Optical cable fault early warning method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to an optical cable fault early warning method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring sensor data for a to-be-detected optical cable; mapping the sensor data into a space-time heterogeneous graph through a space-time graph neural network, and carrying out feature evolution analysis on the space-time heterogeneous graph to generate fault information; and performing fault early warning on the to-be-detected optical cable based on the fault information. By adopting the method, the sensor data of the to-be-detected optical cable can be acquired, the sensor data is mapped into the space-time heterogeneous graph by means of the space-time diagram neural network, and the fault information is generated by carrying out feature evolution analysis, so that potential faults can be identified in advance, fault early warning is further realized, maintenance measures are actively taken before the faults occur, and the fault detection efficiency is improved. Operation and maintenance cost is effectively reduced, reliability and stability of an optical cable network are improved, and optical cable operation and maintenance cost is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, and in particular to an optical cable fault early warning method and device, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] With the popularity and development of network communication, the optical cable laying network between cities, rural areas and countries is expanding to support the high-speed and large-capacity data transmission requirements. The laying environment of optical cables is relatively complex, not only needs to pass through various complex terrains, but also may be affected by road construction, pipeline settlement, animal gnawing and the like, and thus is easily damaged by external forces.

[0003] At present, optical cable faults can only be found after data transmission is affected, and then the fault causes are located by manually pulling or using optical cable surveying instruments, OTDR (Optical Time Domain Reflectometer) and the like, and then the faults are repaired, but there is no effective method for early warning of optical cable faults, and thus the operation and maintenance cost is high. SUMMARY

[0004] Therefore, it is necessary to provide an optical cable fault early warning method, device, computer equipment, computer readable storage medium and computer program product capable of early warning of optical cable faults and reducing operation and maintenance cost in view of the above technical problems.

[0005] In a first aspect, the present application provides an optical cable fault early warning method, comprising:

[0006] obtaining sensor data for a to-be-detected optical cable;

[0007] mapping the sensor data into a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, and performing feature evolution analysis on the spatio-temporal heterogeneous graph to generate fault information;

[0008] based on the fault information, performing fault early warning on the to-be-detected optical cable.

[0009] In one embodiment, the mapping of the sensor data into a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, and the performing of feature evolution analysis on the spatio-temporal heterogeneous graph to generate fault information, comprises:

[0010] mapping the sensor data into a spatio-temporal heterogeneous graph, wherein the nodes of the spatio-temporal heterogeneous graph represent preset spatial positions on the to-be-detected optical cable, and the edges between the nodes represent connection relationships between the preset spatial positions;

[0011] For each node, the sensor data corresponding to the node and other nodes adjacent to the node are analyzed in the spatial dimension to generate spatial correlation features;

[0012] The sensor data corresponding to each node is analyzed in the time dimension to generate time evolution features;

[0013] According to the spatial correlation features and the time evolution features, fault analysis is performed to generate fault information.

[0014] In one embodiment, the sensor data is mapped into a spatio-temporal heterogeneous graph by the spatio-temporal graph neural network, and the spatio-temporal heterogeneous graph is analyzed to generate fault information, including:

[0015] In the case of receiving a target event processing instruction, the sensor data is mapped into a spatio-temporal heterogeneous graph by a transfer learning model, and the spatio-temporal heterogeneous graph is analyzed to generate fault information;

[0016] The transfer learning model is obtained by transfer learning of a spatio-temporal graph neural network based on target event training samples.

[0017] In one embodiment, the fault information includes fault category, fault time and fault location, and the fault warning of the to-be-detected optical cable based on the fault information includes:

[0018] According to the fault time and the fault location, a to-be-detected vibration signal is read from the sensor data, and a to-be-detected spectrogram of the to-be-detected vibration signal is generated;

[0019] A sample spectrogram corresponding to the fault category is obtained from a preset spectrogram model library;

[0020] The first similarity between the to-be-detected spectrogram and the sample spectrogram is compared, and in the case that the first similarity is higher than a first preset threshold, the to-be-detected optical cable is fault warned.

[0021] In one embodiment, the sensor data for the to-be-detected optical cable is obtained, including:

[0022] The sensor data for at least two to-be-detected optical cables is obtained, and the sensor data is collected based on synchronously transmitted optical pulse signals;

[0023] After the sensor data is mapped into a spatio-temporal heterogeneous graph by the spatio-temporal graph neural network, it further includes:

[0024] Based on the spatio-temporal heterogeneous graph of any two to-be-detected optical cables, feature matching is performed to obtain a second similarity;

[0025] In a case where the second similarity is higher than a second preset threshold, it is determined that the two arbitrary optical cables have a same route risk.

[0026] In one of the embodiments, the obtaining of the sensor data for the at least two optical cables to be detected comprises:

[0027] The control sensor synchronously sends an optical pulse signal to each optical cable to be detected, and receives sensor data corresponding to each optical cable to be detected returned by the sensor.

[0028] In one of the embodiments, the sensor comprises at least one of a distributed optical fiber acoustic wave sensor, a Brillouin optical time domain reflectometer, and a Raman optical time domain reflectometer.

[0029] In one of the embodiments, the fault information comprises a fault position, and the fault warning of the optical cable to be detected based on the fault information comprises:

[0030] Based on the fault position, a warning area of the optical cable to be detected is determined, and the warning area is tracked and monitored.

[0031] In a second aspect, the present application further provides an optical cable fault warning device, comprising:

[0032] An obtaining module is configured to obtain sensor data for an optical cable to be detected;

[0033] An analyzing module is configured to map the sensor data to a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, perform feature evolution analysis on the spatio-temporal heterogeneous graph, and generate fault information;

[0034] A warning module is configured to perform fault warning of the optical cable to be detected based on the fault information.

