Cloud side-end collaborative high-voltage cable intelligent monitoring and early warning method and system

Abstract

The invention relates to the related technical field of cable monitoring, in particular to a cloud-side-end collaborative high-voltage cable intelligent monitoring and early warning method and system, and the method comprises the steps: carrying out the local preprocessing of a data collection result, building the local risk distribution of a cable, and transmitting the local risk distribution to an edge calculation unit; according to historical data and a data acquisition result, extreme state inversion and future risk fluidization prediction are carried out, collaborative risk assessment and global risk assessment are established, multi-scale game optimization is carried out on cable local risk distribution, collaborative risk assessment and global risk assessment, and a monitoring and early warning result is output. The technical problem that the local risk urgency and the global risk diffusion trend cannot be balanced because local risk judgment is carried out only through single node data is solved, local risk distribution, collaborative risk assessment and global risk assessment of the coupling cable are realized, an end-edge-cloud data closed loop is formed, and the reliability of the system is improved. The real-time performance, the accuracy and the global collaboration of high-voltage cable monitoring and early warning are comprehensively improved, and the technical effect of safe and stable operation of the high-voltage cable is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of cable monitoring technology, specifically to a cloud-edge-device collaborative intelligent monitoring and early warning method and system for high-voltage cables. Background Technology

[0002] The safe and stable operation of high-voltage cables is directly related to the reliability of power grid supply and the efficiency of energy transmission. However, the current high-voltage cable monitoring model suffers from insufficient architecture coordination, with gaps in coordination between the end side, edge, and cloud. This makes it impossible to respond to high-voltage cable faults in real time, and monitoring is prone to failure when communication is interrupted in remote areas. In addition, risk assessment is often based on data from a single node, and the risk assessment system lacks multi-scale fusion capabilities. This makes it difficult to meet the core monitoring needs of real-time status perception, accurate risk warning, and rapid fault handling. The risk prediction capability is weak, and it cannot provide effective support for cable condition-based maintenance and fault handling.

[0003] In summary, existing technologies suffer from the problem of relying solely on data from a single node for local risk assessment, which fails to balance the urgency of local risks with the overall trend of risk spread. Summary of the Invention

[0004] This application provides a cloud-edge-device collaborative intelligent monitoring and early warning method and system for high-voltage cables, aiming to solve the technical problem in the prior art that the assessment of local risks based solely on data from a single node cannot balance the urgency of local risks with the global risk diffusion trend.

[0005] In view of the above problems, the technical solution to achieve the present application is as follows: In its first aspect, this application provides a cloud-edge-device collaborative intelligent monitoring and early warning method for high-voltage cables. The method includes: deploying multimodal sensors at the cable nodes of the high-voltage cable and using an end-side sensing unit to control the multimodal sensors to perform event-driven data acquisition; performing local preprocessing on the data acquisition results to establish a local risk distribution for the cable; deploying an edge computing unit in a ring network side cabinet to send the local risk distribution of the cable to the edge computing unit; using the edge computing unit to perform dynamic risk correction under local group evolution characteristics to establish a collaborative risk assessment on the edge side; constructing a digital twin model of the high-voltage cable group in the cloud; using the digital twin model to read the data acquisition results; performing extreme state inversion and future risk fluidization prediction based on historical data and the data acquisition results to establish a global risk assessment; and performing multi-scale game optimization on the local risk distribution, collaborative risk assessment, and global risk assessment of the cable to output monitoring and early warning results.

[0006] Preferably, the local risk distribution of the cable is mapped to a group risk topology on the edge side, where the nodes of the group risk topology represent cable nodes and the edges represent the risk coupling relationship between cables; the node risk evolution curve of each group risk topology is calculated based on historical data stored in the edge computing unit and data acquisition results to establish a group risk evolution matrix; abnormal deviation nodes are identified in the group risk evolution matrix, and dynamic correction of the risk transmission relationship of the neighboring nodes and the group consistency index of the abnormal deviation nodes is performed to update the local risk distribution of the abnormal deviation nodes; the updated local risk distribution is synchronized to the group risk evolution matrix, collaborative risk analysis is performed, and collaborative risk assessment is established.

[0007] Preferably, the risk propagation probability between nodes is constructed based on the synchronized group risk evolution matrix. The risk propagation probability between nodes is constructed by adaptive correction based on the electromagnetic coupling degree between cable nodes, spatial proximity, and historical linkage fault modes. Neighboring environmental factor parameters are collected and embedded, including temperature and humidity, load fluctuation, and partial discharge interference signals. Based on the neighboring environmental factor parameters and the risk propagation probability between nodes, a collaborative risk aggregation analysis of the group risk evolution matrix is ​​performed to establish a collaborative risk assessment.

[0008] Preferably, the historical data and the data acquisition results are fused and compared to establish a historical-real-time state matching index for the cable group; extreme state inversion is performed based on the historical-real-time state matching index, which calculates the virtual limit response of the cable under overload, breakdown, overheating, and strong environmental disturbance conditions to generate potential extreme state evolution trajectories; the future risk fluidized prediction layer in the digital twin model is activated, and the extreme state evolution trajectory is used as the initial boundary condition. Combined with the time series distribution of historical risk events, the dynamic diffusion and aggregation trend of cable group risk in the time and space dimensions is predicted, and external environmental disturbance factors are introduced into the prediction process to establish a global risk assessment.

[0009] Preferably, the local risk distribution, collaborative risk assessment, and global risk assessment of the cable are mapped to local risk weights, group risk matrices, and global risk tensors, respectively, to construct a multi-scale risk state space; a three-layer game unit is established within the multi-scale risk state space, cross-scale feedback coupling of the three-layer game unit is executed, multi-scale game iteration is performed, and monitoring and early warning results are output based on the convergence result of the multi-scale game iteration.

[0010] Preferably, the three-layer game unit includes a local game unit, a cooperative game unit, and a global game unit. The local game unit is used to simulate the non-cooperative game of cable nodes in risk perception, resource competition, and risk transmission, and outputs a local risk equilibrium strategy. The cooperative game unit is used to simulate the cooperative game of cable groups in maintaining consistency and correcting abnormal deviations, and outputs a group cooperative risk balance solution. The global game unit is used to simulate the adversarial game between cable groups and external environmental factors, and outputs a global risk control strategy.

