Submarine cable state monitoring method and device, computer equipment and storage medium

Abstract

The invention relates to a submarine cable state monitoring method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring disturbance data acquired from a plurality of observation points of a submarine cable; processing the disturbance data according to a preset characteristic index to obtain a disturbance characteristic sequence; inputting the disturbance characteristic sequence into a deep learning model to generate target disturbance sequence data; and when the relationship between the target disturbance sequence data of the observation point and the threshold satisfies an alarm condition, determining that the state of the submarine cable is an abnormal state, and generating alarm information. By adopting the method, the interference of external factors such as sea wave scouring and sensor noise on the state judgment of the submarine cable can be avoided, and the accuracy of the state of the submarine cable is improved, so that the accuracy of alarm information is improved.

Description

[0001] This application is a divisional application filed in accordance with Application No. 202111261121.8 (Title of Invention: Submarine Cable Status Monitoring Method, Device, Computer Equipment and Storage Medium, Application Date: October 28, 2021). Technical Field

[0002] This application relates to the field of engineering applications of statistical learning, and particularly to a submarine cable status monitoring method, device, computer equipment and storage medium. Background Art

[0003] The ocean is a treasure trove of resources that is relatively easy for humans to explore. Currently, the construction of various offshore and submarine engineering projects is progressing in an orderly manner. Submarine cables (i.e., undersea cables) are an important part of these marine engineering constructions. Once damaged, they can easily lead to power and communication outages, causing a huge impact on engineering construction. Therefore, it is necessary to monitor the status of submarine cables in a timely manner and issue real-time alarms when abnormal cable conditions occur. However, compared to land, the marine environment is more complex, and the difficulty and danger of manual monitoring are much higher.

[0004] In traditional technologies, the vibration information of submarine cables can be collected through sensors, and the vibration information of the submarine cables can be compared and analyzed with the ship identification system and the submarine cable route corridor map to obtain the cable status, thereby generating alarm information. However, this method directly judges the cable status based on the vibration information collected by sensors, and is easily interfered by external objective factors such as wave erosion and sensor noise, resulting in deviations in the obtained cable status and low accuracy of alarm information. Summary of the Invention

[0005] Based on this, in order to solve the above technical problems, it is necessary to provide a submarine cable status monitoring method, device, computer equipment and storage medium that can improve the accuracy of submarine cable alarm information.

[0006] In a first aspect, an embodiment of this application provides a submarine cable status monitoring method, and the method includes:

[0007] Obtain the disturbance data of the submarine cable;

[0008] Process the disturbance data according to a preset feature index to obtain a disturbance feature sequence;

[0009] Input the disturbance feature sequence into a deep learning model to generate target disturbance sequence data;

[0010] When the relationship between the target disturbance sequence data and a threshold meets the alarm condition, determine that the status of the submarine cable is an abnormal status and generate an alarm information.

[0011] In one embodiment, the target perturbation sequence data includes a plurality of target perturbation data sorted according to the occurrence time; the thresholds include a first threshold and a second threshold;

[0012] When the relationship between the target perturbation sequence data and the thresholds meets the alarm condition, determining that the state of the submarine cable is an abnormal state, includes:

[0013] Comparing the target perturbation sequence data with the first threshold, and when there is target perturbation data greater than the first threshold in the target perturbation sequence data, comparing the target perturbation sequence data with the second threshold, and determining the first number of target perturbation data greater than the second threshold;

[0014] When the first number is greater than or equal to the first number threshold, determining that the alarm condition is met and determining that the state of the submarine cable is an abnormal state.

[0015] In one embodiment, the perturbation data is data collected from a plurality of observation points of the submarine cable;

[0016] Processing the perturbation data according to a preset characteristic index to obtain a perturbation characteristic sequence, and inputting the perturbation characteristic sequence into a deep learning model to generate target perturbation sequence data, includes:

[0017] Processing the perturbation data of each observation point according to the characteristic index to obtain the perturbation characteristic sequence corresponding to each observation point;

[0018] Inputting the perturbation characteristic sequence corresponding to each observation point into the deep learning model to generate the target perturbation sequence data of each observation point.

[0019] In one embodiment, the threshold further includes a third threshold;

[0020] When the first number is greater than or equal to the first number threshold, determining that the alarm condition is met and generating the state of the submarine cable as an abnormal state, includes:

[0021] When the first number is greater than or equal to the first number threshold, comparing the target perturbation data of the observation point and its adjacent observation points with the third threshold, and when the target perturbation data of the observation point and its adjacent observation points is greater than the third threshold, determining that the alarm condition is met and determining that the state of the submarine cable is an abnormal state.

[0022] In one embodiment, the threshold further includes a fourth threshold and / or a fifth threshold;

[0023] When the first number is greater than or equal to the first number threshold, it is determined that the alarm condition is satisfied, and the state of the submarine cable is determined to be an abnormal state, including:

[0024] When the first number is greater than or equal to the first number threshold, when any one of the following situations occurs, the state of the submarine cable is determined to be an abnormal state:

[0025] Compare the target perturbation data with the fourth threshold, and determine the second number of the target perturbation data greater than the fourth threshold, where the second number is less than or equal to the second number threshold;

[0026] Obtain the average value of multiple pieces of the target perturbation data, compare the average value with the fifth threshold, and determine that the average value is greater than the fifth threshold.

[0027] In one embodiment, the obtaining the perturbation data of the submarine cable includes:

[0028] Obtain the original perturbation data collected by multiple different types of sensors;

[0029] Obtain the dimensionless parameter factor;

[0030] Perform dimensionless processing on the original perturbation data according to the dimensionless parameter factor to obtain the perturbation data.

[0031] In one embodiment, the characteristic indicators include multiple;

[0032] The processing the perturbation data according to the preset characteristic indicators to obtain a perturbation characteristic sequence includes:

[0033] Process the perturbation data according to each characteristic indicator to generate a characteristic factor corresponding to each characteristic indicator;

[0034] Input each characteristic factor into a feature selection model to obtain the weight value of each characteristic factor;

[0035] Screen a preset number of target characteristic factors from multiple characteristic factors according to the weight value of each characteristic factor;

[0036] Generate the perturbation characteristic sequence according to the target characteristic factors.

[0037] In one embodiment, multiple characteristic indicators include at least two of a variance indicator, a standard deviation indicator, an average value indicator, a maximum value indicator, and a minimum value indicator.

[0038] In one embodiment, the determining that the state of the submarine cable is an abnormal state and generating an alarm message includes:

[0039] Obtain the current electronic chart;

[0040] When the ship's travel trajectory in the current electronic chart matches the status of the submarine cable, generate the alarm information.

[0041] In one embodiment, the method further includes:

[0042] Generate and display a perturbation curve according to the perturbation data;

[0043] When it is determined that the status of the submarine cable is an abnormal status, mark the perturbation curve as abnormal and display the alarm information.

