A method for dynamic positioning and electronic chart marking of a submarine cable

Abstract

The application relates to the technical field of electronic sea charts, and discloses a submarine cable dynamic positioning and electronic sea chart marking method, which comprises the following steps: S1. A mobile platform carrying a controllable sound source is used as an excitation end, a path navigation is carried out based on a historical route planning of a submarine cable, and a narrow pulse laser signal connection is established between a DAS system and an associated optical fiber of the submarine cable; S2. The excitation end continuously emits a characteristic sound wave signal matched with a response characteristic of the DAS system, vibration strain signals along the submarine cable are collected, and the vibration strain signals are time-synchronized with real-time position information of the controllable sound source; S3. The vibration strain signals are pretreated to obtain effective signals; S4. Spatial coordinates of the submarine cable along the line are solved in combination with an ocean environment sound velocity parameter, the effective signals and the real-time position information; S5. The spatial coordinates are converted into a coordinate system suitable for an electronic sea chart, electronic sea chart elements corresponding to the submarine cable are generated, and dynamic updating is completed, so that submarine cable position information can be mapped to the electronic sea chart in real time and accurately.

Description

Technical Field

[0001] This application relates to the technical field of electronic nautical charts, and in particular to a method for dynamic positioning of submarine cables and annotation of electronic nautical charts. Background Technology

[0002] Submarine cables (hereinafter referred to as "submarine cables") are core infrastructure for transoceanic energy transmission and global communication. They are characterized by high construction costs and high maintenance difficulties. The accuracy of their spatial location is directly related to the safety of navigation and the cable's own operation. In the complex marine environment, submarine cables are susceptible to the combined effects of multiple factors such as seawater erosion, siltation, ocean current disturbances, and seabed geological activities. This can lead to dynamic positional changes such as route deviations, suspension, and changes in burial depth. If these dynamic changes cannot be captured in a timely and accurate manner, they can easily cause ships to accidentally touch the cable during anchoring or navigation, resulting in major safety accidents such as energy transmission interruptions and communication network paralysis. At the same time, they also bring great difficulties and challenges to the operation and maintenance of submarine cables.

[0003] Currently, the marking and monitoring of submarine cable locations mainly rely on two methods: traditional electronic charts and manual detection. Both static marking and manual detection technologies have significant and fundamental flaws.

[0004] Traditional electronic nautical charts mostly use static coordinate input during the engineering construction phase to mark submarine cables. Their update mechanism relies on periodic manual inspections and feedback, which results in long update cycles and significant lag, making it impossible to reflect the dynamic position changes of submarine cables in real time. At the same time, the accuracy of static marking is limited by the initial measurement methods, making it difficult to meet the technical requirements of modern shipping for millimeter-level accuracy of submarine cable positions.

[0005] Manual detection typically involves ships equipped with sonar equipment conducting patrols and surveys. This method not only faces high operational risks due to wind, waves, and complex sea conditions, but also suffers from low detection efficiency and limited coverage. In particular, in deep-sea areas, the cost and technical difficulty of manual detection increase exponentially, making it impossible to meet the routine needs of dynamic monitoring of submarine cables. Summary of the Invention

[0006] In order to provide an electronic nautical chart annotation method that can adapt to the dynamic characteristics of submarine cables and has the advantages of high precision and automation, this application provides a submarine cable dynamic positioning and electronic nautical chart annotation method.

[0007] A method for dynamic positioning of submarine cables and electronic nautical chart marking includes the following steps: S1. A mobile platform equipped with a controllable sound source is used as the excitation end. The detection path is planned based on the historical route of the submarine cable. A narrow pulse laser signal connection is established between the DAS system and the associated optical fiber of the submarine cable. The associated optical fiber is the spare optical fiber or communication optical fiber of the submarine cable. S2. The excitation end continuously emits a characteristic acoustic wave signal that matches the response characteristics of the DAS system. After collecting the vibration strain signal along the submarine cable, the vibration strain signal is synchronized with the real-time position information of the controllable sound source. The emission frequency of the characteristic acoustic wave signal is dynamically adjusted according to the characteristics of the vibration strain signal fed back by the DAS system, so that the emission frequency is matched with the sensitive frequency band of the DAS system. S3. Preprocess the vibration strain signal to obtain an effective signal; S4. Combining the marine environmental sound velocity parameters, the effective signal, and the real-time location information, calculate the spatial coordinates along the submarine cable; S5. Convert the spatial coordinates into a coordinate system adapted to the electronic nautical chart, generate the electronic nautical chart elements corresponding to the submarine cable, and complete the dynamic update.