[0011] Preferably, the system performs early warning signal matching based on the monitoring and early warning results to establish early warning signal matching results; it then uses the early warning signal matching results to issue early warnings and performs time-related enhanced early warning reporting management.

[0012] In a second aspect, this application provides a cloud-edge-device collaborative intelligent monitoring and early warning system for high-voltage cables. The system comprises: a data acquisition module: deploying multimodal sensors at the cable nodes of the high-voltage cable and using an end-side sensing unit to control the multimodal sensors to perform event-driven data acquisition; performing local preprocessing on the data acquisition results to establish a local risk distribution for the cable; a risk correction module: deploying an edge computing unit in a ring network side cabinet; sending the local risk distribution of the cable to the edge computing unit; using the edge computing unit to perform dynamic risk correction under local group evolution characteristics to establish a collaborative risk assessment at the edge; a risk assessment module: constructing a digital twin model of the high-voltage cable group in the cloud; using the digital twin model to read the data acquisition results; performing extreme state inversion and future risk fluidization prediction based on historical data and the data acquisition results to establish a global risk assessment; and a monitoring and early warning module: performing multi-scale game optimization on the local risk distribution, collaborative risk assessment, and global risk assessment of the cable, and outputting monitoring and early warning results.

[0013] In summary, one or more technical solutions provided in this application achieve the following: the end-side event-driven acquisition mechanism samples only when a critical event is triggered; the edge-side group evolution dynamic correction logic constructs a group risk topology and evolution matrix; the cloud-based digital twin model combines extreme state inversion and risk fluidization prediction; and the local risk distribution, collaborative risk assessment, and global risk assessment of the cable are coupled to form an end-edge-cloud data closed loop. This comprehensively improves the real-time performance, accuracy, and global synergy of monitoring and early warning, and ensures the safe and stable operation of high-voltage cables. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0015] Figure 1 This application provides a flowchart illustrating a cloud-edge-device collaborative intelligent monitoring and early warning method for high-voltage cables.

[0016] Figure 2 This application provides a structural schematic diagram of a cloud-edge-device collaborative intelligent monitoring and early warning system for high-voltage cables.

[0017] Explanation of reference numerals in the attached diagram: Data acquisition module M100, risk correction module M200, risk assessment module M300, monitoring and early warning module M400. Detailed Implementation

[0018] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0019] Example 1: The present application will be described in detail below with reference to the accompanying drawings, as follows... Figure 1 As shown, this application provides a cloud-edge-device collaborative intelligent monitoring and early warning method for high-voltage cables, wherein the method includes: S1: Deploy multimodal sensors at the cable nodes of the high-voltage cable, and use the end-side sensing unit to control the multimodal sensors to perform event-driven data acquisition. Perform local preprocessing on the data acquisition results to establish the local risk distribution of the cable. S2: Deploy an edge computing unit in the ring network side cabinet, send the local risk distribution of the cable to the edge computing unit, and use the edge computing unit to perform dynamic risk correction under the local group evolution characteristics to establish a collaborative risk assessment on the edge side.

[0020] Specifically, multimodal sensors refer to a combination of sensors capable of simultaneously collecting multiple types of data, such as temperature, humidity, partial discharge, and electromagnetic field strength, providing cable operating status information. Compared to single-modal sensors, they can more accurately reflect the cable's health condition. End-side sensing units are data acquisition and preprocessing devices located near cable nodes, responsible for controlling the triggering and data acquisition of multimodal sensors and performing preliminary processing on the acquired data. The role of end-side sensing units is to reduce data transmission volume and improve data processing efficiency. Event-driven data acquisition refers to triggering data acquisition only when specific events such as partial discharge or abnormal temperature increases occur. Compared to continuous acquisition, this effectively reduces data redundancy and improves the relevance and value of the data. Local preprocessing refers to the preliminary processing of the acquired data, including data cleaning and feature extraction, to remove noise and redundant information, extract useful data features, and provide a more accurate data foundation for subsequent risk assessment. Local risk distribution in cables refers to analyzing pre-processed data to determine the risk level and distribution at cable nodes, providing local risk information for subsequent risk assessment. Edge computing units are computing devices deployed in the ring network side cabinet, responsible for processing data from end-side sensing units and performing more complex calculations and analyses. The role of edge computing units is to reduce dependence on the cloud and improve system response speed and reliability. Dynamic risk correction under local group evolution characteristics refers to dynamically adjusting local risks by analyzing the risk evolution characteristics of cable groups, considering the mutual influence and risk propagation between cables, and improving the accuracy of risk assessment. Collaborative risk assessment refers to comprehensively considering the risk information of multiple cable nodes and establishing a more comprehensive risk assessment model through collaborative analysis to reflect the overall risk status of the cable group and provide support for global risk assessment.

[0021] Execution steps: Multimodal sensors are deployed at the cable nodes of the high-voltage cable. End-side sensing units control these sensors to perform event-driven data acquisition, collecting data only when critical events are triggered. Compared to traditional continuous sampling, this not only reduces the burden of data transmission and storage but also improves the relevance and value of the data. Local preprocessing is performed on the collected data to establish a local risk distribution for the cable. This local preprocessing includes data cleaning and feature extraction, which removes noise and redundant information and extracts useful data features. Through local preprocessing, noise signals in the collected temperature data are removed, and accurate temperature change trends are extracted, providing a more accurate data foundation for subsequent risk assessment and improving the accuracy of the risk assessment.

[0022] Edge computing units are deployed in the ring network side cabinet to send the local risk distribution of cables to the edge computing units. Furthermore, the edge computing units utilize dynamic risk correction logic based on the local group evolution characteristics to dynamically adjust local risks. By analyzing the risk evolution characteristics of the cable group, abnormal deviation nodes are identified and corrected according to the risk transmission relationship of neighboring nodes, breaking the limitations of traditional single-node risk judgment and improving the accuracy and real-time performance of risk assessment. The edge computing units establish collaborative risk assessment at the edge, providing more comprehensive risk information for global risk assessment, forming a collaborative decision-making link from the end side to the edge side, and improving the overall effectiveness of high-voltage cable monitoring and early warning.

