Electric power communication optical cable reliability evaluation method and drive route optimization method

Abstract

The invention relates to an electric power communication optical cable reliability evaluation method and a drive route optimization method, and the method comprises the following steps: collecting historical multi-source data of a target electric power communication network, carrying out the preprocessing, and carrying out the calculation of the historical multi-source data based on the preprocessed historical multi-source data; constructing historical space-time diagram data with electric power communication network sites as nodes of the diagram and optical cables as edges; constructing a network correction coefficient calculation model of a fusion graph attention layer and a gating circulation unit, and pre-training the network correction coefficient calculation model through historical space-time graph data; constructing an optical cable reliability scoring model, wherein the model calculates the reliability score of the optical cable through the optical cable characteristics and the network correction coefficient; and obtaining multi-source data of the target electric power communication network at the current moment, performing preprocessing, generating time-space diagram data, inputting the time-space diagram data into the trained network correction coefficient calculation model, outputting a network correction coefficient of each section of optical cable, and calculating a reliability score.

Description

Technical Field

[0001] This invention relates to a reliability assessment method and a drive route optimization method for power communication optical cables, belonging to the field of power communication optical cable optimization technology. Background Technology

[0002] Communication systems are a crucial foundation for power grid production and control operations, and their operational stability is vital to improving the reliability of the large power grid. With the advancement of new power system construction, power communication networks face numerous challenges. In coastal areas, power communication optical cables must not only withstand conventional service loads but also cope with the impact of extreme weather conditions such as typhoons and torrential rains.

[0003] Currently, the operation and maintenance management of power communication optical cables faces numerous challenges. Traditional reliability analysis methods suffer from limitations such as a narrow range of indicator selection and manual weighting, making it difficult to accurately assess the actual reliability of optical cables under complex environments. In terms of service routing planning, existing technologies often separate optical cable reliability analysis from service routing, failing to achieve an organic integration of the two. This results in the inability to promptly optimize and adjust service routes when optical cables experience failures or performance degradation, thus impacting service continuity. Furthermore, traditional models cannot quantify topology cascading risks, making it difficult to predict the successive failure of multiple optical cable segments under extreme weather conditions, leading to insufficient disaster resilience of the communication network. In addition, existing digital twin systems in power communication applications have not formed a "prediction-decision" closed loop, resulting in low operation and maintenance efficiency and failing to meet the higher requirements of new power systems for communication network support capabilities. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention proposes a reliability-driven routing optimization method and system for power communication optical cables.

[0005] The technical solution of the present invention is as follows: On the one hand, this invention proposes a reliability assessment method for power communication optical cables, comprising the following steps: Collect and preprocess historical multi-source data of the target power communication network. Based on the preprocessed historical multi-source data, construct a historical spatiotemporal graph with power communication network sites as nodes and optical cables as edges. A network correction coefficient calculation model integrating a graph attention layer and a gated recurrent unit is constructed, and the model is pre-trained using historical spatiotemporal graph data; the network correction coefficients are used to correct the reliability of optical cables. A reliability scoring model for optical cables is constructed, which calculates the reliability score of optical cables based on optical cable characteristics and network correction coefficients. The system acquires multi-source data of the target power communication network at the current moment and preprocesses it to generate spatiotemporal map data, which is then input into the trained network correction coefficient calculation model. Finally, it outputs the network correction coefficient for each optical cable segment and calculates the reliability score.

[0006] In a preferred embodiment, the multi-source data includes: Topology data, optical cable performance data, meteorological data, and service data of power communication networks; The topology data of the power communication network includes site data, optical cable connection relationships, number of nodes, and adjacency matrix; The optical cable performance data includes optical cable type, optical power attenuation value, transmission delay, load rate, and service life; The meteorological data includes temperature, humidity, and wind speed data obtained from meteorological stations; The business data includes business scheduling priorities and historical fault data; The preprocessing step specifically involves normalizing the multi-source data.

[0007] In a preferred embodiment, the network correction coefficient calculation model includes: The graph attention layer is used to input spatiotemporal graph data, calculate the edge attention weight of each optical cable segment based on the spatiotemporal graph data, and output the updated feature vector of each graph node based on the calculated edge attention weight. The gated recurrent unit is used to input the updated feature vector of each graph node, and updates the gate value based on the updated feature vectors of the graph nodes at both ends of each optical cable and the hidden state vector of the corresponding optical cable at the previous time step, which serves as the network correction coefficient.

