of the present application.
[0061] 202. The server trains a relative geological time prediction model based on a plurality of target plot sub-data, where the target plot sub-data is plot sub-data including at least one portion of seismic data of a first layer among the plurality of plot sub-data.
[0062] In an embodiment of the present application, the server can determine multiple target plot sub-data from multiple plot sub-data based on the seismic data of multiple first layers in the target plot, and the target plot sub-data includes partial seismic data of at least one first layer. Since the first layer is a layer with a relative geological time marked, the server can use the relative geological time of the first layer in the target plot sub-data as supervision information to train the relative geological time prediction model. Among them, in addition to the partial seismic data of the first layer with a relative geological time marked, the target plot sub-data also includes partial seismic data of the second layer without a relative geological time marked. Correspondingly, the supervision information in the training process can only indicate the relative geological time of some layers in the target plot. Therefore, the server can adopt a semi-supervised training method to train the relative geological time prediction model based on multiple target plot sub-data.
[0063] 203. The server predicts the multiple plot sub-data based on the trained relative geological time prediction model, and obtains the predicted relative geological time of each layer in the multiple plot sub-data.
[0064] In an embodiment of the present application, the relative geological time prediction model is used to predict the relative geological time of each layer in the target plot. Therefore, the server predicts multiple plot sub-data of the target plot based on the trained relative geological time prediction model, and can obtain prediction results for multiple plot sub-data, which include the predicted relative geological time of each layer in the plot sub-data.
[0065] 204. The server merges the predicted relative geological time of each layer in the multiple plot sub-data to obtain the predicted relative geological time of multiple second layers.
[0066] In the embodiment of the present application, since each plot sub-data includes partial seismic data of at least one horizon, the horizons corresponding to different plot sub-data may be the same or different. Therefore, the predicted relative geological time of each horizon in the multiple plot sub-data predicted based on the relative geological time prediction model is not exactly the same. The server can obtain the predicted relative geological time of each horizon in the plot seismic data by merging the prediction results of the multiple plot sub-data, and also obtain the predicted relative geological time of multiple second horizons in the plot seismic data that are not marked with relative geological time.
[0067] In the embodiment of the present application, based on the relative geological time prediction model, the server can predict the plot seismic data of the target plot with only the relative geological time of some layers marked, and obtain the predicted relative time of the unmarked layers in the target plot, thereby reducing the workload of manually marking the relative geological time and improving the processing efficiency of the plot seismic data of the target plot. Moreover, based on the predicted relative geological time of multiple layers in the target plot, the layers in the target plot can be identified and tracked according to the isochronous surface composed of the same relative geological time, and the faults in the target plot can be detected when the relative geological time is discontinuous, so as to better explain the geological structure of the target plot.
[0068] The embodiment of the present application provides a relative geological time prediction method, by segmenting the predicted plot seismic data including multiple first layers of seismic data with relative geological time in the target plot and multiple second layers of seismic data without relative geological time, it is possible to obtain multiple plot sub-data including partial seismic data of at least one layer in the target plot. Compared with the traditional supervised training method, the present application is based on the target plot sub-data including partial seismic data of at least one first layer with relative geological time, and can perform semi-supervised training on the relative geological time prediction model, without the need to pre-make a large number of labels, reducing the workload of manual interpretation of seismic data. And in the process of predicting the plot seismic data of the target plot, the first layer of the target plot with relative geological time marked in the training process is introduced to constrain the prediction results of the second layer without relative geological time, while improving the accuracy of the prediction results, and improving the prediction efficiency of the relative geological time prediction model for relative geological time.
[0069] Figure 3 is a flow chart of another relative geological time prediction method provided according to an embodiment of the present application, such as Figure 3 As shown, in the embodiment of the present application, the server is used as an example for explanation. The relative geological time prediction method includes the following steps:
[0070] 301. The server obtains the seismic data of the target plot to be predicted, wherein the seismic data includes seismic data of multiple first layers in the target plot with marked relative geological time and seismic data of multiple second layers without marked relative geological time, wherein the relative geological time is used to indicate the relative formation time of the layers in the target plot.
[0071] In an embodiment of the present application, the target plot is an underground rock formation with a certain geological structure. The target plot includes multiple horizons, and the relative geological time of each horizon is different. The relative geological time of the horizon is used to indicate the relative formation time of the horizon in the target plot. The plot seismic data is two-dimensional seismic data or three-dimensional seismic data used to indicate the geological structure of the target plot, such as a two-dimensional seismic profile or a three-dimensional amplitude image. The plot seismic data includes seismic data of multiple horizons in the target plot, and the multiple horizons include multiple first horizons with marked relative geological time and multiple second horizons without marked relative geological time.
[0072] In some embodiments, before obtaining the seismic data of the target plot to be predicted, the server needs to mark the relative geological time of some layers in the target plot. Accordingly, the server can obtain the seismic data of the target plot through the following methods 1 and 2.
