A method for morphological construction of a marine seismic acquisition cable and related apparatus

Abstract

The application discloses a kind of ocean seismic acquisition cable shape construction method and related equipment, method includes: obtaining current time shape prediction data and attitude data;Based on shape prediction data, generate prediction shape network graph by shape prediction model;Based on attitude data, generate initial shape network graph by initial shape generation model;Similarity quantization is carried out to initial shape network graph and prediction shape network graph, and similarity is obtained;When similarity is not less than preset threshold, initial shape network graph is regarded as the first shape network graph of current time;When similarity is less than preset threshold, reconstruct the first shape network graph of current time according to initial shape network graph and prediction shape network graph;First shape network graph is used to assist subsequent time shape prediction data is constituted.The application can significantly improve the precision and reliability of cable shape determination, and can realize continuous optimization prediction effect, and can be widely applied in data processing technical field.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method for constructing the shape of a marine seismic acquisition cable and related equipment. Background Technology

[0002] Navigation and positioning in towed cable exploration are achieved collaboratively using multiple types of equipment. Relying on DGPS (Differential GPS) installed on the ship's hull, differential information from satellite stations is received to obtain the ship's reference position, serving as the starting coordinates for the entire positioning network. Then, RGPS (Relative GPS) configured on the cable stern marker and gun array is used to calculate the baseline vector between itself and the ship's RGPS reference station, thereby determining the position of the gun array and the cable stern. Based on DGPS and RGPS data, the precise coordinates of the cable's head and tail can be obtained. The coordinates of the cable's mid-section are primarily controlled by a combination of acoustic positioning and a compass bird: acoustic positioning equipment deployed on the cable measures the distances between acoustic nodes, constructing an acoustic boundary network; subsequent network adjustment calculations then solve for the coordinates of the acoustic nodes. Since acoustic nodes cannot cover all receiver points, for receiver point positioning between nodes, azimuth information obtained from the compass bird, combined with the theoretical cable distance, is used to determine the cable's spatial shape, and then the corresponding receiver point coordinates are calculated. In practical engineering applications, curve integration or polynomial fitting methods are typically used to determine the cable shape. Because the cable spatial morphology determined by the curve integral model, polynomial fitting model, or improved model based on both methods has not been verified by an effective method, its accuracy cannot be determined. At the same time, the existing models do not consider the influence of environmental factors or the prior knowledge of historical morphologies when determining the cable morphology. Therefore, the determined morphologies may not all meet the guidance requirements for field data collection operations, nor may they all meet the accuracy requirements for data processing. Summary of the Invention

[0003] The main objective of this invention is to provide a method, apparatus, electronic device, storage medium, and program product for constructing the morphology of a marine seismic acquisition cable, aiming to solve at least one problem in the prior art.

[0004] To achieve the above objectives, one aspect of this invention proposes a method for constructing the morphology of a marine seismic acquisition cable, the method comprising: Obtain the current morphology prediction data and attitude data; Based on morphological prediction data, a morphological prediction network diagram is generated through a morphological prediction model. Based on the posture data, an initial morphology network graph is generated through an initial morphology generation model. The similarity measure between the initial morphological network graph and the predicted morphological network graph is quantified to obtain the similarity score. When the similarity is not less than a preset threshold, the initial morphological network graph is used as the first morphological network graph at the current moment. When the similarity is less than a preset threshold, the first morphological network graph at the current moment is reconstructed based on the initial morphological network graph and the predicted morphological network graph. The first morphological network graph is used to assist in constructing morphological prediction data for subsequent time steps.

[0005] In some embodiments, the first morphological network graph is stored in a historical database, and obtaining the morphological prediction data and attitude data at the current moment includes the following steps: Obtain the current moment's shape prediction data from the historical database; The morphological prediction data includes environmental sequence data, work vessel sequence data, and first morphological network sequence diagram. The environmental sequence data includes environmental data from multiple historical moments before the current moment, the work vessel sequence data includes work vessel data from multiple historical moments before the current moment, and the first morphological network sequence diagram includes first morphological network diagrams from multiple historical moments before the current moment. Attitude data at the current moment is obtained through real-time sensor acquisition.

[0006] In some embodiments, the morphological prediction model includes a first encoding network, a second encoding network, a third encoding network, a first feature fusion network, and a first decoding network. Based on morphological prediction data, the morphological prediction model generates a predicted morphological network graph, including the following steps: The environmental sequence data is encoded using a first encoding network to obtain a first encoding vector; The second encoding network is used to perform a second encoding on the work vessel sequence data to obtain a second encoding vector; The first morphological network sequence diagram is encoded using a third encoding network to obtain a third encoding vector; The first feature fusion network is used to perform a first fusion process on the first encoding vector, the second encoding vector, and the third encoding vector to obtain the first fused vector. The first decoding operation is performed on the first fusion vector through the first decoding network to obtain the first prediction data; wherein, the first prediction data includes the spatial distribution data of each depth-oriented device and each data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data; A predictive morphological network diagram is generated based on the spatial coordinate data corresponding to the first predictive data.

[0007] In some embodiments, the third coding network includes multiple parallel first sub-coding networks and multiple parallel second sub-coding networks. The third coding network is used to perform third coding on the first morphological network sequence diagram to obtain a third coding vector, including the following steps: By using the graph cutting model and the depth-director node sequence, each first-morphology network graph in the first-morphology network sequence graph is cut into multiple sub-network graphs; Using the first grouping model, sub-network diagrams with the same node number in the same acquisition cable corresponding to the first morphological network diagram at different times are classified into the same first sub-network diagram group. Using the second grouping model, sub-network diagrams with the same node number in different acquisition cables in the first morphological network diagram at the same time are classified into the same second sub-network diagram group. Each first sub-network graph group is encoded by multiple parallel first sub-encoding networks, and then the encoding results of all first sub-encoding networks are fused into a first fused sub-vector; wherein, the first sub-encoding network corresponds one-to-one with the first sub-network graph group; Each second sub-network graph group is encoded by multiple parallel second sub-encoding networks, and then the encoding results of all second sub-encoding networks are fused into a second fused sub-vector; wherein, the second sub-encoding network corresponds one-to-one with the second sub-network graph group; The first fusion subvector and the second fusion subvector are fused together to form the third encoding vector.

[0008] In some embodiments, attitude data is acquired through real-time sensors. The attitude data includes coordinate data, relative distance data, depth data, and orientation data. The initial morphology generation model employs a polynomial line integral model. Based on the attitude data, an initial morphology network graph is generated using the initial morphology generation model, including the following steps: Retrieve pre-stored cable configuration data; The cable configuration data includes header and tail markers, depth orientation device, and fixed offset of the data acquisition device on the acquisition cable. Based on attitude data and cable configuration data, a polynomial function is constructed using a polynomial curve integral model to fit the spatial coordinates of any point on the acquisition cable and the offset along the acquisition cable, thus obtaining the initial morphological data at the current moment. The initial morphological data includes the spatial distribution data of each depth-oriented device and each data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data. An initial morphological network diagram is generated based on the spatial coordinate data corresponding to the initial morphological data.

[0009] In some embodiments, both the initial morphological network diagram and the predicted morphological network diagram include spatial coordinate data of each node on each acquisition cable. Each node includes a depth-oriented orienter and a data acquisition device. The initial morphological network diagram and the predicted morphological network diagram are quantified to obtain a similarity score, including the following steps: The first acquisition cable was selected as the candidate cable. The first node on the candidate cable is selected as the candidate node; Extract the first coordinate data of candidate nodes from the initial morphological network graph, and extract the second coordinate data of candidate nodes from the predicted morphological network graph; The spatial distances of candidate nodes in the initial and predicted morphological network graphs are obtained by quantization based on the first and second coordinate data. Take the next node on the candidate cable as the candidate node, return to the step of extracting the first coordinate data of the candidate node from the initial morphological network graph, and continue until all nodes on the candidate cable have been traversed. The average spatial distance of the candidate cable is obtained by averaging the spatial distances of all nodes on the candidate cable. Based on the average spatial distance, the cable similarity corresponding to the candidate cable is obtained by combining the distance threshold transformation. The expression for cable similarity is as follows: ; In the formula, Indicates the first Cable similarity of the acquisition cables This indicates the operation of retrieving the maximum value. Indicates the first Average spatial distance of the data acquisition cables Indicates the distance threshold; The next acquisition cable is selected as a candidate cable, and the process of selecting the first node on the candidate cable as a candidate node is repeated until all acquisition cables have been traversed. The minimum similarity among all collected cables is taken as the final similarity between the initial morphological network graph and the predicted morphological network graph.