[0023] S3: Construct a digital twin model of the high-voltage cable group in the cloud. After reading the data collection results using the digital twin model, perform extreme state inversion and future risk fluidization prediction based on historical data and data collection results to establish a global risk assessment. S4: Perform multi-scale game optimization on the local risk distribution, collaborative risk assessment, and global risk assessment of the cable, and output monitoring and early warning results.

[0024] Specifically, in high-voltage cable monitoring, digital twin models can simulate the cable's operating status, fault modes, and risk diffusion trends, providing support for global risk assessment; extreme state inversion refers to generating potential extreme state evolution trajectories by simulating the cable's response under extreme conditions such as overload, breakdown, overheating, and strong environmental disturbances, which can help identify the cable's risk points under extreme conditions and provide a basis for risk prevention; future risk fluidization prediction refers to using historical and real-time data to predict the dynamic diffusion and aggregation trends of cable group risks in time and space, which can provide early warning of potential future risks and provide a time window for risk prevention and control.

[0025] Global risk assessment refers to the comprehensive risk assessment system established by taking into account the local risks, collaborative risks, and future risks of a cable group. Global risk assessment can provide the overall risk status of the cable group and support decision-making. Multi-scale game optimization refers to the establishment of game units at different scales corresponding to the local risk distribution, collaborative risk assessment, and global risk assessment of the cable, so as to realize cross-scale feedback coupling and iterative optimization. It can balance the urgency of local risks and the global risk diffusion trend, and achieve accurate output of monitoring and early warning results.

[0026] Execution steps: First, historical and real-time data acquisition results are fused and compared to establish a historical-real-time state matching index for the cable group. Specifically, abnormal change points are identified by comparing temperature change trends in historical data and temperature changes in real-time data. Second, based on the historical-real-time state matching index, the cable response under extreme conditions such as overload, breakdown, overheating, and strong environmental disturbances is simulated to generate potential extreme state evolution trajectories. Specifically, the failure modes of the cable under extreme conditions are predicted by simulating temperature changes and insulation performance degradation under overload conditions. Third, the future risk fluidized prediction layer within the digital twin model is activated. Using the extreme state evolution trajectory as initial boundary conditions and combining the time-series distribution of historical risk events, the dynamic diffusion and aggregation trends of cable group risks in time and space are predicted. Finally, considering the local risks, collaborative risks, and future risks of the cable group, a global risk assessment system is established. This global risk assessment provides an overall risk status of the cable group, supporting decision-making.

[0027] The local risk distribution, collaborative risk assessment, and global risk assessment of cables are mapped to local risk weights, a group risk matrix, and a global risk tensor, respectively, to construct a multi-scale risk state space. Here, the local risk weight represents the risk level of each cable node, and the group risk matrix represents the risk coupling relationship of the cable group. Within this multi-scale risk state space, three layers of game units are established: local game units, collaborative game units, and global game units. The local game units simulate the non-cooperative game of cable nodes in risk perception, resource competition, and risk transmission, outputting local risk equilibrium strategies. The collaborative game units simulate the cooperative game of the cable group in maintaining consistency and correcting abnormal deviations, outputting a group collaborative risk equilibrium solution. The global game units simulate the adversarial game between the cable group and external environmental factors, outputting global risk control strategies.

[0028] The system employs cross-scale feedback coupling across three layers of game units to conduct multi-scale game iterations. Further, through iterative optimization of local game units, the risk strategy of each cable node is adjusted; through iterative optimization of collaborative game units, the risk balance of the cable group is adjusted; and through iterative optimization of global game units, the risk control strategy between the cable group and the external environment is adjusted. Based on the convergence results of the multi-scale game iterations, monitoring and early warning results are output, coupling local cable risk distribution, collaborative risk assessment, and global risk assessment to form an end-edge-cloud data closed loop, improving the real-time performance, accuracy, and global synergy of high-voltage cable monitoring and early warning.

[0029] Furthermore, by utilizing the edge computing unit to perform dynamic risk correction under local population evolution characteristics and establishing a collaborative risk assessment at the edge, the method of this application includes: The local risk distribution of the cable is mapped to a group risk topology on the edge side. The nodes of the group risk topology represent cable nodes, and the edges represent the risk coupling relationship between cables. Based on historical data stored in the edge computing unit and data acquisition results, the node risk evolution curve of each group risk topology is calculated to establish a group risk evolution matrix. The group risk evolution matrix is ​​used to identify abnormal deviation nodes, and dynamic correction is performed on the risk transmission relationship of the neighboring nodes and the group consistency index of the abnormal deviation nodes to update the local risk distribution of the abnormal deviation nodes. The updated local risk distribution is synchronized to the group risk evolution matrix, and collaborative risk analysis is performed to establish a collaborative risk assessment.

[0030] Specifically, the group risk topology is used to represent the risk distribution and interrelationships of a cable group. Nodes represent cable nodes, and edges represent the risk coupling relationships between cables, i.e., how the risk of one cable node affects other nodes. Node risk evolution curves describe the changes in cable node risk over time, calculated by analyzing historical and real-time data acquisition results, reflecting the risk status of cable nodes at different points in time. The group risk evolution matrix stores and analyzes the risk evolution curve of each node in the group risk topology, reflecting the overall dynamic changes in the risk of the cable group. Anomaly deviation node identification refers to identifying nodes whose risk status deviates significantly from the group average state by analyzing the group risk evolution matrix. Anomaly deviation nodes have higher risks due to local faults or other abnormal conditions.

[0031] The neighbor node risk transmission relationship describes how the risk of a node is transmitted to its neighboring nodes through risk coupling relationships, and is used to dynamically correct the risk status of abnormally deviating nodes; the group consistency index measures the degree of consistency of the risk status of each node in the group risk evolution matrix, and through dynamic correction, it can reduce the impact of abnormally deviating nodes on group consistency; collaborative risk analysis refers to the comprehensive consideration of the risk status of each node in the group risk evolution matrix and their interrelationships to conduct risk assessment, which can provide an overall risk assessment of the cable group, rather than just the risk status of a single node.