[0035] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor realizes the following steps when executing the computer program:

[0036] Obtaining sensor data for an optical cable to be detected;

[0037] Mapping the sensor data to a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, performing feature evolution analysis on the spatio-temporal heterogeneous graph, and generating fault information;

[0038] Performing fault warning of the optical cable to be detected based on the fault information.

[0039] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which realizes the following steps when executed by a processor:

[0040] obtain sensor data of the optical cable to be detected;

[0041] map the sensor data into a spatio-temporal heterogeneous graph by using a spatio-temporal graph neural network, perform feature evolution analysis on the spatio-temporal heterogeneous graph, and generate fault information;

[0042] based on the fault information, perform fault early warning on the optical cable to be detected.

[0043] The optical cable fault early warning method, device, computer device, computer readable storage medium and computer program product can obtain sensor data of the optical cable to be detected, map the sensor data into a spatio-temporal heterogeneous graph by using a spatio-temporal graph neural network, perform feature evolution analysis on the spatio-temporal heterogeneous graph, and generate fault information, which can identify potential faults in advance, thereby realizing fault early warning, proactively taking maintenance measures before the fault occurs, effectively reducing operation and maintenance costs, and improving the reliability and stability of the optical cable network, and reducing the operation and maintenance costs of the optical cable. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0045] Figure 1 a flowchart of the optical cable fault early warning method in one embodiment;

[0046] Figure 2 a module architecture diagram of the optical cable fault early warning system in one embodiment;

[0047] Figure 3 a flowchart of feature analysis on sensor data in one embodiment;

[0048] Figure 4 a structural block diagram of the optical cable fault early warning device in one embodiment;

[0049] Figure 5 an internal structure diagram of the computer device in one embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0051] It should be noted that the terms "first", "second", etc. used in the present application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "include" and "have" and any variations thereof used in the present application are intended to cover non-exclusive inclusion. The term "a plurality of" used in the present application refers to two or more. The term "and / or" used in the present application refers to one of the options or any combination of the options.

[0052] In the related art, the optical cable fault can be found only after the data transmission is affected, and then the fault reason is manually located and repaired, and there is a lack of effective method for early warning of the optical cable fault, so the operation and maintenance cost is high. Based on this, the present application provides an optical cable fault early warning method, and the present embodiment takes the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is realized through the interaction of the terminal and the server.

[0053] In one exemplary embodiment, as shown in Figure 1 The present application provides an optical cable fault early warning method, comprising the following steps:

[0054] Step 101, acquiring sensor data for the optical cable to be detected.

[0055] In the present application, first, the sensor data for the optical cable to be detected is acquired, wherein the sensor data is physical parameter data that can reflect the real-time condition of the optical cable to be detected, such as vibration frequency, strain value, temperature change and other information of different spatial positions of the optical cable to be detected. The sensor data also has a corresponding time stamp, which ensures that the subsequent analysis can simultaneously associate the spatial dimension and the time dimension characteristics, and provides data support for subsequent construction of a space-time heterogeneous graph.

[0056] Exemplarily, the collection and analysis of the sensor data can focus on the fiber core inside the optical cable to be detected. The optical pulse signal sent by the sensor is directly transmitted inside the fiber core. When the fiber core is disturbed externally or its state changes, the optical pulse signal will produce scattering effect in the fiber core. Then, the sensor receives and analyzes these scattered light signals from the fiber core, and converts them into corresponding sensor data.

[0057] The sensor data can be real-time data continuously collected by the pre-deployed sensor, so as to realize real-time fault early warning of the optical cable to be detected. The sensor data can also be batch data obtained from the server or the cloud, so as to periodically detect the fault of the optical cable to be detected, in order to repair the fault of the optical cable to be detected before the data transmission is affected.

[0058] The sensors for collecting sensor data include one or more of vibration sensors, temperature sensors, stress sensors, etc., which can capture the physical state of the optical cable and environmental changes from different dimensions. For example, the vibration sensor is used to perceive abnormal vibration signals caused by external force activities such as road construction, mechanical excavation, or geological subsidence; the temperature sensor is used to monitor local temperature abnormalities caused by factors such as local overload of the optical cable, aging of the insulation layer, or poor contact; and the stress sensor is used to accurately collect stress changes caused by external extrusion, heavy pressure, or pipeline deformation.

[0059] In step 102, the sensor data is mapped into a spatio-temporal heterogeneous graph by a spatio-temporal graph neural network, and feature evolution analysis is performed on the spatio-temporal heterogeneous graph to generate fault information.

[0060] In this step, the acquired sensor data can be processed by a spatio-temporal graph neural network. First, the sensor data is mapped into a spatio-temporal heterogeneous graph by a spatio-temporal graph neural network, and the node and edge structure in the spatio-temporal heterogeneous graph is used to represent the connection relationship between different spatial positions of the optical cable to be detected, and different attribute information is assigned to each node based on the sensor data.

[0061] Subsequently, the spatio-temporal graph neural network performs feature evolution analysis on the spatio-temporal heterogeneous graph. Through a message passing mechanism, the information of each node and other nodes connected thereto can be aggregated to perceive the propagation and influence range of the event along the optical fiber. Meanwhile, an internal time sequence processing unit can be used to analyze the sequence of the change of the feature of each node over time, so as to accurately capture the dynamic evolution pattern of the event, and finally generate accurate fault information.

[0062] In this way, a direct and reliable decision basis is provided for predictive maintenance and intelligent operation and maintenance of the optical cable to be detected, realizing a fundamental change from passive response to active early warning.

[0063] Specifically, in an exemplary embodiment, the analysis process of the spatio-temporal graph neural network in step 102 includes the following steps:

[0064] The sensor data is mapped into a spatio-temporal heterogeneous graph, wherein the nodes of the spatio-temporal heterogeneous graph represent preset spatial positions on the optical cable to be detected, and the edges between the nodes represent the connection relationship between the preset spatial positions;

[0065] For each node, spatial dimension feature analysis is performed on the sensor data corresponding to the node and other nodes adjacent to the node to generate spatial correlation features;

[0066] For each node, time dimension feature evolution analysis is performed on the sensor data corresponding to the node to generate time evolution features;

[0067] According to the spatial correlation characteristics and the time evolution characteristics, fault analysis is performed to generate fault information.