[0044] In a second aspect, an embodiment of the present application further provides a submarine cable status monitoring device, and the device includes:

[0045] A data acquisition module for acquiring perturbation data of the submarine cable;

[0046] A data processing module for processing the perturbation data according to preset characteristic indexes to obtain a perturbation characteristic sequence, inputting the perturbation characteristic sequence into a deep learning model, and generating target perturbation sequence data;

[0047] An alarm generation module for determining that the status of the submarine cable is an abnormal status and generating alarm information when the relationship between the target perturbation sequence data and a threshold meets an alarm condition.

[0048] In a third aspect, an embodiment of the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, it implements the submarine cable status monitoring method according to any one of the embodiments in the first aspect.

[0049] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the submarine cable status monitoring method according to any one of the embodiments in the first aspect.

[0050] For the above-mentioned submarine cable status monitoring method, device, computer device and storage medium, the perturbation data is processed according to preset characteristic indexes to obtain a perturbation characteristic sequence, the perturbation characteristic sequence is input into a deep learning model to generate target perturbation sequence data, and the relationship between the target perturbation sequence data and a threshold is compared and analyzed with an alarm condition to determine the status of the submarine cable. When the status of the submarine cable is an abnormal status, alarm information is generated, which can avoid interference caused by external factors such as sea wave scouring and sensor noise to the status judgment of the submarine cable, improve the accuracy of the submarine cable status, and thus improve the accuracy of the alarm information. Description of the Drawings

[0051] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following briefly introduces the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0052] Figure 1 It is an application environment diagram of the submarine cable status monitoring method in an embodiment;

[0053] Figure 2 It is a schematic flowchart of the submarine cable status monitoring method in an embodiment;

[0054] Figure 3 It is a schematic flowchart of the step of comparing the target disturbance sequence data with the threshold in an embodiment;

[0055] Figure 4 It is a schematic flowchart of the step of obtaining disturbance data in an embodiment;

[0056] Figure 5 It is a schematic flowchart of the step of generating a disturbance feature sequence in an embodiment;

[0057] Figure 6 It is a schematic flowchart of the submarine cable status monitoring method in another embodiment;

[0058] Figure 7 It is a structural block diagram of the submarine cable status monitoring method in an embodiment;

[0059] Figure 8 It is a schematic flowchart of the step of determining the status of the submarine cable in an embodiment;

[0060] Figure 9 It is a structural block diagram of the submarine cable status monitoring device in an embodiment;

[0061] Figure 10a It is an internal structure diagram of a computer device in an embodiment;

[0062] Figure 10b It is an internal structure diagram of a computer device in another embodiment. Detailed Embodiments

[0063] In order to make the objectives, technical solutions and advantages of this application more clear and understandable, the following further elaborates on this application in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0064] The submarine cable status monitoring method provided by this application can be applied to a computer device, where the computer device can be any one of a terminal, a server, and a computer device cluster composed of a terminal and a server. As Figure 1 An application environment diagram is provided as shown. Among them, the terminal 102 communicates with the server 104 through a communication network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or can be deployed on the cloud or other network computer devices. The characteristic index data and the threshold value of the alarm condition determination logic are pre-stored in the data storage system. At least one trained deep learning model and the alarm condition determination logic are deployed in the server 104. The server 104 obtains the disturbance data of the submarine cable, stores the disturbance data of the submarine cable in the data storage system, and displays it on the terminal 102. The server 104 obtains the characteristic index data in the data storage system, processes the disturbance data according to the preset characteristic index, and obtains a disturbance characteristic sequence. The server 104 inputs the disturbance characteristic sequence into the trained deep learning model to generate target disturbance sequence data, and stores the target disturbance sequence data in the data storage system. The server 104 obtains the threshold value of the alarm condition determination logic in the data storage system, compares the target disturbance sequence data with the threshold value according to the alarm condition determination logic. When the server 104 monitors that the relationship between the target disturbance sequence data and the threshold value satisfies the alarm condition, it determines that the status of the submarine cable is an abnormal status, generates an alarm message, and displays it on the terminal 102.

[0065] Among them, the terminal 102 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart TVs, smart vehicle-mounted devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0066] In one embodiment, as Figure 2 shown, a submarine cable status monitoring method is provided. Taking the computer device in Figure 1 as an example, the method includes the following steps:

[0067] Step S202, obtain the disturbance data of the submarine cable.

[0068] Among them, the disturbance data can be used to reflect the current disturbance state of the submarine cable. When external interference affects the state of the submarine cable, the disturbance data of the submarine cable changes from zero to non-zero. When the external interference intensifies, the value of the disturbance data decreases from large to small. The disturbance data can be obtained by performing data preprocessing on the original disturbance data collected by the sensor. The data preprocessing can include, but is not limited to, dimensionless processing of the original disturbance data. The dimensionless processing can include, but is not limited to, centering processing and scaling processing.

[0069] Specifically, the computer device responds to the submarine cable status monitoring request and obtains the disturbance data of the submarine cable. Among them, the submarine cable status monitoring request can be manually triggered by the user. For example, the user clicks the corresponding submarine cable status monitoring button on the page of the computer device; it can also be automatically triggered by the computer device. For example, after the computer device detects the existence of the disturbance data of the submarine cable, it automatically triggers the submarine cable status monitoring request; or, the computer device automatically triggers the submarine cable status monitoring request at a preset moment to obtain the disturbance data of the submarine cable; or, the computer device automatically triggers the submarine cable status monitoring request within each preset time period to obtain the disturbance data of the submarine cable.

[0070] Step S204, process the disturbance data according to the preset characteristic indicators to obtain a disturbance characteristic sequence.

[0071] Among them, the characteristic indicators can be used to describe the data distribution characteristics of the disturbance data in terms of statistics. The characteristic indicators can include, but are not limited to, average indicators, variation indicators, and kurtosis indicators. The average indicator can be used to reflect the general level of the overall disturbance data or an indicator of the central tendency of the distribution, such as the median, average, etc. The variation indicator can be used to reflect the variation condition or dispersion degree of the overall distribution of the disturbance data, such as variance, standard deviation, etc. The kurtosis indicator can be used to reflect the distribution state of the overall disturbance data, such as the maximum value, minimum value, etc. The disturbance characteristic sequence can be the time series data of the disturbance characteristic data, and the disturbance characteristic data is the data obtained by processing the disturbance data according to the preset characteristic indicators.

[0072] Specifically, the computer device obtains the pre-stored characteristic indicators, processes the disturbance data collected in multiple preset time periods according to the characteristic indicators, obtains the disturbance characteristic data corresponding to each preset time period for the characteristic indicators, and sorts the disturbance characteristic data in the order of the disturbance data collection time period to obtain the disturbance characteristic sequence.

[0073] Step S206, input the disturbance characteristic sequence into the deep learning model to generate target disturbance sequence data.

[0074] Among them, the deep learning model can be any one of models such as Deep Belief Network (DBN), Stacked AutoEncoder (SAE), Long Short-Term Memory (LSTM), etc. The target perturbation sequence data can be the time series data output by the deep learning model, which is used to compare and analyze with the alarm condition to determine the state of the current submarine cable.