[0073] Method 1: The server determines multiple layers associated with the well logging position in the target plot based on the well logging position of the target plot; based on the well logging interpretation information of the target plot, the multiple layers are marked to obtain the plot seismic data, and the well logging interpretation information is used to indicate the relative geological time of the multiple layers. Among them, the well logging interpretation information is used to interpret the multiple layers associated with the well logging position in the target plot, such as the position of the layer, the relative geological time of the layer, etc. The server can mark the multiple layers based on the relative geological time of the multiple layers associated with the well logging position in the well logging interpretation information, thereby obtaining the plot seismic data including the seismic data of multiple first layers with marked relative geological time.
[0074] Method 2: Based on the horizon interpretation information of the target plot, the server determines the multiple horizons interpreted by the horizon interpretation information, and the horizon interpretation information is used to indicate the relative geological time of the multiple horizons; based on the horizon interpretation information, the multiple target horizons are marked to obtain the plot seismic data. Among them, the horizon interpretation information is used to interpret the attributes of the horizons in the target plot, such as the position of the horizon, the relative geological time of the horizon, etc. Based on the relative geological time of the multiple horizons in the horizon interpretation information, the server can mark the multiple horizons explained by the horizon interpretation information, thereby obtaining the plot seismic data including the seismic data of multiple first horizons with marked relative geological time.
[0075] 302. The server divides the seismic data of the target plot to obtain a plurality of sub-plot data of the target plot, where the sub-plot data includes partial seismic data of at least one layer in the target plot.
[0076] In the embodiment of the present application, the server can segment the land parcel seismic data of the target land parcel according to the size of the pre-set land parcel sub-data, and can obtain multiple land parcel sub-data. Each land parcel sub-data includes partial seismic data of at least one layer in the target land parcel, and the layer can be the first layer or the second layer, which is not limited in the embodiment of the present application.
[0077] For example, when the parcel seismic data is three-dimensional seismic data, the server can set the size of the parcel sub-data input to the relative geological time prediction model to d x ×d y ×d t Among them, d x is the length of the data range of the plot sub-data in the horizontal direction x, d y is the length of the data range of the plot sub-data in the vertical direction y, d t is the length of the data range of the plot sub-data in the time direction t. The size of the plot sub-data can be selected according to the computing power of the server.
[0078] 303. For any target plot sub-data among the multiple target plot sub-data, the server predicts the target plot sub-data based on the relative geological time prediction model to obtain prediction information of the target plot sub-data, where the prediction information is used to indicate the predicted relative geological time of each layer in the target plot sub-data.
[0079] In an embodiment of the present application, the server can determine multiple target plot sub-data from multiple plot sub-data based on the seismic data of multiple first layers in the target plot, and the target plot sub-data includes partial seismic data of at least one first layer. Since the first layer is a layer with a marked relative geological time, the server can use the relative geological time of the first layer in the target plot sub-data as supervision information to train a relative geological time prediction model. In the process of training a relative geological time prediction model based on multiple target plot sub-data, for any target plot sub-data, the server can predict the target plot sub-data based on the relative geological time prediction model and obtain prediction information of the target plot sub-data, and the prediction information is used to indicate the predicted relative geological time of each layer in the plot sub-data.
[0080] For example, Figure 4 A schematic diagram of a method for obtaining sub-data of a target land parcel provided in an embodiment of the present application. Figure 4 As shown, the position of the triangle is the well logging position (x0, y0) in the seismic data of the plot, and the surface composed of solid lines is the first layer with marked relative geological time. Figure 4 The cube composed of solid lines is the seismic data of the plot, and the cube composed of dotted lines is the sub-data of the target plot. Based on the well logging position and the first layer, the server can split the seismic data of the plot into a cube of size d x ×d y ×d t The target plot sub-data includes partial seismic data of a layer associated with the well logging position or partial seismic data of the first layer.
[0081] For example, Figure 5 Schematic diagram of a plot of seismic data provided according to an embodiment of the present application. Figure 5 As shown, Figure 5 (1) is the block seismic data, that is, the two-dimensional seismic profile of the target block. Figure 5 (2) is a schematic diagram of the relative geological time of multiple layers in the target plot, with one color representing the relative geological time of one layer. Figure 5 (2) Multiple isochrones are obtained based on the relative geological time of multiple strata. The relative geological time of the strata on the same isochrone is the same.
[0082] 304. The server determines a training loss of the relative geological time prediction model based on the target plot sub-data and the prediction information of the target plot sub-data, where the training loss is used to indicate a difference between the prediction information and the relative geological time marked in the target plot sub-data.
[0083] In an embodiment of the present application, for any target plot sub-data, the server can determine the training loss of the relative geological time prediction model based on the target plot sub-data and the prediction information of the target plot sub-data, and the training loss is used to indicate the difference between the predicted relative geological time in the prediction information and the relative geological time marked in the target plot sub-data. The training loss is inversely correlated with the accuracy. The smaller the training loss of the relative geological time prediction model, the smaller the difference between the predicted relative geological time of the stratum and the pre-marked relative geological time, and the more accurate the predicted relative geological time obtained by the relative geological time prediction model based on the target plot sub-data prediction.
[0084] In some embodiments, the server can determine the structural loss and constraint loss of the target plot sub-data based on the target plot sub-data and the prediction information of the target plot sub-data, and then determine the training loss according to the structural loss and the constraint loss. Accordingly, the server can determine the training loss of the relative geological time prediction model through the following steps (1)-(4).