[0010] In some embodiments, reconstructing the first morphological network graph at the current moment based on the initial morphological network graph and the predicted morphological network graph includes the following steps: Obtain real-time environmental data and real-time operational vessel data at the current moment; The initial morphological network diagram, the predicted morphological network diagram, real-time environmental data, and real-time operational vessel data are input into the first morphological generation model so that the first morphological generation model can process and output the first morphological network diagram at the current moment. The first morphology generation model includes a fourth encoding network, a fifth encoding network, a sixth encoding network, a second feature fusion network, and a second decoding network; the first morphology generation model specifically performs the following operations: The real-time environmental data is encoded using a fourth encoding network to obtain a fourth encoding vector; The real-time operational vessel data is encoded using a fifth-level encoding network to obtain a fifth-level encoding vector. The initial morphological network graph and the predicted morphological network graph are encoded using a sixth encoding network to obtain a sixth encoding vector. The second feature fusion network is used to perform a second fusion process on the fourth, fifth, and sixth coding vectors to obtain the second fused vector. The second decoding operation is performed on the second fusion vector through the second decoding network to obtain the first morphological data; wherein, the first morphological data includes the spatial distribution data of each depth-oriented and data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data; The first-morphology network diagram is generated based on the spatial coordinate data corresponding to the first-morphology data.

[0011] To achieve the above objectives, another aspect of the present invention provides a morphological construction device for a marine seismic acquisition cable, the device comprising: The first module is used to acquire the current morphology prediction data and attitude data; The second module is used to generate a predicted morphology network diagram based on morphology prediction data and through a morphology prediction model. The third module is used to generate an initial morphology network graph based on the posture data and through the initial morphology generation model. The fourth module is used to measure the similarity between the initial morphological network graph and the predicted morphological network graph to obtain the similarity score. The fifth module is used to take the initial morphological network graph as the first morphological network graph at the current time when the similarity is not less than a preset threshold. The sixth module is used to reconstruct the first morphological network graph at the current moment based on the initial morphological network graph and the predicted morphological network graph when the similarity is less than a preset threshold. The first morphological network graph is used to assist in constructing morphological prediction data for subsequent time steps.

[0012] To achieve the above objectives, another aspect of the present invention provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method.

[0013] To achieve the above objectives, another aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method.

[0014] To achieve the above objectives, another aspect of the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0015] The embodiments of this invention include at least the following beneficial effects: This invention provides a method, apparatus, electronic device, storage medium, and program product for constructing the morphology of a marine seismic acquisition cable. This solution acquires morphology prediction data and attitude data at the current moment; generates a predicted morphology network graph using a morphology prediction model based on the morphology prediction data; generates an initial morphology network graph using an initial morphology generation model based on the attitude data; performs similarity quantification on the initial morphology network graph and the predicted morphology network graph to obtain a similarity score; when the similarity score is not less than a preset threshold, the initial morphology network graph is used as the first morphology network graph at the current moment; when the similarity score is less than the preset threshold, the first morphology network graph at the current moment is reconstructed based on the initial morphology network graph and the predicted morphology network graph; wherein, the first morphology network graph is used to assist in constructing morphology prediction data for subsequent moments. The embodiments of this invention, by generating the predicted morphology network graph and the initial morphology network graph in parallel and introducing a similarity quantification comparison mechanism, can effectively verify the output results of the lightweight model, effectively solving the deficiency of traditional methods in being unable to assess their own accuracy. When the similarity is insufficient, the current morphology is reconstructed using both the predicted morphology network graph and the initial morphology network graph. This approach integrates historical prior knowledge while also considering the current real-time attitude, significantly improving the accuracy and reliability of cable morphology determination. Furthermore, the final output of the first morphology network graph is used to feed back morphology prediction data to subsequent time steps, forming a closed-loop iteration that continuously optimizes the prediction results. This better meets the high-precision requirements for field data collection guidance and processing. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of an implementation environment for the method of constructing the morphology of a marine seismic acquisition cable provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the method for constructing the morphology of a marine seismic acquisition cable according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the overall process of constructing the morphology of a marine seismic acquisition cable according to an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating an example of the data processing principle architecture of the morphological prediction model provided in this embodiment of the invention; Figure 5 This is a schematic diagram of the morphological construction device for marine seismic acquisition cables provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of this invention; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this invention as detailed in the appended claims.

[0018] It is understood that the terms "first," "second," etc., used in this invention may be used to describe various concepts, but unless specifically stated otherwise, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of embodiments of this invention, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words "if" or "when" as used herein may be interpreted as "when," "in response to determination," or "in the event of a determination."

[0019] The terms “at least one,” “multiple,” “each,” “any,” etc., used in this invention, “at least one” includes one, two, or more than two; “multiple” includes two or more than two; “each” refers to each of the corresponding multiple; and “any” refers to any one of the multiple.

[0020] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this invention is for descriptive purposes only and is not intended to limit the invention.

[0021] In related technologies, there is an urgent need for an efficient and accurate method for determining the shape of seismic acquisition cables, and this method should be able to simultaneously achieve both lightweight and high-accuracy determination of seismic acquisition cable shape.

[0022] In view of this, this invention provides a method and related equipment for constructing the morphology of a marine seismic acquisition cable. This method acquires morphology prediction data and attitude data at the current moment; generates a predicted morphology network diagram using a morphology prediction model based on the morphology prediction data; generates an initial morphology network diagram using an initial morphology generation model based on the attitude data; performs similarity quantification on the initial morphology network diagram and the predicted morphology network diagram to obtain a similarity score; when the similarity score is not less than a preset threshold, the initial morphology network diagram is used as the first morphology network diagram at the current moment; when the similarity score is less than the preset threshold, the first morphology network diagram at the current moment is reconstructed based on the initial morphology network diagram and the predicted morphology network diagram; wherein, the first morphology network diagram is used to assist in constructing morphology prediction data for subsequent moments. This invention, by generating the predicted morphology network diagram and the initial morphology network diagram in parallel and introducing a similarity quantification comparison mechanism, can effectively verify the output results of the lightweight model, effectively solving the deficiency of traditional methods in being unable to assess their own accuracy. Specifically, when the similarity score is insufficient, the morphology at the current moment is reconstructed using the predicted morphology network diagram and the initial morphology network diagram together, integrating historical prior knowledge while also considering the current real-time attitude, which can significantly improve the accuracy and reliability of cable morphology determination. Furthermore, the final output of the first morphological network graph is used to feed back the morphological prediction data at subsequent time steps to form a closed-loop iteration, which can continuously optimize the prediction effect and thus better meet the high-precision requirements of field collection operation guidance and data processing.

[0023] It is understood that the method for constructing the marine seismic acquisition cable provided by this invention can be applied to any computer device with data processing and computing capabilities, and this computer device can be various terminals or servers. When the computer device in the embodiment is a server, the server 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 communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Optionally, the terminal can be a smartphone, tablet, laptop, or desktop computer, but it is not limited to these.

[0024] like Figure 1 The diagram shown is a schematic representation of an implementation environment provided by an embodiment of the present invention. (Refer to...) Figure 1 The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected via a network, either wirelessly or via a wired connection, to complete data transmission and exchange.

[0025] Server 101 can be a standalone physical server, a server cluster or distributed system consisting 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 communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0026] Additionally, server 101 can also be a node server in a blockchain network. Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms.

[0027] Terminal 102 can be a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. Terminal 102 and server 101 can be directly or indirectly connected via wired or wireless communication, and this embodiment of the invention does not impose any limitations.

[0028] For example, based on Figure 1 The implementation environment shown in this embodiment of the invention provides a method for constructing the shape of a marine seismic acquisition cable. The following description uses the application of this method in server 101 as an example. It can be understood that this method can also be applied in terminal 102.

[0029] Reference Figure 2 , Figure 2 This is an optional flowchart of a method for constructing the morphology of a marine seismic acquisition cable provided in an embodiment of the present invention. The execution subject of this method can be any of the aforementioned computer devices (including servers or terminals). Figure 2 The method may include, but is not limited to, steps S100 to S600.

[0030] Step S100: Obtain the shape prediction data and attitude data at the current moment; It should be noted that the first morphology network graph is stored in a historical database. In some embodiments, step S100 may include the following steps: obtaining morphology prediction data at the current moment from the historical database; wherein, the morphology prediction data includes environmental sequence data, workboat sequence data, and the first morphology network sequence graph, the environmental sequence data includes environmental data from multiple historical moments prior to the current moment, the workboat sequence data includes workboat data from multiple historical moments prior to the current moment, and the first morphology network sequence graph includes the first morphology network graph from multiple historical moments prior to the current moment; and obtaining attitude data at the current moment through real-time sensor acquisition.

[0031] For example, in some specific implementations, morphological prediction data at the current moment is obtained, wherein the morphological prediction data includes environmental sequence data, work vessel sequence data, and a first morphological network sequence diagram, and the environmental sequence data, work vessel sequence data, and first morphological network sequence diagram respectively indicate environmental data, work vessel data, and first morphological network diagram at multiple historical moments prior to the current moment.