[0008] As a preferred embodiment, the gated loop unit also incorporates meteorological influencing factors; The meteorological impact factor is calculated using meteorological data and participates in the gating value update step, specifically as follows: The gated recurrent unit takes the updated feature vector of each graph node as input, and updates the gate value based on the updated feature vectors of the graph nodes at both ends of each optical cable and the hidden state vector of the corresponding optical cable at the previous time step. This value serves as the network correction coefficient, where: The input to the control loop unit is: ; in, Indicates the current time node The feature vector updated after the graph attention layer Indicates the current time node The feature vector updated after the graph attention layer; Indicates the relationship between nodes and The edge feature matrix of the optical cable; These are meteorological influencing factors.

[0009] In a preferred embodiment, the optical cable reliability scoring model extracts the optical cable features of the target optical cable from multi-source data and calculates a basic score value based on the optical cable features; then, a weighted calculation is performed based on the basic score value and the network correction coefficient to obtain the reliability score of the corresponding optical cable.

[0010] On the other hand, the present invention also proposes a reliability-driven routing optimization method for power communication optical cables, comprising the following steps: The reliability score of each optical cable segment in the target power communication network at the current moment is obtained by the power communication optical cable reliability assessment method as described in any embodiment of the present invention. A routing cost function is constructed based on the reliability score of the optical cable and service data; The optimal optical cable route path is obtained by solving the routing cost function using Dijkstra's algorithm.

[0011] In a preferred embodiment, the routing cost function is specifically: ; in, Indicate candidate path The routing cost function, Indicates optical cable Reliability rating This indicates the service level coefficient of the corresponding optical cable. This is the priority coefficient for service scheduling.

[0012] In a preferred embodiment, the method further includes: Constructing a digital twin of the power communication network; The generated optimized optical cable routing path is input into the digital twin of the power communication network for simulation verification. If the service interruption probability obtained from the simulation exceeds the set threshold, the network correction coefficient calculation model is retrained; if it does not exceed the set threshold, the communication strategy is executed according to the optimized optical cable routing path.

[0013] In another aspect, the present invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the power communication optical cable reliability assessment method as described in any embodiment of the present invention.

[0014] In another aspect, the present invention also proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the power communication optical cable reliability assessment method as described in any embodiment of the present invention.

[0015] The beneficial effects of this invention are as follows: This invention significantly improves the stability and transmission efficiency of power communication networks by comprehensively considering the reliability score of optical cables and service requirements. Simultaneously, it can dynamically adapt to changes in network conditions, ensuring optimal routing selection under various operating conditions. Furthermore, through in-depth mining of historical data and continuous model optimization, the system's predictive capabilities and decision-making accuracy are further enhanced, providing strong technical support for the safe and reliable operation of power communication networks.

[0016] Additional aspects and advantages of the invention will be set forth in the following description, and some of them will be obvious from the description, or may be learned by practicing the invention. Furthermore, various aspects and advantages of the invention may be realized and obtained by means of method steps and combinations particularly pointed out in the appended claims. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the method flow of Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the training of the network correction coefficient calculation model in an embodiment of the present invention; Figure 3 This is a schematic diagram of the method flow of Embodiment 2 of the present invention; Figure 4 This is an overall flowchart of routing optimization driven by optical cable reliability scoring in an embodiment of the present invention. Detailed Implementation

[0018] 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.

[0019] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0020] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0022] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0023] Example 1: See Figure 1 This embodiment proposes a reliability assessment method and a drive routing optimization method for power communication optical cables, including the following steps: S100. Collect historical multi-source data of the target power communication network, including network topology data, optical cable performance data, meteorological data, and service data; among which: The topology data of power communication networks includes sites, optical cable connections, number of nodes, and adjacency matrix; The site data includes the highest voltage level of the substation and the number of protection and security control services for site equipment loads above 220kV. The optical cable performance data includes optical cable type, optical power attenuation value, transmission delay, load rate, and service life; The meteorological data includes temperature, humidity, and wind speed data obtained from meteorological stations; The service data includes service scheduling priorities (classified according to line level, such as 10kV, 500kV, etc.) and historical fault data (such as the number of historical interruptions). The collected multi-source data is preprocessed, specifically by normalizing each source and mapping it to the [0,1] interval using Min-Max standardization. Based on the preprocessed historical multi-source data, a historical spatiotemporal graph is constructed, with power communication network sites as nodes and optical cables as edges, containing feature matrices of site node sets, optical cable edge sets, nodes, and edges; including: Node set Each node corresponds to a power communication network site; Edge set: Each edge corresponds to a segment of optical fiber, connecting the two nodes. ; Node feature matrix: for Tensors of dimension Let M be the node feature space, and M be the number of stations. This represents the number of time sampling points (once every 15 minutes). For site feature dimensions, The dimensions of the site data are determined by standardizing the site data, and the feature data in the node feature matrix is ​​also obtained from the standardized site data.