[0085] (1) The server obtains the relative geological time of at least one first layer based on partial seismic data of at least one first layer in the target block sub-data.
[0086] In the embodiment of the present application, the target land parcel sub-data includes partial seismic data of multiple horizons, and the multiple horizons include at least one first horizon. Since the first horizon is a horizon in the target land parcel with a relative geological time marked, the server can obtain the relative geological time of at least one first horizon based on the partial seismic data of at least one first horizon.
[0087] (2) The server determines a constraint loss of the target plot sub-data based on the predicted relative geological time of at least one first horizon and the relative geological time of at least one first horizon, where the constraint loss is used to indicate a difference between the predicted relative geological time of at least one first horizon and the relative geological time of at least one first horizon.
[0088] In an embodiment of the present application, the prediction information of the target plot sub-data is used to indicate the relative geological time of each horizon in the target plot sub-data, and the target plot sub-data includes partial seismic data of at least one first horizon, so the server determines the constraint loss of the target plot sub-data based on the predicted relative geological time of at least one first horizon in the prediction information and the relative geological time of at least one first horizon marked in advance. The constraint loss is used to indicate the difference between the predicted relative geological time of at least one first horizon and the relative geological time of at least one first horizon. The smaller the constraint loss, the smaller the difference between the predicted relative geological time of the first horizon in the target plot sub-data and the relative geological time marked in the target plot sub-data.
[0089] In some embodiments, the server determines the constraint loss of the target plot sub-data based on the predicted relative geological time of at least one first horizon and the relative geological time of at least one first horizon by using the following formula 1.
[0090] Formula 1:
[0091]
[0092] Among them, Loss control is the constraint loss of the target plot sub-data, K is the number of well logging in the target plot sub-data, is the target plot sub-data corresponding to the well logging position (x k ,y k ) is the relative geological time of the first horizon associated with is the target plot sub-data corresponding to the well logging position (x k ,y k )The predicted relative geological time of the first horizon associated with it.
[0093] In some embodiments, the server determines the constraint loss of the target plot sub-data based on the predicted relative geological time of at least one first horizon and the relative geological time of at least one first horizon by using the following formula 2.
[0094] Formula 2:
[0095]
[0096] Among them, Loss control is the constraint loss of the target plot sub-data, K is the number of first-layer bits in the target sub-data, τ k is the relative geological time of the first horizon marked in the target plot sub-data, p(x k ,y k ,t k ) is the predicted relative geological time of the first horizon in the target plot sub-data, (x k ,yk ,t k ) is any point in the first layer.
[0097] (3) The server determines the structural loss of the target plot sub-data based on the gradient vector of the target plot sub-data and the predicted gradient vectors of the multiple layers relative to the geological time, and the structural loss is used to indicate the difference between the gradient vector of the predicted relative geological time of the multiple layers and the gradient vector of the target plot sub-data.
[0098] In an embodiment of the present application, when the target plot sub-data is three-dimensional seismic data, the server can obtain the gradient vector of the target plot sub-data by determining the gradient of the target plot sub-data in the three directions of x, y, and t. Similarly, the server can determine the gradient vectors of the predicted relative geological time of multiple horizons by determining the gradients of the predicted relative geological time of multiple horizons in the target plot sub-data in the three directions of x, y, and t. Based on the gradient vectors of the target plot sub-data and the gradient vectors of the predicted relative geological time of multiple horizons, the server can determine the structural loss of the target plot sub-data. The smaller the structural loss, the smaller the difference between the gradient vectors, and the closer the gradient vector of the predicted relative geological time is to the gradient vector of the target seismic sub-data.
[0099] In some embodiments, the server can determine the tensor matrix of the target plot sub-data based on the gradient vector of the target plot sub-data; then perform singular value decomposition on the tensor matrix to obtain the normal vector of the target plot sub-data and the quality parameter of the normal vector, the quality parameter is used to indicate the accuracy of the normal vector; perform weighted summation on the normal vector, the quality parameter, and the gradient vector predicted relative to geological time to obtain the structural loss. Wherein, the tensor matrix is a symmetric matrix, and the server performs singular value decomposition on the symmetric matrix to obtain multiple eigenvectors and eigenvalues of the eigenvector of the tensor matrix. Wherein, the normal vector of the target plot sub-data is the eigenvector of the tensor matrix, and the server performs weighted summation on the eigenvalues of multiple eigenvectors to obtain the quality parameter of the normal vector. The server can obtain the structural loss of the target plot sub-data by weighted summation of the normal vector, the quality parameter, and the gradient vector predicted relative to geological time.
[0100] In some embodiments, when the target plot sub-data is three-dimensional seismic data, the tensor matrix of the target plot sub-data is as shown in Formula 3. The server performs singular value decomposition on the tensor matrix through the following Formula 4 to obtain the normal vector of the target plot sub-data and the quality parameter of the normal vector.