[0068] That is, first, the space-time graph neural network will map different preset spatial positions of the optical cable to be detected into nodes of a space-time heterogeneous graph according to the characteristics of the sensor data, take the sensor data corresponding to different preset spatial positions as node attributes, and construct edges between nodes according to the connection relationship between the preset spatial positions, forming a space-time heterogeneous graph that can fuse spatial structure and sensor data.

[0069] Among them, the preset spatial position can refer to the specific position of each monitoring point on the optical cable to be detected, for example, a monitoring point can be set every 10 meters or 20 meters on the optical cable to be detected; the connection relationship between the preset spatial positions refers to the adjacent preset spatial positions in the spatial position, because there is a direct physical connection, so the data can be directly transmitted.

[0070] Subsequently, the space-time graph neural network can perform spatial dimension feature analysis on the sensor data corresponding to each node and other nodes adjacent to the node, capture the correlation mode and collaborative change of attributes between different nodes through graph convolution and other operations, and generate spatial correlation characteristics. The spatial correlation characteristics can reflect how the sensor data of a preset spatial position affects its adjacent preset spatial positions through physical connection.

[0071] For example, assume that at a certain time, vibration data is collected for preset spatial position A, adjacent preset spatial position B, and adjacent preset spatial position C of a certain section of optical cable. Under normal conditions, if there is no local external force interference, the vibration frequencies of A, B, and C are stable at about 5Hz, and the data difference does not exceed 0.5Hz, which reflects good spatial correlation; if the vibration frequency of A suddenly rises to 30Hz at this time, while nodes B and C still maintain at about 5Hz, the data difference far exceeds the normal threshold, then through the spatial correlation characteristics between A and B and C, it can be preliminarily inferred that the optical cable at A may be hit by a local external force.

[0072] On the other hand, the space-time graph neural network can also perform time dimension feature evolution analysis on the sensor data corresponding to each node, and use the time series modeling module to mine the dynamic evolution law of the attributes of the same node over time to generate time evolution characteristics. The time evolution characteristics can reflect how the sensor data on a unified preset spatial position develops and propagates over time.

[0073] For example, for a preset spatial position D of a certain optical cable, the temperature at D will present periodic changes with the ambient temperature under normal conditions: it is maintained at 25-28°C during the day and 20-23°C at night, with a small fluctuation range and stable trend; if the temperature at D starts to rise from 26°C at 9:00 am and rises to 38°C at 15:00 pm, and the rising trend of 1°C per hour is maintained for the next 12 hours, which completely deviates from the normal periodic fluctuation rule, then through the time evolution characteristics at D, it can be preliminarily inferred that the optical cable at D may be caused by poor heat dissipation due to pipeline blockage or cable short circuit heating.

[0074] Further, combined with the spatial correlation characteristics and the time evolution characteristics, the fault analysis can identify abnormal patterns with spatio-temporal correlation from massive and noisy sensor data, thereby generating fault information. The fault information can include semantic information such as probability of fault occurrence, fault type, fault time, and fault location, and can provide data support for subsequent operation and maintenance decisions.

[0075] This two-dimensional feature mining and fault analysis based on spatial and temporal dimensions can realize cross-validation and complementary reasoning of the two types of features, provide more accurate and more instructive early warning basis for operation and maintenance personnel, and improve the timeliness and reliability of optical cable fault early warning.

[0076] Step 103, based on the fault information, performing fault early warning on the optical cable to be detected.

[0077] In this step, after obtaining the fault information output by the spatio-temporal graph neural network, the generated fault information can be further used to automatically perform fault early warning on the optical cable to be detected.

[0078] For example, the fault information can be converted into intuitive early warning signals and pushed to relevant operation and maintenance personnel, so that the operation and maintenance personnel can timely master the specific situation of possible faults of the optical cable to be detected, and then take targeted troubleshooting and protection measures in advance to avoid actual impact on data transmission caused by optical cable faults, and realize the change from post-repair to pre-warning.

[0079] Among them, the early warning signals can be pushed in the form of popping up an early warning window on a large screen of an operation and maintenance platform, pushing early warning messages to the mobile phones or smart watches of operation and maintenance personnel, etc.; or, the network work order system and geographic information system can be linked, and when the fault information is generated, the order generation logic is automatically triggered to integrate the fault position, fault time, and fault type information into a standardized protection work order. The early warning signals can include the aforementioned fault position, fault time, and fault type information.

[0080] In one implementation, the geographic information system can highlight the risk area of the optical cable to be detected according to the fault, accurately mark the risk area on the electronic map, intuitively present the geographic distribution and surrounding environment of the fault position, and facilitate the planning of the route by the operation and maintenance personnel.

[0081] In addition, for the area where the video monitoring device is deployed, the camera near the fault position can also be automatically called to return the live picture in real time, so that the operation and maintenance personnel remotely confirm whether there is visible interference such as construction excavation and equipment anomaly, and provide visual support for quickly judging the cause of the fault.

[0082] The early warning mechanism can make full use of the long-term stored sensor data, go beyond the limitation of simply dealing with external damage, and deeply analyze the health status of the optical cable infrastructure and the optical cable itself. For example, for the key points such as joint boxes and manholes where the optical cable runs, the long-term change trend of the sensor data can be continuously tracked to help the operation and maintenance personnel to make maintenance plans in advance, avoid non-sudden failures caused by infrastructure aging and optical cable performance degradation, and further reduce the overall cost and risk of optical cable operation and maintenance.

