Methods, devices, servers, and storage media for detecting the burial depth of submarine cables.

Abstract

This application provides a method, apparatus, server, and storage medium for detecting the burial depth of submarine cables. The method includes: acquiring center frequency data of the optical fiber inside the submarine cable at the burial depth to be detected within a preset time period; calculating the temperature data of the optical fiber based on the center frequency data; labeling the corresponding temperature range based on the temperature data of the optical fiber to obtain labeled temperature data; inputting the labeled temperature data into a trained neural network model for processing to obtain a corresponding predicted burial depth label; determining the corresponding submarine cable burial depth value based on the predicted burial depth label; and outputting the submarine cable burial depth value, thereby making the detection of submarine cable burial depth more accurate.

Description

Technical Field

[0001] This application relates to the field of submarine cable technology, and in particular to a method, device, server and storage medium for detecting the burial depth of submarine cables. Background Technology

[0002] Submarine cable communication has become a primary technology for communication between islands due to its advantages of long transmission distance, low loss, and good confidentiality. To protect submarine cables from the natural environment, they are usually buried beneath the seabed. However, cables buried beneath the seabed are exposed to the seabed and subjected to friction damage when subjected to tides or ocean currents, leading to wear and even breakage. Therefore, measuring the burial depth of submarine cables is essential.

[0003] In existing technologies, traditional methods for detecting the burial depth of submarine cables mainly involve operators using marine magnetometers to detect the burial depth of submarine cables. By clearly identifying the coordinates of the submarine cable, the burial depth of the cable can be displayed.

[0004] However, in existing technologies, traditional methods for detecting the burial depth of submarine cables show clear signals from marine magnetometers when the water depth is less than 50 meters, allowing for a clear identification of the cable's coordinates. But when the water depth is greater than 50 meters, the signal from the marine magnetometer is not clear, which can easily lead to inaccurate measurements of the cable's burial depth. Summary of the Invention

[0005] This application provides a method, device, server, and storage medium for detecting the burial depth of submarine cables, in order to solve the problem that the signal of the marine magnetometer is not obvious when the water depth is greater than 50 meters, which easily leads to inaccurate measurement of the burial depth of submarine cables.

[0006] Firstly, this application provides a method for detecting the burial depth of submarine cables, applied to a server, including:

[0007] Acquire the center frequency data of the internal optical fiber of the submarine cable at the depth to be tested within a preset time period;

[0008] Calculate the temperature data of the optical fiber based on the center frequency data;

[0009] The temperature range is marked according to the temperature data of the optical fiber to obtain the marked temperature data;

[0010] The labeled temperature data is input into a trained neural network model for processing to obtain the corresponding predicted burial depth label.

[0011] The corresponding submarine cable burial depth value is determined based on the predicted burial depth label;

[0012] Output the burial depth value of the submarine cable.

[0013] In one possible design, the training process of the trained neural network model includes: acquiring temperature data of the submarine cable optical fiber within a preset time period from a preset sample library; controlling a sonar device to acquire submarine cable burial depth data within the preset time period; labeling the temperature data of the optical fiber in different temperature ranges to obtain labeled temperature data; matching the corresponding submarine cable burial depth data according to the labeled temperature data to obtain a sample dataset, and determining a portion of the data in the sample dataset as the training set; creating a neural network model and setting initial weight combination values.

[0014] The training set is input into the neural network model for iterative looping. Each time an actual weight combination value is output, a preset feedback network processes the actual weight combination value and a preset expected weight to correct the actual weight combination value until the final actual weight combination value is output. It is then determined whether the difference between the final actual weight combination value and the expected weight is less than a preset error. If the difference is less than the preset error, the predicted embedding depth label is output to complete the training of the neural network model. If the difference is not less than the preset error, the neural network model is modified to obtain a modified neural network model, and the training set is input into the modified neural network model for iterative looping again until the training of the neural network model is completed.

[0015] In one possible design, the method further includes: determining a portion of the data in the sample dataset as a test set, and determining the true depth label corresponding to the test set; inputting the test set into a trained neural network model for prediction processing to obtain the corresponding predicted depth label; determining the evaluation value corresponding to the predicted depth label based on the true depth label; and determining that the trained neural network model test is complete if the evaluation value meets a preset accuracy rate.