[0101] Formula 3:
[0102]
[0103] Formula 4:
[0104] T=λ u uu T +λ v vv T +λ w ww T
[0105] Where T is the tensor matrix of the target plot sub-data, g is the gradient vector of the target plot sub-data, g = [g x ,g y ,g t ] T , g x is the unit vector of the gradient vector in the horizontal direction x, g y is the unit vector of the gradient vector in the vertical direction y, g t is the unit vector of the gradient vector in the time direction t. u, v, w are multiple eigenvectors of the tensor matrix, λ u , v , w are multiple eigenvalues of the tensor matrix. Among them, u is the normal vector of the target land block sub-data, and the quality parameter of the normal vector is
[0106] In some embodiments, the server determines the structural loss of the target plot sub-data by performing a weighted summation of the normal vector, the quality parameter, and the gradient vector predicted relative to the geological time using the following formula five.
[0107] Formula 5:
[0108]
[0109] Among them, Loss structure is the structural loss of the target plot sub-data, w is the normal vector [n x ,n y ,n t ] T The quality parameter, n x is the unit vector of the normal vector in the horizontal direction x, n y is the unit vector of the normal vector in the vertical direction y, n t is the unit vector of the normal vector in the time direction t, as well as To predict the gradient of τ(x,y,t) in three directions relative to geological time, ||.|| L2 is the L2 norm.
[0110] (4) The server performs a weighted summation of the structural loss and constraint loss to obtain the training loss.
[0111] In the embodiment of the present application, for any target plot sub-data, the server performs a weighted summation of the structural loss and the constraint loss of the target plot sub-data, and can obtain the training loss generated in the relative geological time prediction model trained based on the target plot sub-data. The training loss is used to indicate the difference between the predicted relative geological time in the prediction information of the target plot sub-data and the relative geological time marked in the target plot sub-data.
[0112] In some embodiments, the server performs weighted summation of the structural loss and the constraint loss using the following formula six to obtain the training loss.
[0113] Formula 6:
[0114] LOSS all =LOSS structure +εLOSS control
[0115] Among them, Loss all is the training loss of the relative geological time prediction model, Loss structure is the structural loss of the target plot sub-data, Loss control is the constraint loss of the target plot sub-data, and ε is the weight of the constraint loss.
[0116] For example, Figure 6 FIG. 1 is a block diagram of a training process of a relative geological time prediction model provided in an embodiment of the present application. Figure 6 As shown, the server predicts the target plot sub-data based on the relative geological time prediction model to obtain the predicted relative geological time of each horizon in the target plot sub-data, and then obtains the constraint loss of the target plot sub-data based on the predicted relative time of at least one first horizon in the target plot sub-data and the relative geological time of at least one first horizon marked in the target plot sub-data. The server processes the target plot sub-data and can determine the normal vector and the quality parameter of the normal vector of the target plot sub-data based on the tensor matrix of the target plot sub-data. Then, based on the normal vector of the target plot sub-data, the quality parameter of the normal vector and the predicted relative geological time of each horizon in the predicted target plot sub-data, the structural loss of the target plot sub-data is determined. Finally, the server performs a weighted summation of the constraint loss and the structural loss to obtain the training loss of the relative geological time prediction model.
[0117] 305. The server updates the model parameters of the relative geological time prediction model based on the training loss.
[0118] In an embodiment of the present application, the server can update the model parameters of the relative geological time prediction model based on the training loss to reduce the training loss of the relative geological time prediction model, and train to obtain an updated relative geological time prediction model. If the updated relative geological time prediction model meets the training end conditions, such as the number of training times is the target number of times, or the training loss of the relative geological time prediction model is within the target range, the updated relative geological time prediction model is used as the relative geological time prediction model that has completed the training.
[0119] 306. The server predicts the multiple plot sub-data based on the trained relative geological time prediction model, and obtains the predicted relative geological time of each layer in the multiple plot sub-data.
[0120] In an embodiment of the present application, the relative geological time prediction model is used to predict the relative geological time of each layer in the target plot. Therefore, the server predicts multiple plot sub-data of the target plot based on the trained relative geological time prediction model, and can obtain prediction results for multiple plot sub-data, which include the predicted relative geological time of each layer in the plot sub-data.
[0121] For example, Figure 7 is a schematic diagram of relative geological time provided according to an embodiment of the present application. Figure 7 As shown, Figure 7 (1) is a plot sub-data including partial seismic data of two first horizons A and B with relative geological time annotated, Figure 7 (2) The server predicts the relative geological time of other layers in the plot sub-data that are not marked with relative geological time, based on the relative geological time prediction model.
[0122] For example, Figure 8 is another schematic diagram of relative geological time provided according to an embodiment of the present application. Figure 8 As shown in the figure, the relative geological time of multiple layers predicted by the relative geological time model is shown. As the number of iterations increases, the relative geological time output by the relative geological time prediction model becomes more accurate.
[0123] 307. The server merges the predicted relative geological time of each layer in the multiple plot sub-data to obtain the predicted relative geological time of multiple second layers.
[0124] In the embodiment of the present application, since each plot sub-data includes partial seismic data of at least one horizon, the horizons corresponding to different plot sub-data may be the same or different. Therefore, the predicted relative geological time of each horizon in the multiple plot sub-data predicted based on the relative geological time prediction model is not exactly the same. The server can obtain the predicted relative geological time of each horizon in the plot seismic data by merging the prediction results of the multiple plot sub-data, and also obtain the predicted relative geological time of multiple second horizons in the plot seismic data that are not marked with relative geological time.
