A method and system for submarine cable risk assessment and early warning

Abstract

This invention relates to the field of submarine cable monitoring technology, specifically disclosing a method and system for submarine cable risk assessment and early warning. The method includes: geological exploration and topographic analysis, deployment of an environmental sensor network, model construction, and unmanned submersible inspection. Through geological exploration and topographic analysis, this invention identifies risk areas, such as geological faults and volcanic craters; deploys an environmental sensor network to monitor the marine environment and cable status in real time; processes sensor data using machine learning and artificial intelligence algorithms to predict cable faults and risks; constructs a dynamic risk model; assesses cable risk levels in real time; and, based on the risk model predictions, uses automated submersibles to inspect and verify cables in high-risk areas, confirming the submarine cable status and responding promptly to phenomena such as damage, corrosion, or loosening to reduce the occurrence of faults.

Description

Technical Field

[0001] This invention belongs to the field of submarine cable monitoring technology, specifically relating to a method and system for submarine cable risk assessment and early warning. Background Technology

[0002] Submarine cables have become a crucial bridge and focal point for long-distance transoceanic energy transmission and large-scale utilization of new energy sources. Establishing a standard system for submarine cables, developing new manufacturing technologies, construction techniques for high sea state conditions, and high-precision testing technologies are core and key to systematically solving the challenges of future offshore engineering operations involving greater length, depth, and precision. Improving the detection technology for submarine cable risks and faults is a vital component of this process. Therefore, developing submarine cable safety testing technology is of great strategic significance for supporting and ensuring energy security and sustainable socio-economic development.

[0003] After submarine cables are laid, they are susceptible to damage from both human and natural factors, making them prone to operational safety issues. Submarine cable projects are often crucial channels connecting power grids on both sides of a strait or supplying power from the mainland to islands. They are typically expensive to construct and extremely complex to repair. A failure can lead to system disconnection and severe consequences. Therefore, the safe and reliable operation of submarine cables is paramount. In the early stages of construction, domestic and international submarine cable projects often reference experience from projects like oil pipelines, adopting methods such as full mechanical protection or no protection at all. This approach fails to consider actual risk conditions or quantify risk scenarios and probabilities. Consequently, although several submarine cable projects have been completed both domestically and internationally, very little work has been done on genuine risk assessment for submarine cables, leaving almost no directly applicable experience.

[0004] In response, the inventors proposed a method and system for risk assessment and early warning of submarine cables to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for risk assessment and early warning of submarine cables, in order to solve the problem that in the early stages of construction of submarine cable projects at home and abroad, the experience of projects such as oil pipelines is often referred to, and the method of full mechanical protection or no protection is adopted. This approach does not take into account the actual risk situation, nor does it quantify the risk situation and risk probability.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method and system for risk assessment and early warning of submarine cables, comprising the following steps:

[0008] Geological exploration and topographic analysis, including geological and topographic exploration, including seafloor topography, seismic activity, water depth, and seafloor geological factors;

[0009] Deploy an environmental sensor network, placing multiple environmental sensors near the cable to acquire sensor data for real-time monitoring of the marine environment and cable status, including ocean temperature, water quality, ocean currents, tides, cable current, voltage, and signal strength.

[0010] A model is built that uses machine learning and artificial intelligence algorithms to process the aforementioned sensor data, including marine environmental data and cable status data, to predict faults and risks, construct a dynamic risk model, and assess the cable risk level in real time.

[0011] Unmanned submersible inspections, based on risk model predictions, utilize automated submersibles equipped with cameras and sensors to inspect and verify cables in high-risk areas. This confirms the condition of submarine cables and allows for timely responses to any issues such as damage, corrosion, or loosening, thereby reducing the occurrence of malfunctions.

[0012] Preferably, the environmental sensors include a temperature sensor, a water depth sensor, a water flow rate sensor, and a cable status sensor.

[0013] Preferably, the machine learning and artificial intelligence algorithms are specifically implemented as follows:

[0014] Data extraction involves collecting marine environmental data and cable status data from the aforementioned environmental sensors, and cleaning, transforming, and standardizing the data to ensure consistency and quality. Features in the data are extracted, including the mean, fluctuation, and rate of change of temperature. The data is then divided into training and testing sets for training and validating the model.