[0075] Specifically, the computer device inputs the perturbation feature sequence into the pre-deployed deep learning model to generate the target perturbation sequence data. Among them, the deep learning model can be trained through a number of perturbation feature sequence samples.

[0076] In an example, the deep learning model is taken as the Long Short-Term Memory (LSTM) model for illustration:

[0077] First, the computer device obtains a number of perturbation feature sequence samples to generate a training set. The generation method of each perturbation feature sequence sample can refer to the generation method of the perturbation feature sequence in step S204 above, and will not be elaborated here. Define the data type and data size of the input data and output data of the LSTM model according to the perturbation feature sequence samples, and construct the LSTM model framework. Input the perturbation feature sequence samples in the training set into the LSTM model framework, perform multiple forward calculations and multiple backward calculations, and finally obtain the weights of each row of perturbation feature sequence data. Obtain the predicted perturbation sequence data according to the weights of the perturbation feature sequence data. Evaluate the predicted perturbation sequence data output by the model according to the preset recall rate and precision rate criteria, and adjust the LSTM model parameters until the predicted perturbation sequence data output by the model meets the preset criteria, and obtain the trained LSTM model. The training process of the above LSTM model can be executed once every preset time period.

[0078] Step S208, when the relationship between the target perturbation sequence data and the threshold meets the alarm condition, determine that the state of the submarine cable is an abnormal state and generate an alarm message.

[0079] Among them, the threshold can be obtained through the alarm condition determination logic pre-deployed in the computer device. The alarm condition determination logic can include but is not limited to: comparing the target perturbation sequence data with the threshold of the alarm condition, and determining whether the alarm condition is met according to the comparison result; or, determining whether the alarm condition is met by the number of times the threshold that meets the alarm condition appears in the target perturbation sequence data; or, comparing the average value in the target perturbation sequence data within a preset period with the threshold of the alarm condition, and determining whether the alarm condition is met according to the comparison result.

[0080] Specifically, the computer device obtains the threshold value in the alarm condition determination logic, compares the target disturbance sequence data with the threshold value, and determines whether the target disturbance sequence data meets the alarm condition through the alarm condition determination logic. When the computer device determines that the target disturbance sequence data meets the alarm condition, it determines that the state of the submarine cable is an abnormal state and generates an alarm message. When the computer device determines that the target disturbance sequence data does not meet the alarm condition, it determines that the state of the submarine cable is a normal state and does not generate an alarm message.

[0081] In the above submarine cable status monitoring method, the disturbance data is processed through the characteristic indicators preset by the computer device to obtain a disturbance characteristic sequence, the disturbance characteristic sequence is input into the deep learning model to generate target disturbance sequence data, and the relationship between the target disturbance sequence data and the threshold value is compared and analyzed with the alarm condition to determine the state of the submarine cable. When the state of the submarine cable is an abnormal state, an alarm message is generated, which can avoid the interference caused by external factors such as sea wave scouring and sensor noise to the state judgment of the submarine cable, improve the accuracy of the submarine cable state, and thus improve the accuracy of the alarm message.

[0082] In one embodiment, the threshold value includes a first threshold value and a second threshold value. As Figure 3 shown, step S208, when the relationship between the target disturbance sequence data and the threshold value meets the alarm condition, determining that the state of the submarine cable is an abnormal state and generating an alarm message, includes:

[0083] Step S302, comparing the target disturbance sequence data with the first threshold value.

[0084] Among them, the target disturbance sequence data includes multiple target disturbance data sorted according to the occurrence time. The first threshold value can be used to characterize the critical value of the target disturbance data when the state of the submarine cable is an abnormal state. The first threshold value can be determined by the grid search method.

[0085] Specifically, the computer device obtains the pre-stored first threshold value, and sequentially compares each target disturbance data in the target disturbance sequence data with the first threshold value. When there is target disturbance data greater than the first threshold value, steps S304~S308 are executed; when there is no target disturbance data greater than the first threshold value, step S310 is executed.

[0086] In an example, the inventor found in practical applications that the target disturbance sequence data of the submarine cable disturbance caused by sea waves is small and has a long duration; the target disturbance sequence data of the submarine cable disturbance caused by suspension span scouring is periodic; the target disturbance data of the submarine cable disturbance caused by anchor hanging has multiple peaks; the duration of the submarine cable disturbance caused by anchor pounding is short, and there is only one peak in the target disturbance sequence data of the submarine cable, and this peak is much greater than the first threshold value. Therefore, the first threshold value, the second threshold value, and the first number threshold value are set.

[0087] In one example, when there is no data greater than the first threshold in the target disturbance sequence data of the submarine cable, the disturbance state of the submarine cable may be caused by ocean waves. Identify the current submarine cable disturbance event as an ocean wave event. At this time, the submarine cable is in a normal state and no alarm information is generated.

[0088] In one example, when there is no data greater than the first threshold in the target disturbance sequence data of the submarine cable, the disturbance state of the submarine cable may be caused by span scour. Identify the current submarine cable disturbance event as a span scour event. At this time, the submarine cable is in a normal state and no alarm information is generated.

[0089] Step S304: Compare the target disturbance sequence data with the second threshold to determine the first occurrence number of target disturbance data greater than the second threshold.

[0090] Specifically, the computer device obtains the pre-stored second threshold, compares each target disturbance data in the target disturbance sequence data with the second threshold, and determines the first occurrence number of target disturbance data greater than the second threshold in the target disturbance sequence data.

[0091] Step S306: Compare the first occurrence number with the first occurrence number threshold.

[0092] Among them, the first occurrence number threshold can be used to represent the minimum number of times the second threshold appears when the state of the submarine cable is an abnormal state.

[0093] Specifically, the computer device obtains the pre-stored first occurrence number threshold, compares the first occurrence number with the first occurrence number threshold. When the first occurrence number is greater than or equal to the first occurrence number threshold, execute step S308; when the first occurrence number is less than the first occurrence number threshold, execute step S310.

[0094] Step S308: Determine that the alarm condition is met, determine that the state of the submarine cable is an abnormal state, and generate alarm information.

[0095] Among them, the alarm condition may include that when there is target disturbance data greater than the first threshold in the target disturbance sequence data, the first occurrence number of target disturbance data greater than the second threshold in the target disturbance sequence data is greater than the first occurrence number threshold.

[0096] Specifically, the computer device determines that the current target disturbance sequence data meets the alarm condition, determines that the current state of the submarine cable is an abnormal state, generates alarm information and displays it.

[0097] In one example, the second threshold may be equal to the peak value of the target disturbance data that appears multiple times in the target disturbance sequence data when the state of the submarine cable is in an abnormal state. When the first count is greater than the first count threshold, the disturbance state of the submarine cable may be caused by snagging, and the current submarine cable disturbance event is identified as a snagging event. At this time, the submarine cable is in an abnormal state and an alarm needs to be issued in a timely manner.