[0032] Obtain the morphological prediction data at the current time (time t). The morphological prediction data at time t includes the first morphological network diagrams of the K time points prior to time t, as well as the corresponding K environmental data and K operational vessel data. The first morphological network diagrams of the K time points constitute the first morphological network sequence diagram. ,in , This represents the first-order network diagram at the k-th time point before time t. K environmental data points and K operational vessel data points constitute the environmental sequence data. and work vessel sequence data The K environmental data include K ocean current velocity data and ocean current direction data, represented as: This represents the velocity of the k-th ocean current. This indicates the direction of the k-th ocean current; the K data points for the work vessels include the positions of the K work vessels. { },in This represents the bearing data of the kth working vessel; K is a preset value that can be obtained by taking into account both calculation efficiency and accuracy.

[0033] First-form network diagram The sequence includes the first node of at least one acquisition cable. } and the second node sequence { The first node sequence indicates the depth-oriented device on the acquisition cable. This represents the first-mode network diagram of the i-th acquisition cable. The first in A depth-oriented sensor, and a second node sequence indicator for the data acquisition device on the acquisition cable. This represents the first-mode network diagram of the i-th acquisition cable. The Middle A data acquisition device. The data acquisition device is a seismic data acquisition device, usually a seismic detector; the depth and orientation device is an acoustic bird and / or a compass bird.

[0034] In addition, the attitude data at the current moment is obtained, which includes at least coordinate data, relative distance data, depth data, and orientation data.

[0035] During marine seismic exploration operations, multiple depth and orientation sensors (such as compass birds and acoustic birds) are typically deployed on the seismic acquisition cable to control its attitude. The compass birds and acoustic birds use their depth and orientation sensors to collect depth and orientation data. Simultaneously, a head marker is placed on the seismic acquisition cable, and a tail marker is placed at the stern. This network of the seismic vessel, head marker, multiple compass birds, multiple acoustic birds, and tail markers forms the basis for determining the cable's morphology.

[0036] Navigation and positioning for the acquisition cable typically employs Differential Global Positioning System (DGPS) and Relative GPS. DGPS and an gyrocompass are mounted on the work vessel, providing high-precision absolute position and bow azimuth in a geographic coordinate system, respectively. The DGPS position includes latitude, longitude, and geodetic height or elevation, with an accuracy typically in decimeters. Since the work vessel is assumed to be a rigid body, the relative relationships between equipment are fixed, allowing the calculation of coordinates for other points on the vessel. The underwater depth sensor is deployed on the cable and gun array, thus the gun array utilizes two elevations, unifying the elevation of the entire towed cable system. RGPS measures the distance and true azimuth between two points, transmitting the absolute position from the geophysical vessel to buoys (head / stern) located before and after the gun array and cable. Acoustic bird and compass bird sensors provide the acoustic distance between two points and the cable tangent magnetic azimuth, respectively, using edge and angle observations to transmit coordinate information along the cable.

[0037] Therefore, the attitude data Z collected in real time is: ; In the formula, Let be the attitude data of the i-th acquisition cable at time t. The depth data of the i-th cable at time t includes at least the depth data measured by the depth sensor on the gun array at time t and the depth data measured by multiple depth-fixing orientation devices. The azimuth data of the i-th cable at time t includes at least the azimuth data of the working vessel at time t, the azimuth data measured by the head marker, the azimuth data measured by the stern marker, and the azimuth data measured by multiple depth-oriented and orientation-determining devices. The coordinate data includes at least the coordinate data of the working vessel, the coordinate data of the head marker, and the coordinate data of the stern marker. The data is relative distance data, including at least the distances between multiple adjacent acoustic birds.

[0038] Step S200: Based on the morphology prediction data, generate a predicted morphology network diagram using the morphology prediction model; It should be noted that the morphological prediction model includes a first encoding network, a second encoding network, a third encoding network, a first feature fusion network, and a first decoding network. In some embodiments, step S200 may include the following steps: using the first encoding network to perform a first encoding on the environmental sequence data to obtain a first encoding vector; using the second encoding network to perform a second encoding on the working vessel sequence data to obtain a second encoding vector; using the third encoding network to perform a third encoding on the first morphological network sequence diagram to obtain a third encoding vector; using the first feature fusion network to perform a first fusion process on the first encoding vector, the second encoding vector, and the third encoding vector to obtain a first fusion vector; and using the first decoding network to perform a first decoding operation on the first fusion vector to obtain first prediction data. The first prediction data includes the spatial distribution data of each depth-oriented device and each data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data. A predicted morphological network diagram is generated based on the spatial coordinate data corresponding to the first prediction data.

[0039] For example, in some specific implementations, generating a predictive morphology network graph based on morphology prediction data can be achieved as follows: Based on the morphological prediction data and the morphological prediction model, the first prediction data is obtained, and a prediction morphological network diagram is constructed based on the first prediction data. The morphological prediction model adopts an encoder-decoder architecture, which includes a first encoder network, a second encoder network, a third encoder network, a first feature fusion network, and a first decoder network, wherein: The first encoding network is used to encode the environmental sequence data to obtain the first encoding vector: ; The second encoding network is used to encode the vessel sequence data to obtain the second encoding vector: ; The third encoding network is used to encode the sequence diagram of the first morphological network to obtain the third encoding vector: ; The first feature fusion network is used to fuse the first encoding vector, the second encoding vector, and the third encoding vector to obtain the first fused vector: ; The first decoding network is used to decode the first fused vector to obtain the first predicted data: . The first prediction data consists of the spatial distribution data of the depth-oriented devices and data acquisition devices, including the spatial coordinate data of multiple depth-oriented devices and multiple data acquisition devices. A prediction morphology network diagram can be constructed based on the first prediction data, specifically based on the first node sequence { } and the second node sequence { Generate a predictive morphological network diagram based on the corresponding spatial coordinate data. .

[0040] This invention, when predicting the morphology of seismic acquisition cables, simultaneously considers environmental sequence data, operational vessel sequence data, and the first morphological network sequence diagram prior to time t. This allows for the acquisition of a relatively accurate predicted morphological network diagram, which is then displayed to the user before data acquisition at time t. The user can intuitively see the predicted morphological network diagram at time t, aiding in decision-making. The first and second encoders can employ domain dictionary models, which are trained using domain data on a BERT pre-trained language model, specifically historical domain data. The third encoder employs a lightweight convolutional neural network model, a convolutional neural network model, or a graph convolutional neural network model. The decoder employs a generative model such as a bidirectional long short-term memory neural network. Other neural network models or deep learning models can be used for the first, second, third, and decoder; no specific limitations are imposed here.

[0041] It should be noted that the third coding network includes multiple parallel first sub-coding networks and multiple parallel second sub-coding networks. In some embodiments, the third coding network is used to perform third coding on the first morphological network sequence graph to obtain the third coding vector, which may include the following steps: cutting each first morphological network graph in the first morphological network sequence graph into multiple sub-network graphs using a graph cutting model and a depth-director node sequence; classifying sub-network graphs with the same node number in the same acquisition cable in the first morphological network graphs at different times into the same first sub-network graph group using a first grouping model; and classifying the first morphological network graphs at the same time into the same first sub-network graph group using a second grouping model. Sub-network graphs with the same node number in different acquisition cables are classified into the same second sub-network graph group in the network diagram; each first sub-network graph group is encoded by multiple parallel first sub-coding networks, and then the encoding results of all first sub-coding networks are fused into a first fused sub-vector; wherein, the first sub-coding network corresponds one-to-one with the first sub-network graph group; each second sub-network graph group is encoded by multiple parallel second sub-coding networks, and then the encoding results of all second sub-coding networks are fused into a second fused sub-vector; wherein, the second sub-coding network corresponds one-to-one with the second sub-network graph group; the first fused sub-vector and the second fused sub-vector are fused into a third encoding vector.

[0042] For example, in some specific implementations, the graph cutting model divides the first-morphology network graph according to the depth-oriented node sequence. Divide into multiple sub-network graphs, that is, according to the first node sequence, the first-form network graph. Two adjacent depth-oriented nodes on the same acquisition cable i Segmented into a sub-network graph ,in, This represents the sub-network diagram formed by the j1st and j2nd depth-oriented devices in the i-th acquisition cable within the k-th first-morphology network diagram.

[0043] Assuming each first-morphology network diagram has I cables and J depth-directioners are deployed on each cable, then each first-morphology network diagram can be divided into I×(J-1) sub-graphs, and the first-morphology network sequence diagram can be divided into K×I×(J-1) sub-graphs.