[0016] In one possible design, the different temperature ranges include abnormal temperature ranges and normal temperature ranges. Correspondingly, the step of annotating the temperature data of the optical fiber in different temperature ranges to obtain annotated temperature data includes: annotating the temperature data of the optical fiber in the abnormal temperature range to obtain first-annotated temperature data; and annotating the temperature data of the optical fiber in each normal temperature range to obtain second-annotated temperature data. Correspondingly, the step of matching the corresponding submarine cable burial depth data based on the annotated temperature data to obtain a sample dataset includes: matching the corresponding submarine cable burial depth data based on the first-annotated temperature data and the second-annotated temperature data to obtain a sample dataset.

[0017] In one possible design, the weight combination value is a weight and a threshold; correspondingly, the calculation formula for processing the actual weight combination value and the preset expected weight through a preset feedback network to correct the actual weight combination value is as follows:

[0018]

[0019] In the formula, These are the corrected actual weights; It is a preset expected weight; The learning rate is set to 0.01; X is the gradient of the loss function with respect to the weights in the preset feedback network.

[0020] In one possible design, the evaluation value is the accuracy rate; correspondingly, the calculation formula for determining the evaluation value corresponding to the predicted burial depth label based on the actual burial depth label is:

[0021]

[0022] In the formula, To predict the accuracy of the burial depth label; TP is the number of correctly predicted positive examples within the predicted burial depth label based on the actual burial depth label; TN is the number of correctly predicted negative examples within the predicted burial depth label based on the actual burial depth label; FP is the number of incorrectly predicted positive examples within the predicted burial depth label based on the actual burial depth label; FN is the number of incorrectly predicted negative examples within the predicted burial depth label based on the actual burial depth label.

[0023] In one possible design, the formula for calculating the temperature data of the optical fiber based on the center frequency data is as follows:

[0024] ,in

[0025] In the formula, The temperature change corresponding to the temperature data of the optical fiber; For varying wavelengths; The initial wavelength; is the thermo-optic coefficient; c is the speed of light; The center frequency is denoted as .

[0026] Secondly, this application provides a device for detecting the burial depth of submarine cables, applied to a server, comprising:

[0027] The first acquisition module is used to acquire the center frequency data of the internal optical fiber of the submarine cable at the burial depth to be detected within a preset time period.

[0028] The calculation module is used to calculate the temperature data of the optical fiber based on the center frequency data;

[0029] The first labeling module is used to label the corresponding temperature range based on the temperature data of the optical fiber, so as to obtain the labeled temperature data.

[0030] The first processing module is used to input the labeled temperature data into a trained neural network model for processing in order to obtain the corresponding predicted burial depth label.

[0031] The first determining module is used to determine the corresponding submarine cable burial depth value based on the predicted burial depth label;

[0032] The first output module is used to output the burial depth value of the submarine cable.

[0033] Thirdly, this application provides a server, including: at least one processor and a memory;

[0034] The memory stores computer-executed instructions;

[0035] The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method for detecting the burial depth of submarine cables as described in the first aspect and various possible designs of the first aspect.

[0036] Fourthly, this application provides a computer storage medium storing computer execution instructions, which, when executed by a processor, implement the method for detecting the burial depth of submarine cables as described in the first aspect and various possible designs of the first aspect.

[0037] Fifthly, embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, implements the method for detecting the burial depth of submarine cables as described in the first aspect and various possible designs of the first aspect.

[0038] The method, apparatus, server, and storage medium for detecting the burial depth of submarine cables provided in this application acquire center frequency data of the optical fiber inside the submarine cable at the burial depth to be detected within a preset time period; calculate the temperature data of the optical fiber based on the center frequency data; label the corresponding temperature range based on the temperature data of the optical fiber to obtain labeled temperature data; input the labeled temperature data into a trained neural network model for processing to obtain a corresponding predicted burial depth label; determine the corresponding burial depth value of the submarine cable based on the predicted burial depth label; and output the burial depth value of the submarine cable, making the detection of the burial depth of submarine cables more accurate. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 A schematic diagram illustrating an application scenario of the submarine cable burial depth detection method provided in this application embodiment;

[0041] Figure 2 Schematic diagram of the method for detecting the burial depth of submarine cables provided in this application embodiment Figure 1 ;

[0042] Figure 3 Schematic diagram of the method for detecting the burial depth of submarine cables provided in this application embodiment Figure 2 ;

[0043] Figure 4 A schematic diagram of the structure of the submarine cable burial depth detection device provided in the embodiments of this application;