[0015] Data processing involves building a logistic regression machine learning model, using training set data to train the model, establishing the relationship between cable status and environmental factors, and using test set data to evaluate the model's performance, including precision, recall, and F1 score.

[0016] Prediction and alarm triggering: Based on the model, the probability or risk level of cable failure is predicted. If the model predicts that the probability of cable failure exceeds the threshold of 80% and the temperature threshold is 30°C, an alarm is triggered. The system automatically sends an alarm message, which includes the alarm type, timestamp, and relevant temperature data, and notifies the submarine cable maintenance team via email.

[0017] Preferably, the logistic regression is as follows:

[0018] Define whether the cable is faulty (1 indicates fault, 0 indicates normal).

[0019]

[0020] Where P(y=1) is the probability of cable fault, x1, x2, ..., xn is the feature, β0, β1, β2,…,β n These are model parameters.

[0021] A submarine cable risk assessment and early warning system includes sensors, cameras, and computers as described in the submarine cable risk assessment and early warning method above, wherein the sensors, cameras, and computers are communicatively connected.

[0022] Compared with the prior art, the beneficial effects of the present invention are:

[0023] This invention identifies risk areas, such as geological faults and volcanic craters, through geological exploration and topographic analysis. It then deploys an environmental sensor network to monitor the marine environment and cable status in real time. The sensor data is processed using machine learning and artificial intelligence algorithms to predict cable faults and risks, construct a dynamic risk model, and assess the cable risk level in real time. Based on the results of the risk model prediction, automated submersibles are used to inspect and verify the cables in high-risk areas to confirm the condition of the submarine cables. The invention also responds promptly to any damage, corrosion, or loosening of the cables to reduce the occurrence of faults. Attached Figure Description

[0024] Figure 1 This invention provides a method and flowchart for risk assessment and early warning of submarine cables. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Example:

[0027] Please see Figure 1 As shown, a method for risk assessment and early warning of submarine cables includes the following steps:

[0028] Geological exploration and topographic analysis: Conducting geological and topographic exploration, including seafloor topography, seismic activity, water depth, seafloor geological factors, and identifying risk areas such as geological faults and volcanic craters;

[0029] Deploy an environmental sensor network, placing multiple environmental sensors near the cable to acquire sensor data for real-time monitoring of the marine environment and cable status, including ocean temperature, water quality, ocean currents, tides, cable current, voltage, and signal strength. Use the sensor data for accurate risk assessment and early warning.

[0030] A model is built that utilizes machine learning and artificial intelligence algorithms to process the aforementioned sensor data, including marine environmental data and cable status data, to predict faults and risks. A dynamic risk model is constructed to assess the cable risk level in real time, enabling early detection of potential problems and the implementation of appropriate measures.

[0031] Unmanned submersible inspections, based on risk model predictions, utilize automated submersibles equipped with cameras and sensors to inspect and verify cables in high-risk areas. This confirms the condition of submarine cables and allows for timely responses to any issues such as damage, corrosion, or loosening, thereby reducing the occurrence of malfunctions.

[0032] Specifically, the environmental sensors include a temperature sensor, a water depth sensor, a water flow rate sensor, and a cable status sensor.

[0033] Specifically, the machine learning and artificial intelligence algorithms are implemented as follows:

[0034] Data extraction involves collecting marine environmental data and cable status data from the aforementioned environmental sensors, and cleaning, transforming, and standardizing the data to ensure consistency and quality. Features in the data are extracted, including the mean, fluctuation, and rate of change of temperature. The data is then divided into training and testing sets for training and validating the model.

[0035] Data processing involves building a logistic regression machine learning model. The training set data is used to train the model, establishing the relationship between cable condition and environmental factors. Test set data is used to evaluate the model's performance, including precision, recall, and F1 score. TP, FP, FN, and TN are defined as follows:

[0036] TP: Predicted value was 1, actual value was 1, prediction was correct;

[0037] FP: Predicted value is 1, actual value is 0, prediction error;

[0038] FN: Predicted value was 0, actual value was 1, prediction error;

[0039] TN: Predicted value was 0, actual value was 0, prediction was correct;

[0040] Accuracy:

[0041] Recall rate:

[0042] F1 score: ;

[0043] Prediction and alarm triggering: Based on the model, the probability or risk level of cable faults is predicted. If the model predicts that the probability of cable faults exceeds a threshold of 80% (temperature threshold is 30°C), an alarm is triggered. The system automatically sends an alarm message including the alarm type, timestamp, and relevant temperature data, notifying the submarine cable maintenance team via email. If an alarm is triggered when the cable temperature rises above a threshold exceeding the normal operating range, the following formula can be used:

[0044] Temperature > Threshold → Trigger Alarm

[0045] The above formula indicates that an alarm is triggered when the cable temperature exceeds a predefined threshold.