[0008] Optionally, the time synchronization error between the vibration strain signal and the real-time position information in step S2 does not exceed a preset error threshold, and the characteristic acoustic wave signal is processed by encoding modulation.

[0009] Optionally, in step S3, the preprocessing includes: S31. Noise Reduction: The original strain signal is processed using the JDAS self-supervised deep learning model; S32. Data Compression: Compression processing is performed on the noise-reduced signal using compressed sensing (CS) technology combined with signal envelope extraction. S33. Boundary Interference Shielding: Set a signal threshold for DAS sensor data within a specified range at both ends of the submarine cable. When the signal strength is lower than the threshold, it is determined to be end-face reflection interference and is shielded and removed.

[0010] Optionally, in step S4, marine environmental sound velocity parameters are obtained through real-time on-site acquisition, and a layered sound velocity profile model is constructed based on the marine environmental sound velocity parameters to correct the sound wave propagation path. The submarine cable plane coordinates were calculated by combining time-reversal focusing with the ESPRIT super-resolution algorithm. Calculation of submarine cable elevation based on signal arrival time difference and layered sound velocity profile model; The positioning error is corrected by a dual mechanism of multipath error model compensation and filtering algorithm smoothing, and the three-dimensional coordinates of the submarine cable are output.

[0011] Optionally, the layered sound velocity profile model is constructed in the following way: based on historical sound velocity data and real-time measurement data of the sea area where the submarine cable is deployed, the seawater layer is divided into a sound velocity layer with equal gradient every 10m in depth, and the seabed sedimentary layer is divided into silt layer, sand layer and bedrock layer according to lithology. Each layer is assigned a fixed sound velocity value, and a dynamically updated layered sound velocity profile model is constructed.

[0012] Optionally, in step S5, the dynamic update includes incremental updates and predictive updates; Incremental updates only update the submarine cable segments that have a preset offset from the historical coordinates; non-offset segments retain the original labeled data. Predictive updates predict and label future location trajectories based on the changing patterns of submarine cable coordinates.

[0013] Optionally, in step S5, the coordinate transformation adopts the seven-parameter Bursa model to convert the submarine cable coordinates under the WGS-84 geodetic coordinate system to the Mercator coordinate system or Gauss-Kruger coordinate system used in electronic charts. The electronic chart elements also include attribute information, which includes at least the data source, update time, location uncertainty, and monitoring period.

[0014] Optionally, the triggering mechanism of the method includes two modes: periodic triggering and event triggering; The periodic triggering automatically executes the positioning and labeling process according to a preset cycle; Event triggering occurs when an abnormal event is detected, and the monitoring system automatically initiates location and label updates.

[0015] Optionally, the training steps of the JDAS self-supervised deep learning model include: A1. Dataset Construction: DAS signal data were collected under different marine environments and divided into clean signal samples and noise samples. The sample duration was 10s and the sampling frequency was 30kHz. A2. Model initialization: Set the input layer dimension to 300000×1, the feature extraction layer uses a 3-layer convolutional neural network (CNN) with kernel sizes of 3×1, 5×1 and 7×1, and the activation function is ReLU; A3. Training process: Using noise samples as input, the model generates simulated noise signals, which are then superimposed with clean signal samples to construct training sample pairs; with the objective function of minimizing signal reconstruction error, the Adam optimizer is used, the learning rate is set to 0.001, and the number of iterations is 500 to complete model training.

