Routing method of electric power communication network based on path risk quantification

Abstract

The invention discloses an electric power communication network routing method based on path risk quantification, and belongs to the technical field of electric power communication. The method comprises the following steps: constructing an electric power communication network topological graph according to a network structure of an electric power communication network; calculating a node risk value of each power node according to the historical fault rate and the traffic load factor of each power node; determining the fault probability of each power node according to the node risk value; and determining a target routing path of the power network by taking the minimum path cost of the routing path as a target. According to the scheme, when the target routing path is selected, the transmission efficiency, the security and the stability are ensured, and the reliability and the security of data transmission are improved.

Description

Technical Field

[0001] This invention relates to the field of power communication technology, and in particular to a routing method for power communication networks based on path risk quantification. Background Technology

[0002] Power communication networks are dedicated communication networks used to support various information transmission, control, and monitoring within power systems. They are a core component of smart grids, undertaking tasks such as real-time data transmission, remote monitoring, and equipment control to ensure the efficient, safe, and reliable operation of the power system.

[0003] The routing algorithms used in existing power communication networks are mainly divided into traditional routing algorithms, heuristic routing algorithms, and deep reinforcement learning-based routing algorithms. Traditional routing algorithms calculate paths based on predefined metrics such as hop count, bandwidth, and latency, offering simplicity but lacking adaptability to dynamic network changes. Heuristic routing algorithms incorporate biological inspiration and swarm intelligence concepts, enabling them to adapt to network environment changes to some extent, but still primarily focus on performance optimization rather than security considerations. Deep reinforcement learning-based routing algorithms possess powerful learning and adaptability capabilities, but still have shortcomings in risk identification and security awareness.

[0004] Therefore, power communication networks face increasingly complex security threats, but existing algorithms lack the ability to perceive and respond to dynamic security threats, resulting in low reliability and security of data transmission. Summary of the Invention

[0005] Therefore, it is necessary to provide a routing method for power communication networks based on path risk quantification to address the above problems, thereby improving the reliability and security of data transmission.

[0006] On the one hand, this application provides a routing method for power communication networks based on path risk quantification, the method comprising: Based on the network structure of the power communication network, a power communication network topology diagram is constructed; the power communication network topology diagram includes several power nodes and edges connecting the power nodes; the power communication network topology diagram also stores the network parameters of each power node and each edge; Based on the historical failure rate and traffic load factor of each power node, the node risk value of each power node is calculated; the traffic load factor is used to characterize the ratio between the degree of the power node and the link bandwidth of the power node. Based on the node risk value, determine the failure probability of each power node; The target routing path of the power network is determined with the goal of minimizing the path cost of the routing path; the path cost is used to indicate the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

[0007] In one optional implementation, the path cost is a weighted sum of the original path cost, the maximum risk cost, the total risk cost, and the failure probability cost; The original path cost is used to characterize the delay and number of nodes traversed in the routing path; the maximum risk cost is used to characterize the maximum node risk value on the routing path; the total risk cost is used to characterize the sum of node risk values ​​of power nodes on the routing path; and the failure probability cost is the sum of failure probabilities of power nodes on the routing path.

[0008] In one alternative implementation, the original path cost is a weighted sum of the edge delay cost of each edge in the routing path and the number of nodes; The edge delay cost for each edge is a weighted sum of the edge's original transmission delay, the predicted failure probability of the target node toward which the edge is directed, and the node risk value of the target node.

[0009] In one optional implementation, calculating the node risk value of each power node based on its historical failure rate and flow load factor includes: For any power node in the power communication network topology, obtain the neighboring nodes of the power node; the distance between the neighboring nodes and the power node is less than k hops; The fault impact radius of the power node is obtained based on the importance score of the power node in the power communication network topology, the path quality weight between each neighboring node and the power node, and the node distance. The node risk value of the power node is calculated based on the fault impact radius of the power node, the load flow factor of the power node, and the historical failure rate of the power node.

[0010] In one alternative implementation, the node risk value of the power node is positively correlated with the fault impact radius of the power node, the degree of the power node, and the fault probability of the power node.

