Online edge communication method for power transmission network

Abstract

The invention relates to the technical field of power transmission network online edge communication, in particular to a power transmission network online edge communication method. According to the method, meteorological data and a network topology space are superposed, threatened nodes are identified, alternative paths are pre-calculated, service migration and gateway role transfer are actively completed before weather comes, and a robust communication path is established. When the network is divided into islands by extreme weather, the key monitoring data is ensured not to be lost by electing temporary sink nodes and implementing local hierarchical storage. And after the network is recovered, orderly return of historical backlog data and real-time data is realized by adopting a metadata abstract exchange and bandwidth negotiation mechanism, and channel congestion is avoided. According to the method, self-adaptive survival and flexible operation of the network under severe conditions are realized, and complete delivery of monitoring data, especially key alarms, is ensured.

Description

A method for online edge communication in power transmission networks Technical Field

[0001] This invention relates to the field of online edge communication technology for power transmission networks, and more specifically, to an online edge communication method for power transmission networks. Background Technology

[0002] In high-voltage power transmission networks in remote mountainous areas, transmission towers are typically located in areas with complex terrain and inconvenient transportation, making it extremely difficult and costly to achieve full coverage using public network base stations. To address this issue, existing technologies utilize communication modules deployed on some towers to construct wireless ad hoc networks, enabling the transmission of monitoring data.

[0003] However, the extreme dynamism of the mountainous environment presents severe challenges to this communication network. First, the strong electromagnetic environment generated by high-voltage power lines continuously interferes with the wireless link, leading to fluctuations in the signal-to-noise ratio, increased bit error rate, and unstable link quality. Second, the complex and changeable local climate in mountainous areas, with weather factors such as heavy rain and dense fog not only further deteriorating link quality but also causing frequent changes in the connection status between communication modules and public network base stations—nodes that were originally acting as gateways may lose connection due to weather conditions, while other nodes may reconnect to the public network as the weather improves. This dynamic drift of node roles means the network topology is constantly changing. More seriously, when a local area experiences extreme weather, multiple nodes may fail simultaneously, disrupting all communication links between that area and the outside world, creating a network "island," and posing a risk of losing critical monitoring data collected during this period. Summary of the Invention

[0004] The purpose of this invention is to provide an online edge communication method for power transmission networks, so as to solve the problem that existing traditional communication methods cannot meet the communication reliability requirements of remote mountainous power grid areas under severe weather conditions.

[0005] To achieve the above objectives, this application provides the following technical solution: This application provides an online edge communication method for power transmission networks, applicable to power transmission communication networks. The power transmission communication network includes communication modules installed on multiple power transmission towers. Each communication module is dynamically divided into a first communication module or a second communication module based on its connection status with a public network base station. The first communication module is connected to the public network base station, and the second communication module is not connected to the public network base station. The method includes: responding to meteorological warning data, retrieving a digital topology map and node geographical location information of the power transmission network, and spatially overlaying the meteorological warning data with the digital topology map to obtain a set of threatened nodes. The meteorological warning data includes spatial range and intensity information; based on the set of threatened nodes and the current link status of the power transmission communication network, a corresponding reconstruction plan containing alternative communication path information is generated and encrypted and sent to relevant nodes to form a network reconstruction plan; within a preset advance time window before the meteorological event, the data flow path switching of non-critical services and the synchronous transmission of critical alarm data are initiated, while the link quality between each first communication module and the public network base station is continuously monitored, and when the link quality is continuously lower than a first threshold, a new first communication module is elected through a distributed gateway election mechanism, enabling it to establish a connection with the public network base station, and the routing and forwarding tables of intermediate nodes are adjusted accordingly to ensure that the service flow always points to an actually available gateway.

[0006] Optionally, generating a corresponding reconstruction plan containing alternative communication path information based on the threatened node set and the current link status of the power transmission communication network includes: calculating the expected availability probability of each node in the threatened node set during the warning period based on historical fault data and the current link status of the power transmission communication network; identifying all critical paths passing through the threatened node set based on the data flow backbone in the current power transmission communication network; and calculating one or more alternative communication paths for each critical path with the goal of avoiding threat areas and maximizing the expected reliability of the paths, thereby generating a reconstruction plan containing the alternative communication path information. The optimization objective function for calculating the alternative communication paths comprehensively considers the product of the expected availability probabilities of each hop node in the path, the total number of hops in the path, the current load of each relay node, and the remaining energy. The constraint condition of the optimization objective function is that the alternative communication paths must bypass the core area of ​​the threatened node set.