[0032] Execution Steps: A group risk topology is constructed to visually represent the risk distribution and interrelationships of a cable group. Nodes in the group risk topology represent cable nodes, and edges represent the risk coupling relationships between cables. Specifically, in a system containing multiple cable nodes, each node has its corresponding risk state, and edges represent the risk transmission paths between these nodes. Through the group risk topology on the edge side, it is clear how the risk of one node affects other nodes through coupling relationships. By analyzing the historical and real-time data of each cable node, its risk change curve over time is calculated and denoted as the node risk evolution curve. Based on the node risk evolution curves of each group risk topology, a group risk evolution matrix is ​​established. Furthermore, by analyzing the temperature changes and partial discharge data of a cable node over the past year, its risk evolution curve is obtained. The group risk evolution matrix reflects the overall dynamic changes in the risk of the cable group, providing data support for subsequent risk assessment.

[0033] By identifying nodes whose risk status deviates significantly from the group average, and dynamically correcting them based on the risk transmission relationship with their neighboring nodes, specifically, if a cable node has an abnormally high risk status, the risk assessment of that node can be adjusted to be closer to the group average by analyzing the risk status and coupling relationship of its neighboring nodes. Simultaneously, dynamic correction of the group consistency index reduces the impact of abnormally deviating nodes on group consistency, improving the accuracy of risk assessment. Furthermore, by resynchronizing the updated local risk distribution to the group risk evolution matrix, collaborative risk analysis is performed. Further, by analyzing the updated group risk evolution matrix, collaborative risk analysis assesses the overall risk status of the cable group. Collaborative risk assessment provides more comprehensive risk information, offering more accurate basis for decision-making, breaking the limitations of traditional single-node risk judgment, and improving the accuracy and real-time performance of risk assessment.

[0034] Furthermore, the updated local risk distribution is synchronized to the population risk evolution matrix, and collaborative risk analysis is performed. The method of this application includes: The risk propagation probability between nodes is constructed based on the synchronized group risk evolution matrix. The risk propagation probability between nodes is constructed by adaptive correction based on the electromagnetic coupling degree between cable nodes, spatial proximity, and historical linkage fault modes. Neighboring environmental factor parameters are collected and embedded, including temperature and humidity, load fluctuation, and partial discharge interference signals. Based on the neighboring environmental factor parameters and the risk propagation probability between nodes, a collaborative risk aggregation analysis of the group risk evolution matrix is ​​performed to establish a collaborative risk assessment.

[0035] Specifically, the risk propagation probability between nodes represents how the risk state of one cable node is transmitted to other nodes with a specific probability. The risk propagation probability between nodes is affected by a variety of factors, including the degree of electromagnetic coupling, spatial proximity, and historical linkage failure modes. The degree of electromagnetic coupling refers to the degree to which cable nodes influence each other through electromagnetic fields. Specifically, the stronger the electromagnetic coupling, the higher the probability of risk propagation. Spatial proximity refers to the physical distance between cable nodes. Specifically, the closer the distance, the higher the probability of risk propagation.

[0036] Historical linkage failure modes refer to the correlation patterns when failures occurred between cable nodes in the past. If two nodes have failed simultaneously or successively multiple times in the past, the probability of risk propagation between them is high. Proximity environmental factor parameters refer to external environmental factors that affect the operating status of cables, including temperature and humidity, load fluctuations, partial discharge interference signals, etc., which can affect the risk status and risk propagation of cables. Collaborative risk aggregation analysis refers to the analysis of the group risk evolution matrix by comprehensively considering multiple factors to assess the overall risk status of the cable group. It can provide a more comprehensive risk assessment, rather than just the risk status of a single node.

[0037] Execution steps: By analyzing the electromagnetic coupling degree, spatial proximity, and historical linkage failure modes between cable nodes, the risk propagation probability between nodes is adaptively corrected. For example, if the electromagnetic coupling degree between two cable nodes is 0.8, the spatial proximity is 0.7, and the historical linkage failure mode is 0.6, the initial weight coefficients corresponding to the electromagnetic coupling degree, spatial proximity, and historical linkage failure mode are all set to 1 / 3, and the risk propagation probability between nodes is calculated to be 0.75. The adaptive correction of the risk propagation probability between nodes can more accurately reflect the risk propagation relationship between nodes and improve the accuracy of risk assessment.

[0038] Immediate environmental factors reflect the external environmental conditions of cable operation and are crucial for risk assessment. Furthermore, changes in temperature and humidity can affect cable insulation performance, load fluctuations can impact thermal stability, and partial discharge interference signals may indicate insulation aging. By collecting these parameters and embedding them into the risk assessment model, a more comprehensive assessment of the cable's risk status can be achieved. By comprehensively considering the risk propagation probability between nodes and the parameters of the immediate environmental factors, the group risk evolution matrix can be analyzed. Preferably, through collaborative risk aggregation analysis, the overall risk status of the node and its neighboring nodes can be assessed, providing a comprehensive risk assessment of the cable group. This collaborative risk assessment can more effectively identify high-risk areas, enabling early maintenance and repair, and ensuring the safe and stable operation of high-voltage cables.

[0039] Furthermore, after reading the data collection results using the digital twin model, extreme state inversion and future risk fluidization prediction are performed based on historical data and the data collection results to establish a global risk assessment. The method of this application includes: The historical data and data acquisition results are fused and compared to establish a historical-real-time state matching index for the cable group. Based on this index, extreme state inversion is performed. This inversion calculates the virtual limit response of the cable under overload, breakdown, overheating, and strong environmental disturbance conditions, generating potential extreme state evolution trajectories. The future risk fluidized prediction layer within the digital twin model is activated. Using these extreme state evolution trajectories as initial boundary conditions, and combined with the time-series distribution of historical risk events, the dynamic diffusion and aggregation trends of cable group risks in both time and space are predicted. External environmental disturbance factors are introduced into the prediction process to establish a global risk assessment.