[0024] Edge feature matrix: for Tensors of dimension Let Fe be the edge feature space, N be the number of optical cables, and Fe be the feature dimension of the optical cables. This represents the number of time sampling points, synchronized with the nodes. The dimensions of the optical cable performance data are determined by the standardized optical cable performance data, and the feature data in the edge feature matrix are also obtained by the standardized optical cable performance data.

[0025] S200, for details, see S200. Figure 2 A network correction coefficient calculation model (ST-GATGRU model) integrating graph attention layer and gated recurrent unit was constructed, and the network correction coefficient calculation model was pre-trained using historical spatiotemporal graph data.

[0026] The graph attention layer is used to input spatiotemporal graph data and calculates the edge attention weights for each segment of the optical cable based on the spatiotemporal graph data. The specific formula is as follows: ; in For nodes The node feature matrix at the current moment, Let be the edge feature matrix of the optical cable between the two nodes. and To make the weight matrix learnable and uniformly aligned to the same dimension, Attention coefficient For normalization function, This is the activation function.

[0027] Based on the calculated attention weights for each side, the features of the neighboring nodes of a node are weighted and aggregated to output the updated feature vector for each node in the graph. ; The gated recurrent unit (GRU) is used to input the updated feature vector of each graph node. It updates the gate value based on the updated feature vectors of the graph nodes at both ends of each optical cable and the hidden state vector of the corresponding optical cable from the previous time step. This value serves as a network correction coefficient. The GRU time-series update mechanism mainly includes the following: The input vector is: ; in, Indicates the current time node The feature vector updated after the graph attention layer Indicates the current time node The feature vector updated after the graph attention layer; Indicates the relationship between nodes and The edge feature matrix of the optical cable; These are meteorological influencing factors.

[0028] The formula for resetting the door is: ; This is the activation function (usually the sigmoid function). The hidden state vector of the optical cable selected at the previous moment (i.e., the timing state maintained by the GRU unit). To reset the trainable weight matrix of the gate, To reset the door offset; Used to control historical states The degree of forgetting.

[0029] The updated gate formula is: ; As the network correction coefficient output To update the trainable weight matrix of the gate, To update the door offset.

[0030] The candidate state formula is: ; tanh is the hyperbolic tangent activation function. Its role in the candidate states is to filter out valid information from historical states. The trainable weight matrix computed for the candidate states. The bias is calculated for the candidate state.

[0031] The formula for updating the hidden state is: ; Balanced historical state With candidate state Final output This will be the hidden state for the next moment.

[0032] This embodiment introduces a meteorological impact factor into the gated loop unit. Calculated using meteorological data, this factor effectively learns long-term degradation patterns and addresses the impact of complex weather conditions on optical cable performance. The specific calculation steps for the meteorological impact factor are as follows: First, acquire temperature, humidity, and wind speed data, and then perform standardization processing on the single-dimensional meteorological data. (For example, wind speed) is standardized using Min-Max: ,in , These are the minimum and maximum values ​​of this parameter for the same historical period.

[0033] The standardized meteorological data are weighted and fused to obtain the meteorological impact factor: ; in, , and These are the preset weighting factors.

[0034] This embodiment incorporates meteorological influencing factors into the gating decision, enhancing the model's sensitivity to optical power attenuation and physical damage to optical cables during sudden events such as typhoons. For example, when... When it rises sharply, It will suppress historical state The impact of meteorological changes on optical cable reliability should be studied first, so as to capture the coupling relationship between short-term failure risk and long-term degradation pattern.

[0035] S300. Construct an optical cable reliability scoring model. This model calculates the reliability score of the optical cable using optical cable characteristics and network correction coefficients. This reliability score is mapped to the failure rate of the optical cable.