[0007] Optionally, the step of initiating the data flow path switching for non-critical services and the synchronous transmission of critical alarm data, while continuously monitoring the link quality between each first communication module and the public network base station, and electing a new first communication module through a distributed gateway election mechanism when the link quality is continuously lower than a first threshold, includes: initiating the data flow path switching for non-critical services within a preset advance time window before the weather event, switching the transmission path of periodic status data from the original critical path to the optimal alternative communication path; for critical alarm data, sending the same data packet through the original critical path and the optimal alternative communication path; wherein the endpoint of the optimal alternative communication path points to a currently available first communication module; and continuously monitoring the link quality between each first communication module and the public network base station within the preset advance time window before the weather event. Regarding the link quality of a station, when a first communication module detects that its public network link quality is continuously lower than a first threshold, it actively broadcasts a degradation warning message containing its identity and link quality. Candidate nodes receiving the degradation warning message, based on their link quality with the public network base station, remaining node energy, and network topology centrality index, perform a distributed gateway election. The distributed gateway election adopts a voting-based consensus mechanism. Each node, based on the received election message, votes for candidate nodes according to the weighted sum of their public network link quality score, remaining energy score, and topology centrality score. The candidate node with the highest number of votes exceeding a preset number is confirmed as the winner. The winning node establishes a connection with the public network base station, its identity is upgraded to a new first communication module, and it broadcasts its gateway identity and routing information, simultaneously triggering a routing table update.

[0008] Optionally, adjusting the routing table of the intermediate nodes accordingly to ensure that the service flow always points to the actually available gateway includes: in response to the generation of the new first communication module, updating the path endpoint of all alternative communication paths that have been switched to or are about to switch to the original first communication module to the new first communication module, and adjusting the routing table of the intermediate nodes accordingly to ensure that the service flow always points to the actually available gateway.

[0009] Optionally, after electing a new first communication module, establishing a connection with the public network base station, and adjusting the routing table of intermediate nodes accordingly, the process further includes: when a communication module cannot communicate with a known first communication module within a preset time period, it automatically identifies itself as entering a partitioned island state. Nodes within the island elect a unique temporary aggregation node through a distributed consensus algorithm. This temporary aggregation node is responsible for collecting monitoring data uploaded by all other nodes within the island and storing it in local non-volatile memory according to data priority, forming a structured partitioned data set. The temporary aggregation nodes first exchange partition summary reports, enabling the first communication module, as the receiver, to... The module learns the type, priority distribution, and data volume of the data to be transmitted back from the sender. Based on the obtained data backlog and its currently available backhaul bandwidth, the first communication module negotiates with the sender's temporary aggregation node to determine an initial backhaul bandwidth quota and reaches a transmission rate control protocol. According to the transmission rate control protocol, the sender prioritizes transmitting all unreported critical alarm data. After the critical alarm data transmission is completed, the sender transmits compressed periodic status data or its abnormal subset according to the negotiated bandwidth quota. Under the premise that the real-time service bandwidth is guaranteed, the remaining bandwidth is used to schedule the backhaul of large-capacity inspection data stored in the system to complete the fusion of partitioned data.

[0010] Optionally, the temporary aggregation node performs lossy compression or anomaly detection storage on the collected periodic status data, completely saves key alarm data, and generates a partition summary report containing alarm statistics; wherein, the partition summary report includes the number, type, first and last occurrence time of key alarm events in the island, the numerical range of periodic status data, and the number of anomalies.

[0011] Optionally, the negotiation adopts a bandwidth allocation protocol based on the token bucket model. The first communication module allocates a sustainable data token rate to the temporary aggregation node, and the temporary aggregation node must adjust its data transmission rate according to the data token rate.