[0011] In an optional implementation, before obtaining the fault impact radius of the power node based on the importance score of the power node in the power communication network topology graph, the path quality weights between each neighboring node and the power node, and the node distance, the method further includes: The power communication network topology is input into a graph neural network to obtain the importance score of each power node in the power communication network topology.

[0012] In one optional implementation, determining the failure probability of each power node based on the node risk value includes: For each power node, a candidate probability is generated according to the node risk value of the power node in a first proportion; When the node risk value is less than the first threshold, the minimum value between the candidate probability and the first probability is determined as the failure probability of the power node. When the node risk value of the power node is greater than or equal to the first threshold and less than or equal to the second threshold, the minimum value between the candidate probability and the second probability is determined as the failure probability of the power node. When the node risk value of the power node is greater than the second threshold, the minimum value of the candidate probability and the third probability shall be determined as the failure probability of the power node.

[0013] In another aspect, this application provides a routing device for a power communication network based on path risk quantification, the device comprising: The topology graph construction module is used to construct a power communication network topology graph based on the network structure of the power communication network; the power communication network topology graph includes several power nodes and edges connecting the power nodes; The risk acquisition module is used to calculate the node risk value of each power node based on the historical failure rate and traffic load factor of each power node; the traffic load factor is used to characterize the ratio between the degree of the power node and the link bandwidth of the power node. The fault probability acquisition module is used to determine the fault probability of each power node based on the node risk value. The routing path determination module is used to determine the target routing path of the power network with the goal of minimizing the path cost of the routing path; the path cost is used to indicate the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

[0014] On the other hand, an electronic device is provided, comprising: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes these computer instructions to perform the aforementioned routing method for power communication networks based on path risk quantification.

[0015] In another aspect, a computer-readable storage medium is provided, on which computer instructions are stored, the computer instructions being used to cause a computer to execute the above-described routing method for power communication networks based on path risk quantification.

[0016] Compared with the prior art, the technical solution provided in this application has the following advantages: When routing data in a power communication network, the controller constructs a power communication network topology map, which includes several power nodes and edges connecting them. The controller then obtains the historical failure rate and load factor for each power node. The load factor is the ratio of the node's degree to the bandwidth of the connection link between the node and other nodes; in other words, the load factor characterizes the potential load pressure on the power node. Therefore, based on the historical failure rate and load pressure of each power node, a node risk value can be defined for each node. The failure probability of each power node is then predicted based on its risk value. After obtaining the above-mentioned data for each node, path planning can be used to determine the target routing path of the power network, aiming to minimize the path cost. The target routing path obtained by this scheme considers the node length, the delay of the traversed edges, the failure probability of the traversed nodes, and the node risk value. In selecting the target routing path, transmission efficiency, security, and stability are simultaneously guaranteed, improving the reliability and security of data transmission. Attached Figure Description

[0017] Figure 1 A flowchart of a routing method for a power communication network based on path risk quantification according to an embodiment of the present invention is shown. Figure 2 A flowchart illustrating a routing method for a power communication network based on path risk quantification according to an embodiment of the present invention is shown. Figure 3 This illustration shows a schematic diagram of a key node importance scoring prediction model according to an embodiment of this application; Figure 4 The results of the path influence rate comparison experiment are shown in the figure; Figure 5 The results of the packet loss rate performance comparison experiment are shown in the figure; Figure 6 The experimental results comparing delay performance are shown in the figure. Figure 7 The experimental results of the path risk assessment are shown in the figure; Figure 8 The results of the network throughput comparison experiment are shown in the figure; Figure 9The results of the path reliability comparison experiment are shown in the figure; Figure 10 This paper illustrates a structural block diagram of a routing device for a power communication network based on path risk quantification, according to an embodiment of this application. Figure 11 This is a schematic diagram of the structure of an electronic device provided in an optional embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0019] With the deepening of digital transformation, power communication networks are facing increasingly complex security threats. Traditional routing algorithms mainly select paths based on static topology information and performance indicators, lacking the ability to perceive and respond to dynamic security threats. When faced with network attacks such as DDoS attacks and node hijacking, it is difficult to quickly adjust routing strategies, making it difficult to guarantee the reliability and security of data transmission.