[0044] Figure 5 This is a schematic diagram of the hardware structure of the server provided in an embodiment of this application. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] Submarine cable communication has become a primary technology for communication between continents, between continents and islands, and between islands due to its advantages of long transmission distance, large capacity, low loss, high reliability, and good anti-interference and confidentiality. To protect the cables from natural environmental and human factors, they are usually buried beneath the seabed. However, due to the complex underwater environment, over time, sections of the cable originally buried beneath the seabed become exposed. When subjected to tides or ocean currents, the exposed cable suffers repeated friction damage, or is suspended and subjected to periodic swinging or twisting due to ocean currents. This long-term friction damages the optical fibers inside the cable, leading to loss or even breakage. Therefore, measuring the burial depth of submarine cables is essential. In existing technologies, traditional methods for detecting submarine cable burial depth mainly involve operators using marine magnetometers to determine the cable's burial depth by clearly identifying its coordinates. However, in existing technologies, traditional methods for detecting the burial depth of submarine cables show clear signals from marine magnetometers when the water depth is less than 50 meters, allowing for a clear identification of the cable's coordinates. But when the water depth is greater than 50 meters, the signal from the marine magnetometer is not clear, which can easily lead to inaccurate measurements of the cable's burial depth.

[0047] To address the aforementioned technical problems, this application proposes the following technical concept: Considering the inaccuracy of existing marine magnetometers in measuring the burial depth of submarine cables, the inventors label temperature ranges based on the temperature data of the optical fibers inside the submarine cable, obtain labeled temperature data, and process the labeled temperature data using a trained neural network model to obtain corresponding predicted burial depth labels. The submarine cable burial depth is then determined and output based on the predicted burial depth labels, making the detection of submarine cable burial depth more accurate.

[0048] Figure 1 This is a schematic diagram illustrating an application scenario of the submarine cable burial depth detection method provided in this application embodiment.

[0049] like Figure 1 As shown, the scenario includes a display terminal 101 and a server 102.

[0050] The display terminal 101 can be a display screen or a personal computer or other terminal.

[0051] Server 102 can be a standalone server or a cluster of multiple servers.

[0052] Server 102 retrieves the center frequency data of the internal optical fiber of the submarine cable to be tested within a preset time period from the database. Based on the center frequency data, it calculates the temperature data of the optical fiber and labels the temperature range. The labeled temperature data is then input into a trained neural network model for processing to obtain the corresponding predicted burial depth label. The corresponding submarine cable burial depth value is determined based on the predicted burial depth label and then output to the display terminal 101 via a wireless network for display. A detailed embodiment is described below.

[0053] Figure 2 Schematic diagram of the method for detecting the burial depth of submarine cables provided in this application embodiment Figure 1 The execution entity in this embodiment can be Figure 1 The server in the illustrated embodiment is not specifically limited in this embodiment. Figure 2 As shown, the method includes:

[0054] S201: Obtain the center frequency data of the internal optical fiber of the submarine cable at the depth to be tested within a preset time period.

[0055] In this embodiment, the preset time can be any time within one month, six months, or one year, or it can be any other time.

[0056] In this embodiment, the center frequency data will change accordingly as the burial depth of the submarine cable varies at different locations.

[0057] S202: Calculate the temperature data of the optical fiber based on the center frequency data.

[0058] In this embodiment, the formula for calculating the temperature data of the optical fiber based on the center frequency data is as follows:

[0059] ,in

[0060] In the formula, The temperature change corresponding to the temperature data of the optical fiber; For varying wavelengths; The initial wavelength; is the thermo-optic coefficient; c is the speed of light; The center frequency.

[0061] S203: Mark the corresponding temperature range based on the temperature data of the optical fiber to obtain the marked temperature data.

[0062] In this embodiment, the temperature range can be one or more temperature spans of 2°C, 4°C, or 6°C, or other temperature spans.

[0063] For example, the temperature ranges of 2°C and 4°C are marked according to the temperature data of the optical fiber, so that the marked temperature data are used as the first temperature data and the second temperature data.

[0064] S204: Input the labeled temperature data into the trained neural network model for processing to obtain the corresponding predicted burial depth label.

[0065] In this embodiment, the trained neural network model can be trained using an AI model or a database.

[0066] In this embodiment, the predicted burial depth label is a mapping relationship between temperature and burial depth.