[0083] The scheme provided in the application can obtain sensor data of an optical cable to be detected, map the sensor data into a spatio-temporal heterogeneous graph by means of a spatio-temporal graph neural network, and perform feature evolution analysis to generate fault information, which can identify potential faults in advance, and thus realize fault early warning, so as to actively take maintenance measures before the fault occurs, effectively reduce the operation and maintenance cost, and improve the reliability and stability of the optical cable network, and reduce the optical cable operation and maintenance cost.

[0084] The spatio-temporal graph neural network can effectively analyze the sensor data after being pre-trained by a large amount of training data, but it is difficult to obtain sufficient training data for rare events or new events, so the analysis effect of the spatio-temporal graph neural network cannot be guaranteed. Based on this, in one exemplary embodiment, step 102 includes:

[0085] In the case of receiving a target event processing instruction, the sensor data is mapped into a spatio-temporal heterogeneous graph by means of a transfer learning model, and feature evolution analysis is performed on the spatio-temporal heterogeneous graph to generate fault information.

[0086] The transfer learning model is obtained by transfer learning of the spatio-temporal graph neural network by using a target event training sample.

[0087] In this exemplary embodiment, the spatio-temporal graph neural network and the transfer learning model are cooperated to realize comprehensive coverage of regular events and rare events.

[0088] Specifically, when receiving the target event processing instruction, it can be judged that the current sensor data may correspond to a rare or new target event, then a special transfer learning model can be called to perform data processing and fault analysis, and the construction of the transfer learning model depends on the transfer learning training process, that is, using a limited number of target event training samples to fine-tune the spatio-temporal graph neural network based on the spatio-temporal graph neural network with regular event analysis capability, and transfer the spatio-temporal feature extraction capability learned by the spatio-temporal graph neural network on regular events to the target event scene, forming a transfer learning model that can accurately identify target events.

[0089] In actual application scenarios, the spatio-temporal graph neural network and the transfer learning model are in a state of synchronous cooperation: after receiving the sensor data, the event type corresponding to the sensor data can be determined through a pre-set feature matching mechanism or directly according to the user's operation instruction, if it is a regular event, then the sensor data is directly mapped into a spatio-temporal heterogeneous graph by the spatio-temporal graph neural network according to the regular process, and the spatial correlation feature and the time evolution feature are analyzed and fault information is generated; if it is determined to be a target event, a target event processing instruction is triggered, and the transfer learning model is switched to for processing.

[0090] The transfer learning model will follow the same spatio-temporal heterogeneous graph mapping logic as the spatio-temporal graph neural network to convert the sensor data into a spatio-temporal heterogeneous graph containing pre-set spatial position nodes and connection edges, and then combine it with the target event feature pattern it has mastered through transfer learning to carry out targeted feature evolution analysis on the spatio-temporal heterogeneous graph, accurately capture the unique spatio-temporal correlation and dynamic change rule of the target event, and then generate fault information containing specific information of this special event.

[0091] In this way, the efficiency and accuracy of regular event analysis are guaranteed, and the recognition problem of target events caused by sample scarcity is solved, greatly improving the adaptability and coverage of the optical cable fault early warning system to various complex events.

[0092] In one exemplary embodiment, the fault information includes a fault category, a fault time, and a fault location, and step 103 includes:

[0093] According to the fault time and the fault location, a to-be-detected vibration signal is read from the sensor data, and a to-be-detected spectrogram of the to-be-detected vibration signal is generated;

[0094] A sample spectrogram corresponding to the fault category is obtained from a pre-set spectrogram model library;

[0095] The first similarity between the to-be-detected spectrogram and the sample spectrogram is compared, and in the case that the first similarity is higher than a first pre-set threshold, a fault warning is performed on the to-be-detected optical cable.

[0096] In this example embodiment, the fault information identified by the spatio-temporal graph neural network is secondarily verified by acoustic spectrum comparison, forming a double safeguard logic.

[0097] Specifically, after obtaining the fault information containing the fault category, fault time and fault location, the time window of data collection can be first locked according to the fault time, and the corresponding monitoring point can be determined in combination with the fault location, so as to accurately read the to-be-detected vibration signal in the spatio-temporal range from the sensor data. The to-be-detected vibration signal can directly reflect the physical state of the to-be-detected optical cable affected, such as impact vibration of external force impact, friction vibration of structural looseness, etc.

[0098] Subsequently, the read to-be-detected vibration signal can be processed, and the to-be-detected vibration signal can be converted from time domain to frequency domain by Fourier transform algorithm, etc., to obtain a to-be-detected acoustic spectrum. The color depth or gray scale change on the acoustic spectrum represents the energy intensity of different frequency vibrations, and the frequency characteristic pattern of the fault vibration can be clearly presented.

[0099] On this basis, a sample acoustic spectrum corresponding to the fault category can be called from a preset acoustic spectrum model library. The acoustic spectrum model library is constructed in advance through a large number of historical fault cases, and for each fault category, a sample acoustic spectrum graph of its typical vibration signal is stored. These sample acoustic spectrum graphs are labeled with sample acoustic spectrum graphs of various faults, and can represent the characteristic frequency features of various faults.

[0100] Further, a first similarity between the to-be-detected acoustic spectrum and the sample acoustic spectrum can be calculated. By comparing the coincidence degree of the energy distribution, peak position, waveform contour and other characteristics of the two, the first similarity between the to-be-detected acoustic spectrum and the sample acoustic spectrum is output. For example, the first similarity can be calculated by using any one or more of structural similarity index, cosine similarity or feature point-based matching algorithm, etc., and the specific implementation is not limited.