[0016] In summary, this application includes at least one of the following beneficial technical effects: This application overcomes the limitations of traditional submarine cable positioning, which relies on static labeling and manual detection, by constructing a collaborative positioning link of "active sound source excitation - DAS passive sensing." Utilizing the fiber optic resources inherent in the submarine cable itself as the sensing medium, it eliminates the need for additional dedicated sensors, significantly reducing monitoring costs and implementation difficulty. The distributed sensing capability of the DAS system, combined with the directional excitation characteristics of the active sound source, enables continuous and dynamic monitoring of the submarine cable route. Coupled with a hierarchical sound velocity model and super-resolution fusion algorithm, positioning accuracy is improved to the centimeter level, effectively addressing the technical bottleneck of insufficient accuracy in traditional methods. The data processing stage employs a collaborative scheme of the JDAS self-supervised learning model and compressed sensing technology, efficiently eliminating ocean noise while reducing data transmission and processing pressure, ensuring low latency in the positioning process. The combination of incremental electronic chart updates and dynamic predictive labeling not only avoids redundant calculations caused by full-segment updates and improves the response speed of navigation equipment, but also provides forward-looking obstacle avoidance references for ship navigation planning, filling the gap in the foresight of dynamic labeling in existing technologies. The application of multipath error correction model and Kalman filter algorithm further improves the stability and reliability of positioning results, enabling submarine cable location information to be mapped to electronic charts in real time and accurately. This effectively resolves the core contradiction of the disconnect between traditional static labeling and dynamic submarine cable routing, providing strong technical support for shipping safety and submarine cable operation and maintenance. The application of unmanned tugboats also reduces the risks of operations in complex sea areas and enhances the normalization of monitoring capabilities. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the dynamic positioning and electronic nautical chart marking method for submarine cables in this application. Detailed Implementation

[0018] The embodiments of this application are described in detail below, and examples of the embodiments are shown in the accompanying drawings.

[0019] In the description of this specification, the references to "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples" refer to specific features, structures, materials, or characteristics described in connection with the described embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0020] This embodiment uses the dynamic positioning and electronic nautical chart annotation of a transoceanic communication submarine cable as an application scenario to describe in detail the implementation process of the present invention. The submarine cable is approximately 120km long and is deployed in waters with a depth of 50m-200m. It contains 4 communication optical fibers and 2 spare optical fibers. The method of the present invention is used to perform dynamic positioning and annotation on the cable. A method for dynamic positioning of submarine cables and marking on electronic nautical charts is disclosed. See [link to relevant documentation]. Figure 1 This includes the following steps: S1. A mobile platform equipped with a controllable sound source is used as the excitation end. The detection path is planned based on the historical route of the submarine cable. A narrow pulse laser signal connection is established between the DAS system and the associated optical fiber of the submarine cable. The associated optical fiber is the spare optical fiber or communication optical fiber of the submarine cable. S2. The excitation end continuously emits a characteristic acoustic wave signal that matches the response characteristics of the DAS system. After collecting the vibration strain signal along the submarine cable, the vibration strain signal is synchronized with the real-time position information of the controllable sound source. S3. Preprocess the vibration strain signal to obtain an effective signal; S4. Combining the marine environmental sound velocity parameters, the effective signal, and the real-time location information, calculate the spatial coordinates along the submarine cable; S5. Convert the spatial coordinates into a coordinate system adapted to the electronic nautical chart, generate the electronic nautical chart elements corresponding to the submarine cable, and complete the dynamic update.

[0021] The specific steps in this embodiment are as follows: Excitation end deployment and system connection: A 500-ton autonomous unmanned tugboat is selected as the mobile platform (excitation end) and equipped with a 500W directional controllable sound source; based on the historical route data of the submarine cable, the detection path is planned on the electronic nautical chart. The detection path extends 100m on each side of the historical route of the submarine cable, and the distance between adjacent paths is set to 150m (i.e., 1.5 times the estimated route width of the submarine cable of 100m), with a total detection path length of 150km.