[0020] Existing routing algorithms are mainly divided into traditional routing algorithms, heuristic routing algorithms, and deep reinforcement learning-based routing algorithms. Traditional routing algorithms calculate paths based on predefined metrics such as hop count, bandwidth, and latency, which are simple to implement but lack adaptability to dynamic network changes. Heuristic routing algorithms incorporate biological inspiration and swarm intelligence concepts, enabling them to adapt to changes in the network environment to some extent, but they still primarily focus on performance optimization rather than security considerations. Deep reinforcement learning-based routing algorithms have powerful learning and adaptive capabilities, but they still have shortcomings in risk identification and security awareness.

[0021] Most existing algorithms focus on performance metrics such as path optimization, load balancing, and QoS guarantees, lacking consideration for adjusting routing policies in the event of network attacks. When faced with security threats such as malicious attacks, node failures, or link interruptions, they often fail to identify risky nodes and adjust routing paths in a timely manner. In addition, traditional fault tolerance mechanisms mainly rely on redundant paths and backup mechanisms, which not only increase system costs in resource-constrained environments but may also introduce new points of failure due to increased management complexity.

[0022] To address the aforementioned issues, this application provides a routing method for power communication networks based on path risk quantification. Figure 1This diagram illustrates a method flowchart for a routing method in a power communication network based on path risk quantification, according to an embodiment of the present invention. This method can be executed by a controller in the power communication network, such as... Figure 1 As shown, the routing method for power communication networks based on path risk quantification provided in this embodiment of the invention includes the following steps: Step 101: Construct a power communication network topology diagram based on the network structure of the power communication network.

[0023] The power communication network topology diagram includes several power nodes and the edges connecting them. The diagram also stores the network parameters for each power node and each edge.

[0024] In this step, the system first constructs a power communication network topology map for routing analysis based on the actual structural information of the target power communication network.

[0025] Specifically, the controller can collect structural information of various communication devices, substation communication nodes, fiber optic links, and wireless links in the power communication network to establish a topology graph G=(V, E), where: V represents the set of power communication nodes in the network, and each node can correspond to the dispatch center, substation terminal, aggregation equipment, etc.; E represents the set of edges between power communication nodes, and each edge corresponds to an actual communication link, such as a fiber optic link, carrier link, or wireless communication link.

[0026] In this embodiment of the application, the power communication network topology diagram further records the network parameters of each edge, including parameters such as bandwidth, transmission delay, and link type, as well as the network parameters of each node, such as failure probability, for subsequent node risk calculation and routing cost modeling.

[0027] Step 102: Calculate the node risk value of each power node based on its historical failure rate and flow load factor.

[0028] The traffic load factor is used to characterize the degree of the power node and the ratio between the link bandwidth of the power node and other power nodes. The link bandwidth of a power node is essentially a combination of the bandwidths of the links connecting the power node to other nodes.

[0029] In this embodiment of the application, in order to achieve path risk quantification, it is necessary to perform risk assessment on each node in the topology graph to obtain the node risk value R(v).

[0030] In order to obtain the node risk value, this embodiment of the application needs to first obtain the historical failure rate and traffic load factor of the power node, and then use the historical failure rate and traffic load factor of the power node to evaluate the node risk value of the power node.

[0031] The historical failure rate can be obtained by analyzing the historical operation records, alarm information and maintenance logs of power communication nodes. Nodes with higher historical failure rates are usually less stable and should be given a higher risk weight in path selection.

[0032] The traffic load factor is used to characterize the relative magnitude of the communication pressure carried by a node. It is defined as the ratio of the node's degree to the total bandwidth of the node's connection links. The node's degree is used to represent the node's connection complexity, that is, how many other nodes it is connected to.

[0033] The link bandwidth of a power node represents the sum of the bandwidths of the links connecting the power node to each other, which in turn represents the total communication capacity that the power node can provide. Therefore, the larger the traffic load factor, the more connections the node has and the limited link bandwidth, making it more likely to become a high-load, bottleneck node, and thus a higher level of risk.