[0098] In one example, the second threshold may be equal to the value of the target disturbance data that appears multiple times in the target disturbance sequence data when the state of the submarine cable is in an abnormal state. When the first count is greater than the first count threshold and there is only one peak in the target disturbance sequence data, the disturbance state of the submarine cable may be caused by an anchor strike. The current submarine cable disturbance event is identified as an anchor strike event. At this time, the submarine cable is in an abnormal state and an alarm message is generated.

[0099] Step S310, determine that the alarm condition is not met, and determine that the state of the submarine cable is in a normal state.

[0100] Specifically, the computer device determines that the current target disturbance sequence data does not meet the alarm condition, determines that the current state of the submarine cable is in a normal state, and does not generate an alarm message.

[0101] In one example, when the first count of the target disturbance data greater than the second threshold in the target disturbance sequence data is less than the first count threshold, the disturbance state of the submarine cable may be caused by sensor noise. At this time, the submarine cable is in a normal state and no alarm message is generated.

[0102] In this embodiment, the inventor sets alarm conditions that may include but are not limited to threshold parameters such as the first threshold, the second threshold, and the first count threshold according to the characteristics of the target disturbance sequence data when the submarine cable is affected by events such as waves, anchor strikes, snagging, and scour of suspension spans and appears in a disturbed state. By comparing the target disturbance sequence data with the thresholds in this embodiment and determining the state of the submarine cable according to the comparison results, an interpretation mechanism between the target disturbance sequence data of the submarine cable and the state of the submarine cable can be established, the submarine cable disturbance event can be identified, the informatization of the submarine cable facilities can be realized, and the accuracy of determining the state of the submarine cable can be improved.

[0103] In one embodiment, the disturbance data is data collected from multiple observation points of the submarine cable. The disturbance data is processed according to a preset characteristic index to obtain a disturbance characteristic sequence, and the disturbance characteristic sequence is input into a deep learning model to generate target disturbance sequence data, including: processing the disturbance data of each observation point according to the characteristic index to obtain a disturbance characteristic sequence corresponding to each observation point, and inputting the disturbance characteristic sequence corresponding to the observation point into the deep learning model to generate the target disturbance sequence data of each observation point.

[0104] Specifically, the computer device obtains the disturbance data collected from multiple observation points of the submarine cable, and performs the following operations on the disturbance data of each observation point: obtaining the characteristic indexes prestored in the computer device, processing the disturbance data of the observation point according to the preset characteristic indexes to obtain a disturbance characteristic sequence corresponding to the observation point, and inputting the disturbance characteristic sequence corresponding to the observation point into the deep learning model to generate the target disturbance sequence data of the observation point.

[0105] In this embodiment, by setting multiple observation points on the submarine cable, collecting and processing the disturbance data of multiple observation points of the submarine cable, and generating the target disturbance sequence data corresponding to each observation point, the coverage range of the obtained disturbance data can be made wider, and the range of monitoring the state of the submarine cable can be expanded.

[0106] In one embodiment, the threshold further includes a third threshold. When the first number is greater than or equal to the first number threshold, it is determined that the alarm condition is satisfied, and the state of the submarine cable is generated as an abnormal state, including: when the first number is greater than or equal to the first number threshold, comparing the target disturbance data of the observation point and the adjacent observation points of the observation point with the third threshold, and when it is determined that the target disturbance data of the observation point and the adjacent observation points are greater than the third threshold, it is determined that the alarm condition is satisfied, and the state of the submarine cable is determined as an abnormal state.

[0107] Specifically, the computer device obtains the prestored third threshold, compares the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points of the observation point with the third threshold respectively. When both the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points of the observation point are greater than the third threshold, it is determined that the target disturbance sequence data of the observation point satisfies the alarm condition, the state of the submarine cable corresponding to the observation point is determined as an abnormal state, and an alarm message is generated.

[0108] In one example, the inventor found in practical applications that the disturbance of the submarine cable caused by anchoring has a certain influence range, and both the anchoring point and the surrounding points will be affected. When there are a series of abnormal values in the target disturbance sequence data of only a certain point, it may be due to sensor signal problems rather than anchoring damage. Therefore, a third threshold related to the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points of the observation point is set.

[0109] In one example, when both the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points of the observation point are greater than the third threshold, the disturbance state of the submarine cable may be caused by anchoring. Identify the current submarine cable disturbance event as an anchoring event. At this time, the submarine cable is in an abnormal state, and an alarm message is generated.

[0110] In this embodiment, the inventor sets a third threshold related to the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points of the observation point according to the characteristics of the submarine cable disturbance caused by anchoring. By comparing the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points around the observation point with the third threshold respectively to determine the state of the submarine cable, the accuracy of determining the state of the submarine cable in the event of an anchoring event can be improved.

[0111] In one embodiment, the threshold further includes a fourth threshold and / or a fifth threshold. When the first number is greater than or equal to the first number threshold, it is determined that the alarm condition is satisfied, and the state of the submarine cable is determined to be an abnormal state, including: when the first number is greater than or equal to the first number threshold, when any of the following situations occurs, the state of the submarine cable is determined to be an abnormal state:

[0112] (1) Compare the target disturbance data with the fourth threshold to determine the second number of target disturbance data greater than the fourth threshold, and the second number is less than or equal to the second number threshold.

[0113] Among them, the fourth threshold can be used to characterize the numerical value of the target disturbance data that appears multiple times in the target disturbance sequence data of the submarine cable in the case of the submarine cable disturbance caused by the sensor. The second number threshold can be used to characterize the minimum number of times the fourth threshold appears in the target disturbance sequence data of the submarine cable in the case of the submarine cable disturbance caused by the sensor noise.

[0114] Specifically, the computer device obtains the pre-stored fourth threshold, compares the target disturbance sequence data within a preset time period with the fourth threshold to determine the second number of target disturbance data greater than the fourth threshold in the target disturbance sequence data. When the second number is less than or equal to the second number threshold, the computer device determines that the state of the submarine cable is an abnormal state and generates an alarm message.

[0115] In one example, the inventor found in practical applications that in the case of the submarine cable disturbance caused by the sensor noise, the target disturbance sequence data of the submarine cable will have multiple target disturbance data with relatively small thresholds, so a fourth threshold related to the sensor noise is set.

[0116] In one example, when the second number is less than the second number threshold, it can be explained that the disturbance state of the submarine cable is not caused by the sensor noise. At this time, the submarine cable is in an abnormal state and an alarm message is generated.

[0117] (2) Obtain the average value of multiple target disturbance data, compare the average value with the fifth threshold, and determine that the average value is greater than the fifth threshold.

[0118] Specifically, the computer device obtains a pre-stored fifth threshold value and the average value within a preset time period of the target disturbance sequence data, compares the average value of the target disturbance sequence data with the fifth threshold value, and when the average value is greater than the fifth threshold value, the computer device determines that the state of the submarine cable is an abnormal state and generates an alarm message.

[0119] In one example, the inventor found in practical applications that in the case of the submarine cable being disturbed by an anchor hanging, the target disturbance sequence data of the submarine cable increases from small to large and has a long duration. When there are a series of abnormal values at only a certain point, it may be due to problems with the sensor signal rather than an anchor hazard occurring. Therefore, a fifth threshold value related to the duration of the anchor hanging is set.