[0044] The first grouping model classifies multiple sub-network graphs into multiple first sub-network graph groups according to a first preset rule. Specifically, the first grouping model classifies sub-network graphs divided into the same node index in the same acquisition cable within the first-form network graphs at different times into the same first sub-network graph group. In this way, I×(J-1) first sub-network graph groups can be obtained, and each first sub-network graph group is represented as follows: } The second grouping model, according to the second preset rule, classifies multiple sub-network graphs into multiple second sub-network graph groups. Specifically, it classifies sub-network graphs with the same node index in different acquisition cables within the first-mode network graph at the same time into the same second sub-network graph group. This yields K×(J-1) second sub-network graph groups, each represented as: }

[0045] The third coding network in the morphological prediction model includes multiple parallel first sub-coding networks, multiple parallel second sub-coding networks, a first fusion sub-network, a second fusion sub-network, and a third fusion sub-network, wherein: Multiple parallel first sub-encoding networks are used to encode multiple first sub-network graph groups in parallel, respectively, to obtain multiple first sub-encoding vectors: ; The number of first sub-encoding networks is the same as the number of first sub-network graphs, and the number of first sub-encoding vectors obtained is also the same as the number of first sub-network graphs, both being I×(J-1); the first sub-encoding network adopts the lightweight convolutional neural network model LightCNN.

[0046] The first fusion sub-network is used to fuse multiple first sub-encoded vectors to obtain a first fused sub-vector: ; Multiple parallel second sub-encoding networks are used to encode multiple second sub-network graph groups in parallel, respectively, to obtain multiple second sub-encoding vectors: ; The number of second sub-encoding networks is the same as the number of second sub-network graph groups, and the number of second sub-encoding vectors obtained is also the same as the number of second sub-network graph groups, both being K×(J-1); the second sub-encoding networks also use the lightweight convolutional neural network model LightCNN. The second fusion sub-network is used to fuse multiple second sub-encoding vectors to obtain a second fused sub-vector: ; The third fusion sub-network is used to fuse the first fusion sub-vector and the second fusion sub-vector to obtain the third encoding vector: ; In this embodiment, the first-morphology network subgraphs, segmented from K first-morphology network graphs, are classified according to both temporal and spatial dimensions to obtain multiple first-sub-network graph groups and multiple second-sub-network graph groups. Multiple parallel sub-encoding networks are then used to encode these first-sub-network graph groups and second-sub-network graph groups, thereby obtaining a third encoding vector. Because both temporal and spatial dimensions are integrated, prediction accuracy is further improved. The multiple parallel sub-encoding networks employ a lightweight convolutional neural network model, improving prediction efficiency.

[0047] Step S300: Based on the pose data, generate an initial morphology network diagram using an initial morphology generation model; It should be noted that the attitude data is acquired through real-time sensors and includes coordinate data, relative distance data, depth data, and orientation data. The initial morphology generation model adopts a polynomial curve integral model. In some embodiments, step S300 may include the following steps: acquiring pre-stored cable configuration data; wherein, the cable configuration data includes head markers, tail markers, depth-oriented devices, and fixed offsets of the data acquisition devices on the acquisition cable; based on the attitude data and cable configuration data, constructing a polynomial function between the spatial coordinates of any point on the acquisition cable and the offset along the acquisition cable using a polynomial curve integral model, and fitting it to obtain the initial morphology data at the current moment; wherein, the initial morphology data includes the spatial distribution data of each depth-oriented device and each data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data; generating an initial morphology network diagram based on the spatial coordinate data corresponding to the initial morphology data.

[0048] For example, in some specific implementations, constructing an initial morphological network graph based on pose data can be achieved as follows: After obtaining the attitude data, an initial shape network diagram of the cable is generated based on the initial shape generation model. The initial morphology generation model can employ a polynomial line integral model. Since polynomial line integral models are lightweight models, they can quickly calculate and obtain the morphology of the seismic acquisition cable.

[0049] Generate the initial shape diagram of the cable based on the initial shape generation model. This includes: acquiring cable configuration data; generating initial shape data based on the cable configuration data, attitude data, and initial shape model; and generating an initial shape diagram based on the initial shape data. Since the offsets of the header, tail, and depth-oriented pointers on the cable are fixed, these offsets can be easily obtained from the cable configuration data. This is achieved by constructing the coordinates of any point on the cable. Initial shape data is obtained by fitting a polynomial function of the offset along the cable. This initial shape data includes spatial distribution data of multiple depth-oriented geophones and multiple data acquisition units, including spatial coordinate data. The data acquisition units are seismic detectors, whose positions within the cable are fixed; therefore, the offset of each seismic detector can also be obtained using the cable configuration data.

[0050] Step S400: Measure the similarity between the initial morphological network graph and the predicted morphological network graph to obtain the similarity score. It should be noted that both the initial morphological network diagram and the predicted morphological network diagram include the spatial coordinate data of each node on each acquisition cable. The nodes include depth-oriented orienters and data acquisition devices. In some embodiments, step S400 may include the following steps: taking the first acquisition cable as a candidate cable; taking the first node on the candidate cable as a candidate node; extracting the first coordinate data of the candidate node from the initial morphological network diagram and the second coordinate data of the candidate node from the predicted morphological network diagram; quantizing the spatial distance of the candidate node in the initial morphological network diagram and the predicted morphological network diagram based on the first coordinate data and the second coordinate data; taking the next node on the candidate cable as a candidate node and returning to the step of extracting the first coordinate data of the candidate node from the initial morphological network diagram until all nodes on the candidate cable have been traversed; averaging the spatial distances corresponding to all nodes on the candidate cable to obtain the average spatial distance of the candidate cable; obtaining the cable similarity of the candidate cable based on the average spatial distance and combined with a distance threshold conversion; taking the next acquisition cable as a candidate cable and returning to the step of taking the first node on the candidate cable as a candidate node until all acquisition cables have been traversed; taking the minimum value of the cable similarity among all acquisition cables as the final similarity between the initial morphological network diagram and the predicted morphological network diagram.

[0051] For example, in some specific implementations, the similarity between the initial morphological network graph and the predicted morphological network graph is calculated in the following way: ; in, This indicates the similarity between the initial morphological network graph and the predicted morphological network graph. This represents the similarity between the i-th cable in the initial morphological network graph and the i-th cable in the predicted morphological network graph;

[0052] in, The first morphological network in the initial network diagram represents the... The cable and the predicted morphology network diagram The average spatial distance between corresponding nodes in the cable, where each node includes multiple depth-oriented directional sensors and multiple seismic acquisition devices. The preset distance threshold;

[0053] in, Indicates the first The first of the cables Each node is either a depth-oriented or data acquisition device. The total number of nodes is equal to the sum of the depth orientation and data acquisition. , Represents nodes in the predictive morphology network graph Spatial coordinate data, , Represents nodes in the initial form of the network graph. Spatial coordinate data.

[0054] Step S500: When the similarity is not less than a preset threshold, the initial morphological network graph is used as the first morphological network graph at the current moment. For example, in some specific implementations, the initial morphological network graph is used as the first morphological network graph at the current moment. The initial morphological network graph is generated using a lightweight initial morphological generation model. Compared with the predicted morphology network diagram generated by the morphology prediction model If the similarity between the two is not less than a preset first threshold, it indicates that the initial morphological network graph is relatively accurate. Therefore, the initial morphological network graph can be used as the first morphological network graph at time t. The network diagram of the cable configuration at multiple subsequent time points can also serve as the first configuration network diagram at time t. Participate in data post-processing, etc.

[0055] Step S600: When the similarity is less than a preset threshold, reconstruct the first morphological network graph at the current moment based on the initial morphological network graph and the predicted morphological network graph. The first morphological network graph is used to assist in constructing morphological prediction data for subsequent time steps. It should be noted that in some embodiments, reconstructing the first morphological network graph at the current moment based on the initial morphological network graph and the predicted morphological network graph may include the following steps: acquiring real-time environmental data and real-time work vessel data at the current moment; inputting the initial morphological network graph, the predicted morphological network graph, the real-time environmental data, and the real-time work vessel data into the first morphological generation model, so that the first morphological generation model processes and outputs the first morphological network graph at the current moment; wherein, the first morphological generation model includes a fourth encoding network, a fifth encoding network, a sixth encoding network, a second feature fusion network, and a second decoding network; the first morphological generation model specifically performs the following operations: using the fourth encoding network to perform a fourth encoding on the real-time environmental data to obtain the first morphological network graph at the current moment. The system employs a fourth encoding vector; a fifth encoding network is used to encode the real-time operational vessel data, resulting in a fifth encoding vector; a sixth encoding network is used to encode the initial morphological network diagram and the predicted morphological network diagram, resulting in a sixth encoding vector; a second feature fusion network is used to perform a second fusion process on the fourth, fifth, and sixth encoding vectors, resulting in a second fusion vector; a second decoding network is used to perform a second decoding operation on the second fusion vector, resulting in first morphological data; wherein, the first morphological data includes the spatial distribution data of each depth-orientation device and each data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data; a first morphological network diagram is generated based on the spatial coordinate data corresponding to the first morphological data.