[0034] The node risk value R(v) can be calculated by fusing historical failure rates and traffic load factors according to a preset model (such as product or weighted summation), and is used as a key input for subsequent failure probability prediction.

[0035] Step 103: Determine the failure probability of each power node based on the node risk value.

[0036] In this embodiment, the controller may include a node failure prediction model to map the node risk value to the probability that the node may experience a communication failure in the future. For example, the likelihood of a node experiencing a link interruption, communication anomaly, or device disconnection.

[0037] In one optional implementation, high, medium, and low risk thresholds can be set according to different risk value ranges, and the node failure probability can be calculated through a hierarchical prediction method. For example, when the node risk value is higher than the high risk threshold, a higher failure probability is output; when the node risk value is in the medium risk range, a medium failure probability is output; and when the node risk value is low, a low failure probability is output.

[0038] Step 104: Determine the target routing path for the power network with the goal of minimizing the path cost of the routing path.

[0039] The path cost is used to indicate the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

[0040] In this embodiment, the controller needs to search the topology graph for a path that meets the business requirements and has the minimum path cost as the target routing path. The path cost function comprehensively considers network performance indicators and node risk indicators, namely the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

[0041] Among them, path length is used to reflect the number of nodes traversed by the path. More hops mean a longer path and greater cumulative latency. Transmission delay at each edge is used to describe the basic transmission performance of the link. Higher latency means a worse transmission capacity of the link and a lower priority should be given. Failure probability of each node in the path is used to measure the risk of path interruption due to node failure. Node risk value of each node in the path is used to further reflect the potential risk exposure of the path. For example, the more high-risk nodes there are, the worse the path stability.

[0042] The controller calculates the path cost for all candidate paths and selects the path with the minimum path cost as the target routing path, thereby achieving a balance between risk awareness and optimal performance and providing a stable, reliable, and secure routing strategy for the power communication network.

[0043] In summary, when routing power communication networks, the controller constructs a power communication network topology diagram, which includes several power nodes and edges connecting them. The controller then obtains the historical failure rate and load factor of each power node. The load factor is the ratio of the node's degree to the bandwidth of the connection link between the node and other nodes; in other words, the load factor characterizes the potential load pressure on a power node. Therefore, based on the historical failure rate and load pressure of each power node, a node risk value can be defined for each node. The failure probability of each power node is then predicted based on its risk value. After obtaining the aforementioned data for each node, path planning can be used to determine the target routing path of the power network, aiming to minimize the path cost. The target routing path obtained by this scheme considers the node length, the delay of traversed edges, the failure probability of traversed nodes, and the node risk value. It simultaneously ensures transmission efficiency, security, and stability when selecting the target routing path, improving the reliability and security of data transmission.

[0044] Figure 2 A flowchart illustrating a routing method for a power communication network based on path risk quantification according to an embodiment of the present invention is shown. Figure 2 As shown, the method includes: Step 201: Based on the network structure of the power communication network, construct a power communication network topology diagram. This power communication network topology diagram includes several power nodes and the edges connecting these power nodes.

[0045] In this embodiment, the controller first constructs a network topology diagram G = (V, E) based on the actual architecture of the target power communication network. Wherein: V represents several power communication nodes, including dispatch center nodes, substation communication nodes, aggregation and switching units, etc.; E represents the communication links connecting the power communication nodes, such as fiber optic links, power line carrier links, wireless links, etc.

[0046] During the process of building the topology map, the controller can also record basic parameters of the links, such as bandwidth, physical distance, transmission delay, and link type, so that they can be used in subsequent steps to calculate path quality weight and path cost.

[0047] Step 202: For any power node in the power communication network topology, obtain the neighboring nodes of that power node.

[0048] In this embodiment of the application, the distance between the neighboring node and the power node is less than k hops.

[0049] In other words, the controller can obtain the set of neighboring nodes N of the target node v in the network based on the topology graph. k (v). Where: neighboring nodes are all nodes whose distance from node v is less than k hops.