[0012] The beneficial effects of this invention are as follows: The online edge communication method for power transmission networks described in this invention integrates meteorological early warning information with network topology analysis and constructs a closed-loop processing flow including predictive reconstruction, regional autonomy, and orderly recovery. This systematically solves the technical problem of reliable communication in mountainous high-voltage tower communication networks under extreme dynamic environments, achieving adaptive survival and resilient operation of the network under harsh conditions. It ensures the complete delivery of monitoring data, especially critical alarm data. Specifically, it first performs spatial overlay analysis of meteorological early warning data and network topology, identifies threatened nodes, and pre-calculates alternative paths. This allows the network to proactively adjust resource allocation before weather events occur, smoothly migrate service traffic, and dynamically transfer gateway roles, thereby establishing robust communication paths before link quality deteriorates. When extreme weather causes the network to irreversibly split into multiple isolated islands, this invention uses the election of temporary aggregation nodes within the islands and a local hierarchical storage mechanism to form autonomous units in the disconnected areas, effectively preserving monitoring data, including critical alarms, and preventing data loss. When network conditions are restored, this invention achieves the orderly fusion and backhaul of historical backlog data and real-time data through metadata digest exchange and bandwidth negotiation mechanisms, avoiding channel congestion and secondary crashes caused by a large number of nodes sending data simultaneously in traditional methods.

[0013] The aforementioned technical methods are interconnected and together constitute a complete communication solution encompassing proactive prevention, fault self-management, and orderly recovery. This solution enables the communication network of high-voltage transmission towers in mountainous areas to maintain reliable acquisition and transmission of monitoring data in the face of dynamic topology changes, link fluctuations, and sudden malfunctions, providing a fundamental guarantee for the safe operation of transmission lines.

[0014] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 is a schematic flowchart of an online edge communication method for power transmission networks according to an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0018] It should be noted that similar reference numerals or letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0019] Example 1:

[0020] As shown in Figure 1, this embodiment provides an online edge communication method for power transmission networks, applicable to power transmission communication networks. The power transmission communication network includes communication modules installed on multiple power transmission towers. The communication modules are dynamically divided into a first communication module or a second communication module according to their connection status with public network base stations. The first communication module is the communication module connected to the public network base station, and the second communication module is the communication module not connected to the public network base station. The method includes: Step S1, in response to meteorological warning data, retrieving the digital topology map and node geographical location information of the power transmission network, and spatially overlaying the meteorological warning data with the digital topology map to obtain a set of threatened nodes. The meteorological warning data includes spatial range and intensity information. The spatial range refers to the geographical area expected to be affected by meteorological events (such as rainstorms, lightning, strong winds, etc.), usually represented in polygon or grid coordinate form. The intensity information refers to the severity of the meteorological event, such as quantitative or graded indicators such as rainfall (mm / hour), wind force level, and lightning activity.

[0021] Spatial extent is used for spatial overlay analysis with the geographical location of transmission towers to accurately identify which tower nodes will be directly exposed to severe weather impacts, forming the basis for subsequent assessments. Intensity information is used to quantify the impact of weather on communication links. Higher intensity indicates a greater probability of communication link quality deterioration or even interruption. By correlating intensity information with historical fault data, the availability probability of nodes in future periods can be predicted more accurately, providing a basis for subsequent path recalculation and network reconfiguration decisions.

[0022] The process of identifying threatened nodes includes: Step S11, obtaining the spatial range (polygonal area) in the meteorological warning data and the precise latitude and longitude coordinates of all communication modules in the network, making an inclusion judgment between the geographical coordinates of each communication module and the spatial range of the meteorological warning, and marking the node as "directly affected by the weather" if the coordinates of a certain communication module are located inside the polygon of the warning area or less than the preset buffer distance from the warning boundary (considering the possibility of weather spread); forming a "set of threatened nodes" from all the marked communication modules as the input object for subsequent vulnerability assessment. It should be noted that this identification process is dynamic and is recalculated as the meteorological warning is updated.

[0023] Step S2: Based on historical fault data and the current link status of the power transmission communication network, calculate the expected availability probability of each node in the threatened node set during the warning period; based on the data flow backbone in the current power transmission communication network, identify all critical paths passing through the threatened node set, where the data flow backbone refers to the set of paths in the network that undertake the main data backhaul task, typically including paths from various second communication modules to the first communication module, and interconnection paths between the first communication modules. The backbone is the part of the network topology with the densest traffic and most important for network connectivity. The critical path refers to those paths in the backbone where some or all of their nodes / links are located within the threatened node set. That is, the critical path is a subset of the backbone that is threatened by weather. A backbone may contain multiple paths, of which only those paths passing through the threatened area are identified as critical paths and need to be prioritized for reconstruction; the expected availability probability can be calculated as follows: Step S21: Extract the fault records of the node under similar weather intensities from the historical database. For example, by statistically analyzing the probability (number of failures / total number of failures) of a node communication interruption when the precipitation in the region reaches a certain threshold over the past three years, the historical failure probability is obtained; Step S22: Obtain the node's current real-time link quality indicators (such as signal-to-noise ratio SNR, bit error rate BER, received signal strength RSSI, etc.), compare them with the baseline values ​​when the node is working normally, and calculate the current health coefficient, where the health coefficient ranges from 0 to 1, with 1 indicating complete health; Step S23: Based on the intensity information of this warning (such as precipitation), combined with the impact of the node's terrain (such as mountain tops or valleys) on communication, generate an attenuation factor (range 0 to 1, with smaller values ​​indicating greater impact). The attenuation factor can be based on an empirical model to establish a relationship curve between precipitation and signal attenuation.