[0056] The specific logical principle of the sixth encoding can be implemented as follows: the initial morphological network graph and the predicted morphological network are encoded by two sets of parallel encoding sub-networks respectively, and then fused to obtain the sixth encoding vector.

[0057] For example, in some specific implementations, reconstructing the first morphological network graph based on the initial morphological network graph and the predicted morphological network graph specifically includes: converting the initial morphological network graph... Predictive morphological network diagram Environmental data at time t and working vessel data Input the first morphology generation model, and generate the first morphology network graph based on the first morphology generation model.

[0058] The first morphology generation model also adopts an encoder-decoder architecture, which includes a fourth encoder network, a fifth encoder network, a sixth encoder network, a second feature fusion network, and a second decoder network. The fourth encoder network is used to process the environmental data at time t. Encode to obtain the fourth encoded vector: ; The fifth encoding network is used for the operational vessel data at time t. Encode the vector to obtain the fifth encoded vector: ; The sixth coding network is used for the initial morphological network diagram. And predictive morphological network diagram Encode to obtain the sixth encoded vector: ; The second feature fusion network is used to fuse the fourth, fifth, and sixth encoding vectors to obtain the second fused vector: ; The second decoding network is used to decode the second fused vector to obtain the first morphological data: . The first morphological data consists of spatial distribution data of depth-oriented devices and data acquisition devices generated based on the first morphological model. This spatial distribution data includes the spatial coordinate data of multiple depth-oriented devices and multiple data acquisition devices. A first morphological network diagram can be constructed based on the first morphological data, specifically based on the first node sequence { } and the second node sequence { The corresponding spatial coordinate data are used to generate the first-form network graph. The fourth and fifth encoders use the same models as the first and second encoders, and can employ a domain dictionary model. This domain dictionary model is obtained by training the BERT pre-trained language model using domain data, specifically by training the BERT pre-trained language model using historical domain data. The sixth encoder uses a lightweight convolutional neural network model, a convolutional neural network model, or a graph convolutional neural network model. The decoder uses a generative model such as a bidirectional long short-term memory neural network.

[0059] When training the first morphology generation model, positive and negative sample data are constructed based on the similarity between the initial morphology network graph and the predicted morphology network graph. The initial morphology network graph and the predicted morphology network graph with a similarity greater than a preset second threshold are selected as inputs. The initial morphology data or the first predicted data are used as positive sample labels to obtain positive sample data. Noise is added to the initial morphology data or the first predicted data and used as negative sample labels to obtain negative sample data. The first morphology generation model is trained based on the positive and negative sample data, where the second threshold is greater than the first threshold.

[0060] This invention generates a first morphological network graph based on a first morphological generation model when the similarity between the initial morphological network graph and the predicted morphological network graph is less than a preset first threshold. The first morphological generation model uses the initial morphological network graph, the predicted morphological network graph, environmental data at time t, and work vessel data. Since the initial morphological network graph is constructed based on the attitude data at time t, and the predicted morphological network graph is constructed based on environmental data, work vessel data, and the first morphological network graph corresponding to multiple times prior to time t, the first morphological network graph generated based on the first morphological generation model considers not only the environmental data, work vessel data, and attitude data at the current time but also environmental data, work vessel data, and attitude data from multiple historical times. Therefore, the generated first morphological network graph is more accurate. The first morphological network graph generated based on the first morphological generation model can be used to predict the morphological network graph after time t and can also be used for subsequent data processing.

[0061] To explain in detail the principle of the technical solution of the present invention, the overall process of the present invention will be described below with reference to some specific embodiments. It is easy to understand that the following is an explanation of the technical principle of the present invention and should not be regarded as a limitation of the present invention.

[0062] To address the shortcomings of existing technologies, this invention proposes a method for constructing the morphology of a marine seismic acquisition cable, such as... Figure 3 As shown, the method includes the following steps: S1: Obtain the morphological prediction data at the current moment. The morphological prediction data includes environmental sequence data, work vessel sequence data, and the first morphological network sequence diagram. The environmental sequence data, work vessel sequence data, and the first morphological network sequence diagram respectively indicate the environmental data, work vessel data, and the first morphological network diagram at multiple historical moments before the current moment.

[0063] Obtain the morphological prediction data at the current time (time t). The morphological prediction data at time t includes the first morphological network diagrams of the K time points prior to time t, as well as the corresponding K environmental data and K operational vessel data. The first morphological network diagrams of the K time points constitute the first morphological network sequence diagram. ,in , This represents the first-order network diagram at the k-th time point before time t. K environmental data points and K operational vessel data points constitute the environmental sequence data. and work vessel sequence data The K environmental data include K ocean current velocity data and ocean current direction data, represented as: This represents the velocity of the k-th ocean current. This indicates the direction of the k-th ocean current; the K data points for the work vessels include the positions of the K work vessels. { },in This represents the bearing data of the kth working vessel; K is a preset value that can be obtained by taking into account both calculation efficiency and accuracy.

[0064] First-form network diagram The sequence includes the first node of at least one acquisition cable. } and the second node sequence { The first node sequence indicates the depth-oriented device on the acquisition cable. This represents the first-mode network diagram of the i-th acquisition cable. The first in A depth-oriented sensor, and a second node sequence indicator for the data acquisition device on the acquisition cable. This represents the first-mode network diagram of the i-th acquisition cable. The Middle A data acquisition device. The data acquisition device is a seismic data acquisition device, usually a seismic detector; the depth and orientation device is an acoustic bird and / or a compass bird.

[0065] S2: Generate a predictive morphology network diagram based on morphology prediction data.

[0066] Based on the morphological prediction data and the morphological prediction model, the first prediction data is obtained, and a prediction morphological network diagram is constructed based on the first prediction data. .like Figure 4 As shown, the morphological prediction model adopts an encoder-decoder architecture, which includes a first encoder network, a second encoder network, a third encoder network, a first feature fusion network, and a first decoder network, wherein: The first encoding network is used to encode the environmental sequence data to obtain the first encoding vector: ; The second encoding network is used to encode the vessel sequence data to obtain the second encoding vector: ; The third encoding network is used to encode the sequence diagram of the first morphological network to obtain the third encoding vector: ; The first feature fusion network is used to fuse the first encoding vector, the second encoding vector, and the third encoding vector to obtain the first fused vector: ; The first decoding network is used to decode the first fused vector to obtain the first predicted data: . The first prediction data consists of the spatial distribution data of the depth-oriented devices and data acquisition devices, including the spatial coordinate data of multiple depth-oriented devices and multiple data acquisition devices. A prediction morphology network diagram can be constructed based on the first prediction data, specifically based on the first node sequence { } and the second node sequence { Generate a predictive morphological network diagram based on the corresponding spatial coordinate data. .

[0067] This invention, when predicting the morphology of seismic acquisition cables, simultaneously considers environmental sequence data, operational vessel sequence data, and the first morphological network sequence diagram prior to time t. This allows for the acquisition of a relatively accurate predicted morphological network diagram, which is then displayed to the user before data acquisition at time t. The user can intuitively see the predicted morphological network diagram at time t, aiding in decision-making. The first and second encoders can employ domain dictionary models, which are trained using domain data on a BERT pre-trained language model, specifically historical domain data. The third encoder employs a lightweight convolutional neural network model, a convolutional neural network model, or a graph convolutional neural network model. The decoder employs a generative model such as a bidirectional long short-term memory neural network. Other neural network models or deep learning models can be used for the first, second, third, and decoder; no specific limitations are imposed here.

[0068] In another embodiment, before obtaining the first prediction data based on the morphological prediction data and the morphological prediction model, the method further includes: cutting each first morphological network graph in the first morphological network sequence graph into multiple sub-network graphs using a graph cutting model and a depth-oriented node sequence; classifying the multiple sub-network graphs into multiple first sub-network graph groups using a first grouping model and a first preset rule; and classifying the multiple sub-network graphs into multiple second sub-network graph groups using a second classification model and a second preset rule.

[0069] The graph cutting model divides the first-morphology network graph according to the depth-oriented directional node sequence. Divide into multiple sub-network graphs, that is, according to the first node sequence, the first-form network graph. Two adjacent depth-oriented nodes on the same acquisition cable i Segmented into a sub-network graph ,in, This represents the sub-network diagram formed by the j1st and j2nd depth-oriented devices in the i-th acquisition cable within the k-th first-morphology network diagram.