[0040] Specifically, the historical-real-time state matching index is used to fuse and compare historical and real-time data acquisition results to quickly find and match similar states, helping to identify the similarities and differences between the current state and historical states; extreme state inversion refers to generating potential extreme state evolution trajectories by simulating the virtual limit response of cables under extreme conditions such as overload, breakdown, overheating, and strong environmental disturbances, which can help identify risk points of cables under extreme conditions and provide a basis for risk prevention; virtual limit response refers to the limit states that cables may exhibit under extreme conditions, including temperature limits and voltage limits, etc. By calculating these limit responses, the behavior of cables under extreme conditions can be predicted; the future risk fluidized prediction layer is a module in the digital twin model used to predict future risks, predicting the dynamic changes of cable group risks in time and space dimensions by analyzing historical and real-time data; external environmental disturbance factors refer to external environmental factors that may affect the operating status of cables, including temperature changes and humidity changes.

[0041] Execution steps: By comparing historical data and data acquisition results, the similarities and differences between the current state and historical states are identified. Specifically, by analyzing temperature change trends in historical data and temperature changes in real-time data, a historical-real-time state matching index is established to quickly find historical states similar to the current state. This index helps to quickly identify abnormal states and improve the efficiency of risk assessment. Furthermore, by simulating the virtual limit response of high-voltage cables under extreme conditions including overload, breakdown, overheating, and strong environmental disturbances, potential extreme state evolution trajectories are generated. Further, by simulating the temperature changes and insulation performance degradation of cables under overload conditions, the potential extreme state evolution trajectory is predicted. The system measures cable failure modes under extreme conditions; identifies risk points in cables under extreme conditions through extreme state inversion, providing a basis for risk prevention; activates the future risk fluidized prediction layer within the digital twin model, and predicts potential future risk areas and times by analyzing the time-series distribution of historical risk events corresponding to failure start and end times, and the spatial sequence distribution corresponding to risk areas, using the extreme state evolution trajectory as initial boundary conditions. Introducing external environmental disturbance factors, such as temperature and humidity changes, during the prediction process improves prediction accuracy; and establishes a global risk assessment based on the prediction results, providing comprehensive risk information for decision-making.

[0042] Furthermore, the method of this application involves multi-scale game optimization of the local risk distribution, collaborative risk assessment, and global risk assessment of the cable, and outputs monitoring and early warning results. The local risk distribution, collaborative risk assessment, and global risk assessment of the cable are mapped to local risk weights, group risk matrices, and global risk tensors, respectively, to construct a multi-scale risk state space. A three-layer game unit is established within the multi-scale risk state space, and cross-scale feedback coupling of the three-layer game unit is executed to perform multi-scale game iteration. The monitoring and early warning results are output based on the convergence result of the multi-scale game iteration.

[0043] Specifically, the local risk weight represents the quantitative value of the local risk distribution of the cable, reflecting the risk level of each cable node. The higher the local risk weight, the greater the risk of that node. The group risk matrix represents the collaborative risk assessment result of the cable group, storing the risk coupling relationship and collaborative risk status between each node in matrix form. The global risk tensor represents the global risk assessment result of the cable group, storing the overall risk status of the cable group in time and space in tensor form. The multi-scale risk state space refers to a multi-dimensional space that integrates the local, group, and global risk assessment results, used to comprehensively represent the risk status of the cable group.

[0044] The three-layer game unit includes local game units, cooperative game units, and global game units. Furthermore, the local game unit simulates the non-cooperative game of cable nodes in risk perception, resource competition, and risk transmission, outputting a local risk equilibrium strategy; the cooperative game unit simulates the cooperative game of cable groups in maintaining consistency and correcting abnormal deviations, outputting a group cooperative risk equilibrium solution; the global game unit simulates the adversarial game between the cable group and external environmental factors, outputting a global risk control strategy; cross-scale feedback coupling refers to the information exchange and state updates between different game units through feedback mechanisms to achieve collaborative optimization of multi-scale risk assessment; multi-scale game iteration refers to the iterative calculation to gradually optimize the strategies of each game unit, ultimately reaching the optimal solution for risk assessment and decision-making.

[0045] Execution steps: By quantifying and integrating risk assessment results at different scales, a unified framework is provided for game optimization. Furthermore, the local risk weights corresponding to the local risk distribution of cables represent the risk level of each cable node, the group risk matrix corresponding to the collaborative risk assessment represents the risk coupling relationship of the cable group, and the global risk tensor corresponding to the global risk assessment represents the overall risk state of the cable group in the time and space dimensions. By mapping the local risk distribution of cables to local risk weights, the collaborative risk assessment to a group risk matrix, and the global risk assessment to a global risk tensor, a multi-scale risk state space is constructed to comprehensively represent the risk state of the cable group and provide more comprehensive information for decision-making.

[0046] By simulating risk decision-making processes at different scales, risk assessment and decision-making strategies are gradually optimized. Furthermore, in local game units, by simulating the non-cooperative game of cable nodes in risk perception, resource competition, and risk transmission, a local risk equilibrium strategy is output. Specifically, by analyzing the risk status and resource allocation of each node, the risk response strategy of each node can be optimized. In cooperative game units, by simulating the cooperative game of cable groups in maintaining consistency and correcting abnormal deviations, a group cooperative risk balance solution is output. Specifically, by analyzing the group risk matrix, abnormal deviation nodes can be identified, and the accuracy of group risk assessment can be improved through cooperative correction. In global game units, by simulating the adversarial game between the cable group and external environmental factors, a global risk control strategy is output. Specifically, by analyzing the global risk tensor, high-risk areas are identified, and resource allocation is optimized through global control strategies.