[0125] In the embodiment of the present application, based on the relative geological time prediction model, the server can predict the plot seismic data of the target plot with only the relative geological time of some layers marked, and obtain the predicted relative time of the unmarked layers in the target plot, thereby reducing the workload of manually marking the relative geological time and improving the processing efficiency of the plot seismic data of the target plot. Moreover, based on the predicted relative geological time of multiple layers in the target plot, the layers in the target plot can be identified and tracked according to the isochronous surface composed of the same relative geological time, and the faults in the target plot can be detected when the relative geological time is discontinuous, so as to better explain the geological structure of the target plot.
[0126] The embodiment of the present application provides a relative geological time prediction method, by segmenting the predicted plot seismic data including multiple first layers of seismic data with relative geological time in the target plot and multiple second layers of seismic data without relative geological time, it is possible to obtain multiple plot sub-data including partial seismic data of at least one layer in the target plot. Compared with the traditional supervised training method, the present application is based on the target plot sub-data including partial seismic data of at least one first layer with relative geological time, and can perform semi-supervised training on the relative geological time prediction model, without the need to pre-make a large number of labels, reducing the workload of manual interpretation of seismic data. And in the process of predicting the plot seismic data of the target plot, the first layer of the target plot with relative geological time marked in the training process is introduced to constrain the prediction results of the second layer without relative geological time, while improving the accuracy of the prediction results, and improving the prediction efficiency of the relative geological time prediction model for relative geological time.
[0127] Fig. 9 is a block diagram of a relative geological time prediction device provided according to an embodiment of the present application. The device is used to execute the above relative geological time prediction method, see Fig. 9 The device includes: a segmentation module 901, a training module 902, a prediction module 903 and a merging module 904.
[0128] A segmentation module 901 is used to segment the block seismic data of the target block to be predicted to obtain a plurality of block sub-data of the target block, wherein the block seismic data includes seismic data of a plurality of first layers in the target block with marked relative geological time and seismic data of a plurality of second layers without marked relative geological time, wherein the relative geological time is used to indicate the relative formation time of the layers in the target block, and the block sub-data includes partial seismic data of at least one layer in the target block;
[0129] A training module 902 is used to train a relative geological time prediction model based on a plurality of target plot sub-data, where the target plot sub-data is plot sub-data including at least one partial seismic data of a first horizon among the plurality of plot sub-data;
[0130] A prediction module 903 is used to predict the multiple plot sub-data based on the relative geological time prediction model obtained through training, and obtain the predicted relative geological time of each layer in the multiple plot sub-data;
[0131] The merging module 904 is used to merge the predicted relative geological time of each layer in the multiple plot sub-data to obtain the predicted relative geological time of multiple second layers.
[0132] In some embodiments, Fig.10 is a block diagram of another relative geological time prediction device provided according to an embodiment of the present application, see Fig.10 , the training module 902 comprises:
[0133] The prediction unit 1001 is used to predict any target land parcel sub-data among the multiple target land parcel sub-data based on the relative geological time prediction model to obtain prediction information of the target land parcel sub-data, where the prediction information is used to indicate the predicted relative geological time of each horizon in the target land parcel sub-data;
[0134] A determining unit 1002 is used to determine a training loss of a relative geological time prediction model based on the target plot sub-data and the prediction information, where the training loss is used to indicate a difference between the prediction information and the relative geological time marked in the target plot sub-data;
[0135] The updating unit 1002 is used to update the model parameters of the relative geological time prediction model based on the training loss.
[0136] In some embodiments, the target block sub-data includes partial seismic data of a plurality of horizons, the plurality of horizons including at least one first horizon;
[0137] Continue to see Fig.10 , the determining unit 1002 includes:
[0138] An acquisition subunit 10021 is used to acquire a relative geological time of at least one first horizon based on partial seismic data of at least one first horizon in the target block subdata;
[0139] A first determining subunit 10022 is used to determine the constraint loss of the target plot sub-data based on the predicted relative geological time of at least one first horizon and the relative geological time of at least one first horizon, wherein the constraint loss is used to indicate the difference between the predicted relative geological time of at least one first horizon and the relative geological time of at least one first horizon;
[0140] The second determination subunit 10023 is used to determine the structural loss of the target plot subdata based on the gradient vector of the target plot subdata and the gradient vector of the predicted relative geological time of the multiple horizons, wherein the structural loss is used to indicate the difference between the gradient vector of the predicted relative geological time of the multiple horizons and the gradient vector of the target plot subdata;
[0141] The third determination subunit 10024 is used to perform weighted summation on the structural loss and the constraint loss to obtain the training loss.
[0142] In some embodiments, the second determination subunit 10023 is used to determine the tensor matrix of the target plot sub-data based on the gradient vector of the target plot sub-data; perform singular value decomposition on the tensor matrix to obtain the normal vector of the target plot sub-data and the quality parameter of the normal vector, the quality parameter being used to indicate the accuracy of the normal vector; perform weighted summation on the normal vector, the quality parameter and the gradient vector of the predicted relative geological time to obtain the structural loss.