[0120] In one example, when the average value of the target disturbance sequence data within the preset time period is greater than the fifth threshold value, the disturbance state of the submarine cable may be caused by an anchor hanging. Identify the current submarine cable disturbance event as an anchor hanging event. At this time, the submarine cable is in an abnormal state and an alarm message is generated.

[0121] In this embodiment, the inventor sets a fourth threshold value related to sensor noise according to the characteristics of the submarine cable disturbance caused by sensor noise; and sets a fifth threshold value related to the duration of the anchor hanging according to the characteristics of the submarine cable disturbance caused by the anchor hanging. By comparing the target disturbance sequence data within the preset time period with the threshold value in this embodiment, the accuracy of determining the state of the submarine cable can be improved.

[0122] In one embodiment, as Figure 4 shown, step S202 of obtaining the disturbance data of the submarine cable includes:

[0123] Step S402 of obtaining the original disturbance data collected by multiple different types of sensors.

[0124] Step S404 of obtaining the dimensionless parameter factor.

[0125] Step S406 of performing dimensionless processing on the original disturbance data according to the dimensionless parameter factor to obtain the disturbance data.

[0126] Among them, the types of sensors can include but are not limited to temperature sensors, disturbance sensors, and current sensors. The original disturbance data can be used to represent the data directly collected by the sensors. The dimensionless parameter factor refers to a constant without physical units and is used to perform dimensionless processing on the data.

[0127] In one example, the inventors found in actual applications that each sensor has certain functions and measurement ranges. The data of a single sensor reflects the status of the submarine cable from a certain dimension, which has certain limitations. Therefore, a relatively large variety of sensors can be deployed inside the submarine cable to collect disturbance data in multiple dimensions, reduce the inaccuracy of information, and increase the credibility of information.

[0128] Specifically, the computer device obtains the original disturbance data collected by multiple different types of sensors, as well as the dimensionless parameter factors.

[0129] In one example, each original disturbance data can be subtracted by the dimensionless parameter factor, so as to translate the data to a preset position to obtain the disturbance data.

[0130] In another example, each original disturbance data can be divided by the dimensionless parameter factor, or the logarithm of each original disturbance data can be taken with the dimensionless parameter factor, so as to fix the data within a preset range to obtain the disturbance data.

[0131] In this embodiment, obtaining data through multiple sensors can avoid the limitations of using a single sensor and the problem that the data collected by a single sensor is inaccurate due to being easily interfered by sensor noise. Performing dimensionless processing on the original disturbance data can avoid the influence of a feature with a particularly large value range on the distance calculation of the deep learning model, which helps to improve the model accuracy and thus improve the accuracy of the alarm information.

[0132] In one embodiment, there are multiple preset feature indicators, such as Figure 5 As shown, in step S204, the disturbance data is processed according to the preset feature indicators to obtain a disturbance feature sequence, including:

[0133] In step S502, the disturbance data is processed according to each feature indicator to generate a feature factor corresponding to each feature indicator.

[0134] Among them, the feature factor can be used to characterize the data distribution characteristics of the disturbance data. The multiple feature indicators can include but are not limited to at least two of the average value indicator, variance indicator, standard deviation indicator, maximum value indicator, and minimum value indicator.

[0135] Specifically, the computer device obtains multiple feature indicators, and processes the disturbance data collected in multiple preset time periods according to each feature indicator to generate a feature factor corresponding to each feature indicator.

[0136] In one example, the computer device processes the disturbance data according to the variance indicator, standard deviation indicator, and average value indicator to generate the variance factor, standard deviation factor, and average value factor of the disturbance data.

[0137] In one example, the computer device processes the perturbation data according to variance metrics, standard deviation metrics, mean metrics, maximum metrics, and minimum metrics to generate a variance factor, a standard deviation factor, a mean factor, a maximum factor, and a minimum factor of the perturbation data.

[0138] Step S504: Input each feature factor into the feature selection model to obtain the weight value of each feature factor.

[0139] Step S506: Screen and obtain a preset number of target feature factors from multiple feature factors according to the weight value of each feature factor.

[0140] Among them, the feature selection model is a model with the ability to screen feature factors. The feature selection model can be any one of a support vector machine model (Support Vector Machine, SVM), a regression model, a logistic regression model, etc.

[0141] In one example, taking the feature selection model as a logistic regression model as an example, the training process of the feature selection model is described as follows:

[0142] First, the computer device obtains a number of feature factors, takes the set of the number of feature factors as the initial feature training set and inputs it into the feature selection model to be trained. The weight value of each feature factor is obtained through the objective function in the feature selection model, and the feature factor with the smallest weight value is deleted from the initial feature training set to obtain a new feature set. The model adjusts the parameters of the objective function through multiple forward searches and backward searches to obtain a new feature selection model. The new feature set is used as the training set and input into the new feature selection model for training. The above operations are repeated until the obtained feature set includes a preset number of feature factors, and the trained feature selection model is obtained.

[0143] Specifically, the computer device obtains the trained feature selection model, inputs each feature factor into the feature selection model to obtain the weight value of each feature factor, and screens and obtains a preset number of target feature factors from multiple feature factors according to the weight value of each feature factor.

[0144] Step S508: Generate a perturbation feature sequence according to the target feature factors.

[0145] Specifically, the computer device sorts multiple target feature factors according to the order of the acquisition time periods corresponding to the target feature factors to generate a perturbation feature sequence.

[0146] In this embodiment, the feature selection model is used to select the feature factors generated from the perturbation data to generate a perturbation feature sequence, which can reduce the amount of data processed by the deep learning model, lower the processing cost, and improve the processing efficiency.

[0147] In one embodiment, when it is determined that the state of the submarine cable is an abnormal state, an alarm message is generated, including: obtaining the current electronic nautical chart, and generating an alarm message when the ship's travel trajectory in the current electronic nautical chart matches the state of the submarine cable.

[0148] Among them, the electronic nautical chart can be a digital nautical chart of the maritime vessel traffic management system. The maritime vessel traffic management system can be used to monitor the positions and travel trajectories of maritime vessels.

[0149] Specifically, when the computer device determines that the state of the submarine cable is an abnormal state, the computer device obtains the current electronic nautical chart and obtains the positional relationship between the ship's travel trajectory in the current electronic nautical chart and the observation point area of the abnormal state. When there is an overlapping area between the ship's travel trajectory in the current electronic nautical chart and the observation point area of the abnormal state, it is determined that the state of the submarine cable corresponding to the ship's travel trajectory matches the observation point, and an alarm message is generated; when there is no overlapping area between the ship's travel trajectory in the current electronic nautical chart and the observation point area of the abnormal state, it is determined that the state of the submarine cable corresponding to the ship's travel trajectory does not match, and it is determined that the state of the submarine cable corresponding to the observation point is a normal state.