[0036] Specifically, the optical cable reliability scoring model extracts the optical cable features of the target optical cable from multi-source data and calculates a basic score value based on these features. Then, a weighted calculation is performed based on the basic score value and a network correction coefficient to obtain the reliability score of the corresponding optical cable. In this embodiment, the weighting formula is: ; The base score is out of 100 points and is generated by taking into account the optical cable type (OPGW, ADSS), years of operation, and number of historical failures through a preset method. In this embodiment, the default value of the base score is set to 100 points. When the operating years are more than 10 years, 40 points are deducted; when the operating years are 5-10 years, 20 points are deducted; and no points are deducted within 5 years. If the number of failures in a year exceeds 5, 20 points are deducted; if the number of failures is less than 5, 10 points are deducted; and if there are no failures, no points are deducted. If the optical cable type is OPGW, the deduction value of the optical cable is multiplied by a weight of 1.0; if the optical cable type is ADSS, the deduction value of the optical cable is multiplied by a weight of 1.25.

[0037] The network correction coefficients are output from the network correction coefficient calculation model in subsequent steps.

[0038] The formula for mapping reliability score to optical cable failure rate is as follows: ; in, This represents the failure rate. By mapping failure probabilities, a basis for subsequent decision-making can be provided.

[0039] S400: Acquire multi-source data of the target power communication network at the current moment and preprocess it to generate spatiotemporal map data and input it into the trained network correction coefficient calculation model. Output the network correction coefficient of each optical cable segment and calculate the reliability score.

[0040] By integrating multi-source data, including the physical state of optical cables, communication quality indicators, and external environmental information, the model can dynamically adjust the weight distribution of network correction coefficients. In practical applications, this method effectively improves the adaptability and accuracy of the scoring system. For example, when a section of optical cable has a high historical failure rate and current weather conditions are severe, its reliability score will significantly decrease, thus providing a more reliable basis for route optimization. Furthermore, the model supports comparative analysis of data from different time periods to further uncover potential risk patterns.

[0041] This embodiment significantly improves the stability and transmission efficiency of the power communication network by comprehensively considering the reliability score of the optical cable and service requirements. Simultaneously, it can dynamically adapt to changes in network conditions, ensuring optimal routing selection under various operating conditions. Furthermore, through in-depth mining of historical data and continuous model optimization, the system's predictive capabilities and decision-making accuracy are further enhanced, providing strong technical support for the safe and reliable operation of the power communication network.

[0042] Example 2: See Figure 3 and Figure 4 This embodiment provides a reliability-driven routing optimization method for power communication optical cables. A100. Obtain the reliability score of each optical cable segment in the target power communication network at the current moment by means of the power communication optical cable reliability assessment method as described in any embodiment of the present invention. A200. Construct a routing cost function based on the optical cable's reliability score and service data. The routing cost function is specifically as follows: ; in, Indicate candidate path The routing cost function, Indicates optical cable Reliability rating This indicates the service level coefficient of the corresponding optical cable (1.5 for 500kV and above, 1.2 for 220kV, and 1.0 for 110kV and below). This is the priority coefficient for business scheduling (1.0 for security control services and 0.8 for other services).

[0043] The A300 uses the classic Dijkstra's shortest path algorithm to find the optimal optical cable route based on a specific routing cost function. The implementation process of this method mainly includes the following steps: (1) Constructing the cost matrix With node set As a dimension, construct Cost adjacency matrix Matrix elements Defined as: if node and Directly through fiber optic cable Connect, then If node and If there is no direct fiber optic connection, then (Indicates not directly reachable); diagonal elements (The cost from a node to itself is 0).

[0044] (2) Initialize parameters Set the starting point of the path and the end point ; Initialize the distance record structure: for each node Maintain a list Used to store data from the starting point To the node The cost values ​​of the first K lowest cost paths (initially) The remaining nodes (List length is K). Initialize the predecessor record structure: for each node And the cost of each path, recording the nodes in the path. The predecessor node is used to trace the complete path.