[0067] For example, the predicted burial depth label is the submarine cable burial depth corresponding to the first temperature data and the submarine cable burial depth corresponding to the second temperature data.

[0068] S205: Determine the corresponding submarine cable burial depth value based on the predicted burial depth label.

[0069] For example, the burial depth of the submarine cable at the location corresponding to the first temperature data is 1m; the burial depth of the submarine cable at the location corresponding to the second temperature data is 2m.

[0070] S206: Output the burial depth of the submarine cable.

[0071] For example, the submarine cable burial depth value at the location corresponding to the first temperature data is output as 1m on the display interface, and the submarine cable burial depth value at the location corresponding to the second temperature data is output as 2m.

[0072] In summary, the submarine cable burial depth detection method provided in this embodiment acquires the center frequency data of the optical fiber inside the submarine cable at the burial depth to be detected within a preset time period; calculates the temperature data of the optical fiber based on the center frequency data; labels the corresponding temperature ranges based on the temperature data of the optical fiber to obtain labeled temperature data; inputs the labeled temperature data into a trained neural network model for processing to obtain the corresponding predicted burial depth label; determines the corresponding submarine cable burial depth value based on the predicted burial depth label; and outputs the submarine cable burial depth value, making the detection of submarine cable burial depth more accurate.

[0073] In addition, this embodiment directly collects the center frequency data of the optical fiber inside the submarine cable for subsequent temperature and burial depth analysis, making it simpler to obtain optical fiber temperature data.

[0074] Figure 3 Schematic diagram of the method for detecting the burial depth of submarine cables provided in this application embodiment Figure 2 In the embodiments of this application, in Figure 2 Based on the provided embodiments, the specific training process of the trained neural network model in S204 is described in detail. For example... Figure 3As shown, the method includes:

[0075] S301: Obtain temperature data of submarine cable optical fiber within a preset time period from the preset sample library.

[0076] In this embodiment, the preset sample library can be a MySQL database or other databases.

[0077] In this embodiment, the preset time can be any time within one month, six months, or one year, or it can be any other time.

[0078] S302: Control the sonar equipment to acquire data on the burial depth of the submarine cable within a preset time period.

[0079] In this embodiment, the sonar device can be a side-scan sonar device or other sonar devices.

[0080] S303: Label the temperature data of optical fibers in different temperature ranges to obtain the labeled temperature data.

[0081] In this embodiment, the different temperature ranges include abnormal temperature ranges and various normal temperature ranges; correspondingly, step S303 specifically includes:

[0082] S3031: The temperature data of the optical fiber in the abnormal temperature range is labeled accordingly to obtain the temperature data after the first labeling.

[0083] In this embodiment, the abnormal temperature range is the range in which the temperature difference changes from small to large and then from large to small within a preset time period.

[0084] For example, the temperature data of the optical fiber in the abnormal temperature range is marked with 0, so that the temperature data after the first marking is the temperature data marked with 0.

[0085] S3032: The temperature data of the optical fiber in each normal temperature range is labeled accordingly to obtain the temperature data after the second labeling.

[0086] For example, the normal temperature ranges can be [0, 2), [2, 4] and (4, +∞).

[0087] For example, the temperature data of [0, 2), [2, 4] and (4, +∞) are labeled accordingly to obtain the temperature data after the second labeling, which are temperature data with label 1, temperature data with label 2 and temperature data with label 3 respectively.

[0088] S304: Match the corresponding submarine cable burial depth data with the temperature data after each annotation to obtain a sample dataset, and determine a portion of the data in the sample dataset as the training set.

[0089] In this embodiment, a portion of the data in the sample dataset can be any one of 70%, 80%, or 90%, or it can be other data.

[0090] Specifically, step S304 involves matching the corresponding submarine cable burial depth data with the temperature data after the first and second annotations to obtain a sample dataset.

[0091] For example, temperature data marked with 0 corresponds to cable burial depth data where the cable is above the seabed; temperature data marked with 1 corresponds to cable burial depth data where the cable is 1m below the seabed on average; temperature data marked with 2 corresponds to cable burial depth data where the cable is 2m below the seabed on average; and temperature data marked with 3 corresponds to cable burial depth data where the cable is 3m below the seabed on average.

[0092] S305: Create a neural network model and set the initial weight combination values.

[0093] S306: Input the training set into the neural network model for iterative loop, and each time the actual weight combination value is output, process the actual weight combination value and the preset expected weight through a preset feedback network to correct the actual weight combination value until the final actual weight combination value is output.