[0022] Meanwhile, at the submarine cable shore-based relay station, the optical transmitter and receiver of the DAS system are connected to the spare optical fiber of the submarine cable to complete the narrow pulse laser signal connection. The laser emission parameters of the DAS system are configured as follows: laser type is narrow pulse laser, pulse width is 10ns, repetition frequency is 10kHz, and output power is 100mW.

[0023] The unmanned tugboat traveled along the detection path at a speed of 10 knots (approximately 5.14 m / s). The initial transmission frequency of the controllable sound source was set to 1 kHz, which was the initial value of the sensitive frequency band of the DAS system determined in previous tests. During the voyage, the DAS system provided real-time feedback on the vibration strain signal characteristics, and the controllable sound source dynamically adjusted its transmission frequency according to these characteristics to ensure that it always matched the sensitive frequency band of the DAS system. The characteristic sound wave signal was encoded and modulated using a 1024-bit pseudo-random code and continuously transmitted into the sea area. The DAS system collected vibration strain signals along the submarine cable at a sampling frequency of 30 kHz. At the same time, the high-precision GNSS positioning system (positioning accuracy ±1 cm) carried by the unmanned tugboat collected real-time location information (latitude and longitude coordinates, heading, and speed) at a sampling frequency of 10 Hz. Both types of data were synchronously transmitted to the shore-based data processing center through a satellite communication link. A timestamp matching algorithm was used to achieve data synchronization, and the synchronization error was controlled within 0.8 ms, which is lower than the preset error threshold of 1 ms.

[0024] Preprocessing steps for the acquired DAS strain signals: (1) Noise reduction: The pre-trained and converged JDAS self-supervised deep learning model was called. The input of the model was a 10-second strain signal (containing 300,000 data points) at a sampling frequency of 30kHz. The model extracted signal features through a 3-layer CNN network of feature extraction layer. The noise recognition layer identified wave noise (frequency range 0.1Hz-10Hz) and marine biological activity noise (frequency range 2kHz-5kHz) based on the noise feature library formed by training. The signal reconstruction layer output the denoised pure strain signal. Specifically, JDAS (j-invariance Distributed Acoustic Sensing denoising model) is a self-supervised deep learning denoising model designed for distributed acoustic sensing (DAS) data. Its core principle is based on a spatiotemporal coherence separation mechanism: spatiotemporally coherent effective signals can be recovered through interpolation of adjacent channel data, while spatiotemporally incoherent noise (such as ocean waves, marine biological activity, etc.) cannot be interpolated, thus achieving accurate separation of noise and signal. This model can be trained without relying on clean labeled data, making it suitable for DAS signal processing needs in complex environments such as the ocean. In this embodiment, the training steps of the JDAS self-supervised deep learning model include: A1. Dataset Construction: DAS signal data were collected under different marine environments (wind and waves, calm, biological activity), and divided into clean signal samples (including active sound source signals) and noise samples. The sample duration was 10s and the sampling frequency was 30kHz. A2. Model initialization: Set the input layer dimension to 300000×1 (corresponding to 10s×30kHz data), the feature extraction layer uses a 3-layer convolutional neural network (CNN) with kernel sizes of 3×1, 5×1, and 7×1, and the activation function is ReLU; A3. Training process: Using noise samples as input, the model generates simulated noise signals, which are then superimposed with clean signal samples to construct training sample pairs; with the objective function of minimizing signal reconstruction error, the Adam optimizer is used, the learning rate is set to 0.001, and the number of iterations is 500 to complete model training.

[0025] (2) Data compression: Envelope extraction is performed on the denoised signal to retain the signal peak and phase characteristics. Compressed sensing technology is used to compress the signal at a compression ratio of 8:1. The Gaussian random matrix is ​​used as the measurement matrix in the compression process. Subsequently, the orthogonal matching pursuit (OMP) algorithm is used to achieve high-precision reconstruction of the signal with a reconstruction accuracy of ≥98%. (3) Boundary interference shielding: Set the signal strength threshold to 0.1mV, perform threshold detection on DAS data within 500m of both ends of the submarine cable, determine the signals below the threshold as end face reflection interference and shield and remove them, and finally retain the effective strain data of the middle 119km.