[0150] In an example, the state of the submarine cable at the observation point is affected by sensor noise, and the computer device determines that the state of the submarine cable corresponding to the observation point is an abnormal state. The computer device compares the observation point area of the abnormal state with the ship's travel trajectory in the current electronic nautical chart, determines that no ship has passed through the observation point area, identifies it as a sensor noise event, and determines that the state of the submarine cable is a normal state.

[0151] In this embodiment, the inventor found in actual application that the appearance of a ship's travel trajectory near the submarine cable is a necessary but not sufficient condition for submarine cable early warning. Sometimes, the sensor signal may have problems. Although the probability is low, there may also be a situation where no ship appears and a false alarm occurs. Therefore, a secondary confirmation process for the abnormal state of the submarine cable is set. In this embodiment, by obtaining the positional relationship between the ship's travel trajectory in the current electronic nautical chart and the observation point area, and comparing the positional relationship with the determined state of the submarine cable to confirm the state of the submarine cable, the accuracy of the determined state of the submarine cable can be improved, and the probability of false alarms can be reduced.

[0152] In one embodiment, the computer device can generate and display a perturbation curve based on the perturbation data. When it is determined that the state of the submarine cable is an abnormal state, the perturbation curve is marked as abnormal and the alarm message is displayed.

[0153] In one embodiment, such as Figure 6As shown, a method for monitoring the state of a submarine cable is provided, including:

[0154] Step S602, obtaining the original disturbance data collected by multiple different types of sensors.

[0155] Specifically, the computer device obtains the original disturbance data of multiple observation points collected by a temperature sensor, a disturbance sensor, and a current sensor, and performs all the following steps on the original disturbance data collected at each observation point.

[0156] Step S604, obtaining a dimensionless parameter factor, and performing dimensionless processing on the original disturbance data according to the dimensionless parameter factor to obtain disturbance data.

[0157] Specifically, the computer device obtains the dimensionless parameter factor, and subtracts the dimensionless parameter factor from the original disturbance data to obtain the disturbance data.

[0158] Step S606, processing the disturbance data according to a preset plurality of characteristic indexes to generate a characteristic factor corresponding to each characteristic index.

[0159] Specifically, the computer device processes the disturbance data according to a plurality of characteristic indexes to generate characteristic factors of the disturbance data. The characteristic factors include a variance factor, a standard deviation factor, an average value factor, a maximum value factor, and a minimum value factor.

[0160] Step S608, inputting each characteristic factor into a feature selection model to obtain a preset number of target characteristic factors, and generating a disturbance characteristic sequence.

[0161] Specifically, the computer device inputs each characteristic factor into a logistic regression model to obtain a preset number of target characteristic factors, and further generates a disturbance characteristic sequence. The training method of the specific logistic regression model and the generation method of the disturbance characteristic sequence refer to the method provided in the above embodiment and will not be specifically described here.

[0162] In one example, the inventor found that in practical applications, the amount of data of the generated characteristic factors is large. To improve the data processing efficiency, it is necessary to select some characteristic factors that have a greater impact on the model output result from all the characteristic factors. Commonly used feature selection methods mainly include filtering methods, embedding methods, wrapper methods, etc.

[0163] The filtering method, generally used as a data preprocessing step, does not require any machine learning algorithms, and only selects features according to the scores in various statistical tests and various indexes of correlation. The filtering method uses variance for preliminary filtering, and then uses methods based on chi-square, F-test, and mutual information to obtain correlation for secondary filtering to achieve feature selection.

[0164] The embedding method is trained through machine learning algorithms and models to obtain the weight coefficients of each feature factor, and features are selected according to the weight coefficients from large to small. These weight coefficients often represent the contribution or importance of the feature factors to the model. Compared with the filtering method, the result of the embedding method is more precise to the utility of the model itself, and it has a better effect on improving the model effectiveness. Moreover, because the contribution of features to the model is considered, irrelevant features (features that need to be filtered by correlation) and features without discrimination (features that need to be filtered by variance) will be deleted due to the lack of contribution to the model.

[0165] The wrapper method is also a method that performs feature selection and algorithm training simultaneously. The difference from the embedding method is that it uses an objective function as a black box to select features, rather than the user inputting a threshold of an evaluation metric or statistic. The wrapper method trains an estimator (i.e., a logistic regression model) on the initial feature set (i.e., the set composed of all feature factors), and then prunes the least important features from the current set of features (i.e., deletes the feature factor with the lowest weight from the feature set). This process is recursively repeated on the pruned set until finally reaching the required number of features to be selected (i.e., obtaining the preset number of target feature factors). Different from the filtering method and the embedding method that solve all problems in one training, the wrapper method needs to perform multiple trainings using feature subsets, so the computational cost it requires is the highest among the several methods. However, the feature factors it obtains have the highest confidence in the model for generating target perturbation sequence data. Therefore, the inventor finally chooses to use the wrapper method for feature selection.

[0166] Step S610: Input the perturbation feature sequence into the deep learning model to generate target perturbation sequence data. The target perturbation sequence data includes multiple target perturbation data sorted by occurrence time.

[0167] Specifically, the computer device inputs the perturbation feature sequence into the LSTM model to generate target perturbation sequence data. The specific training method of the LSTM model refers to the method provided in the above embodiments and will not be elaborated here.

[0168] Step S612: When there is target perturbation data greater than the first threshold, compare the target perturbation sequence data with the second threshold to determine the first occurrence number of target perturbation data greater than the second threshold.

[0169] Specifically, the computer device compares the target perturbation sequence data with the first threshold. When there is target perturbation data greater than the first threshold in the target perturbation sequence data, compare the target perturbation sequence data with the second threshold to determine the first occurrence number of target perturbation data greater than the second threshold.

[0170] Step S614, when the first number is greater than or equal to the first threshold, determine whether the alarm condition is satisfied according to the relationship between the target disturbance sequence data and the third threshold, the fourth threshold, and the fifth threshold.

[0171] Specifically, when the first number is greater than or equal to the first threshold, the computer device compares the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points of the observation point with the third threshold; determines the second number of target disturbance data greater than the fourth threshold in the target disturbance sequence data of the observation point, and compares the second number with the second number threshold; obtains the average value of the target disturbance sequence data within a preset time period, and compares the average value with the fifth one. When the target disturbance sequence data of the observation point and the target disturbance sequence data of the adjacent observation points of the observation point are respectively greater than the third threshold, and the second number is less than or equal to the second number threshold, and the average value is greater than the fifth threshold, it is determined that the alarm condition is satisfied.

[0172] Step S616, determine that the state of the submarine cable is an abnormal state and generate an alarm message.

[0173] Specifically, after the computer device determines that the alarm condition is satisfied, it determines that the state of the submarine cable is an abnormal state. The computer device obtains the current electronic chart, compares the ship's travel trajectory in the current electronic chart with the observation point area, and determines the state of the current submarine cable. When the computer device determines that the state of the submarine cable is an abnormal state, it generates an alarm message.