[0045] (3) Iterative calculation of path cost Use a priority queue (min-heap) to store the pending "node-cost" pairs, initially... Join the queue (indicating the starting point) The corresponding initial cost is 0). Retrieve the node-cost pair with the lowest current cost from the queue. ( For the current node, From the starting point to Path cost); Traversing nodes All adjacent nodes Calculate the distance from the starting point arrive New path cost: ; like Smaller than node of If the Kth largest cost value in the list (or the list length is less than K), then... insert And keep the list ordered (in ascending order of cost), while recording The predecessor node is ; Will Add to the priority queue and continue iterating until the queue is empty or the endpoint is reached. of The costs of K valid paths have been recorded.

[0046] (4) Select K feasible paths with the lowest cost From the finish line of The K smallest cost values ​​are selected from the list, and the corresponding paths are the candidate path set. Tracing back the complete node sequence of each path using the predecessor record structure (from the endpoint) Tracing back to the starting point Finally, we obtain K lowest-cost feasible paths from the starting point to the ending point.

[0047] Finally, to ensure the reliability of the routing scheme, an N-1 security check mechanism is introduced to verify the robustness of the selected path under network failure scenarios. It should be noted that Dijkstra's algorithm, as a mature graph theory algorithm, has been extensively studied in related fields regarding its basic principles and standard implementation process; therefore, this embodiment will not repeat the specific details of the algorithm.

[0048] In a preferred embodiment of this invention, the method further includes: Constructing a digital twin of the power communication network; The generated optimized optical cable routing path is input into the digital twin of the power communication network for simulation verification. If the service interruption probability obtained from the simulation exceeds the set threshold, the network correction coefficient calculation model is retrained; if it does not exceed the set threshold, the communication strategy is executed according to the optimized optical cable routing path.

[0049] The simulation verification process described above effectively assesses the feasibility and reliability of the optimized path in practical applications. Furthermore, combining digital twin technology with routing optimization methods not only enhances the scientific rigor of the solution but also provides intuitive decision-making support for network planning and operation. In addition, analysis of the simulation results allows for the identification of potential risks and the targeted adjustment of relevant parameters or strategies to ensure the communication network maintains efficient and stable operation. This closed-loop optimization mechanism significantly enhances the system's adaptability and robustness, laying the foundation for the long-term reliable operation of the power communication network.

[0050] To help those skilled in the art quickly understand the principles of this embodiment, a specific example is provided below: Taking a 500kV power communication optical cable network in a coastal area as an example, the topology data of the optical cable network in this area was collected, including the connection relationships of 12 site nodes and the adjacency matrix; real-time optical cable performance data was acquired, such as the maximum standard deviation of optical power of the AB section optical cable in the past month being 2.51%, the current average transmission delay being 12ms, and the load rate being 78%; meteorological data was collected, as it was currently typhoon season, and the typhoon path passed through the AB section area, with a wind speed of 15m / s and humidity of 92%; in terms of service data, the network carries 3 protection service channels, among which the stable service scheduling has the highest priority, and the current path redundancy is 2 backup paths. After performing Min-Max standardization preprocessing on these data, spatiotemporal map data was constructed.

[0051] The preprocessed spatiotemporal map data was input into the constructed ST-GATGRU model for training. Historical data from the past three typhoon seasons (June-September) were used as training samples, including 20 cases of fiber optic cable failures caused by typhoons. A loss function was set. ,in =0.3, set during iterative optimization. =400, learning rate 0.01, using The optimizer was used to bring the loss to 0.028, completing the model training.

[0052] Based on the trained model, reliability scores are assigned to each optical cable segment. Taking segment AB as an example, its basic reliability score is calculated, and combined with spatiotemporal attenuation coefficients such as those caused by typhoons, the final reliability score is obtained. The score is 38, which is used to map the failure probability according to the formula. The reliability score was 40.62%; the CD segment reliability score was 75 points, and the failure probability was 32.09%.

[0053] Based on optical cable reliability score and service level coefficient (protection service) =1.5) and service scheduling priority (security control services) =1.2) Construct the routing cost function. Dijkstra's algorithm is used to calculate the candidate paths, and the optimal path is finally obtained as AECD, with a total cost of 0.86.

[0054] A digital twin of the regional power communication network was constructed, and the optimal path scheme AECD was input for simulation verification. The simulation results showed that the service interruption probability met the requirements, and the final decision was output: switch the stable service to the AECD path.

[0055] Example 3: This embodiment proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the power communication optical cable reliability assessment method as described in any embodiment of the present invention.