[0047] Through cross-scale feedback coupling, information exchange and state updates are conducted between different game units, gradually optimizing risk assessment and decision-making strategies. Furthermore, the strategy updates of local game units are fed back to collaborative game units, and the strategy updates of collaborative game units are fed back to the global game unit. Multi-scale game iteration is performed to gradually optimize the strategies of each game unit under multi-scale game to achieve the optimal solution for risk assessment and decision-making. Furthermore, through multiple iterations, local risk weights, group risk matrices, and global risk tensors are gradually adjusted to output accurate monitoring and early warning results, thereby balancing the urgency of local risks with the global risk diffusion trend and achieving collaborative optimization of multi-scale risk assessment.

[0048] Furthermore, the method of this application includes: The three-layer game unit includes a local game unit, a cooperative game unit, and a global game unit. The local game unit is used to simulate the non-cooperative game of cable nodes in risk perception, resource competition, and risk transmission, and outputs a local risk equilibrium strategy. The cooperative game unit is used to simulate the cooperative game of cable groups in maintaining consistency and correcting abnormal deviations, and outputs a group cooperative risk equilibrium solution. The global game unit is used to simulate the adversarial game between cable groups and external environmental factors, and outputs a global risk control strategy.

[0049] Specifically, the local game unit is used to analyze the risk assessment and decision-making process of a single cable node, considering the node's own risk perception, resource competition, and risk transmission, and optimizing the node's risk response strategy through a non-cooperative game model; the cooperative game unit is used to analyze the risk collaborative assessment and decision-making process within the cable group, maintaining consistency within the group through a cooperative game model, correcting abnormally deviating nodes, and outputting a group collaborative risk balance solution; the global game unit is used to analyze the interaction between the cable group and external environmental factors, optimizing global risk control strategies through an adversarial game model to cope with the uncertainty of the external environment. Furthermore, external environmental factors include weather, power grid load, etc.

[0050] Execution Steps: In the local game unit, each cable node may face different risk factors, such as partial discharge and abnormal temperature. The local game unit analyzes these risk factors and, combined with the node's resource conditions, including cooling system capacity and insulation material status, formulates an optimal risk response strategy. Furthermore, the node may reduce the risk of abnormal temperature by reducing load, or increase monitoring frequency to detect partial discharge problems earlier. The local game unit allows each node to independently optimize its own risk strategy, improving the system's flexibility and adaptability through non-cooperative game theory.

[0051] In a cable group, the risk status of a single node may affect other nodes. By analyzing the risk coupling relationship within the group, the cooperative game unit ensures that the risk strategies of all nodes remain consistent. Furthermore, if the risk status of a node is abnormally high, the cooperative game unit can identify this anomaly and correct it by adjusting the resource allocation or risk strategies of neighboring nodes. This non-global cooperative game model helps to improve the overall risk response capability of the group and reduce the chain reaction caused by the anomaly of a single node.

[0052] External environmental factors, such as weather changes and power grid load fluctuations, can significantly impact the risk status of cable groups. The global game unit simulates the adversarial game between the cable group and external environmental factors to formulate global risk control strategies. Furthermore, in hot weather, the global game unit may suggest increasing the operating time of the cooling system; during peak power grid load periods, it may adjust the operating parameters of the cables to avoid overload. This global adversarial game model helps the cable group maintain stable operation and reduce the risk of failure when facing external uncertainties.

[0053] In the above steps, through the collaborative work of three layers of game units—local game units, cooperative game units, and global game units—risk optimization from local to global is achieved, thereby improving the real-time performance, accuracy, and global synergy of the high-voltage cable monitoring and early warning system.

[0054] Furthermore, the method for outputting monitoring and early warning results in this application includes: Based on the monitoring and early warning results, early warning signals are matched to establish early warning signal matching results; early warnings are issued using the early warning signal matching results, and time-related enhanced early warning reporting management is implemented.

[0055] Specifically, early warning signal matching refers to comparing the monitoring and early warning results with preset early warning signal patterns to determine the early warning signal corresponding to the current risk status. Early warning signals typically include different levels and types; the early warning signal matching result refers to the result obtained after comparison, clearly indicating the early warning signal corresponding to the current risk status, providing a basis for subsequent early warning issuance; early warning issuance refers to issuing early warning information based on the early warning signal matching result, notifying the current risk status; time-related enhanced early warning reporting management refers to adjusting the intensity or method of early warning based on time factors including the duration and frequency of the early warning signal. Specifically, if the duration of a certain risk early warning signal exceeds the average duration under the same risk level, it may trigger a higher-level early warning response.

[0056] Execution steps: By comparing the monitoring and early warning results with preset early warning signal patterns, the early warning signal corresponding to the current risk status is determined. Specifically, if the monitoring and early warning results show that the temperature of a certain cable node rises abnormally and exceeds the preset temperature threshold, the early warning signal matching result will indicate that this is a temperature anomaly early warning signal of a certain level. This matching process can quickly and accurately identify the risk status and provide a clear basis for subsequent early warning issuance. The early warning issuance is executed using the early warning signal matching result. Specifically, an alarm SMS is sent to the operation and maintenance personnel, and an alarm icon is displayed on the monitoring interface. At the same time, time-related early warning enhancement reporting management is executed. The intensity or method of the early warning is adjusted according to the duration and frequency of the early warning signal. If the duration exceeds a certain threshold, the early warning level will be automatically upgraded, triggering a higher level of response measures. This time-related early warning enhancement reporting management can dynamically adjust the early warning strategy according to the urgency and persistence of the risk, ensuring timely and effective response to risks and guaranteeing the safe operation of high-voltage cables.

[0057] In summary, the beneficial effects of the embodiments of this application are: This application provides a cloud-edge-end collaborative intelligent monitoring and early warning method and system for high-voltage cables. It employs multimodal sensors deployed at cable nodes of the high-voltage cable and uses end-side sensing units to control these sensors for event-driven data acquisition. The acquired data undergoes local preprocessing to establish a local risk distribution for the cable. An edge computing unit is deployed in the ring network side cabinet to transmit the local risk distribution to the edge computing unit, which performs dynamic risk correction based on local population evolution characteristics, establishing a collaborative risk assessment at the edge. A digital twin model of the high-voltage cable population is constructed in the cloud. After reading the acquired data using the digital twin model, extreme state inversion and future risk fluidization prediction are performed based on historical data and the acquired data to establish a global risk assessment. Multi-scale game optimization is applied to the local risk distribution, collaborative risk assessment, and global risk assessment, outputting monitoring and early warning results. The system implements an end-side event-driven acquisition mechanism that samples only when critical events are triggered. The edge-side group evolution dynamic correction logic constructs a group risk topology and evolution matrix, and combines a cloud-based digital twin model with extreme state inversion and risk fluidization prediction. This couples local cable risk distribution, collaborative risk assessment, and global risk assessment to form an end-edge-cloud data closed loop, comprehensively improving the real-time performance, accuracy, and global synergy of monitoring and early warning, and ensuring the safe and stable operation of high-voltage cables.