[0094] In this embodiment, the weight combination value is a weight and a threshold; correspondingly, the actual weight combination value and the preset expected weight are processed by a preset feedback network to correct the actual weight combination value, and the calculation formula is as follows:

[0095]

[0096] In the formula, These are the corrected actual weights; It is a preset expected weight; The learning rate is set to 0.01; X is the gradient of the loss function with respect to the weights in the preset feedback network.

[0097] S307: Determine whether the difference between the final actual weight combination value and the expected weight is less than the preset error.

[0098] In this embodiment, the preset error can be any one of 5%, 10%, or 15%, or other error values.

[0099] In addition, the preset error can also be a preset threshold. When the model's loss function (such as mean squared error) is lower than a preset threshold, training stops; or, when the final calculated loss function is higher than a preset threshold and the model has difficulty converging, a maximum number of iterations is set. After reaching the maximum number of iterations, the model training ends and the label of the unknown data is predicted.

[0100] S308: If the difference between the final actual weight combination value and the expected weight is less than the preset error, then output the predicted embedding depth label of the training to complete the training of the neural network model.

[0101] S309: If the difference between the final actual weight combination value and the expected weight is not less than the preset error, then modify the neural network model to obtain the modified neural network model, and input the training set into the modified neural network model for re-iteration loop until the training of the neural network model is completed.

[0102] Furthermore, the testing process for a trained neural network model specifically includes steps a~d:

[0103] Step a: Select a portion of the data from the sample dataset as the test set, and determine the actual burial depth label corresponding to the test set.

[0104] In this embodiment, a portion of the data in the sample dataset can be any of the remaining 30%, 20%, or 10%, or other corresponding remaining data.

[0105] Step b: Input the test set into the trained neural network model for prediction processing to obtain the corresponding predicted embedment depth label.

[0106] Step c: Determine the evaluation value corresponding to the predicted burial depth label based on the actual burial depth label.

[0107] In this embodiment, the evaluation value is the accuracy rate; correspondingly, the calculation formula for determining the evaluation value corresponding to the predicted burial depth label based on the actual burial depth label is as follows:

[0108]

[0109] In the formula, The accuracy of the predicted burial depth label is denoted as follows: TP is the number of correctly predicted positive examples within the predicted burial depth label based on the actual burial depth label; TN is the number of correctly predicted negative examples within the predicted burial depth label based on the actual burial depth label; FP is the number of incorrectly predicted positive examples within the predicted burial depth label based on the actual burial depth label; and FN is the number of incorrectly predicted negative examples within the predicted burial depth label based on the actual burial depth label.

[0110] In addition, the evaluation value can also be recall rate, F1 score, or other calculated values.

[0111] Step d: If the evaluation value meets the preset accuracy, then the trained neural network model test is complete.

[0112] In this embodiment, the preset accuracy rate can be any one of 70%, 80%, or 90%, or other accuracy rates.

[0113] In summary, the submarine cable burial depth detection method provided in this embodiment obtains the temperature data of submarine cable optical fibers within a preset time period from a preset sample library; controls a sonar device to obtain submarine cable burial depth data within the preset time period; labels the temperature data of optical fibers in different temperature ranges accordingly, and matches the corresponding submarine cable burial depth data based on the labeled temperature data, determining a portion of the data in the sample dataset as the training set; inputs the training set into a pre-created neural network model for iterative loops, and processes the actual weight combination value and the preset expected weight through a preset feedback network to correct the actual weight combination value; determines whether the difference between the final actual weight combination value and the expected weight is less than a preset error; if so, outputs the trained predicted burial depth label; if not, modifies the neural network model to obtain a modified neural network model, and inputs the training set into the modified neural network model for iterative loops again until the training of the neural network model is completed, which can make the subsequent detection of submarine cable burial depth more accurate.

[0114] In addition, this embodiment achieves online detection of submarine cable burial depth and real-time online monitoring of multiple channels and links by training the corresponding neural network model, while also reducing the detection cost of submarine cable burial depth.

[0115] Figure 4 This is a schematic diagram of the structure of the submarine cable burial depth detection device provided in an embodiment of this application. Figure 4 As shown, the device for detecting the burial depth of the submarine cable includes: a first acquisition module 401, a calculation module 402, a first annotation module 403, a first processing module 404, a first determination module 405, and a first output module 406.