[0026] After data preprocessing, the DAS strain signal is associated with the tugboat's GNSS trajectory data through a timestamp matching algorithm to ensure that each strain signal corresponds accurately with the tugboat's real-time position and navigation status.

[0027] By collecting real-time sound velocity data at different depths in the sea area using a sound velocity meter mounted on an unmanned tugboat, and combining this data with historical sound velocity data from the past six months, a layered sound velocity profile model was constructed.

[0028] The seawater layer was divided into equal-gradient sound velocity layers every 10m in depth: 1500m / s for water depths of 50m-60m, 1505m / s for water depths of 60m-100m, and 1510m / s for water depths of 100m-200m. The seabed sedimentary layer was divided into silt layer (sound velocity 1600m / s), sand layer (sound velocity 1800m / s), and bedrock layer (sound velocity 2500m / s) according to lithology. The model was updated every hour to ensure it matched the actual marine environment.

[0029] Planar coordinate calculation: The preprocessed effective signal is input into the GPU parallel computing unit (using NVIDIA A100 GPU), and the time-reversal focusing and ESPRIT super-resolution fusion algorithm is run. Each focusing unit is assigned an independent computing core, and the focusing interval is set to 1m. The signal is propagated backward to the detection plane through the time-reversal algorithm, and the multi-source interference signal is separated by the ESPRIT algorithm to extract the planar coordinates (x,y) corresponding to the signal energy peak.

[0030] Elevation calculation: Extract the TDOA parameters of the time-reversal focusing peak and combine them with v1 (seawater layer sound velocity) and v2 (sediment layer sound velocity) in the layered sound velocity model.

[0031] The burial depth or suspension height of the submarine cable is calculated using the formula z=(v1×v2×Δt) / (v2 - v1); Δt is the time difference between the sound wave reaching the same sensing point through the seawater layer and the sediment layer, in seconds.

[0032] For example, at a certain monitoring point, v1=1505m / s, v2=1600m / s, Δt=0.002s, substituting into the formula, we get z=(1505×1600×0.002) / (1600-1505)=4816 / 95≈50.7m, that is, the burial depth of the submarine cable at this point is about 50.7m.

[0033] The hemispherical multipath model (MHM) constructed based on historical data of the sea area is invoked to compensate for multipath errors in the calculated coordinates. The compensated data is then input into the Kalman filter algorithm for smoothing optimization. The filtered state equation is X(k)=A×X(k-1)+B×u(k)+w(k), and the observation equation is Z(k)=H×X(k)+v(k), where X(k) is the state vector at time k (containing x, y, z coordinates). Random errors are eliminated through filtering, and a standardized three-dimensional coordinate set is output.

[0034] The Kalman filter algorithm is used to smooth the calculated 3D coordinates. The filtered state equation is X(k) = A × X(k-1) + B × u(k) + w(k), and the observation equation is Z(k) = H × X(k) + v(k), where X(k) is the state vector at time k (containing x, y, z coordinates), A is the state transition matrix, B is the control matrix, u(k) is the control variable, w(k) is the process noise, Z(k) is the observation vector, H is the observation matrix, and v(k) is the observation noise. The filtered result is used to fit and generate segmented routing curves for the submarine cable, outputting a standardized 3D coordinate set with a coordinate point interval of 1m.