[0058] Example 2, based on the same inventive concept as the cloud-edge-device collaborative high-voltage cable intelligent monitoring and early warning method in the aforementioned examples, such as... Figure 2 As shown in the figure, this application embodiment provides a cloud-edge-device collaborative intelligent monitoring and early warning system for high-voltage cables, wherein the system includes: Data acquisition module M100: Multimodal sensors are deployed at the cable nodes of high-voltage cables, and the end-side sensing unit controls the multimodal sensors to perform event-driven data acquisition. The data acquisition results are then subjected to local preprocessing to establish the local risk distribution of the cable.

[0059] Risk Correction Module M200: Deploys an edge computing unit in the ring network side cabinet, sends the local risk distribution of the cable to the edge computing unit, and uses the edge computing unit to perform dynamic risk correction under the local group evolution characteristics to establish a collaborative risk assessment on the edge side.

[0060] Risk Assessment Module M300: Constructs a digital twin model of the high-voltage cable group in the cloud. After reading the data collection results using the digital twin model, it performs extreme state inversion and future risk fluidization prediction based on historical data and data collection results to establish a global risk assessment.

[0061] Monitoring and early warning module M400: Performs multi-scale game optimization on the local risk distribution, collaborative risk assessment and global risk assessment of the cable, and outputs monitoring and early warning results.

[0062] Furthermore, the risk correction module M200 is used to perform the following method: The local risk distribution of the cable is mapped to a group risk topology on the edge side. The nodes of the group risk topology represent cable nodes, and the edges represent the risk coupling relationship between cables. Based on historical data stored in the edge computing unit and data acquisition results, the node risk evolution curve of each group risk topology is calculated to establish a group risk evolution matrix. The group risk evolution matrix is ​​used to identify abnormal deviation nodes, and dynamic correction is performed on the risk transmission relationship of the neighboring nodes and the group consistency index of the abnormal deviation nodes to update the local risk distribution of the abnormal deviation nodes. The updated local risk distribution is synchronized to the group risk evolution matrix, and collaborative risk analysis is performed to establish a collaborative risk assessment.

[0063] Furthermore, the risk correction module M200 is also used to perform the following method: The risk propagation probability between nodes is constructed based on the synchronized group risk evolution matrix. The risk propagation probability between nodes is constructed by adaptive correction based on the electromagnetic coupling degree between cable nodes, spatial proximity, and historical linkage fault modes. Neighboring environmental factor parameters are collected and embedded, including temperature and humidity, load fluctuation, and partial discharge interference signals. Based on the neighboring environmental factor parameters and the risk propagation probability between nodes, a collaborative risk aggregation analysis of the group risk evolution matrix is ​​performed to establish a collaborative risk assessment.

[0064] Furthermore, the risk assessment module M300 is used to perform the following methods: The historical data and data acquisition results are fused and compared to establish a historical-real-time state matching index for the cable group. Based on this index, extreme state inversion is performed. This inversion calculates the virtual limit response of the cable under overload, breakdown, overheating, and strong environmental disturbance conditions, generating potential extreme state evolution trajectories. The future risk fluidized prediction layer within the digital twin model is activated. Using these extreme state evolution trajectories as initial boundary conditions, and combined with the time-series distribution of historical risk events, the dynamic diffusion and aggregation trends of cable group risks in both time and space are predicted. External environmental disturbance factors are introduced into the prediction process to establish a global risk assessment.

[0065] Furthermore, the monitoring and early warning module M400 is used to perform the following methods: The local risk distribution, collaborative risk assessment, and global risk assessment of the cable are mapped to local risk weights, group risk matrices, and global risk tensors, respectively, to construct a multi-scale risk state space. A three-layer game unit is established within the multi-scale risk state space, and cross-scale feedback coupling of the three-layer game unit is executed to perform multi-scale game iteration. The monitoring and early warning results are output based on the convergence result of the multi-scale game iteration.

[0066] Furthermore, the monitoring and early warning module M400 is also used to perform the following methods: The three-layer game unit includes a local game unit, a cooperative game unit, and a global game unit. The local game unit is used to simulate the non-cooperative game of cable nodes in risk perception, resource competition, and risk transmission, and outputs a local risk equilibrium strategy. The cooperative game unit is used to simulate the cooperative game of cable groups in maintaining consistency and correcting abnormal deviations, and outputs a group cooperative risk equilibrium solution. The global game unit is used to simulate the adversarial game between cable groups and external environmental factors, and outputs a global risk control strategy.

[0067] Furthermore, the monitoring and early warning module M400 is also used to perform the following methods: Based on the monitoring and early warning results, early warning signals are matched to establish early warning signal matching results; early warnings are issued using the early warning signal matching results, and time-related enhanced early warning reporting management is implemented.

[0068] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1The cloud-edge-end collaborative high-voltage cable intelligent monitoring and early warning method and specific examples in Example 1 are also applicable to the cloud-edge-end collaborative high-voltage cable intelligent monitoring and early warning system in this example. Through the foregoing detailed description of the cloud-edge-end collaborative high-voltage cable intelligent monitoring and early warning method, those skilled in the art can clearly understand the cloud-edge-end collaborative high-voltage cable intelligent monitoring and early warning system in this example. Therefore, for the sake of brevity, it will not be described in detail here.