[0024]

[0025] Secondly This represents the expected availability probability. For health index, Historical failure probability As a decay factor, the expected availability probability reflects the likelihood that a node can maintain normal communication during the warning period. To determine the confidence level of historical data, if the amount of data for that node is small, it can be... Lower it to reduce the influence of historical experience. To improve weather sensitivity, the trigger slope of weather conditions for faults is adjusted for different terrains (such as windward slopes and leeward slopes).

[0026] Historical failure probabilities cannot be used directly because historical data includes past performance under various weather conditions. We must use the intensity of the upcoming weather to "activate" or "filter" this historical experience. Specifically, this can be achieved as follows: Used to characterize the severity of weather; the worse the weather ( The smaller the value, the closer the difference is to 1.

[0027] This is used to characterize the multiplication of historical failure probabilities by the severity of current weather; the worse the weather, the higher the amplification of historical failures. If the weather is extremely good... ,but Approaching 0, regardless of how high the historical failure rate is, this value always approaches 0, meaning that historical failures do not pose a threat when the weather is good.

[0028] Subtracting this product from 1 gives us a "weather-corrected expected normality rate," which indicates how likely this node is not to fail, based on current weather conditions and historical experience.

[0029] This is used to characterize the final correction based on the "current state," meaning that history is only a reference, and the current specific state of the communication module itself is the most important. The specific principle is: if a node is currently about to fail (e.g., ...), ... Very low (e.g., 0.3), even though it is quite resilient in this weather based on historical experience (first-level result). If the probability is very high (e.g., 0.9), its final usability probability will be lowered to 0.27. If a node is currently in excellent condition... Therefore, the final result will mainly depend on the extent to which the weather triggers historical faults.

[0030] The core of the above conditional probability expression is: don't just look at whether a node is good or bad now, or whether it was bad or bad in the past, but predict what probability it has of surviving in the specific environment to come. This prediction result will directly serve as the underlying input for "maximizing the expected reliability of the path" in subsequent steps, thus enabling the entire network's path selection to predict the future.

[0031] Step S3: With the goal of avoiding threat areas and maximizing the expected reliability of paths, calculate one or more alternative communication paths for each critical path, and then generate a reconstruction plan containing the information of the alternative communication paths. This plan is then encrypted and distributed to relevant nodes to form a network reconstruction plan. Maximizing the expected reliability of paths means that for a candidate path, its overall reliability is characterized by the product of the expected availability probabilities of each link segment on the path (or a combination considering the number of hops). Maximizing this product means selecting a path that is most likely to remain unobstructed during the warning period. The optimization objective function for calculating the alternative communication paths comprehensively considers the product of the expected availability probabilities of each hop node, the total number of hops, the current load and remaining energy of each relay node, and the constraint requires that the alternative communication paths must bypass the core area of ​​the threatened node set. Specifically, for an alternative path... Suppose it is made by Jump composition, the first The expected availability probability of the hop link is (This can be derived by combining the expected availability probability of the two endpoints and the link quality, such as...) Then the expected reliability of the path. The optimization objective is to maximize... Simultaneously, other factors (such as hop count, node load, and remaining energy) are considered to form a multi-objective optimization function, for example: ;in, and As weight, This is the normalized value of the minimum remaining energy at each node along the path. This is the normalized value of the maximum current load of each node on the path.

[0032] The constraints are: the path must avoid the core area of ​​the threatened node set (e.g., completely bypass nodes marked as high threat), or the nodes on the path cannot all come from the threatened set.

[0033] Then, the K-shortest path algorithm in graph theory (such as Yen's algorithm) is used to calculate the total cost of each hop link (e.g., ...). Using other weighted costs as edge weights, calculate the top K optimal paths that satisfy the constraints. Then, use the above objective function to score and rank these K paths, selecting the top 1-3 paths as candidate paths.