[0116] The first acquisition module 401 is used to acquire the center frequency data of the internal optical fiber of the submarine cable at the burial depth to be detected within a preset time period.

[0117] Calculation module 402 is used to calculate the temperature data of the optical fiber based on the center frequency data;

[0118] The first annotation module 403 is used to annotate the corresponding temperature range according to the temperature data of the optical fiber to obtain the annotated temperature data.

[0119] The first processing module 404 is used to input the labeled temperature data into a trained neural network model for processing in order to obtain the corresponding predicted burial depth label.

[0120] The first determining module 405 is used to determine the corresponding submarine cable burial depth value based on the predicted burial depth label.

[0121] The first output module 406 is used to output the burial depth value of the submarine cable.

[0122] In one possible implementation, the device further includes:

[0123] The second acquisition module 407 is used to acquire temperature data of the submarine cable optical fiber within a preset time period from a preset sample library;

[0124] The third acquisition module 408 is used to control the sonar equipment to acquire the submarine cable burial depth data within the preset time period;

[0125] The second annotation module 409 is used to annotate the temperature data of the optical fiber in different temperature ranges to obtain the annotated temperature data.

[0126] The matching module 410 is used to match the corresponding submarine cable burial depth data according to the temperature data after each annotation to obtain a sample dataset, and to determine a portion of the data in the sample dataset as a training set.

[0127] Create module 411 to create a neural network model and set the initial weight combination values;

[0128] The first loop module 412 is used to input the training set into the neural network model for iterative looping, and to process the actual weight combination value and the preset expected weight through a preset feedback network each time the actual weight combination value is output, so as to correct the actual weight combination value until the final actual weight combination value is output.

[0129] The judgment module 413 is used to determine whether the difference between the final actual weight combination value and the expected weight is less than a preset error;

[0130] The second output module 414 is used to output the predicted embedding depth label of the training if the difference between the final actual weight combination value and the expected weight is less than a preset error, so as to complete the training of the neural network model.

[0131] The second loop module 415 is used to modify the neural network model to obtain a modified neural network model if the difference between the final actual weight combination value and the expected weight is not less than a preset error, and input the training set into the modified neural network model for re-iteration loop until the training of the neural network model is completed.

[0132] In one possible implementation, the device further includes:

[0133] The second determining module 416 is used to determine a portion of the data in the sample dataset as a test set, and to determine the true burial depth label corresponding to the test set;

[0134] The second processing module 417 is used to input the test set into the trained neural network model for prediction processing to obtain the corresponding predicted burial depth label.

[0135] The third determining module 418 is used to determine the evaluation value corresponding to the predicted burial depth label based on the actual burial depth label.

[0136] The fourth determining module 419 is used to determine that the trained neural network model test is complete if the evaluation value meets the preset accuracy rate.

[0137] In one possible implementation, the different temperature ranges include abnormal temperature ranges and various normal temperature ranges; correspondingly, the second labeling module 409 specifically includes:

[0138] The first annotation unit 4091 is used to annotate the temperature data of the optical fiber in the abnormal temperature range to obtain the temperature data after the first annotation.

[0139] The second annotation unit 4092 is used to annotate the temperature data of the optical fiber in each normal temperature range to obtain the temperature data after the second annotation.

[0140] Accordingly, the matching module 410 is specifically used to: match the corresponding submarine cable burial depth data according to the first labeled temperature data and the second labeled temperature data to obtain a sample dataset.

[0141] In one possible implementation, the weight combination value is a weight and a threshold; correspondingly, the calculation formula for processing the actual weight combination value and the preset expected weight through a preset feedback network to correct the actual weight combination value is as follows:

[0142]

[0143] In the formula, These are the corrected actual weights; It is a preset expected weight; The learning rate is set to 0.01; X is the gradient of the loss function with respect to the weights in the preset feedback network.

[0144] In one possible implementation, the evaluation value is the accuracy rate; correspondingly, the calculation formula for determining the evaluation value corresponding to the predicted burial depth label based on the actual burial depth label is:

[0145]

[0146] In the formula, To predict the accuracy of the burial depth label; TP is the number of correctly predicted positive examples within the predicted burial depth label based on the actual burial depth label; TN is the number of correctly predicted negative examples within the predicted burial depth label based on the actual burial depth label; FP is the number of incorrectly predicted positive examples within the predicted burial depth label based on the actual burial depth label; FN is the number of incorrectly predicted negative examples within the predicted burial depth label based on the actual burial depth label.