[0050] By limiting the neighborhood range k, the scope of risk propagation can be effectively restricted, allowing subsequent risk quantification models to focus on the local structure around the node without obscuring the influence of key nodes in the local communication area.

[0051] Step 203: Based on the importance score of the power node in the power communication network topology, the path quality weights between each neighboring node and the power node, and the node distance, obtain the fault impact radius of the power node.

[0052] In this embodiment of the application, the fault influence radius The calculation formula is as follows:

[0053] in, Let k be the set of all nodes that are no more than k hops away from node v (i.e., the neighboring nodes mentioned above), where k represents the range of hops. The path quality weight from source node u to target node v (in this embodiment, the path quality weight can be obtained based on the bandwidth between the source node and the target node), λ=0.8 is the attenuation factor, and S(v) refers to the importance score of the power node.

[0054] Optionally, in this embodiment of the application, the importance score of each power node can be obtained in the following way: Input the power communication network topology into a graph neural network to obtain the importance score of each power node in the topology.

[0055] Specifically, Figure 3 This diagram illustrates a key node importance scoring prediction model according to an embodiment of this application. This prediction model can be used as the aforementioned graph neural network, such as... Figure 3 As shown, in this embodiment of the application, a key node identification method (AGNN) that combines an autoencoder and a graph neural network (GNN) architecture is used to construct a key node scoring prediction model that adapts to the structure of the power communication backbone network, thereby scoring each power node to effectively identify key nodes in the power communication backbone network.

[0056] Specifically, such as Figure 3 As shown, the key node identification model provided in this application includes an autoencoder and a ranking prediction module based on a graph neural network; the autoencoder module is used to learn the latent structural features of nodes from the network topology; the ranking prediction module is used to rank the importance of each node according to the latent features, thereby realizing key node identification.

[0057] In an autoencoder, the input layer is used to perform one-hot encoding on each node in the graph to form an initial feature matrix; each node has only its own position set to 1, and the rest are 0.

[0058] Then the initial feature matrix is ​​output as a multi-layer GCN (i.e., a graph convolutional network), for example... Figure 3 The algorithm employs a two-layer GCN. The first-layer GCN extracts local neighborhood features of nodes to generate the first-stage potential representation. The second-layer GCN further integrates multi-hop neighborhood information to generate the final potential representation.

[0059] Then the reconstructed output layer decodes the final latent representation to obtain the final reconstructed output Z; then it is processed by the loss function. The loss between the reconstructed output Z and the node truth value degree is calculated, thereby enabling the model to learn potential structural features that can characterize the degree of node topological contribution.

[0060] In the model, the latent representation of the first stage of the autoencoder is input into the ranking prediction module based on the graph neural network to train and obtain the node importance scores and ranking.

[0061] Specifically, in the ranking prediction module, the latent representation is further enhanced with a graph convolutional network (GCN) to enhance its relationships, enabling the model to still perceive the dependencies between nodes during the ranking stage. Then, a fully connected (FC) layer maps the output of the GCN to a one-dimensional score space, thereby representing the model's prediction of the importance of each node.

[0062] Finally, supervised training is performed using the ranking scores and the real labels (i.e., the real key node influence) from the SIR model. The ListMLE loss function is used to optimize the overall ranking of the predicted ranking sequence, so that the ranking prediction module can stably output the importance scores and ranking results of each node.

[0063] Step 204: Calculate the node risk value of the power node based on its fault impact radius, load flow factor, and historical failure rate.

[0064] In this embodiment, the node risk value of the power node is positively correlated with the fault impact radius of the power node, the degree of the power node, and the historical failure rate of the power node.

[0065] Specifically, the formula for calculating the risk value R(v) of a node is as follows:

[0066] in, For flow load factor, Let v be the degree of node v. For the edge bandwidth, This is the sum of the bandwidths of the links directly connected to node v. The failure probability of node v based on historical data. Let V be the radius of the failure impact of node v.

[0067] Step 205: Determine the failure probability of each power node based on the node risk value.