[0174] In this embodiment, by obtaining the original disturbance data, performing dimensionless processing on the original disturbance data to obtain the disturbance data, processing the disturbance data according to the preset characteristic indicators to obtain the characteristic factors of the disturbance data, inputting the characteristic factors into the logistic regression model for screening to generate the disturbance characteristic sequence, inputting the disturbance characteristic sequence into the LSTM model to generate the target disturbance sequence data, analyzing and comparing the relationship between the target disturbance sequence data and the threshold with the alarm condition to obtain the state of the submarine cable, and then comparing the state of the submarine cable with the electronic chart to determine the state of the submarine cable. When the state of the submarine cable is an abnormal state, an alarm message is generated, which can avoid the interference caused by external factors such as wave scouring and sensor noise to the judgment of the state of the submarine cable, make full use of the multi-sensor data resources in multiple time and space, improve the accuracy of the submarine cable state, and thus improve the accuracy of the alarm message.

[0175] In one embodiment, a structural block diagram of a submarine cable state monitoring method is further provided. As Figure 7As shown, the computer device acquires disturbance data and stores it in the database. The background processing module processes the disturbance data, and stores the characteristic indicators and the corresponding characteristic factors during the processing as characteristic data in the database, stores the target disturbance sequence data as target data in the database, stores the thresholds of the alarm conditions as status data in the database, and stores the generated alarm information as alarm data in the database. The front end reads the disturbance data in the database to generate a disturbance curve and display it, and reads the alarm data to generate and display the alarm information.

[0176] In this embodiment, by storing data in the database, processing data by the background processing module, and displaying data by the front end, the data processing efficiency can be improved, and the visualization function is available.

[0177] In one embodiment, a schematic diagram of the data processing flow for submarine cable status monitoring is also provided, as Figure 8 shown. Multiple characteristic factors are used as the data of the input layer. The computer device generates a disturbance characteristic sequence according to the characteristic factors in the input layer, and generates target disturbance sequence data according to the disturbance characteristic sequence. Multiple thresholds of the alarm conditions are used as the data of the hidden layer. The computer device compares the target disturbance sequence data with the multiple thresholds of the alarm conditions in the hidden layer, and performs global fusion on multiple comparison results (that is, when each comparison result meets the alarm condition, it is determined that the status of the submarine cable is abnormal; when at least one of the multiple comparison results does not meet the alarm condition, it is determined that the status of the submarine cable is normal), and obtains the status data of the submarine cable. The status data of the submarine cable is used as the data of the output layer. When the computer device determines that the relationships between the target disturbance sequence data and the multiple thresholds of the alarm conditions all meet the alarm condition, it outputs that the status of the submarine cable is 0 abnormal; when the computer device determines that the relationships between the target disturbance sequence data and the multiple thresholds of the alarm conditions do not meet the alarm condition, it outputs that the status of the submarine cable is 1 normal.

[0178] In this embodiment, the inventor sets an operation of performing global fusion on multiple comparison results according to the characteristic that in actual use, due to different judgment dimensions of each threshold in the alarm condition, the confidence levels of the comparison results of multiple thresholds in the alarm condition are also different. By performing global fusion on the relationships between the target disturbance sequence data and the multiple thresholds of the alarm condition in this embodiment, the accuracy of the determined status of the submarine cable can be improved.

[0179] It should be understood that although the steps in the above flowcharts are shown sequentially according to the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise clearly stated in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above flowcharts may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0180] Based on the description of the embodiments of the submarine cable status monitoring method described above, the present disclosure also provides a submarine cable status monitoring device. The device may include a system (including a distributed system), software (application), module, component, computer device, client, etc. that uses the method described in the embodiments of this specification and combines the necessary implementation hardware. Based on the same inventive concept, the devices in one or more embodiments provided by the embodiments of the present disclosure are as described in the following embodiments. Since the implementation solutions for the device to solve problems are similar to the method, the implementation of the specific device in the embodiments of this specification can refer to the implementation of the foregoing method, and the repeated parts will not be described again. As used hereinafter, the term "unit" or "module" may be a combination of software and / or hardware that can implement a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0181] In one embodiment, as Figure 9 shown, a submarine cable status monitoring device 900 is provided, including: a data acquisition module 902, a data processing module 904, and an alarm generation module 906, where:

[0182] The data acquisition module 902 is configured to acquire disturbance data of the submarine cable.

[0183] The data processing module 904 is configured to process the disturbance data according to preset characteristic indicators to obtain a disturbance characteristic sequence, input the disturbance characteristic sequence into a deep learning model, and generate target disturbance sequence data.

[0184] The alarm generation module 906 is configured to determine that the status of the submarine cable is an abnormal status and generate an alarm message when the relationship between the target disturbance sequence data and the threshold satisfies the alarm condition.

[0185] In one embodiment, the target disturbance sequence data includes a plurality of target disturbance data sorted according to the occurrence time, and the thresholds include a first threshold and a second threshold. The alarm generation module 906 includes: a first number determination unit, configured to compare the target disturbance sequence data with the first threshold, and when there is target disturbance data greater than the first threshold in the target disturbance sequence data, compare the target disturbance sequence data with the second threshold, and determine the first number of target disturbance data greater than the second threshold. The first number alarm unit is configured to determine that the alarm condition is satisfied and determine that the state of the submarine cable is an abnormal state when the first number is greater than or equal to the first number threshold.

[0186] In one embodiment, the disturbance data is data collected from multiple observation points of the submarine cable. The data processing module 904 includes: a disturbance feature sequence generation unit, configured to process the disturbance data of each observation point according to the feature index to obtain a disturbance feature sequence corresponding to each observation point. The target disturbance sequence data generation unit is configured to input the disturbance feature sequence corresponding to each observation point into the deep learning model to generate the target disturbance sequence data of each observation point.

[0187] In one embodiment, the threshold further includes a third threshold, and the alarm generation module 906 is further configured to: when the first number is greater than or equal to the first number threshold, compare the target disturbance data of the observation point and its adjacent observation points with the third threshold, and when it is determined that the target disturbance data of the observation point and its adjacent observation points is greater than the third threshold, determine that the alarm condition is satisfied and determine that the state of the submarine cable is an abnormal state.

[0188] In one embodiment, the threshold further includes a fourth threshold and / or a fifth threshold. When the first number is greater than or equal to the first number threshold, when any of the following situations occurs, the alarm generation module 906 further includes: a fourth threshold comparison unit, configured to compare the target disturbance data with the fourth threshold to determine the second number of target disturbance data greater than the fourth threshold, and the second number is less than or equal to the second number threshold. The fifth threshold comparison unit is configured to obtain the average value of a plurality of target disturbance data, compare the average value with the fifth threshold, and determine that the average value is greater than the fifth threshold.

[0189] In one embodiment, the data acquisition module 902 includes: a raw disturbance data acquisition unit, configured to acquire raw disturbance data collected by a plurality of different types of sensors. The data non-dimensionalization unit is configured to obtain a non-dimensionalization parameter factor; perform non-dimensionalization processing on the raw disturbance data according to the non-dimensionalization parameter factor to obtain disturbance data.