[0101] If the first similarity is higher than a first preset threshold, it means that the frequency characteristics of the to-be-detected vibration signal are highly matched with the typical characteristics of the target fault category, and the fault judgment result is reliable, so that the fault warning of the to-be-detected optical cable can be triggered. If the first similarity is lower than the first preset threshold, it means that the frequency characteristics of the to-be-detected vibration signal are not highly matched with the typical characteristics of the target fault category, i.e. the fault judgment result is not reliable, so that the subsequent warning process can be suspended to avoid false warning caused by single signal fluctuation, and the rigor of the final warning result is ensured.

[0102] The same route risk refers to a potential risk that two or more optical cables may simultaneously fail due to the same external interference or environmental factor, because the two or more optical cables have path overlap, adjacent parallel or share key infrastructure in the physical laying path. Once this risk occurs, it is likely to cause regional and systematic network interruption, and the consequences are much more serious than single optical cable failure. Therefore, it is very important to detect the same route risk between multiple optical cables to be detected.

[0103] In an exemplary embodiment, step 101 comprises:

[0104] Obtaining sensor data for at least two optical cables to be detected, the sensor data being collected based on synchronously transmitted optical pulse signals;

[0105] Then, in step 102, after mapping the sensor data into a spatio-temporal heterogeneous graph by the spatio-temporal graph neural network, it further comprises:

[0106] Performing feature matching based on the spatio-temporal heterogeneous graphs of any two optical cables to be detected to obtain a second similarity;

[0107] In the case where the second similarity is higher than a second preset threshold, it is determined that any two optical cables to be detected have the same route risk.

[0108] In this exemplary embodiment, in step 101, not only the sensor data of a single optical cable to be detected is collected, but also the sensor data of at least two optical cables to be detected is obtained, and the collection of these sensor data is based on synchronously transmitted optical pulse signals, thereby ensuring that the sensor data of different optical cables to be detected has consistency in time dimension, avoiding feature comparison deviation caused by collection time difference, laying a time synchronization foundation for subsequent cross-cable risk analysis, and making the state features of different optical cables in the same time window comparable.

[0109] Then, in step 102, the sensor data of each optical cable to be detected can be mapped into a corresponding spatio-temporal heterogeneous graph by using a spatio-temporal graph neural network, and feature matching is performed based on the spatio-temporal heterogeneous graphs of any two optical cables to be detected to obtain a second similarity.

[0110] Specifically, in an implementation, for any two optical cables to be detected, a graph structure similarity algorithm can be used to map the nodes of the two spatio-temporal heterogeneous graphs to the same vector space, calculate the cosine similarity of the node embedding vectors, and compare the adjacency matrix similarity of the edge connection relationship to obtain a similarity score in the spatial dimension; then, for the time series of the sensor data, the dynamic time warping distance of the corresponding sequences of the two optical cables to be detected is calculated to measure the similarity of their time trends, and then a similarity score in the time dimension is obtained; then, according to the importance of space and time in the same route risk judgment, the two types of similarity scores are weighted and summed to obtain the second similarity. Alternatively, in another implementation, the second similarity can also be calculated by comparing the spatial correlation features and time evolution features of the two spatio-temporal heterogeneous graphs. In addition, other feature matching algorithms can also be used, which are not limited in the present application.

[0111] If the second similarity is higher than the second preset threshold, it is determined that the two optical cables to be detected have the same routing risk, that is, there are likely to be overlapping or adjacent areas in the physical laying path of the two optical cables to be detected, and when external interference occurs in the area, it may cause both optical cables to be detected to fail at the same time, causing regional paralysis of the communication network.

[0112] In this way, the feature matching based on the spatio-temporal heterogeneous graph can actively identify the optical cable combination with the same routing risk, rather than passively discovering it after a failure occurs, so that the operation and maintenance team can develop special protection strategies for these risky optical cables, effectively reduce the risk of large-scale communication interruption caused by the same routing failure, and improve the disaster tolerance and fault tolerance of the entire optical cable network.

[0113] In an implementation, sensor data for at least two optical cables to be detected is obtained, including:

[0114] The sensor is controlled to synchronously send an optical pulse signal to each optical cable to be detected, and receive sensor data corresponding to each optical cable to be detected returned by the sensor.

[0115] In this implementation, by controlling the sensor with multiple ports, the consistency of the sensor data of multiple optical cables to be detected in the time dimension is ensured, providing an accurate time sequence basis for subsequent cross-optical cable feature matching.

[0116] Specifically, the device can send a synchronization control instruction to the sensor, and after the sensor receives the synchronization control instruction, it uses its multiple independent ports to simultaneously send optical pulse signals to the corresponding multiple optical cables to be detected, ensuring that the sending time, pulse frequency, pulse intensity, and other parameters of the multiple optical pulse signals are consistent, and avoiding deviations in subsequent sensor data comparison caused by differences in signal transmission.

[0117] In the process of transmitting the optical pulse signals along the optical cables to be detected, corresponding modulation changes will be generated due to the physical state of the optical cables to be detected, for example, the vibration of a certain place of the optical cables to be detected will cause the phase shift of the optical pulse, the strain will change the transmission delay of the optical pulse, the temperature fluctuation will affect the attenuation degree of the optical pulse, etc. These optical pulse signals carrying the state information of the optical cables will return to the sensor along the original path, and the multi-path port of the sensor receives the return signals of the corresponding optical cables and converts them into electrical signals for preliminary processing such as filtering, amplification, analog-to-digital conversion, etc., to generate sensor data.

[0118] Due to the synchronous transmission and synchronous reception of the multi-path optical pulse signals, the multi-path sensor data generated naturally have timestamp consistency, that is, each set of sensor data can accurately correspond to the running state of two optical cables to be detected at the same time point, which provides a reliable basis in the time dimension for subsequent feature matching and judgment of the same route risk based on the space-time heterogeneous graph, effectively avoiding the misjudgment or omission of the same route risk caused by the asynchronous data collection.

[0119] Among them, the sensor includes at least one of a distributed fiber acoustic wave sensor, a Brillouin optical time domain reflectometer, and a Raman optical time domain reflectometer.