[0068] In one alternative implementation, for each power node, candidate probabilities are generated according to a first ratio based on the node risk value of that power node; When the risk value of the node is less than the first threshold, the minimum value of the candidate probability and the first probability is determined as the failure probability of the power node. When the node risk value of a power node is greater than or equal to the first threshold and less than or equal to the second threshold, the minimum value between the candidate probability and the second probability is determined as the failure probability of the power node. When the node risk value of a power node is greater than the second threshold, the minimum value of the candidate probability and the third probability shall be determined as the failure probability of the power node.

[0069] Specifically, in this embodiment, for each node v in the network, the system calculates the corresponding failure probability based on its risk value R(v). :

[0070] in This is the high-risk threshold (i.e., the second threshold mentioned above). The threshold is set to the medium risk threshold (i.e., the first threshold mentioned above). At this threshold, the first probability is 0.1, the second probability is 0.3, and the third probability is 0.8. In this embodiment, by setting this hierarchical prediction mechanism, nodes with different risk levels can be effectively distinguished, providing an accurate failure probability assessment for subsequent path selection.

[0071] Step 206: Determine the target routing path for the power network with the goal of minimizing the path cost of the routing path.

[0072] The path cost is used to indicate the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

[0073] In one alternative implementation, the path cost is a weighted sum of the original path cost, the maximum risk cost, the total risk cost, and the failure probability cost. The original path cost is used to characterize the delay and number of nodes traversed by the routing path; the maximum risk cost is used to characterize the maximum node risk value on the routing path; the total risk cost is used to characterize the sum of node risk values ​​of the power nodes on the routing path; and the failure probability cost is the sum of failure probabilities of the power nodes on the routing path.

[0074] Specifically, this application embodiment designs a multi-dimensional path cost function, which comprehensively considers traditional routing metrics and risk perception factors. The path cost C is calculated using the following formula:

[0075] in, This represents the original path cost based on latency and hop count. and These represent the maximum node risk value and the sum of node risk values ​​on the path, respectively. This represents the sum of predicted failure probabilities for all nodes along the path. The weighting parameters α=0.3, β=0.25, γ=0.2, and δ=0.25 reflect the policy's emphasis on risk perception, ensuring that lower-risk transmission paths are prioritized while maintaining basic transmission performance.

[0076] Furthermore, the original path cost is a weighted sum of the edge delay cost of each edge in the routing path and the number of nodes; the edge delay cost of each edge is a weighted sum of the edge's original transmission delay, the failure prediction probability of the target node toward which the edge is directed, and the node risk value of the target node.

[0077] In the specific path search process, the strategy of this application embodiment also adopts a dynamic edge weight calculation method based on a prediction model. For each edge (u,v) in the network, the system not only considers its basic transmission delay, but also introduces a penalty mechanism based on the predicted probability of target node failure, and the dynamic edge cost of edge (u,v) is calculated. The calculation formula is as follows:

[0078] in, Let (u,v) be the original transmission delay. Let be the failure prediction probability of the target node v of edge (u,v). Let be the node risk value of the target node v of edge (u,v). This design allows the algorithm to automatically avoid nodes and links with high failure risk during path search.

[0079] To verify the effectiveness of the proposed risk-aware security routing algorithm, a simulation experimental environment was constructed. The experiment used Python 3.9 as the development language, implemented the algorithm model based on the NetworkX network analysis library and the PyTorch deep learning framework, and conducted simulation experiments on a high-performance computer.

[0080] This experiment utilized a city-level power communication backbone network dataset containing 300 nodes and randomly corrupted the dataset, setting five scenarios with node corruption rates of 3%, 6%, 9%, 12%, and 15%. To comprehensively evaluate the algorithm's performance, the experiment randomly selected 5000 different end-to-end paths as test samples, with each path containing three candidate route choices.

[0081] To verify the effectiveness of the method proposed in this application, a multi-dimensional performance evaluation system is also constructed in this application. The evaluation system includes the following five core indicators: Definition 1: Average packet transmission delay. The average packet transmission delay is the ratio of end-to-end delay to path length. Each link in the middle The sum of time delays is an additivity parameter.

[0082]

[0083] in, The end-to-end delay of the path, The latency of the links that form the path.