[0046] Specifically, the logistic regression is as follows:

[0047] Define whether the cable is faulty (1 indicates fault, 0 indicates normal).

[0048]

[0049] Where P(y=1) is the probability of cable fault, x1, x2, ..., x n is the feature, β0, β1, β2,…,β n These are model parameters.

[0050] As can be seen from the above, risk areas such as geological faults and volcanic craters can be identified through geological exploration and topographic analysis. An environmental sensor network can be deployed to monitor the marine environment and cable status in real time. The sensor data can be processed using machine learning and artificial intelligence algorithms to predict cable faults and risks, build a dynamic risk model, and assess the cable risk level in real time. Based on the results of the risk model prediction, automated submersibles can be used to inspect and verify the cables in high-risk areas to confirm the status of submarine cables. Any phenomena such as damage, corrosion, or loosening of the lines can be dealt with in a timely manner to reduce the occurrence of faults.

[0051] A submarine cable risk assessment and early warning system includes sensors, cameras, and computers as described in the submarine cable risk assessment and early warning method above. The sensors and cameras are communicatively connected to the computer. Its beneficial effects are the same as those in the embodiment of the submarine cable risk assessment and early warning method, and will not be repeated here.

[0052] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for risk assessment and early warning of submarine cables, characterized in that, Includes the following steps: Geological exploration and topographic analysis, including geological and topographic exploration, including seafloor topography, seismic activity, water depth, and seafloor geological factors; Deploy an environmental sensor network, placing multiple environmental sensors near the cable to acquire sensor data for real-time monitoring of the marine environment and cable status, including ocean temperature, water quality, ocean currents, tides, cable current, voltage, and signal strength. A model is built that uses machine learning and artificial intelligence algorithms to process the aforementioned sensor data, including marine environmental data and cable status data, to predict faults and risks, construct a dynamic risk model, and assess the cable risk level in real time. Unmanned submersible inspections, based on risk model predictions, utilize automated submersibles equipped with cameras and sensors to inspect and verify cables in high-risk areas. This confirms the condition of submarine cables and allows for timely responses to any issues such as damage, corrosion, or loosening, thereby reducing the occurrence of malfunctions.

2. The method for risk assessment and early warning of submarine cables according to claim 1, characterized in that: The environmental sensors include a temperature sensor, a water depth sensor, a water flow rate sensor, and a cable status sensor.

3. The method for risk assessment and early warning of submarine cables according to claim 1, characterized in that: The specific implementations of the machine learning and artificial intelligence algorithms are as follows: Data extraction involves collecting marine environmental data and cable status data from the aforementioned environmental sensors, and cleaning, transforming, and standardizing the data to ensure consistency and quality. Features in the data are extracted, including the mean, fluctuation, and rate of change of temperature. The data is then divided into training and testing sets for training and validating the model. Data processing involves building a logistic regression machine learning model, using training set data to train the model, establishing the relationship between cable status and environmental factors, and using test set data to evaluate the model's performance, including precision, recall, and F1 score. Prediction and alarm triggering: Based on the model, the probability or risk level of cable failure is predicted. If the model predicts that the probability of cable failure exceeds the threshold of 80% and the temperature threshold is 30°C, an alarm is triggered. The system automatically sends an alarm message, which includes the alarm type, timestamp, and relevant temperature data, and notifies the submarine cable maintenance team via email.

4. The method for risk assessment and early warning of submarine cables according to claim 3, characterized in that: The logistic regression is described in detail below: Define whether the cable is faulty (1 indicates fault, 0 indicates normal). Where P(y=1) is the probability of cable fault, x1, x2, ..., x n is the feature, β0, β1, β2,…,β n These are model parameters.

5. A submarine cable risk assessment and early warning system, characterized in that: The method includes a sensor, a camera, and a computer as described in any one of claims 1-4 above, wherein the sensor, camera, and computer are communicatively connected.

Images