[0034] Step S4: Within a preset lead time window before the meteorological event occurs, initiate a data flow path switching for non-critical services, switching the transmission path of periodic status data from the original critical path to the optimal alternative communication path. For critical alarm data, send the same data packets through both the original critical path and the optimal alternative communication path. The endpoint of the optimal alternative communication path points to a currently available first communication module. Non-critical service data includes periodic status monitoring data (such as routine temperature, humidity, and current values, which can tolerate a certain delay), video surveillance recordings (large data volume, but not requiring high real-time performance), log files, etc. This type of data can be slowed down or paused during periods of network resource scarcity, without posing an immediate threat to power grid safety. Critical warning data refers to real-time alarm information crucial to the safe operation of the power grid, such as conductor temperature exceeding limits, tower tilt exceeding thresholds, and insulator pollution flashover. This type of data requires extremely high reliability and low latency; its loss could have serious consequences, therefore it needs to be prioritized for protection.

[0035] Within a preset lead time window before the meteorological event, the link quality between each first communication module and the public network base station is continuously monitored. When a first communication module detects that its public network link quality is continuously lower than a first threshold, it actively broadcasts a degradation warning message containing its identity and link quality, and notifies relevant nodes that its availability as a path endpoint is about to be lost. Candidate nodes receiving the degradation warning message, based on their link quality with the public network base station, remaining node energy, and network topology centrality indicators, perform a distributed gateway election. The distributed gateway election adopts a consensus mechanism based on voting, where each node, according to the received... The election process involves voting on candidate nodes based on a weighted sum of their public network link quality score, remaining energy score, and topology centrality score. The candidate node receiving the most votes, exceeding a preset number, is confirmed as the winner. The winning node establishes a connection with the public network base station, upgrades its status to the new first communication module, and broadcasts its gateway identity and routing information, simultaneously triggering a routing table update. The public network link quality score reflects the stability and quality of the communication link between the node and the public base station, typically derived from real-time measurements of signal strength (RSRP), signal-to-noise ratio (SINR), and bit error rate. A higher score indicates a stronger potential gateway capability for the node.

[0036] The remaining energy score is used to reflect the node's own power supply capacity (such as the remaining power of the solar battery). The higher the score, the longer the node can work continuously, making it suitable for undertaking the important tasks of a gateway or relay.

[0037] The topology centrality score is used to reflect the importance of a node's position in the network topology, such as its degree (number of connected neighbors) and betweenness (number of shortest paths through the node). The higher the centrality, the more efficiently the node can cover other nodes as a gateway.

[0038] Secondly, the specific voting method described above is as follows: In the distributed gateway election, each participating node (usually the candidate node itself and its neighboring nodes) will conduct a comprehensive evaluation based on the election messages received (including various scores of the candidate nodes), and then vote for the candidate node it supports. Voting can be based on preset voting rules, such as casting votes for the candidate with the highest comprehensive score.

[0039] It should be noted that the reason for not directly selecting the highest value is that in a distributed system, nodes may have inconsistent perceptions of candidate node scores due to information delays or differences in local views. Directly selecting the highest value could lead to multiple nodes simultaneously claiming victory, causing conflict. The voting mechanism achieves consensus through majority rule, avoiding disagreement. The voting process allows nodes to verify each other; if a candidate node overestimates its own score but its neighbors perceive it poorly, other nodes can vote against it. Secondly, sometimes the node with the highest overall score may already be carrying too many tasks. The voting mechanism allows for the introduction of randomness or preference to avoid overloading a single node. The voting process requires more than a quorum (e.g., more than half) to ensure the new gateway is widely accepted and avoids network splits. Voting is a necessary means to achieve distributed consensus, not just a simple ranking.

[0040] Step S5: In response to the generation of the new first communication module, all alternative communication paths that have switched to or are preparing to switch to the original first communication module update their path endpoints to the new first communication module, and adjust the routing tables of intermediate nodes accordingly to ensure that the service flow always points to the actually available gateway. Step S6: When a communication module cannot communicate with a known first communication module within a preset time period, it automatically identifies itself as entering a partitioned island state. Each communication module periodically sends heartbeat messages to the known gateway (first communication module) and listens for the gateway's periodic broadcasts. If no response is received from the gateway or a route cannot be established within a preset timeout period (e.g., 3 consecutive periods), it is determined that it has lost connection with the gateway. Simultaneously, if a node attempts to communicate with other nodes in the network but cannot find a path to the gateway, it can be confirmed that it has been disconnected from the outside world. When a node determines that it may be in an island, it will actively broadcast an "island detection" message. Nodes that receive this message and are also unable to contact the gateway will reply and exchange neighbor information. Through this diffuse neighbor discovery, the entire island's member set can ultimately be outlined. A flood-like approach can be used, where each node maintains a list of isolated members and periodically swaps and merges the lists until convergence.