[0147] In one possible implementation, the formula for calculating the temperature data of the optical fiber based on the center frequency data is as follows:

[0148] ,in

[0149] In the formula, The temperature change corresponding to the temperature data of the optical fiber; For varying wavelengths; The initial wavelength; is the thermo-optic coefficient; c is the speed of light; The center frequency is denoted as .

[0150] The apparatus provided in this embodiment can be used to execute the technical solutions of the above method embodiments. Its implementation principle and technical effects are similar, and will not be described again here.

[0151] Figure 5 This is a schematic diagram of the hardware structure of the server provided in an embodiment of this application. Figure 5 As shown, the server in this embodiment includes a processor 501 and a memory 502; the memory stores computer execution instructions; at least one processor executes the computer execution instructions stored in the memory, causing at least one processor to execute the above-described method for detecting the burial depth of submarine cables.

[0152] Alternatively, the memory 502 can be either standalone or integrated with the processor 501.

[0153] When the memory 502 is set up independently, the server also includes a bus 503 for connecting the memory 502 and the processor 501.

[0154] This application embodiment also provides a computer storage medium storing computer execution instructions. When the processor executes the computer execution instructions, the method for detecting the burial depth of submarine cables as described above is implemented.

[0155] This application also provides a computer program product, including a computer program, which, when executed by a processor, implements the method for detecting the burial depth of submarine cables as described above.

[0156] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0157] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; 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 implement the solution of this embodiment according to actual needs.

[0158] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The unit composed of the above modules can be implemented in hardware or in the form of hardware plus software functional units.

[0159] The integrated modules implemented as software functional modules described above can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application.

[0160] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0161] The memory may include high-speed RAM, and may also include non-volatile storage (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0162] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0163] The aforementioned storage media can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage media can be any available medium accessible to general-purpose or special-purpose computers.

[0164] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Alternatively, the storage medium can be an integral part of the processor. Both the processor and the storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and storage medium can exist as discrete components in an electronic device or host device.

[0165] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for detecting the burial depth of submarine cables, characterized in that, Applied to servers, including: Acquire the center frequency data of the internal optical fiber of the submarine cable at the depth to be tested within a preset time period; Calculate the temperature data of the optical fiber based on the center frequency data; The temperature range is marked according to the temperature data of the optical fiber to obtain the marked temperature data; The labeled temperature data is input into a trained neural network model for processing to obtain the corresponding predicted burial depth label. The neural network model is based on temperature data of submarine cable optical fibers within a preset time period obtained from a preset sample library, combined with submarine cable burial depth data of the same period obtained by sonar equipment. The model is formed by labeling and matching optical fiber temperature data in different temperature ranges. The model is created and an initial weight combination value is set. The training set is input into the model for iterative loop. Each time the actual weight combination value is output, it is corrected by a preset feedback network until the difference between the final actual weight combination value and the expected weight is less than the preset error, and the accuracy is verified by the test set to meet the preset requirements. The different temperature ranges include abnormal temperature ranges and normal temperature ranges. The corresponding submarine cable burial depth value is determined based on the predicted burial depth label; Output the burial depth value of the submarine cable; The formula for calculating the temperature data of the optical fiber based on the center frequency data is as follows: ,in In the formula, The temperature change corresponding to the temperature data of the optical fiber; For varying wavelengths; The initial wavelength; is the thermo-optic coefficient; c is the speed of light; The center frequency is denoted as .

2. The method according to claim 1, characterized in that, The training process of the trained neural network model includes: Temperature data of the submarine cable optical fiber within a preset time period are obtained from a preset sample library; Control the sonar equipment to acquire data on the burial depth of the submarine cable within the preset time period; The temperature data of the optical fiber in different temperature ranges are labeled accordingly to obtain the labeled temperature data; The temperature data after each annotation is matched with the corresponding submarine cable burial depth data to obtain a sample dataset, and a portion of the data in the sample dataset is determined as the training set. Create a neural network model and set the initial weight combination values; The training set is input into the neural network model for iterative looping. Each time the actual weight combination value is output, the actual weight combination value and the preset expected weight are processed by a preset feedback network to correct the actual weight combination value until the final actual weight combination value is output. Determine whether the difference between the final actual weight combination value and the expected weight is less than a preset error; If the difference between the final actual weight combination value and the expected weight is less than the preset error, the predicted embedding depth label of the training is output to complete the training of the neural network model. If the difference between the final actual weight combination value and the expected weight is not less than a preset error, the neural network model is modified to obtain a modified neural network model, and the training set is input into the modified neural network model for re-iteration loop until the training of the neural network model is completed.