[0035] The automatic annotation process during the generation of electronic nautical chart elements corresponding to submarine cables: (1) Coordinate transformation: The three-dimensional coordinates of the submarine cable under the WGS-84 geodetic coordinate system are transformed into the Mercator coordinate system used in the electronic nautical chart of this area using the seven-parameter Bursa model. The transformation parameters are obtained by calibration of the control points in this sea area, and the transformation accuracy is ≤1cm. (2) Dynamic update execution: A combination of incremental updates and predictive updates is used. ① Incremental update: Compare the offset between the current three-dimensional coordinates and the historical coordinates of the electronic chart. Update the three submarine cable segments (total length of about 8km) with an offset greater than 0.5m to the obstacle layer according to the S101 electronic chart standard format. The remaining 112km non-offset segments retain the original labels. The update takes 28s. ② Predictive Update: Extract the positioning coordinates of the last three times (12 hours apart), calculate the rate of change of coordinates for each segment, and use a cubic polynomial fitting algorithm (fitting formula y=ax³+bx²+cx+d, where a, b, c, and d are fitting coefficients) to predict the offset trajectory of each segment of the submarine cable in the next 24 hours. Generate a "prediction range" layer on the electronic chart. High confidence areas (≥95%) are marked with solid green boxes, medium confidence areas (80%-95%) are marked with dashed yellow boxes, and low confidence areas (<80%) are marked with dotted red boxes. (3) Attribute assignment and triggering mechanism: Automatically add attribute information to the updated submarine cable electronic chart elements, including "Data source: DAS dynamic monitoring", "Update time: 10:30 on December 16, 2025", "Location uncertainty: ±4cm" and "Monitoring period: 8:00-10:00 on December 16, 2025"; In this embodiment, the triggering mechanism is set to trigger periodically every 24 hours. At the same time, it is connected to the marine environmental monitoring system. When a strong storm with wind force ≥ level 12 or an earthquake with magnitude ≥ level 4.0 is detected, the event triggering process is automatically started, and the positioning and labeling are updated in real time.

[0036] This embodiment adopts a combination of periodic triggering and event triggering. The periodic triggering cycle is set to 24 hours, and the positioning and labeling are automatically updated once a day. At the same time, it is connected to the marine environmental monitoring system of the sea area. When a strong storm with wind force ≥ level 12 or an earthquake with magnitude ≥ 4.0 is detected, the event triggering process is automatically started, and the positioning and labeling are updated immediately.

[0037] The mobile platform described in this invention is not limited to unmanned tugboats, but can also be a manned monitoring vessel or an autonomous underwater vehicle (AUV). The core requirement is to meet the technical requirements of carrying a controllable sound source and achieving high-precision positioning. The optical fiber accessed by the DAS system is not limited to spare optical fiber. Idle channels of communication optical fibers can be used without affecting the submarine cable communication function. The standard formats of electronic charts include S101, S57, etc. The coordinate transformation model can be adapted and adjusted according to the specific coordinate system of the electronic chart. All the above-mentioned technical modifications are within the protection scope of this invention.

[0038] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for dynamic positioning of submarine cables and electronic nautical chart marking, characterized in that, Includes the following steps: S1. A mobile platform equipped with a controllable sound source is used as the excitation end. The detection path is planned based on the historical route of the submarine cable. A narrow pulse laser signal connection is established between the DAS system and the associated optical fiber of the submarine cable. The associated optical fiber is the spare optical fiber or communication optical fiber of the submarine cable. S2. The excitation end continuously emits a characteristic acoustic wave signal that matches the response characteristics of the DAS system. After collecting the vibration strain signal along the submarine cable, the vibration strain signal is synchronized with the real-time position information of the controllable sound source. The emission frequency of the characteristic acoustic wave signal is dynamically adjusted according to the characteristics of the vibration strain signal fed back by the DAS system, so that the emission frequency is matched with the sensitive frequency band of the DAS system. S3. Preprocess the vibration strain signal to obtain an effective signal; S4. Combining the marine environmental sound velocity parameters, the effective signal, and the real-time location information, calculate the spatial coordinates along the submarine cable; S5. Convert the spatial coordinates into a coordinate system adapted to the electronic nautical chart, generate the electronic nautical chart elements corresponding to the submarine cable, and complete the dynamic update.