[0069] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0070] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A cloud-edge collaborative high-voltage cable intelligent monitoring and early warning method, characterized in that, The method comprises: Deploying a multi-modal sensor at a cable node of a high-voltage cable, and using an end-side perception unit to control the multi-modal sensor to perform event-driven data acquisition, perform local preprocessing on the data acquisition result, and establish a cable local risk distribution; Deploying an edge computing unit at a ring network side cabinet, sending the cable local risk distribution to the edge computing unit, using the edge computing unit to perform dynamic risk correction under local group evolution characteristics, and establishing a collaborative risk assessment on the edge side; Building a digital twin model of the high-voltage cable group in the cloud, using the digital twin model to read the data acquisition result, performing extreme state inversion and future risk fluidization prediction according to historical data and the data acquisition result, and establishing a global risk assessment; Multi-scale game optimization is performed on the cable local risk distribution, collaborative risk assessment, and global risk assessment, and a monitoring and early warning result is output. 2.The cloud-edge collaboration high-voltage cable intelligent monitoring and early warning method of claim 1, wherein, Using the edge computing unit to perform dynamic risk correction under local group evolution characteristics, and establishing a collaborative risk assessment on the edge side, comprises: Mapping the cable local risk distribution to a group risk topology structure on the edge side, wherein the nodes of the group risk topology structure represent the cable nodes, and the edges represent the risk coupling relationship between the cables; Calculating the node risk evolution curve of each group risk topology structure according to the historical data and the data acquisition result stored in the edge computing unit, and establishing a group risk evolution matrix; Abnormal deviation node identification is performed on the group risk evolution matrix, and dynamic correction of the adjacent node risk transmission relationship and group consistency index of the abnormal deviation node is performed, and the local risk distribution of the abnormal deviation node is updated; The updated local risk distribution is synchronized to the group risk evolution matrix, a collaborative risk analysis is performed, and a collaborative risk assessment is established. 3.The cloud-edge collaboration high-voltage cable intelligent monitoring and early warning method of claim 2, wherein, Synchronizing the updated local risk distribution to the group risk evolution matrix and performing a collaborative risk analysis comprises: According to the synchronized group risk evolution matrix, a risk propagation probability between nodes is constructed, which is constructed by self-adaptive correction through the degree of electromagnetic coupling between cable nodes, spatial proximity, and historical linkage failure mode; Collecting and embedding adjacent environmental factor parameters, including temperature and humidity, load fluctuation, and local discharge interference signal; According to the adjacent environmental factor parameters and the inter-node risk propagation probability, a collaborative risk aggregation analysis of the group risk evolution matrix is performed, and a collaborative risk assessment is established. 4.The cloud-edge collaboration high-voltage cable intelligent monitoring and early warning method of claim 1, wherein, Using the digital twin model to read the data acquisition result, performing extreme state inversion and future risk fluidization prediction according to historical data and the data acquisition result, and establishing a global risk assessment, comprises: Fusing and comparing the historical data and the data acquisition result to establish a historical-real-time state matching index of the cable group; Performing extreme state inversion according to the historical-real-time state matching index, wherein the extreme state inversion is performed by calculating the virtual limit response of the cable under overload, breakdown, overheating, and environmental strong disturbance conditions to generate a potential extreme state evolution trajectory; Activate the future risk fluidization prediction layer in the digital twin model, use the extreme state evolution trajectory as the initial boundary condition, combine the time series distribution of historical risk events to predict the dynamic diffusion and aggregation trend of cable group risk in the time and space dimensions, and introduce external environmental disturbance factors in the prediction process to establish global risk assessment. 5.The cloud-edge collaboration high-voltage cable intelligent monitoring and early warning method of claim 1, wherein, Multi-scale game optimization is performed on the cable local risk distribution, collaborative risk assessment, and global risk assessment, and monitoring and early warning results are output, including: The cable local risk distribution, collaborative risk assessment, and global risk assessment are respectively mapped to local risk weight, group risk matrix, and global risk tensor to construct a multi-scale risk state space. A three-layer game unit is established in the multi-scale risk state space, cross-scale feedback coupling of the three-layer game unit is performed, multi-scale game iteration is performed, and monitoring and early warning results are output according to the convergence results of multi-scale game iteration. 6.The cloud-edge collaboration high-voltage cable intelligent monitoring and early warning method of claim 5, wherein, The three-layer game unit includes a local game unit, a collaborative game unit, and a global game unit. The local game unit is used to simulate the non-cooperative game of cable nodes in risk perception, resource competition, and risk transmission, and outputs a local risk equilibrium strategy. The collaborative game unit is used to simulate the cooperative game of cable groups in consistency maintenance and abnormal deviation correction, and outputs a group collaborative risk balance solution. The global game unit is used to simulate the antagonistic game of cable groups and external environmental factors, and outputs a global risk regulation strategy.

7. The cloud-edge collaboration high-voltage cable intelligent monitoring and early warning method of claim 1, wherein, The monitoring and early warning results include: According to the monitoring and early warning results, a warning signal matching result is established. The warning signal matching result is used to perform warning reporting and time-related warning enhancement reporting management.

8. A cloud-edge collaborative high-voltage cable intelligent monitoring and early warning system, characterized in that, Steps for implementing the cloud-edge-end collaborative high-voltage cable intelligent monitoring and early warning method according to any one of claims 1-7, the system comprising: Data acquisition module: multi-modal sensors are arranged at cable nodes of high-voltage cables, and event-driven data acquisition is performed by the multi-modal sensors under the control of an end-side perception unit. Local preprocessing is performed on the data acquisition results, and a cable local risk distribution is established. Risk correction module: deploy an edge computing unit at the ring network side cabinet, send the cable local risk distribution to the edge computing unit, and perform dynamic risk correction under the local group evolution characteristics using the edge computing unit to establish collaborative risk assessment on the edge side. Risk assessment module: build a digital twin model of high-voltage cable groups in the cloud, and perform extreme state inversion and future risk fluidization prediction based on historical data and data acquisition results after reading the data acquisition results using the digital twin model to establish global risk assessment. Monitoring and early warning module: multi-scale game optimization is performed on the cable local risk distribution, collaborative risk assessment, and global risk assessment, and monitoring and early warning results are output.

Images