[0120] That is, the device for collecting sensor data of the optical cables to be detected can flexibly select at least one of DAS (Distributed Acoustic Sensing), BOTDR (Brillouin Optical Time Domain Reflectometer), and ROTDR (Raman Optical Time Domain Reflectometer), or realize multi-dimensional sensor data collection through the combination of multiple sensors. Each sensor relies on a unique optical principle to adapt to the monitoring needs of different state parameters of the optical cables to be detected, ensuring that the sensor data can fully reflect the running state of the optical cables to be detected.

[0121] DAS is a distributed optical fiber sensing technology that uses coherent Rayleigh scattering signals as useful information to measure the disturbance of the optical cable to be detected. It uses the optical cable to be detected itself as a sensing medium, captures the phase changes caused by external vibration and acoustic disturbance when the optical pulse signal propagates in the optical cable to be detected, and then converts these changes into high-resolution vibration data. BOTDR is based on the spontaneous Brillouin scattering in the optical fiber of the optical cable to be detected, and the temperature / strain distribution is inverted by measuring the change of Brillouin frequency shift. ROTDR (Raman Optical Time Domain Reflectometer) uses the spontaneous Raman scattering of the optical fiber in the optical cable to be detected, and calculates the temperature distribution by separating the Stokes and Anti-Stokes light intensity ratio.

[0122] In practical applications, a single sensor or a combination of sensors can be selected according to monitoring needs: for example, if only vibration and strain need to be monitored, a combination of DAS and BOTDR can be selected; if temperature, strain and vibration need to be monitored simultaneously, the three sensors can be used in cooperation to synchronously collect multi-dimensional state data of the optical cable, providing a more comprehensive and accurate data basis for subsequent construction of a space-time heterogeneous graph, analysis of fault information and same-route risk.

[0123] In an exemplary embodiment, the fault information includes a fault location, and step 103 includes:

[0124] Based on the fault location, a warning area of the optical cable to be detected is determined, and the warning area is tracked and monitored.

[0125] In this exemplary embodiment, a dynamic tracking and warning monitoring mechanism can be constructed around the fault location to ensure targeted management and control of the risk area of the optical cable to be detected. When the fault information containing the fault location is obtained, the physical routing data of the optical cable to be detected can be combined to define a warning area with the fault location as the core. For example, if it is a local sudden fault such as external impact, the warning area is usually set to a range of 50 meters before and after the fault location, covering the optical cable section that may be directly affected; if it is a fault that may spread, such as structural loosening or temperature anomaly, the warning area will be appropriately expanded to 100-200 meters before and after the fault location, and include the surrounding key infrastructure to avoid the spread of the fault hidden danger to other areas.

[0126] Once the warning area is identified, a tracking and monitoring mechanism can be activated. For example, the data collection frequency of sensors in the area can be automatically increased, changing from the usual 10-minute collection to 1-minute collection. This high-frequency sensor data captures the location of the fault and subtle changes in the surrounding environment, allowing for real-time monitoring of the fault's development trend. Alternatively, it can be linked with surrounding monitoring equipment and maintenance systems. If video surveillance cameras are deployed in the warning area, the camera feed will automatically focus on the warning area, transmitting real-time images to facilitate remote observation by maintenance personnel to check for ongoing interference. Simultaneously, the warning area information can be synchronized to a geographic information system and dynamically highlighted on an electronic map, allowing the maintenance team to intuitively understand the dynamics of the warning area.

[0127] In this way, the precise delineation of early warning areas and dynamic tracking and monitoring based on the fault location allows maintenance personnel to focus on high-risk fault locations, enabling refined and dynamic management of optical cable faults, saving valuable time for subsequent on-site handling, and effectively reducing the impact of faults.

[0128] like Figure 2 The diagram shown is a modular architecture diagram of an optical cable fault early warning system in an exemplary embodiment. The system includes a multi-functional sensing module, a multi-modal feature analysis module, and a control and storage module.

[0129] The multi-functional sensing module integrates technologies such as distributed optical fiber acoustic sensing (DAS), Brillouin optical time domain reflectometer (BOTDR), and Raman optical time domain reflectometer (ROTDR), enabling it to send optical pulse signals to the fiber cores (fiber core 1 and fiber core 2) of two optical cables under test. It can be understood that if fiber core 1 and fiber core 2 traverse the same or very close paths in physical space, they share a common route risk; conversely, they do not share a common route risk.

[0130] The multimodal feature analysis module uses a spatiotemporal graph neural network to perform multidimensional analysis on the sensor data collected by the multifunctional sensing module from the two fiber cores, mining multimodal features such as vibration, strain, and temperature contained in the sensor data. When the sensor data of a certain optical cable under test is analyzed and a situation that may affect the safety of the fiber core, such as construction machinery operation, is identified, it will be determined that there is a risk event in the optical cable under test and fault information will be generated.

[0131] The control and storage module is used to control the operation of the entire system. On the one hand, it can control the multi-functional sensing module to send two optical pulse signals at the same time, supporting the measurement of fiber core 1 and fiber core 2. On the other hand, it can also receive fault information generated by the multi-modal feature analysis module and store the relevant sensing data and fault information for subsequent query and further analysis.

[0132] likeFigure 3 Fig. 1 shows a flowchart of the process of feature analysis on sensor data in the above embodiment. The process includes two parts of fault early warning and same route detection.

[0133] First, the sensor data of the fiber cores 1 and 2 are input into the neural network model library for analysis, which includes a spatio-temporal graph neural network and a transfer learning model trained based on the spatio-temporal graph neural network, and can output fault information. Then, the pre-set acoustic spectrum model library can further verify the fault information. If the secondary verification is passed, the fault information of the fiber cores 1 and 2 can be obtained. Then, based on the obtained spatio-temporal heterogeneous graphs and fault information of the two fiber cores, a feature matching algorithm is used for processing to determine whether the two fiber cores belong to the same route, thereby completing the same route detection.