2. The method for dynamic positioning of submarine cables and electronic nautical chart annotation according to claim 1, characterized in that, In step S2, the time synchronization error between the vibration strain signal and the real-time position information does not exceed a preset error threshold, and the characteristic acoustic wave signal is processed by encoding modulation.

3. The method for dynamic positioning of submarine cables and electronic nautical chart marking according to claim 1, characterized in that, In step S3, the preprocessing includes: S31. Noise Reduction: The original strain signal is processed using the JDAS self-supervised deep learning model; S32. Data Compression: Compression processing is performed on the noise-reduced signal using compressed sensing (CS) technology combined with signal envelope extraction. S33. Boundary Interference Shielding: Set a signal threshold for DAS sensor data within a specified range at both ends of the submarine cable. When the signal strength is lower than the threshold, it is determined to be end-face reflection interference and is shielded and eliminated.

4. The method for dynamic positioning of submarine cables and electronic nautical chart annotation according to claim 1, characterized in that, In step S4, marine environmental sound velocity parameters are obtained through real-time on-site data acquisition, and a layered sound velocity profile model is constructed based on the marine environmental sound velocity parameters to correct the sound wave propagation path. The submarine cable plane coordinates were calculated by combining time-reversal focusing with the ESPRIT super-resolution algorithm. Calculation of submarine cable elevation based on signal arrival time difference and layered sound velocity profile model; The positioning error is corrected by a dual mechanism of multipath error model compensation and filtering algorithm smoothing, and the three-dimensional coordinates of the submarine cable are output.

5. The method for dynamic positioning of submarine cables and electronic nautical chart annotation according to claim 2, characterized in that, The layered sound velocity profile model is constructed in the following way: based on historical sound velocity data and real-time measurement data of the sea area where the submarine cable is deployed, the seawater layer is divided into a sound velocity layer with equal gradient every 10m in depth, and the seabed sedimentary layer is divided into silt layer, sand layer and bedrock layer according to lithology. Each layer is assigned a fixed sound velocity value, and a dynamically updated layered sound velocity profile model is constructed.

6. The method for dynamic positioning of submarine cables and electronic nautical chart annotation according to claim 1, characterized in that, In step S5, the dynamic update includes incremental update and predictive update; Incremental updates only update the submarine cable segments that have a preset offset from the historical coordinates; non-offset segments retain the original labeled data. Predictive updates predict and label future location trajectories based on the changing patterns of submarine cable coordinates.

7. The method for dynamic positioning of submarine cables and electronic nautical chart annotation according to claim 1, characterized in that, In step S5, the coordinate transformation adopts the seven-parameter Bursa model to convert the submarine cable coordinates under the WGS-84 geodetic coordinate system into the Mercator coordinate system or the Gauss-Kruger coordinate system used in electronic charts. The electronic chart elements also include attribute information, which includes at least the data source, update time, location uncertainty, and monitoring period.

8. The method for dynamic positioning of submarine cables and electronic nautical chart marking according to claim 1, characterized in that, The triggering mechanism of the method includes two modes: periodic triggering and event triggering; The periodic triggering automatically executes the positioning and labeling process according to a preset cycle; Event triggering occurs when an abnormal event is detected, and the monitoring system automatically initiates location and label updates.

9. The method for dynamic positioning of submarine cables and electronic nautical chart annotation according to claim 3, characterized in that, The training steps of the JDAS self-supervised deep learning model include: A1. Dataset Construction: DAS signal data were collected under different marine environments and divided into clean signal samples and noise samples. The sample duration was 10s and the sampling frequency was 30kHz. A2. Model initialization: Set the input layer dimension to 300000×1, the feature extraction layer uses a 3-layer convolutional neural network (CNN) with kernel sizes of 3×1, 5×1 and 7×1, and the activation function is ReLU; A3. Training process: Using noise samples as input, the model generates simulated noise signals, which are then superimposed with clean signal samples to construct training sample pairs; with the objective function of minimizing signal reconstruction error, the Adam optimizer is used, the learning rate is set to 0.001, and the number of iterations is 500 to complete model training.

Images