[0070] Assuming each first-morphology network diagram has I cables and J depth-directioners are deployed on each cable, then each first-morphology network diagram can be divided into I×(J-1) sub-graphs, and the first-morphology network sequence diagram can be divided into K×I×(J-1) sub-graphs.

[0071] The first grouping model classifies multiple sub-network graphs into multiple first sub-network graph groups according to a first preset rule. Specifically, the first grouping model classifies sub-network graphs divided into the same node index in the same acquisition cable within the first-form network graphs at different times into the same first sub-network graph group. In this way, I×(J-1) first sub-network graph groups can be obtained, and each first sub-network graph group is represented as follows: } The second grouping model, according to the second preset rule, classifies multiple sub-network graphs into multiple second sub-network graph groups. Specifically, it classifies sub-network graphs with the same node index in different acquisition cables within the first-mode network graph at the same time into the same second sub-network graph group. This yields K×(J-1) second sub-network graph groups, each represented as: }

[0072] The third coding network in the morphological prediction model includes multiple parallel first sub-coding networks, multiple parallel second sub-coding networks, a first fusion sub-network, a second fusion sub-network, and a third fusion sub-network, wherein: Multiple parallel first sub-encoding networks are used to encode multiple first sub-network graph groups in parallel, respectively, to obtain multiple first sub-encoding vectors: ; The number of first sub-encoding networks is the same as the number of first sub-network graphs, and the number of first sub-encoding vectors obtained is also the same as the number of first sub-network graphs, both being I×(J-1); the first sub-encoding network adopts the lightweight convolutional neural network model LightCNN.

[0073] The first fusion sub-network is used to fuse multiple first sub-encoded vectors to obtain a first fused sub-vector: ; Multiple parallel second sub-encoding networks are used to encode multiple second sub-network graph groups in parallel, respectively, to obtain multiple second sub-encoding vectors: ; The number of second sub-encoding networks is the same as the number of second sub-network graph groups, and the number of second sub-encoding vectors obtained is also the same as the number of second sub-network graph groups, both being K×(J-1); the second sub-encoding networks also use the lightweight convolutional neural network model LightCNN. The second fusion sub-network is used to fuse multiple second sub-encoding vectors to obtain a second fused sub-vector: ; The third fusion sub-network is used to fuse the first fusion sub-vector and the second fusion sub-vector to obtain the third encoding vector: ; In this embodiment, the first-morphology network subgraphs, segmented from K first-morphology network graphs, are classified according to both temporal and spatial dimensions to obtain multiple first-sub-network graph groups and multiple second-sub-network graph groups. Multiple parallel sub-encoding networks are then used to encode these first-sub-network graph groups and second-sub-network graph groups, thereby obtaining a third encoding vector. Because both temporal and spatial dimensions are integrated, prediction accuracy is further improved. The multiple parallel sub-encoding networks employ a lightweight convolutional neural network model, improving prediction efficiency.

[0074] S3: Obtain the attitude data at the current moment. The attitude data includes at least coordinate data, relative distance data, depth data, and orientation data.

[0075] During marine seismic exploration operations, multiple depth and orientation sensors (such as compass birds and acoustic birds) are typically deployed on the seismic acquisition cable to control its attitude. The compass birds and acoustic birds use their depth and orientation sensors to collect depth and orientation data. Simultaneously, a head marker is placed on the seismic acquisition cable, and a tail marker is placed at the stern. This network of the seismic vessel, head marker, multiple compass birds, multiple acoustic birds, and tail markers forms the basis for determining the cable's morphology.

[0076] Navigation and positioning for the acquisition cable typically employs Differential Global Positioning System (DGPS) and Relative GPS. DGPS and an gyrocompass are mounted on the work vessel, providing high-precision absolute position and bow azimuth in a geographic coordinate system, respectively. The DGPS position includes latitude, longitude, and geodetic height or elevation, with an accuracy typically in decimeters. Since the work vessel is assumed to be a rigid body, the relative relationships between equipment are fixed, allowing the calculation of coordinates for other points on the vessel. The underwater depth sensor is deployed on the cable and gun array, thus the gun array utilizes two elevations, unifying the elevation of the entire towed cable system. RGPS measures the distance and true azimuth between two points, transmitting the absolute position from the geophysical vessel to buoys (head / stern) located before and after the gun array and cable. Acoustic bird and compass bird sensors provide the acoustic distance between two points and the cable tangent magnetic azimuth, respectively, using edge and angle observations to transmit coordinate information along the cable.

[0077] Therefore, the attitude data Z collected in real time is:

[0078] In the formula, Let be the attitude data of the i-th acquisition cable at time t. The depth data of the i-th cable at time t includes at least the depth data measured by the depth sensor on the gun array at time t and the depth data measured by multiple depth-fixing orientation devices. The azimuth data of the i-th cable at time t includes at least the azimuth data of the working vessel at time t, the azimuth data measured by the head marker, the azimuth data measured by the stern marker, and the azimuth data measured by multiple depth-oriented and orientation-determining devices. The coordinate data includes at least the coordinate data of the working vessel, the coordinate data of the head marker, and the coordinate data of the stern marker. The data is relative distance data, including at least the distances between multiple adjacent acoustic birds.

[0079] S4: Construct an initial morphological network graph based on attitude data.

[0080] After obtaining the attitude data, an initial shape network diagram of the cable is generated based on the initial shape generation model. The initial morphology generation model can employ a polynomial line integral model. Since polynomial line integral models are lightweight models, they can quickly calculate and obtain the morphology of the seismic acquisition cable.

[0081] Generate the initial shape diagram of the cable based on the initial shape generation model. This includes: acquiring cable configuration data; generating initial shape data based on the cable configuration data, attitude data, and initial shape model; and generating an initial shape diagram based on the initial shape data. Since the offsets of the header, tail, and depth-oriented pointers on the cable are fixed, these offsets can be easily obtained from the cable configuration data. This is achieved by constructing the coordinates of any point on the cable. Initial shape data is obtained by fitting a polynomial function of the offset along the cable. This initial shape data includes spatial distribution data of multiple depth-oriented geophones and multiple data acquisition units, including spatial coordinate data. The data acquisition units are seismic detectors, whose positions within the cable are fixed; therefore, the offset of each seismic detector can also be obtained using the cable configuration data.

[0082] S5: Calculate the similarity between the initial morphological network graph and the predicted morphological network graph. If the similarity is not less than the preset first threshold, proceed to S6; otherwise, proceed to S7.

[0083] In one embodiment, the similarity between the initial morphological network graph and the predicted morphological network graph is calculated in the following manner: , in, This indicates the similarity between the initial morphological network graph and the predicted morphological network graph. This represents the similarity between the i-th cable in the initial morphological network graph and the i-th cable in the predicted morphological network graph;

[0084] in, This represents the average spatial distance between the i-th cable in the initial morphological network diagram and the corresponding nodes in the i-th cable in the predicted morphological network diagram. Each corresponding node includes multiple depth-oriented directional sensors and multiple seismic acquisition devices. The preset distance threshold;

[0085] Where m represents the m-th node in the i-th cable, which is either a depth-oriented device or a data acquisition device, and N is the total number of nodes, which is equal to the sum of the depth-oriented device and the data acquisition device. , This represents the spatial coordinate data of node m in the predictive morphology network graph. , This represents the spatial coordinates of node m in the initial network graph.

[0086] S6: Use the initial morphological network graph as the first morphological network graph at the current time step. The initial morphological network graph is generated using a lightweight initial morphological generation model. Compared with the predicted morphology network diagram generated by the morphology prediction model If the similarity between the two is not less than a preset first threshold, it indicates that the initial morphological network graph is relatively accurate. Therefore, the initial morphological network graph can be used as the first morphological network graph at time t. The network diagram of the cable configuration at multiple subsequent time points can also serve as the first configuration network diagram at time t. Participate in data post-processing, etc.

[0087] S7: Reconstruct the first morphological network graph at the current time step based on the initial morphological network graph and the predicted morphological network graph. If the similarity between the initial morphological network graph generated by the lightweight initial morphological generation model and the predicted morphological network graph generated by the morphological prediction model is less than a preset first threshold, it indicates that the initial morphological network graph is inaccurate and cannot be used as the first morphological network graph at time t for calculations at subsequent time steps, or for data post-processing. Therefore, it is necessary to reconstruct a more accurate first morphological network graph.

[0088] Reconstructing the first morphological network graph based on the initial morphological network graph and the predicted morphological network graph specifically includes: reconstructing the initial morphological network graph... Predictive morphological network diagram Environmental data at time t and working vessel data Input the first morphology generation model, and generate the first morphology network graph based on the first morphology generation model.