[0041] Step S7: Nodes within the isolated island elect a unique temporary aggregation node through a distributed consensus algorithm. This temporary aggregation node is responsible for collecting monitoring data uploaded by all other nodes within the island and storing it in local non-volatile memory according to data priority, forming a structured partitioned data set. The temporary aggregation node performs lossy compression or anomaly detection storage on the collected periodic status data, completely saves key alarm data, and generates a partition summary report containing alarm statistics. The partition summary report includes the number, type, first and last occurrence times of key alarm events within the island, the numerical range of periodic status data, and the number of anomalies. The number represents the total number of various alarms occurring within the island, used to assess the severity of the security situation within the island. The type is used to distinguish between different types of alarms. The alarms (such as over-temperature, tilt, and flashover) help the backend quickly locate the nature of the problem. The first and last occurrence times provide the time span of the alarm event, facilitating the determination of whether the event is continuous or intermittent, and whether new alarms have been generated. The numerical range of the periodic status data records the minimum and maximum values ​​of regular data (such as temperature and humidity) collected by each monitoring point within the island, enabling rapid identification of abnormal fluctuations. The number of anomalies represents the number of data points in the periodic data that deviate from the normal threshold, used to quantify the degree of data anomaly. In the scheme described in step S7, when the network recovers, a summary report is exchanged first, allowing the receiver (gateway or central platform) to quickly understand the important events and data overview occurring within the island without transmitting all data, thereby determining the priority and urgency of the data transmission. For example, if the summary shows multiple critical alarms, the complete alarm data is transmitted first; if the periodic data range is normal and there are few anomalies, the transmission of the compressed package can be delayed. This mechanism effectively avoids blind transmission and improves recovery efficiency.

[0042] The specific network election method can be as follows: Using the Bully algorithm, each node decides whether to run for election based on its own ID (or remaining energy, centrality, etc.). A node sends an election message to all nodes "stronger" than itself. If it does not receive a response from a stronger node, it declares victory. The election principle of this algorithm is a conventional technique and will not be further elaborated here. Secondly, the above method is one way to implement this technical solution. This embodiment can also use a priority comparison method for election. Specifically, each node broadcasts its own priority (such as a weighted sum of remaining energy and centrality), and all nodes record the highest priority received and its source. After several rounds, if a node consistently has the highest priority, it nominates itself as the convergence point and broadcasts a confirmation message to the entire network.

[0043] After the election is completed, the temporary aggregation node will broadcast its identity, and other nodes will register with it and begin uploading data.

[0044] Step S8: When the network link is restored and different islands can establish a stable connection, the temporary aggregation nodes first exchange the partition summary report so that the first communication module, as the receiver, can know the type, priority distribution and data size of the data to be transmitted back by the sender. The standard for establishing a stable connection can be: the boundary nodes between the two islands can successfully exchange data packets (such as heartbeat messages) and maintain a packet loss rate below a threshold (such as 5%) within a certain time window (such as 5 consecutive minutes), then the signal strength is considered stable.

[0045] Step S9: Based on the known data backlog and its currently available backhaul bandwidth, the first communication module negotiates with the temporary aggregation node of the sender to determine an initial backhaul bandwidth quota and reach a transmission rate control protocol. The negotiation adopts a bandwidth allocation protocol based on a token bucket model. The first communication module allocates a sustainable data token rate to the temporary aggregation node, and the temporary aggregation node must adjust its data transmission rate according to this data token rate. Step S10: According to the transmission rate control protocol, the sender prioritizes transmitting all unreported critical alarm data. After the critical alarm data transmission is completed, it backhauls compressed periodic status data or its abnormal subset according to the negotiated bandwidth quota.

[0046] Step S11: Under the premise that the real-time service bandwidth is guaranteed, use the remaining bandwidth to schedule the back transmission of large-capacity inspection data to complete the fusion of partition data.