[0143] In some embodiments, see Fig.10 , the device further comprises:
[0144] A determination module 905 is used to determine a plurality of layers associated with the well logging position in the target plot based on the well logging position of the target plot;
[0145] The marking module 906 is used to mark multiple layers based on the well logging interpretation information of the target block to obtain the seismic data of the block. The well logging interpretation information is used to indicate the relative geological time of the multiple layers.
[0146] In some embodiments, the apparatus further comprises:
[0147] The determination module 905 is further used to determine the multiple horizons interpreted by the horizon interpretation information based on the horizon interpretation information of the target land parcel, where the horizon interpretation information is used to indicate the relative geological time of the multiple horizons;
[0148] The labeling module 906 is also used to label multiple target horizons based on the horizon interpretation information to obtain the seismic data of the plot.
[0149] The embodiment of the present application provides a relative geological time prediction device, which can obtain multiple block sub-data including partial seismic data of at least one layer in the target block by segmenting the seismic data of the block to be predicted, including the seismic data of multiple first layers with relative geological time marked in the target block and the seismic data of multiple second layers without relative geological time marked. Compared with the traditional supervised training method, the present application can perform semi-supervised training on the relative geological time prediction model based on the target block sub-data including partial seismic data of at least one first layer with relative geological time marked, without the need to pre-make a large number of labels, reducing the workload of manual interpretation of seismic data. And in the process of predicting the block seismic data of the target block, the first layer of the target block with relative geological time marked in the training process is introduced to constrain the prediction results of the second layer without relative geological time marked, while improving the accuracy of the prediction results, and improving the prediction efficiency of the relative geological time prediction model for relative geological time.
[0150] It should be noted that the relative geological time prediction device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the terminal is divided into different functional modules to complete all or part of the functions described above. In addition, the relative geological time prediction device provided in the above embodiment and the relative geological time prediction method embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.
[0151] In the embodiments of the present application, the computer device can be configured as a terminal or a server. When the computer device is configured as a terminal, the terminal can be used as the execution subject to implement the technical solution provided in the embodiments of the present application. When the computer device is configured as a server, the server can be used as the execution subject to implement the technical solution provided in the embodiments of the present application. The technical solution provided in the present application can also be implemented through interaction between the terminal and the server. The embodiments of the present application are not limited to this.
[0152] Fig.11 It is a structural block diagram of a terminal 1100 provided according to an embodiment of the present application.
[0153] Typically, the terminal 1100 includes a processor 1101 and a memory 1102 .
[0154] The processor 1101 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 1101 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 1101 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 1101 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 1101 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.
[0155] The memory 1102 may include one or more computer-readable storage media, which may be non-transitory. The memory 1102 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 1102 is used to store at least one computer program, which is used to be executed by the processor 1101 to implement the relative geological time prediction method provided in the method embodiment of the present application.
[0156] In some embodiments, the terminal 1100 may further optionally include: a peripheral device interface 1103 and at least one peripheral device. The processor 1101, the memory 1102 and the peripheral device interface 1103 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface 1103 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 1104, a display screen 1105, a camera assembly 1106, an audio circuit 1107 and a power supply 1108.
[0157] The peripheral device interface 1103 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 1101 and the memory 1102. In some embodiments, the processor 1101, the memory 1102, and the peripheral device interface 1103 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 1101, the memory 1102, and the peripheral device interface 1103 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.
[0158] The radio frequency circuit 1104 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 1104 communicates with communication networks and other communication devices through electromagnetic signals. The radio frequency circuit 1104 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals into electrical signals. In some embodiments, the radio frequency circuit 1104 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 radio frequency circuit 1104 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes but is not limited to: the World Wide Web, a metropolitan area network, an intranet, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or WiFi ( WirelessFidelity In some embodiments, the radio frequency circuit 1104 may also include a circuit related to NFC (Near Field Communication), which is not limited in the present application.
[0159] The display screen 1105 is used to display a UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof. When the display screen 1105 is a touch display screen, the display screen 1105 also has the ability to collect touch signals on the surface or above the surface of the display screen 1105. The touch signal can be input to the processor 1101 as a control signal for processing. At this time, the display screen 1105 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen 1105 can be one, set on the front panel of the terminal 1100; in other embodiments, the display screen 1105 can be at least two, respectively set on different surfaces of the terminal 1100 or in a folding design; in other embodiments, the display screen 1105 can be a flexible display screen, set on a curved surface or a folding surface of the terminal 1100. Even, the display screen 1105 can also be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen 1105 can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).
[0160] The camera assembly 1106 is used to capture images or videos. In some embodiments, the camera assembly 1106 includes a front camera and a rear camera. Typically, the front camera is disposed on the front panel of the terminal, and the rear camera is disposed on the back of the terminal. In some embodiments, there are at least two rear cameras, which are any one of a main camera, a depth of field camera, a wide-angle camera, and a telephoto camera, so as to realize the fusion of the main camera and the depth of field camera to realize the background blur function, the fusion of the main camera and the wide-angle camera to realize the panoramic shooting and VR (Virtual Reality) shooting function or other fusion shooting functions. In some embodiments, the camera assembly 1106 may also include a flash. The flash may be a monochrome temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cold light flash, which can be used for light compensation at different color temperatures.