[0089] The first morphology generation model also adopts an encoder-decoder architecture, which includes a fourth encoder network, a fifth encoder network, a sixth encoder network, a second feature fusion network, and a second decoder network. The fourth encoder network is used to process the environmental data at time t. Encode to obtain the fourth encoded vector: ; The fifth encoding network is used for the operational vessel data at time t. Encode the vector to obtain the fifth encoded vector: ; The sixth coding network is used for the initial morphological network diagram. And predictive morphological network diagram Encode to obtain the sixth encoded vector: ; The second feature fusion network is used to fuse the fourth, fifth, and sixth encoding vectors to obtain the second fused vector: ; The second decoding network is used to decode the second fused vector to obtain the first morphological data: . The first morphological data consists of spatial distribution data of depth-oriented devices and data acquisition devices generated based on the first morphological model. This spatial distribution data includes the spatial coordinate data of multiple depth-oriented devices and multiple data acquisition devices. A first morphological network diagram can be constructed based on the first morphological data, specifically based on the first node sequence { } and the second node sequence { The corresponding spatial coordinate data are used to generate the first-form network graph. The fourth and fifth encoders use the same models as the first and second encoders, and can employ a domain dictionary model. This domain dictionary model is obtained by training the BERT pre-trained language model using domain data, specifically by training the BERT pre-trained language model using historical domain data. The sixth encoder uses a lightweight convolutional neural network model, a convolutional neural network model, or a graph convolutional neural network model. The decoder uses a generative model such as a bidirectional long short-term memory neural network.

[0090] When training the first morphology generation model, positive and negative sample data are constructed based on the similarity between the initial morphology network graph and the predicted morphology network graph. The initial morphology network graph and the predicted morphology network graph with a similarity greater than a preset second threshold are selected as inputs. The initial morphology data or the first predicted data are used as positive sample labels to obtain positive sample data. Noise is added to the initial morphology data or the first predicted data and used as negative sample labels to obtain negative sample data. The first morphology generation model is trained based on the positive and negative sample data, where the second threshold is greater than the first threshold.

[0091] This invention generates a first morphological network graph based on a first morphological generation model when the similarity between the initial morphological network graph and the predicted morphological network graph is less than a preset first threshold. The first morphological generation model uses the initial morphological network graph, the predicted morphological network graph, environmental data at time t, and work vessel data. Since the initial morphological network graph is constructed based on the attitude data at time t, and the predicted morphological network graph is constructed based on environmental data, work vessel data, and the first morphological network graph corresponding to multiple times prior to time t, the first morphological network graph generated based on the first morphological generation model considers not only the environmental data, work vessel data, and attitude data at the current time but also environmental data, work vessel data, and attitude data from multiple historical times. Therefore, the generated first morphological network graph is more accurate. The first morphological network graph generated based on the first morphological generation model can be used to predict the morphological network graph after time t and can also be used for subsequent data processing.

[0092] In summary, the morphology prediction and predicted morphology network graph in this invention have three functions: First, they are used to verify the accuracy of the initial morphology network graph. The initial morphology network graph is generated based on a lightweight model (i.e., a polynomial line integral model). This lightweight model only considers the data at the acquisition time and does not consider prior knowledge from previous times, so the accuracy of the initial morphology network graph needs to be verified. Existing technologies also do not verify the initial morphology network graph. This invention calculates the similarity between the initial morphology network graph and the predicted morphology network graph. When the similarity is greater than a preset threshold, it indicates that the initial morphology network graph is relatively accurate, and this initial morphology network graph is directly used as the first morphology network graph. When the similarity is less than the preset threshold, the first morphology network graph is reconstructed, thus improving both accuracy and efficiency. Second, they serve as input to the first morphology generation model to reconstruct the first morphology network graph. The first morphology network graph obtained in this way is more accurate because the initial morphology network graph considers knowledge from the acquisition time, and the morphology prediction network graph considers knowledge from multiple previous times. Third, they assist engineers in decision-making. Before the next data collection time, the predicted morphology network diagram is presented to the engineers, who can then have a visual understanding of the cable morphology and make informed decisions.

[0093] Compared with the prior art, the present invention has at least the following beneficial effects: (1) This invention first uses a lightweight polynomial curve integral model to generate an initial morphology network diagram, and uses a morphology prediction model to generate a predicted morphology network diagram. Based on the similarity between the initial morphology network diagram and the predicted morphology network diagram, it selects to use the initial morphology network diagram as the cable morphology network diagram at the current moment, or selects to reconstruct the cable morphology network diagram at the current moment based on the initial morphology network diagram and the predicted morphology network diagram. Using the predicted morphology network diagram allows users to intuitively see the morphology prediction network diagram at the next moment, helping users make decisions. At the same time, this invention uses the predicted morphology network diagram to verify the initial morphology network diagram. When the similarity is not less than a preset first threshold, the initial morphology network diagram is directly used as the morphology of the acquisition cable at the current moment. When the similarity is less than the preset first threshold, the first morphology generation model is used to reconstruct the morphology of the acquisition cable at the current moment. This takes into account both lightweightness and accuracy, and improves the efficiency and accuracy of cable morphology determination.

[0094] (2) In generating the predictive morphology network, this invention uses environmental sequence data, work vessel sequence data and first morphology network sequence graph before the current time. The first morphology network sequence graph is divided into multiple network subgraphs. The multiple network subgraphs are classified according to the two dimensions of time and space to obtain multiple first sub-network graph groups and multiple second sub-network graph groups. Multiple parallel first coding sub-networks and multiple parallel second coding sub-networks are used to encode the multiple first sub-network graph groups and multiple second sub-network graph groups to obtain the coding vector corresponding to the first morphology network sequence graph. Since the time dimension and the space dimension are integrated at the same time, the accuracy of the predictive morphology network graph is further improved.

[0095] (3) The first morphology generation model of the present invention adopts an initial morphology network diagram, a predicted morphology network diagram, environmental data and work vessel data at the current moment. Since the initial morphology network diagram is constructed based on the attitude data at the current moment, and the predicted morphology network diagram is constructed based on the environmental data, work vessel data and the first morphology network diagram corresponding to multiple moments before the current moment, the first morphology network diagram generated based on the first morphology generation model not only considers the environmental data, work vessel data and attitude data at the current moment, but also considers the environmental data, work vessel data and attitude data at multiple historical moments, so the generated first morphology network diagram is more accurate.

[0096] like Figure 5 As shown, this embodiment of the invention also provides a morphological construction device 900 for a marine seismic acquisition cable, which can implement the above-described method. This device may include: The first module 901 is used to acquire the shape prediction data and attitude data at the current moment; The second module 902 is used to generate a predicted morphology network diagram based on morphology prediction data and through a morphology prediction model. The third module 903 is used to generate an initial morphology network graph based on attitude data through an initial morphology generation model. The fourth module 904 is used to measure the similarity between the initial morphological network graph and the predicted morphological network graph to obtain the similarity score. The fifth module 905 is used to take the initial morphological network graph as the first morphological network graph at the current time when the similarity is not less than a preset threshold. The sixth module 906 is used to reconstruct the first morphological network graph at the current moment based on the initial morphological network graph and the predicted morphological network graph when the similarity is less than a preset threshold. The first morphological network graph is used to assist in constructing morphological prediction data for subsequent time steps.

[0097] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0098] This invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0099] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0100] like Figure 6 As shown, Figure 6 The hardware structure of an electronic device 1000 according to another embodiment is illustrated. The electronic device 1000 includes: The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (aSIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention. The memory 1002 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RaM). The memory 1002 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called and executed by the processor 1001. Input / output interface 1003 is used to implement information input and output; The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004); The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0101] The electronic device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0103] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0104] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0105] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0106] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0107] The present invention provides a method, apparatus, electronic device, storage medium, and program product for constructing the morphology of a marine seismic acquisition cable. This method acquires morphology prediction data and attitude data at the current moment; generates a predicted morphology network diagram using a morphology prediction model based on the morphology prediction data; generates an initial morphology network diagram using an initial morphology generation model based on the attitude data; performs similarity quantification on the initial morphology network diagram and the predicted morphology network diagram to obtain a similarity score; when the similarity score is not less than a preset threshold, the initial morphology network diagram is used as the first morphology network diagram at the current moment; when the similarity score is less than the preset threshold, the first morphology network diagram at the current moment is reconstructed based on the initial morphology network diagram and the predicted morphology network diagram; wherein the first morphology network diagram is used to assist in constructing morphology prediction data for subsequent moments. The present invention, by generating the predicted morphology network diagram and the initial morphology network diagram in parallel and introducing a similarity quantification comparison mechanism, can effectively verify the output results of the lightweight model, effectively solving the deficiency of traditional methods in assessing their own accuracy. Specifically, when the similarity score is insufficient, the morphology at the current moment is reconstructed using the predicted morphology network diagram and the initial morphology network diagram together, integrating historical prior knowledge while also considering the current real-time attitude, which can significantly improve the accuracy and reliability of cable morphology determination. Furthermore, the final output of the first morphological network graph is used to feed back the morphological prediction data at subsequent time steps to form a closed-loop iteration, which can continuously optimize the prediction effect and thus better meet the high-precision requirements of field collection operation guidance and data processing.