[0056] Example 4: This embodiment proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the power communication optical cable reliability assessment method as described in any embodiment of the present invention.

[0057] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0058] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0059] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0060] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0061] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A power communications optical cable reliability evaluation method characterized by, Includes the following steps: Collect and preprocess historical multi-source data of the target power communication network. Based on the preprocessed historical multi-source data, construct a historical spatiotemporal graph with power communication network sites as nodes and optical cables as edges. A network correction coefficient calculation model integrating a graph attention layer and a gated recurrent unit is constructed, and the model is pre-trained using historical spatiotemporal graph data; the network correction coefficients are used to correct the reliability of optical cables. A reliability scoring model for optical cables is constructed, which calculates the reliability score of optical cables based on optical cable characteristics and network correction coefficients. The system acquires multi-source data of the target power communication network at the current moment and preprocesses it to generate spatiotemporal map data, which is then input into the trained network correction coefficient calculation model. Finally, it outputs the network correction coefficient for each optical cable segment and calculates the reliability score.

2. The method of claim 1, wherein, The multi-source data includes: Topology data, optical cable performance data, meteorological data, and service data of power communication networks; The topology data of the power communication network includes site data, optical cable connection relationships, number of nodes, and adjacency matrix; The optical cable performance data includes optical cable type, optical power attenuation value, transmission delay, load rate, and service life; The meteorological data includes temperature, humidity, and wind speed data obtained from meteorological stations; The business data includes business scheduling priorities and historical fault data; The preprocessing step specifically involves normalizing the multi-source data.

3. The method of claim 2, wherein, In the network correction coefficient calculation model: The graph attention layer is used to input spatiotemporal graph data, calculate the edge attention weight of each optical cable segment based on the spatiotemporal graph data, and output the updated feature vector of each graph node based on the calculated edge attention weight. The gated recurrent unit is used to input the updated feature vector of each graph node, and updates the gate value based on the updated feature vectors of the graph nodes at both ends of each optical cable and the hidden state vector of the corresponding optical cable at the previous time step, which serves as the network correction coefficient.

4. The method of claim 3, wherein, Meteorological influencing factors are also introduced into the gated loop unit; The meteorological impact factor is calculated using meteorological data and participates in the gating value update step, specifically as follows: The gated recurrent unit takes the updated feature vector of each graph node as input, and updates the gate value based on the updated feature vectors of the graph nodes at both ends of each optical cable and the hidden state vector of the corresponding optical cable at the previous time step. This value serves as the network correction coefficient, where: The input to the gated loop unit is: ； wherein, represents the current time node the feature vector updated by the graph attention layer, represents the current time node the feature vector updated by the graph attention layer; represents the inter-node and the edge feature matrix of the inter-optical cable; is a meteorological influence factor.

5. The reliability assessment method for power communication optical cables according to claim 1, characterized in that: The optical cable reliability scoring model extracts the optical cable features of the target optical cable from multi-source data and calculates a basic score value based on the optical cable features; then, a weighted calculation is performed based on the basic score value and the network correction coefficient to obtain the reliability score of the corresponding optical cable.

6. A reliability-driven routing optimization method for power communication optical cables, characterized in that, Includes the following steps: The reliability score of each optical cable segment in the target power communication network at the current moment is obtained by the power communication optical cable reliability assessment method as described in any one of claims 1 to 5. A routing cost function is constructed based on the reliability score of the optical cable and service data; The optimal optical cable route path is obtained by solving the routing cost function using Dijkstra's algorithm.

7. The reliability-driven routing optimization method for power communication optical cables according to claim 6, characterized in that, The routing cost function is specifically as follows: ； in, Indicate candidate path The routing cost function, Indicates optical cable Reliability rating This indicates the service level coefficient of the corresponding optical cable. This is the priority coefficient for service scheduling.

8. The reliability-driven routing optimization method for power communication optical cables according to claim 6, characterized in that, The method further includes: Constructing a digital twin of the power communication network; The generated optimized optical cable routing path is input into the digital twin of the power communication network for simulation verification. If the service interruption probability obtained from the simulation exceeds the set threshold, the network correction coefficient calculation model is retrained; if it does not exceed the set threshold, the communication strategy is executed according to the optimized optical cable routing path.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the power communication optical cable reliability assessment method as described in any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the power communication optical cable reliability assessment method as described in any one of claims 1 to 5.

Images