[0161] The audio circuit 1107 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals and input them into the processor 1101 for processing, or input them into the radio frequency circuit 1104 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal 1100. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 1101 or the radio frequency circuit 1104 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 1107 may also include a headphone jack.
[0162] The power supply 1108 is used to power various components in the terminal 1100. The power supply 1108 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 1108 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 charged through a wired line, and a wireless rechargeable battery is a battery charged through a wireless coil. The rechargeable battery can also be used to support fast charging technology.
[0163] In some embodiments, the terminal 1100 further includes one or more sensors 1109 , including but not limited to: an acceleration sensor 1110 , a gyroscope sensor 1111 , a pressure sensor 1112 , an optical sensor 1113 , and a proximity sensor 1114 .
[0164] The acceleration sensor 1110 can detect the magnitude of acceleration on the three coordinate axes of the coordinate system established by the terminal 1100. For example, the acceleration sensor 1110 can be used to detect the components of gravity acceleration on the three coordinate axes. The processor 1101 can control the display screen 1105 to display the user interface in a horizontal view or a vertical view according to the gravity acceleration signal collected by the acceleration sensor 1110. The acceleration sensor 1110 can also be used for collecting game or user motion data.
[0165] The gyro sensor 1111 can detect the body direction and rotation angle of the terminal 1100, and the gyro sensor 1111 can cooperate with the acceleration sensor 1110 to collect the user's 3D actions on the terminal 1100. The processor 1111 can implement the following functions based on the data collected by the gyro sensor 1111: motion sensing (such as changing the UI according to the user's tilt operation), image stabilization during shooting, game control, and inertial navigation.
[0166] The pressure sensor 1112 may be provided at the side frame of the terminal 1100 and / or the lower layer of the display screen 1105. When the pressure sensor 1112 is provided at the side frame of the terminal 1100, it may detect the user's holding signal of the terminal 1100, and the processor 1101 may perform left and right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 1112. When the pressure sensor 1112 is provided at the lower layer of the display screen 1105, the processor 1101 may control the operability control on the UI interface according to the user's pressure operation on the display screen 1105. The operability control includes at least one of a button control, a scroll bar control, an icon control, and a menu control.
[0167] The optical sensor 1113 is used to collect the ambient light intensity. In one embodiment, the processor 1101 can control the display brightness of the display screen 1105 according to the ambient light intensity collected by the optical sensor 1113. Specifically, when the ambient light intensity is high, the display brightness of the display screen 1105 is increased; when the ambient light intensity is low, the display brightness of the display screen 1105 is reduced. In another embodiment, the processor 1101 can also dynamically adjust the shooting parameters of the camera assembly 1106 according to the ambient light intensity collected by the optical sensor 1113.
[0168] The proximity sensor 1114, also called a distance sensor, is usually arranged on the front panel of the terminal 1100. The proximity sensor 1114 is used to collect the distance between the user and the front of the terminal 1100. In one embodiment, when the proximity sensor 1114 detects that the distance between the user and the front of the terminal 1100 is gradually decreasing, the processor 1101 controls the display screen 1105 to switch from the screen-on state to the screen-off state; when the proximity sensor 1104 detects that the distance between the user and the front of the terminal 1100 is gradually increasing, the processor 1101 controls the display screen 1105 to switch from the screen-off state to the screen-on state.
[0169] Those skilled in the art will understand that Fig.11 The structure shown in the figure does not constitute a limitation on the terminal 1100, and the terminal 1100 may include more or fewer components than those shown in the figure, or combine certain components, or adopt a different component arrangement.
[0170] Fig.12It is a structural diagram of a server provided according to an embodiment of the present application. The server 1200 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 1201 and one or more memories 1202, wherein the memory 1202 stores at least one computer program, and the at least one computer program is loaded and executed by the processor 1201 to implement the relative geological time prediction method provided by the above-mentioned various method embodiments. Of course, the server may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output, and the server may also include other components for implementing device functions, which will not be repeated here.
[0171] The embodiment of the present application also provides a computer-readable storage medium, in which at least one computer program is stored, and the at least one computer program is loaded and executed by a processor to implement the relative geological time prediction method in the above embodiment. For example, the computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device.
[0172] An embodiment of the present application also provides a computer program product, including a computer program product, which is executed by a processor to implement the relative geological time prediction method in the above embodiment.
[0173] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.
[0174] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
Claims
1. A relative geological time prediction method, characterized in that: The method comprises: Segmenting the block seismic data of the target block to be predicted to obtain a plurality of block sub-data of the target block, wherein the block seismic data includes seismic data of a plurality of first layers in the target block with marked relative geological time and seismic data of a plurality of second layers without marked relative geological time, wherein the relative geological time is used to indicate the relative formation time of the layers in the target block, and the block sub-data includes partial seismic data of at least one layer in the target block; Based on a plurality of target plot sub-data, a relative geological time prediction model is trained, wherein the target plot sub-data is plot sub-data including at least one portion of seismic data of a first horizon among the plurality of plot sub-data; Based on the relative geological time prediction model obtained through training, the plurality of land parcel sub-data are predicted respectively to obtain the predicted relative geological time of each layer in the plurality of land parcel sub-data; The predicted relative geological time of each layer in the plurality of land parcel sub-data is merged to obtain the predicted relative geological time of the plurality of second layers.