[0108] The embodiments described in this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.

[0109] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present invention, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0110] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0111] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0112] The preferred embodiments of the present invention have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and spirit of the present invention should be within the scope of the claims of the present invention.

Claims

1. A method for constructing the morphology of a marine seismic acquisition cable, characterized in that, The method includes the following steps: Obtain the current morphology prediction data and attitude data; Based on the morphology prediction data, a predicted morphology network diagram is generated using the morphology prediction model. Based on the posture data, an initial morphology network graph is generated using an initial morphology generation model; The similarity measure between the initial morphological network graph and the predicted morphological network graph is performed to obtain the similarity score. When the similarity is not less than a preset threshold, the initial morphological network graph is used as the first morphological network graph at the current moment. When the similarity is less than the preset threshold, the first morphological network graph at the current moment is reconstructed based on the initial morphological network graph and the predicted morphological network graph. The first morphological network graph is used to assist in constructing morphological prediction data for subsequent time periods.

2. The method according to claim 1, characterized in that, The first morphological network graph is stored in a historical database. Obtaining the current morphological prediction data and attitude data includes the following steps: Obtain the current morphological prediction data from the historical database; The morphological prediction data includes environmental sequence data, workboat sequence data, and a first morphological network sequence diagram. The environmental sequence data includes environmental data from multiple historical moments prior to the current moment. The workboat sequence data includes workboat data from multiple historical moments prior to the current moment. The first morphological network sequence diagram includes first morphological network diagrams from multiple historical moments prior to the current moment. The attitude data at the current moment is obtained by real-time sensor acquisition.

3. The method according to claim 2, characterized in that, The morphology prediction model includes a first encoding network, a second encoding network, a third encoding network, a first feature fusion network, and a first decoding network. The step of generating a predicted morphology network graph based on the morphology prediction data using the morphology prediction model includes the following steps: The environmental sequence data is first encoded using the first encoding network to obtain a first encoding vector; The second encoding network is used to perform a second encoding on the work vessel sequence data to obtain a second encoding vector; The first morphological network sequence diagram is encoded using the third encoding network to obtain a third encoding vector. The first feature fusion network is used to perform a first fusion process on the first encoding vector, the second encoding vector, and the third encoding vector to obtain a first fusion vector. The first decoding network performs a first decoding operation on the first fusion vector to obtain first prediction data; wherein, the first prediction data includes the spatial distribution data of each depth-oriented and data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data; The predicted morphological network diagram is generated based on the spatial coordinate data corresponding to the first predicted data.

4. The method according to claim 3, characterized in that, The third coding network includes multiple parallel first sub-coding networks and multiple parallel second sub-coding networks. The step of using the third coding network to perform third coding on the first morphological network sequence diagram to obtain a third coding vector includes the following steps: By using the graph cutting model and the depth-director node sequence, each of the first morphological network graphs in the first morphological network sequence graph is cut into multiple sub-network graphs; Using the first grouping model, the sub-network diagrams corresponding to the same node number in the same acquisition cable in the first morphology network diagram at different times are classified into the same first sub-network diagram group. Using the second grouping model, the sub-network diagrams with the same node number in different acquisition cables in the first morphology network diagram at the same time are classified into the same second sub-network diagram group. Each of the first sub-network graph groups is encoded by multiple parallel first sub-encoding networks, and then the encoding results of all the first sub-encoding networks are fused into a first fused sub-vector; wherein, the first sub-encoding network corresponds one-to-one with the first sub-network graph group; Each second sub-network graph group is encoded by multiple parallel second sub-encoding networks, and then the encoding results of all second sub-encoding networks are fused into a second fused sub-vector; wherein, the second sub-encoding network corresponds one-to-one with the second sub-network graph group; The first fusion sub-vector and the second fusion sub-vector are fused together to form the third encoding vector.

5. The method according to claim 1, characterized in that, The attitude data is acquired through real-time sensors and includes coordinate data, relative distance data, depth data, and orientation data. The initial shape generation model adopts a polynomial line integral model. The process of generating an initial shape network graph based on the attitude data using the initial shape generation model includes the following steps: Retrieve pre-stored cable configuration data; The cable configuration data includes header and tail labels, depth orientation device, and fixed offset of the data acquisition device on the acquisition cable. Based on the attitude data and the cable configuration data, a polynomial function between the spatial coordinates of any point on the acquisition cable and the offset along the acquisition cable is constructed using the polynomial curve integral model, and the initial shape data at the current moment is obtained by fitting. The initial morphological data includes the spatial distribution data of each depth-oriented device and each data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data. The initial morphological network diagram is generated based on the spatial coordinate data corresponding to the initial morphological data.

6. The method according to claim 1, characterized in that, Both the initial morphological network diagram and the predicted morphological network diagram include spatial coordinate data of each node on each acquisition cable. Each node includes a depth-oriented orienter and a data acquisition device. The similarity quantification of the initial morphological network diagram and the predicted morphological network diagram to obtain the similarity score includes the following steps: The first acquisition cable was selected as the candidate cable. The first node on the candidate cable is selected as the candidate node; First coordinate data of the candidate nodes are extracted from the initial morphological network graph, and second coordinate data of the candidate nodes are extracted from the predicted morphological network graph; Based on the first coordinate data and the second coordinate data, the spatial distances of the candidate nodes in the initial morphological network graph and the predicted morphological network graph are quantized. Take the next node on the candidate cable as the candidate node, and return to the step of extracting the first coordinate data of the candidate node from the initial network graph, until all nodes on the candidate cable have been traversed; The average spatial distance of the candidate cable is obtained by averaging the spatial distances corresponding to all nodes on the candidate cable. Based on the average spatial distance, the cable similarity corresponding to the candidate cable is obtained by combining the distance threshold transformation. The expression for the cable similarity is as follows: ； In the formula, Indicates the first Cable similarity of the acquisition cables This indicates the operation of retrieving the maximum value. Indicates the first Average spatial distance of the data acquisition cables Indicates the distance threshold; The next acquisition cable is selected as the candidate cable, and the process returns to the step of selecting the first node on the candidate cable as the candidate node, until all acquisition cables have been traversed. The minimum similarity among all collected cables is taken as the final similarity between the initial morphological network graph and the predicted morphological network graph.

7. The method according to claim 1, characterized in that, The process of reconstructing the first morphological network graph at the current moment based on the initial morphological network graph and the predicted morphological network graph includes the following steps: Obtain real-time environmental data and real-time operational vessel data at the current moment; The initial morphological network diagram, the predicted morphological network diagram, the real-time environmental data, and the real-time work vessel data are input into the first morphological generation model so that the first morphological generation model processes and outputs the first morphological network diagram at the current moment. The first morphology generation model includes a fourth encoding network, a fifth encoding network, a sixth encoding network, a second feature fusion network, and a second decoding network; the first morphology generation model specifically performs the following operations: The real-time environmental data is encoded using the fourth encoding network to obtain a fourth encoding vector; The real-time operation vessel data is encoded using the fifth encoding network to obtain a fifth encoding vector; The initial morphological network graph and the predicted morphological network graph are encoded using the sixth encoding network to obtain the sixth encoding vector; The second feature fusion network is used to perform a second fusion process on the fourth encoding vector, the fifth encoding vector, and the sixth encoding vector to obtain a second fused vector; The second decoding network performs a second decoding operation on the second fusion vector to obtain first morphological data; wherein, the first morphological data includes the spatial distribution data of each depth-oriented device and each data acquisition device on the acquisition cable, and the spatial distribution data includes spatial coordinate data; The first morphological network diagram is generated based on the spatial coordinate data corresponding to the first morphological data.

8. A morphological construction device for a marine seismic acquisition cable, characterized in that, The device includes: The first module is used to acquire the current morphology prediction data and attitude data; The second module is used to generate a predicted morphology network diagram based on the morphology prediction data and through a morphology prediction model. The third module is used to generate an initial morphology network graph based on the posture data using an initial morphology generation model. The fourth module is used to measure the similarity between the initial morphological network graph and the predicted morphological network graph to obtain the similarity score. The fifth module is used to use the initial morphological network graph as the first morphological network graph at the current moment when the similarity is not less than a preset threshold. The sixth module is used to reconstruct the first morphological network graph at the current moment based on the initial morphological network graph and the predicted morphological network graph when the similarity is less than the preset threshold. The first morphological network graph is used to assist in constructing morphological prediction data for subsequent time periods.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images