[0047] In this embodiment, by spatially overlaying meteorological warning data with network topology, this solution can identify threatened nodes and pre-calculate alternative paths before weather events occur, transforming the traditional "passive response" into "proactive prevention." During the predictive reconfiguration phase, a service-specific, phased path switching strategy is adopted. Non-critical services are migrated and verified first, while critical services utilize dual-path concurrency, ensuring service continuity and zero data loss during the switching process. When a gateway node link degrades, a distributed election mechanism quickly completes the transfer of the gateway role, achieving seamless migration of the network aggregation point and avoiding localized network paralysis caused by the failure of a single gateway.

[0048] Secondly, the path calculation comprehensively considers multiple dimensions of indicators, including the expected availability probability of nodes (integrating historical failures, current status, and weather intensity), remaining energy, current load, and topological centrality, constructing a multi-objective optimization function. This ensures that the selected path is not only available at the current moment but also maintains high reliability during the warning period, avoiding the frequent path failures caused by traditional methods that select routes solely based on instantaneous signal strength. Maximizing the expected reliability of the path ensures that data is transmitted along the most stable path possible, significantly reducing packet loss rate and retransmission frequency.

[0049] When a network is fragmented into isolated islands due to factors such as weather, this invention enables the formation of autonomous units within these islands through steps such as island self-identification, temporary aggregation node election, and hierarchical storage of local data. The temporary aggregation node is responsible for collecting and compressing periodic data, fully preserving key alarm data, and generating summary reports, effectively solving the problem of data loss during periods of network outages in traditional methods. This mechanism ensures that even during prolonged periods of network outages, monitoring data collected within the isolated islands can be systematically preserved, providing a complete basis for post-event analysis.

[0050] During the network healing phase, this invention designs a progressive backhaul mechanism based on metadata digest exchange and bandwidth negotiation. First, lightweight partition digest reports are exchanged, enabling the receiver to quickly grasp the data overview within the isolated area. Then, backhaul bandwidth is negotiated via the token bucket protocol, and data is transmitted in stages according to the priority order of critical alarms, compressed status data, and large-capacity inspection data. This mechanism avoids channel congestion and gateway overload caused by a large number of nodes simultaneously sending backlogged data during traditional recovery processes, achieving a smooth integration of historical and real-time data and ensuring stable operation after network recovery.

[0051] This solution incorporates a business priority concept throughout the entire process: critical alarm data is processed via dual-path concurrency during path switching; the integrity of critical alarm data is prioritized when storing data in isolated locations; and strict priority scheduling is followed during recovery and backhaul. This business-aware differentiated processing ensures that alarm data crucial to power grid security always receives the highest priority transmission resources, significantly improving the real-time performance and reliability of power grid monitoring.

[0052] In summary, this embodiment systematically solves the reliability problem of high-voltage tower communication networks in mountainous areas under complex and dynamic environments through innovative means such as weather forecast-driven, multi-objective path optimization, regional autonomy, orderly recovery, and service awareness, significantly improving network resilience, data delivery rate, and resource utilization efficiency.

[0053] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for online edge communication in a power transmission network, applicable to power transmission communication networks, wherein the power transmission communication network includes communication modules installed on multiple power transmission towers, and the communication modules are dynamically divided into a first communication module or a second communication module according to their connection status with a public network base station, wherein the first communication module is a communication module connected to the public network base station, and the second communication module is a communication module not connected to the public network base station, characterized in that... The method includes: responding to meteorological warning data, retrieving the digital topology map and node geographic location information of the power transmission network, and spatially overlaying the meteorological warning data with the digital topology map to obtain a set of threatened nodes, wherein the meteorological warning data includes spatial range and intensity information; based on the set of threatened nodes and the current link status of the power transmission communication network, generating a corresponding reconstruction plan containing alternative communication path information, and encrypting and distributing it to relevant nodes to form a network reconstruction plan; within a preset advance time window before the meteorological event, initiating the switching of data flow paths for non-critical services and the synchronous transmission of critical alarm data, while continuously monitoring the link quality between each first communication module and the public network base station, and when the link quality is continuously lower than a first threshold, electing a new first communication module through a distributed gateway election mechanism, enabling it to establish a connection with the public network base station, and adjusting the routing table of intermediate nodes accordingly to ensure that the service flow always points to an actually available gateway.