3. The method according to claim 2, characterized in that, Also includes: A portion of the data in the sample dataset is selected as the test set, and the actual burial depth label corresponding to the test set is determined. The test set is input into the trained neural network model for prediction processing to obtain the corresponding predicted burial depth label; The evaluation value corresponding to the predicted burial depth label is determined based on the actual burial depth label; If the evaluation value meets the preset accuracy rate, then the trained neural network model is considered to have completed the test.

4. The method according to claim 2, characterized in that, The step of labeling the temperature data of the optical fiber in different temperature ranges to obtain labeled temperature data includes: The temperature data of the optical fiber in the abnormal temperature range is labeled accordingly to obtain the temperature data after the first labeling; The temperature data of the optical fiber in each normal temperature range are labeled accordingly to obtain the second labeled temperature data; Accordingly, the step of matching the corresponding submarine cable burial depth data with the labeled temperature data to obtain a sample dataset includes: Based on the temperature data after the first annotation and the temperature data after the second annotation, the corresponding submarine cable burial depth data are matched to obtain a sample dataset.

5. The method according to claim 2, characterized in that, The weight combination value is a weight and a threshold. Accordingly, the calculation formula for processing the actual weight combination value and the preset expected weight through a preset feedback network to correct the actual weight combination value is as follows: In the formula, These are the corrected actual weights; It is a preset expected weight; The learning rate is set to 0.01; X is the gradient of the loss function with respect to the weights in the preset feedback network.

6. The method according to claim 3, characterized in that, The evaluation value mentioned above is the accuracy rate; Accordingly, the calculation formula for determining the evaluation value corresponding to the predicted burial depth label based on the actual burial depth label is as follows: In the formula, To predict the accuracy of buried depth labels; TP represents the number of correctly predicted positive examples within the predicted depth label determined based on the actual burial depth label; TN represents the number of correctly predicted negative examples within the predicted burial depth label determined based on the actual burial depth label; FP represents the number of incorrectly predicted positive examples within the predicted burial depth label determined based on the actual burial depth label; and FN represents the number of incorrectly predicted negative examples within the predicted burial depth label determined based on the actual burial depth label.

7. A device for detecting the burial depth of submarine cables, characterized in that, Applied to servers, including: The first acquisition module is used to acquire the center frequency data of the internal optical fiber of the submarine cable at the burial depth to be detected within a preset time period. The calculation module is used to calculate the temperature data of the optical fiber based on the center frequency data; the calculation formula for calculating the temperature data of the optical fiber based on the center frequency data is: ,in In the formula, The temperature change corresponding to the temperature data of the optical fiber; For varying wavelengths; The initial wavelength; is the thermo-optic coefficient; c is the speed of light; The center frequency; The first labeling module is used to label the corresponding temperature range based on the temperature data of the optical fiber, so as to obtain the labeled temperature data. The first processing module is used to input the labeled temperature data into a trained neural network model for processing to obtain the corresponding predicted burial depth label. The neural network model is a sample dataset formed by matching and labeling fiber temperature data in different temperature ranges with temperature data of submarine cables within a preset time period obtained from a preset sample library and submarine cable burial depth data of the same period obtained by sonar equipment. The model is created and an initial weight combination value is set. The training set is input into the model for iterative loop. Each time the actual weight combination value is output, it is corrected by a preset feedback network until the difference between the final actual weight combination value and the expected weight is less than the preset error, and the accuracy is verified by the test set to meet the preset requirements. The different temperature ranges include abnormal temperature ranges and normal temperature ranges. The first determining module is used to determine the corresponding submarine cable burial depth value based on the predicted burial depth label; The first output module is used to output the burial depth value of the submarine cable.

8. A server, characterized in that, include: At least one processor and memory; The memory stores computer-executed instructions; The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method for detecting the burial depth of submarine cables as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer execution instructions, and when the processor executes the computer execution instructions, it implements the method for detecting the burial depth of submarine cables as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for detecting the burial depth of submarine cables as described in any one of claims 1 to 6.

Images