[0190] In one embodiment, there are multiple characteristic indicators, and the data processing module 904 includes: a characteristic factor generation unit, configured to process the perturbation data according to each characteristic indicator to generate a characteristic factor corresponding to each characteristic indicator; a characteristic factor screening unit, configured to input each characteristic factor into a feature selection model to obtain the weight of each characteristic factor, and screen a preset number of target characteristic factors from multiple characteristic factors according to the weight of each characteristic factor; and a perturbation characteristic sequence generation unit, configured to generate a perturbation characteristic sequence according to the target characteristic factors.

[0191] In one embodiment, the multiple characteristic indicators include at least two of a variance indicator, a standard deviation indicator, an average value indicator, a maximum value indicator, and a minimum value indicator.

[0192] In one embodiment, the alarm generation module 906 is further configured to: obtain the current electronic chart, and generate an alarm message when the ship's travel trajectory in the current electronic chart matches the state of the submarine cable.

[0193] In one embodiment, the submarine cable status monitoring device 900 further includes: a front-end display module, configured to generate and display a perturbation curve according to the perturbation data; when it is determined that the state of the submarine cable is an abnormal state, mark the perturbation curve as abnormal and display the alarm message.

[0194] For the specific limitations of the submarine cable status monitoring device, reference can be made to the limitations of the submarine cable status monitoring method in the foregoing text, which will not be elaborated herein. Each module in the above submarine cable status monitoring device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above respective modules.

[0195] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as Figure 10a shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store perturbation data, characteristic data, target data, status data, and alarm data. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a submarine cable status monitoring method.

[0196] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structural diagram may be as shown in Figure 10b . The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be implemented through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for monitoring the state of a submarine cable. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device may be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0197] Those skilled in the art can understand that Figure 10a and Figure 10b the structure shown in

[0198] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0199] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, it implements the steps in the above method embodiments.

[0200] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0201] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0202] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for monitoring the status of a submarine cable, characterized in that, the method includes: Obtaining disturbance data of the submarine cable; Processing the disturbance data according to preset characteristic indexes to obtain a disturbance characteristic sequence; Inputting the disturbance characteristic sequence into a deep learning model to generate target disturbance sequence data; When the relationship between the target disturbance sequence data and a threshold meets an alarm condition, determining that the status of the submarine cable is an abnormal status and generating an alarm message; wherein, the target disturbance sequence data includes a plurality of target disturbance data sorted according to the occurrence time; the threshold includes a first threshold and a second threshold; The step of determining that the status of the submarine cable is an abnormal status when the relationship between the target disturbance sequence data and the threshold meets the alarm condition includes: Comparing the target disturbance sequence data with the first threshold. In the case where there is target disturbance data greater than the first threshold in the target disturbance sequence data, comparing the target disturbance sequence data with the second threshold and determining the first occurrence number of the target disturbance data greater than the second threshold; When the first occurrence number is greater than or equal to a first occurrence number threshold, determining that the alarm condition is met and determining that the status of the submarine cable is an abnormal status; wherein, the disturbance data is data collected from multiple observation points of the submarine cable, and the threshold further includes a third threshold; The step of determining that the alarm condition is met and generating the status of the submarine cable as an abnormal status when the first occurrence number is greater than or equal to the first occurrence number threshold includes: When the first occurrence number is greater than or equal to the first occurrence number threshold, comparing the target disturbance data of the observation point and its adjacent observation points with the third threshold. When it is determined that the target disturbance data of the observation point and its adjacent observation points is greater than the third threshold, determining that the alarm condition is met and determining that the status of the submarine cable is an abnormal status.

2. The method according to claim 1, characterized in that, The step of processing the disturbance data according to preset characteristic indexes to obtain a disturbance characteristic sequence, and inputting the disturbance characteristic sequence into a deep learning model to generate target disturbance sequence data includes: Processing the disturbance data of each observation point according to the characteristic indexes to obtain the disturbance characteristic sequence corresponding to each observation point; Inputting the disturbance characteristic sequence corresponding to each observation point into the deep learning model to generate the target disturbance sequence data of each observation point.

3. The method according to claim 1, characterized in that, The characteristic indexes include a plurality of; The step of processing the disturbance data according to preset characteristic indexes to obtain a disturbance characteristic sequence includes: Processing the disturbance data according to each characteristic index to generate a characteristic factor corresponding to each characteristic index; Inputting each characteristic factor into a feature selection model to obtain the weight value of each characteristic factor; Screening a preset number of target characteristic factors from multiple characteristic factors according to the weight value of each characteristic factor; Generating the disturbance characteristic sequence according to the target characteristic factors.

4. The method according to claim 3, wherein, the multiple characteristic indicators include at least two of a variance indicator, a standard deviation indicator, an average value indicator, a maximum value indicator, and a minimum value indicator.

5. The method according to any one of claims 1 to 4, wherein, the method further includes: generating and displaying a perturbation curve according to the perturbation data; when it is determined that the state of the submarine cable is an abnormal state, marking the perturbation curve as abnormal and displaying the alarm information.

6. The method according to any one of claims 1 to 5, wherein, the obtaining of the perturbation data of the submarine cable includes: obtaining original perturbation data collected by multiple different types of sensors; obtaining a dimensionless parameter factor; performing dimensionless processing on the original perturbation data according to the dimensionless parameter factor to obtain the perturbation data.

7. The method according to claim 1, wherein, the determining that the state of the submarine cable is an abnormal state and generating an alarm information includes: obtaining a current electronic nautical chart; when the ship's travel trajectory in the current electronic nautical chart matches the state of the submarine cable, generating the alarm information.

8. A submarine cable status monitoring device, wherein, the device includes: a data acquisition module for acquiring perturbation data of a submarine cable; a data processing module for processing the perturbation data according to preset characteristic indicators to obtain a perturbation feature sequence, and inputting the perturbation feature sequence into a deep learning model to generate target perturbation sequence data; an alarm generation module for determining that the state of the submarine cable is an abnormal state and generating alarm information when the relationship between the target perturbation sequence data and a threshold meets an alarm condition; wherein, the target perturbation sequence data includes a plurality of target perturbation data sorted according to the occurrence time; the threshold includes a first threshold and a second threshold; the alarm generation module includes: a first number determination unit for comparing the target perturbation sequence data with the first threshold, and when there is target perturbation data greater than the first threshold in the target perturbation sequence data, comparing the target perturbation sequence data with the second threshold to determine the first number of target perturbation data greater than the second threshold; a first number alarm unit for determining that the alarm condition is met and determining that the state of the submarine cable is an abnormal state when the first number is greater than or equal to a first number threshold; wherein, the perturbation data is data collected from multiple observation points of the submarine cable, and the threshold further includes a third threshold; the alarm generation module is further configured to: when the first number is greater than or equal to the first number threshold, compare the target perturbation data of the observation point and its adjacent observation points with the third threshold, and determine that the alarm condition is met and the state of the submarine cable is an abnormal state when the target perturbation data of the observation point and its adjacent observation points is greater than the third threshold.

9. A computer device, including a memory and a processor, the memory stores a computer program, wherein, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, wherein, when the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images