2. The method according to claim 1, characterized in that The relative geological time prediction model is trained based on the multiple target plot sub-data, including: For any target land parcel sub-data among the plurality of target land parcel sub-data, predicting the target land parcel sub-data based on the relative geological time prediction model to obtain prediction information of the target land parcel sub-data, wherein the prediction information is used to indicate the predicted relative geological time of each horizon in the target land parcel sub-data; Determining a training loss of the relative geological time prediction model based on the target plot sub-data and the prediction information, wherein the training loss is used to indicate a difference between the prediction information and the relative geological time marked in the target plot sub-data; Based on the training loss, model parameters of the relative geological time prediction model are updated.
3. The method according to claim 2, characterized in that The target block sub-data includes partial seismic data of multiple horizons, and the multiple horizons include at least one first horizon; The determining the training loss of the relative geological time prediction model based on the target plot sub-data and the prediction information includes: Based on partial seismic data of at least one first horizon in the target land block sub-data, obtaining a relative geological time of the at least one first horizon; Determining a constraint loss of the target plot sub-data based on the predicted relative geological time of the at least one first horizon and the relative geological time of the at least one first horizon, the constraint loss being used to indicate a difference between the predicted relative geological time of the at least one first horizon and the relative geological time of the at least one first horizon; Determine a structural loss of the target plot sub-data based on a gradient vector of the target plot sub-data and a gradient vector of the predicted relative geological time of the plurality of horizons, wherein the structural loss is used to indicate a difference between a gradient vector of the predicted relative geological time of the plurality of horizons and a gradient vector of the target plot sub-data; The structural loss and the constraint loss are weightedly summed to obtain the training loss.
4. The method according to claim 3, characterized in that The determining of the structural loss of the target land parcel sub-data based on the gradient vector of the target land parcel sub-data and the gradient vectors of the predicted relative geological time of the plurality of horizons comprises: Determining a tensor matrix of the target land parcel sub-data based on a gradient vector of the target land parcel sub-data; Performing singular value decomposition on the tensor matrix to obtain a normal vector of the target land block sub-data and a quality parameter of the normal vector, wherein the quality parameter is used to indicate the accuracy of the normal vector; The normal vector, the quality parameter and the predicted gradient vector relative to the geological time are weightedly summed to obtain the structural loss.
5. The method according to claim 1, characterized in that The method further comprises: Based on the well logging position of the target plot, determining a plurality of horizons in the target plot associated with the well logging position; Based on the well logging interpretation information of the target land parcel, the multiple layers are marked to obtain the seismic data of the land parcel, and the well logging interpretation information is used to indicate the relative geological time of the multiple layers.
6. The method according to claim 1, characterized in that The method further comprises: Based on the horizon interpretation information of the target land parcel, determining a plurality of horizons interpreted by the horizon interpretation information, wherein the horizon interpretation information is used to indicate a relative geological time of the plurality of horizons; Based on the horizon interpretation information, the multiple target horizons are marked to obtain the seismic data of the land block.
7. A relative geological time prediction device, characterized in that: The device comprises: A segmentation module, used for segmenting the block seismic data of the target block to be predicted to obtain a plurality of block sub-data of the target block, wherein the block seismic data includes seismic data of a plurality of first layers in the target block with marked relative geological time and seismic data of a plurality of second layers without marked relative geological time, wherein the relative geological time is used to indicate the relative formation time of the layers in the target block, and the block sub-data includes partial seismic data of at least one layer in the target block; A training module, for training a relative geological time prediction model based on a plurality of target plot sub-data, wherein the target plot sub-data is plot sub-data including at least one partial seismic data of a first horizon among the plurality of plot sub-data; A prediction module, used to predict the plurality of land parcel sub-data respectively based on the relative geological time prediction model obtained through training, so as to obtain the predicted relative geological time of each layer in the plurality of land parcel sub-data; The merging module is used to merge the predicted relative geological time of each layer in the multiple plot sub-data to obtain the predicted relative geological time of the multiple second layers.
8. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory is used to store at least one computer program, and the at least one computer program is loaded by the processor and executes the relative geological time prediction method described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store at least one computer program, and the at least one computer program is used to execute the relative geological time prediction method described in any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the relative geological time prediction method as claimed in any one of claims 1 to 6 is implemented.
Citation Information
Patent Citations
Frequency domain full-waveform inversion seismic velocity modeling method
CN103207409A
Determination method and apparatus for acquisition footprint
CN105425294A
A method for identifying alluvial fan microfacies through chronostratigraphic facies
CN106547022A
Seismic horizon identification and tracking method and system
CN109709603A
Seismic fault and horizon interpretation method and device based on multi-task deep learning
CN114048676A