2. The online edge communication method for power transmission networks according to claim 1, characterized in that, The step of generating a corresponding reconstruction plan containing alternative communication path information based on the threatened node set and the current link status of the power transmission communication network includes: calculating the expected availability probability of each node in the threatened node set during the warning period based on historical fault data and the current link status of the power transmission communication network; identifying all critical paths passing through the threatened node set based on the data flow backbone in the current power transmission communication network; and calculating one or more alternative communication paths for each critical path with the goal of avoiding threat areas and maximizing the expected reliability of the paths, thereby generating a reconstruction plan containing the alternative communication path information. The optimization objective function for calculating the alternative communication paths comprehensively considers the product of the expected availability probabilities of each hop node in the path, the total number of hops in the path, the current load of each relay node, and the remaining energy. The constraint condition of the optimization objective function is that the alternative communication paths must bypass the core area of ​​the threatened node set.

3. The online edge communication method for power transmission networks according to claim 2, characterized in that, The process involves switching the data flow path for non-critical services and synchronously sending critical alarm data. Simultaneously, it continuously monitors the link quality between each first communication module and the public network base station. When the link quality consistently falls below a first threshold, a new first communication module is selected through a distributed gateway election mechanism. This includes: within a preset lead time window before the weather event, switching the data flow path for non-critical services from the original critical path to the optimal alternative communication path for periodic status data; for critical alarm data, sending the same data packet through both the original critical path and the optimal alternative communication path; wherein the endpoint of the optimal alternative communication path points to a currently available first communication module; and continuously monitoring the link quality between each first communication module and the public network base station within the preset lead time window before the weather event. For link quality, when a first communication module detects that its public network link quality is continuously lower than a first threshold, it actively broadcasts a degradation warning message containing its identity and link quality. Candidate nodes receiving the degradation warning message perform a distributed gateway election based on their link quality with the public network base station, remaining node energy, and network topology centrality. The distributed gateway election adopts a voting-based consensus mechanism. Each node votes on candidate nodes based on the weighted sum of their public network link quality score, remaining energy score, and topology centrality score, according to the received election message. The candidate node with the most votes exceeding a preset number is confirmed as the winner. The winning node establishes a connection with the public network base station, upgrades its identity to a new first communication module, and broadcasts its gateway identity and routing information, while simultaneously triggering a routing table update.

4. The online edge communication method for power transmission networks according to claim 3, characterized in that, The corresponding adjustment of the routing and forwarding tables of intermediate nodes to ensure that the service flow always points to the actually available gateway includes: in response to the generation of the new first communication module, updating the path endpoint of all alternative communication paths that have been switched to or are about to switch to the original first communication module to the new first communication module, and correspondingly adjusting the routing and forwarding tables of intermediate nodes to ensure that the service flow always points to the actually available gateway.

5. The online edge communication method for power transmission networks according to claim 4, characterized in that, After electing a new first communication module, establishing a connection with the public network base station, and adjusting the routing table of intermediate nodes accordingly, the process further includes: when a communication module cannot communicate with a known first communication module within a preset time period, it automatically identifies itself as entering a partitioned island state. Nodes within the island elect a unique temporary aggregation node through a distributed consensus algorithm. This temporary aggregation node is responsible for collecting monitoring data uploaded by all other nodes within the island and storing it in local non-volatile memory according to data priority, forming a structured partitioned data set. The temporary aggregation nodes first exchange partition summary reports, enabling the first communication module, as the receiver, to... The first communication module learns the type, priority distribution, and data volume of the data to be transmitted back from the sender. Based on the obtained data backlog and its currently available backhaul bandwidth, the first communication module negotiates with the sender's temporary aggregation node to determine an initial backhaul bandwidth quota and reaches a transmission rate control protocol. According to the transmission rate control protocol, the sender prioritizes transmitting all unreported critical alarm data. After the critical alarm data transmission is completed, the sender transmits compressed periodic status data or its abnormal subset according to the negotiated bandwidth quota. Under the premise that the real-time service bandwidth is guaranteed, the remaining bandwidth is used to schedule the backhaul of large-capacity inspection data stored in the system to complete the fusion of partitioned data.

6. The online edge communication method for power transmission networks according to claim 5, characterized in that, The temporary aggregation node performs lossy compression or anomaly detection storage on the collected periodic status data, completely saves key alarm data, and generates a partition summary report containing alarm statistics. The partition summary report includes the number, type, first and last occurrence time of key alarm events within the island, the numerical range of periodic status data, and the number of anomalies.

7. The online edge communication method for power transmission networks according to claim 5, characterized in that, The negotiation adopts a bandwidth allocation protocol based on the token bucket model. The first communication module allocates a sustainable data token rate to the temporary aggregation node, and the temporary aggregation node must adjust its data transmission rate according to the data token rate.

Images