[0134] In one exemplary embodiment, the optical cable fault early warning method provided by the embodiment of the present application specifically includes the following steps:

[0135] The control sensor synchronously sends an optical pulse signal to the two optical cables to be detected, and receives sensor data corresponding to each optical cable to be detected returned by the sensor;

[0136] In the case of receiving a target event processing instruction, the sensor data is mapped into a spatio-temporal heterogeneous graph through a transfer learning model, and the spatio-temporal heterogeneous graph is analyzed for feature evolution to generate fault information, wherein the transfer learning model is obtained by transfer learning training of a spatio-temporal graph neural network through a target event training sample;

[0137] In the case of not receiving a target event processing instruction, the sensor data is mapped into a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, and the spatio-temporal heterogeneous graph is analyzed for feature evolution to generate fault information, wherein the fault information includes a fault category, a fault time, and a fault position;

[0138] According to the fault time and the fault position, a to-be-detected vibration signal is read from the sensor data, and a to-be-detected acoustic spectrum graph of the to-be-detected vibration signal is generated;

[0139] A sample acoustic spectrum graph corresponding to the fault category is obtained from a pre-set acoustic spectrum model library;

[0140] The first similarity between the to-be-detected acoustic spectrum graph and the sample acoustic spectrum graph is compared, and in the case that the first similarity is higher than a first pre-set threshold, feature matching is performed based on the spatio-temporal heterogeneous graphs of any two optical cables to be detected to obtain a second similarity;

[0141] In the case that the second similarity is higher than a second pre-set threshold, it is determined that any two optical cables to be detected have a same route risk, and the optical cables to be detected are fault warned.

[0142] The scheme provided in the application can early identify potential faults, and then realize fault early warning, so that maintenance measures are actively taken before the fault occurs, effectively reducing operation and maintenance costs and improving the reliability and stability of the optical cable network, and reducing the optical cable operation and maintenance cost.

[0143] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps. It can be understood that the steps in different embodiments can be freely combined as needed, and various non-contradictory schemes formed by combination are within the scope of protection of the application.

[0144] Based on the same inventive concept, the embodiments of the application also provide an optical cable fault early warning device for implementing the above-mentioned optical cable fault early warning method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more optical cable fault early warning device embodiments provided below can refer to the limitations of the optical cable fault early warning method in the above, which will not be repeated here.

[0145] In one exemplary embodiment, as shown in Figure 4 An optical cable fault early warning device is provided, comprising:

[0146] The acquisition module 401 is configured to acquire sensor data for the optical cable to be detected.

[0147] The analysis module 402 is configured to map the sensor data to a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, and perform feature evolution analysis on the spatio-temporal heterogeneous graph to generate fault information.

[0148] The early warning module 403 is configured to perform fault early warning on the optical cable to be detected based on the fault information.

[0149] In one exemplary embodiment, the analysis module 402 is specifically configured to:

[0150] mapping the sensor data into a spatio-temporal heterogeneous graph, wherein nodes of the spatio-temporal heterogeneous graph represent preset spatial positions on the optical cable to be detected, and edges between the nodes represent connection relationships between the preset spatial positions;

[0151] For each node, performing spatial dimension feature analysis on sensor data corresponding to the node and other nodes adjacent to the node to generate spatial correlation features;

[0152] For each node, performing time dimension feature evolution analysis on sensor data corresponding to the node to generate time evolution features;

[0153] Performing fault analysis according to the spatial correlation features and the time evolution features to generate fault information.

[0154] In an exemplary embodiment, the analysis module 402 is specifically configured to:

[0155] In the case of receiving a target event processing instruction, mapping the sensor data into a spatio-temporal heterogeneous graph through a transfer learning model, and performing feature evolution analysis on the spatio-temporal heterogeneous graph to generate fault information;

[0156] The transfer learning model is obtained by transfer learning training of a spatio-temporal graph neural network through a target event training sample.

[0157] In an exemplary embodiment, the fault information includes a fault category, a fault time, and a fault position, and the early warning module 403 is specifically configured to:

[0158] According to the fault time and the fault position, reading a to-be-detected vibration signal from the sensor data, and generating a to-be-detected spectrogram of the to-be-detected vibration signal;

[0159] Obtaining a sample spectrogram corresponding to the fault category from a preset spectrogram model library;

[0160] Comparing a first similarity between the to-be-detected spectrogram and the sample spectrogram, and in the case that the first similarity is higher than a first preset threshold, performing fault early warning on the optical cable to be detected.

[0161] In an exemplary embodiment, the obtaining module 401 is specifically configured to:

[0162] Obtaining sensor data for at least two optical cables to be detected, the sensor data being collected based on synchronously transmitted optical pulse signals;

[0163] After mapping the sensor data into a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, the method further includes:

[0164] perform feature matching based on the space-time heterogeneous graphs of any two optical cables to be detected to obtain a second similarity;

[0165] In a case where the second similarity is higher than a second preset threshold, it is determined that the any two optical cables to be detected have the same routing risk.

[0166] In an exemplary embodiment, the acquisition module 401 is specifically configured to:

[0167] The sensor is controlled to synchronously send an optical pulse signal to each optical cable to be detected, and sensor data corresponding to each optical cable to be detected returned by the sensor is received.

[0168] In an exemplary embodiment, the sensor includes at least one of a distributed optical fiber acoustic wave sensor, a Brillouin optical time domain reflectometer, and a Raman optical time domain reflectometer.

[0169] In an exemplary embodiment, the fault information includes a fault position, and the early warning module 403 is specifically configured to:

[0170] Based on the fault position, a pre-warning area of the optical cable to be detected is determined, and the pre-warning area is tracked and monitored.

[0171] The above-mentioned modules in the optical cable fault early warning device can be all or partially realized by software, hardware, and combinations thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned modules.

[0172] In an exemplary embodiment, a computer device is provided, which can be a server, and an internal structure diagram thereof can be as shown in Figure 5 The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured 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 operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is configured to store measurement data and / or positioning information. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement an optical cable fault early warning method.

[0173] Those skilled in the art can understand that, Figure 5 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0174] In one exemplary embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the following steps:

[0175] Obtaining sensor data for a to-be-detected optical cable;

[0176] Mapping the sensor data into a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, and performing feature evolution analysis on the spatio-temporal heterogeneous graph to generate fault information;

[0177] Based on the fault information, performing fault warning on the to-be-detected optical cable.

[0178] In one embodiment, a computer readable storage medium is provided, having a computer program stored thereon, and the computer program is executed by a processor to implement the following steps:

[0179] Obtaining sensor data for a to-be-detected optical cable;

[0180] Mapping the sensor data into a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, and performing feature evolution analysis on the spatio-temporal heterogeneous graph to generate fault information;

[0181] Based on the fault information, performing fault warning on the to-be-detected optical cable.

[0182] In one embodiment, a computer program product is provided, comprising a computer program, and the computer program is executed by a processor to implement the following steps:

[0183] Obtaining sensor data for a to-be-detected optical cable;

[0184] Mapping the sensor data into a spatio-temporal heterogeneous graph through a spatio-temporal graph neural network, and performing feature evolution analysis on the spatio-temporal heterogeneous graph to generate fault information;

[0185] Based on the fault information, performing fault warning on the to-be-detected optical cable.

[0186] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0187] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related 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 above-mentioned embodiments of each method. In the embodiments provided in the present application, any reference to memory, database or other medium can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0188] Any technical features in the above embodiments can be combined, and for the sake of brevity, not all possible combinations are described above, however, any combination of these technical features is deemed to be within the scope of the present application.

[0189] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for early warning of optical cable faults, characterized in that, The method includes: Acquire sensor data for the optical cable under test; The sensor data is mapped into a spatiotemporal heterogeneous graph using a spatiotemporal graph neural network, and feature evolution analysis is performed on the spatiotemporal heterogeneous graph to generate fault information. Based on the fault information, a fault warning is issued for the optical cable under test.

2. The method according to claim 1, characterized in that, The process involves mapping the sensor data into a spatiotemporal heterogeneous graph using a spatiotemporal graph neural network, performing feature evolution analysis on the spatiotemporal heterogeneous graph, and generating fault information, including: The sensor data is mapped into a spatiotemporal heterogeneous graph, wherein the nodes of the spatiotemporal heterogeneous graph represent preset spatial locations on the optical cable to be detected, and the edges between the nodes represent the connection relationships between the preset spatial locations; For each node, spatial feature analysis is performed on the sensor data corresponding to that node and other adjacent nodes to generate spatial correlation features; Perform time-dimensional feature evolution analysis on the sensor data corresponding to each node to generate time evolution features; Fault analysis is performed based on the spatial correlation characteristics and the temporal evolution characteristics to generate fault information.

3. The method according to claim 1, characterized in that, The process involves mapping the sensor data into a spatiotemporal heterogeneous graph using a spatiotemporal graph neural network, performing feature evolution analysis on the spatiotemporal heterogeneous graph, and generating fault information, including: Upon receiving a target event processing instruction, the sensor data is mapped into a spatiotemporal heterogeneous graph using a transfer learning model, and feature evolution analysis is performed on the spatiotemporal heterogeneous graph to generate fault information. The transfer learning model is obtained by training a spatiotemporal graph neural network using target event training samples.

4. The method according to claim 1, characterized in that, The fault information includes fault type, fault time, and fault location. Based on this fault information, the step of providing a fault warning for the optical cable under test includes: Based on the fault time and the fault location, the vibration signal to be detected is read from the sensor data, and a detection spectrogram of the vibration signal to be detected is generated. Obtain the sample acoustic spectrum corresponding to the fault category from the preset acoustic spectrum model library; By comparing the first similarity between the spectrum to be detected and the sample spectrum, and if the first similarity is higher than a first preset threshold, a fault warning is issued for the optical cable to be detected.

5. The method according to claim 1, characterized in that, The acquisition of sensor data for the optical cable under test includes: Acquire sensor data for at least two optical cables to be tested, the sensor data being acquired based on synchronously transmitted optical pulse signals; After mapping the sensor data into a spatiotemporal heterogeneous graph using a spatiotemporal graph neural network, the method further includes: Feature matching is performed based on the spatiotemporal heterogeneity maps of any two optical cables to be detected to obtain the second similarity. If the second similarity is higher than the second preset threshold, it is determined that any two optical cables to be detected have the risk of sharing the same route.

6. The method according to claim 5, characterized in that, The acquisition of sensor data for at least two optical cables under test includes: The control sensor synchronously sends optical pulse signals to each optical cable under test and receives sensor data corresponding to each optical cable under test returned by the sensor.

7. The method according to claim 6, characterized in that, The sensor includes at least one of a distributed fiber optic acoustic sensor, a Brillouin optical time-domain reflectometer, and a Raman optical time-domain reflectometer.

8. The method according to claim 1, characterized in that, The fault information includes the fault location, and the fault warning for the optical cable under test based on the fault information includes: Based on the fault location, the warning area of ​​the optical cable to be tested is determined, and the warning area is tracked and monitored.

9. A fiber optic cable fault early warning device, characterized in that, The device includes: The acquisition module is used to acquire sensor data for the optical cable under test; The analysis module is used to map the sensor data into a spatiotemporal heterogeneous graph through a spatiotemporal graph neural network, and to perform feature evolution analysis on the spatiotemporal heterogeneous graph to generate fault information. The early warning module is used to provide early warning of faults to the optical cable under test based on the fault information.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images