[0084] Definition 2: Average throughput. The end-to-end average throughput is the ratio of the total amount of data transmitted in the path to the end-to-end latency.

[0085] in, The average throughput of the path, For link throughput.

[0086] Definition 3: Packet Loss Rate. This invention defines the packet loss rate as the ratio of the total number of (source, destination) node pairs that experience packet loss during routing to the total number of (source, destination) nodes in the routing process. For each (source, destination) node, this invention searches for three paths using each routing strategy. If any of the three paths contain a corrupted node, then packet loss is determined to have occurred when routing that node pair.

[0087] Definition 4: Path Risk Assessment. The formula for calculating path risk assessment is as follows:

[0088] in, For path The risk value of the highest risk node in the middle. , For path Middle node The risk value, For path The number of nodes in = =0.5.

[0089] Definition 5: Path reliability. Path reliability The calculation formula is as follows:

[0090] The above indicators can be used to comprehensively evaluate the balance between performance optimization and risk control of the RASR strategy from different perspectives.

[0091] First, a comparative experiment on path influence rates can be conducted. Figure 4 The results of the path influence rate comparison experiment are shown in the figure. For example... Figure 4As shown, this embodiment compares the path impact rate performance of four routing strategies: Shortest Path Strategy (SPF), Minimum Delay Path Strategy (MDP), Maximum Bandwidth Path Strategy (MBP), and the Risk Awareness Secure Routing Strategy (RASR) proposed in this invention.

[0092] Experimental results show that the path impact rate of all three traditional strategies increases sharply with the increase of node destruction rate. The maximum bandwidth path strategy is most severely affected, with its path impact rate rising sharply from 21.3% at a 3% destruction rate to 50.2% at a 15% destruction rate; the minimum delay path strategy rises from 12.9% to 41.0%; and the shortest path strategy rises from 10.3% to 37.6%.

[0093] In contrast, the Risk-Aware Secure Routing (RASR) strategy maintained the lowest path impact rate throughout the testing process, rising only slightly from 5.1% at a 3% disruption rate to 22.3% at a 15% disruption rate. Under the stringent condition of a 15% disruption rate, the RASR strategy reduced the path impact rate by 15.3, 18.7, and 27.9 percentage points compared to the shortest path strategy, the minimum delay path strategy, and the maximum bandwidth path strategy, respectively.

[0094] This result fully demonstrates that the RASR strategy can effectively predict and avoid high-risk nodes through the risk perception mechanism, thereby significantly improving the stability and fault tolerance of network routing.

[0095] Secondly, a basic performance comparison experiment was also conducted in this embodiment. This embodiment comprehensively evaluates the basic performance of the RASR strategy from five dimensions: packet loss rate, transmission delay, path risk value, path reliability, and network throughput.

[0096] Figure 5 The results of the packet loss rate performance comparison experiment are shown in the figure. For example... Figure 5 As shown, in terms of packet loss rate performance, the RASR strategy consistently maintained the lowest packet loss rate throughout the entire test. Under a high failure rate of 15%, the shortest path strategy achieved a packet loss rate of 11.8%, the minimum delay path strategy achieved 13.5%, and the maximum bandwidth path strategy achieved an even higher rate of 17.4%. Compared to traditional strategies, the RASR strategy achieved packet loss rate reductions of 8.6%, 10.3%, and 14.2%, respectively.

[0097] Figure 6 The experimental results comparing latency performance are shown in the figure. Figure 6As shown, the RASR strategy also performs excellently in terms of latency performance. When the node corruption rate increases to 15%, the latency of the shortest path strategy, the minimum delay path strategy, and the maximum bandwidth path strategy reach 8.349ms, 8.334ms, and 8.418ms, respectively, while the latency of the RASR strategy is only 8.113ms, representing reductions of 2.8%, 2.7%, and 3.6% compared to these three baseline routing strategies.

[0098] Figure 7 The experimental results of the path risk assessment are shown in the figure. (See figure below.) Figure 7 As shown, in terms of path risk assessment, the RASR strategy consistently maintained the lowest path risk value throughout the entire test. Under the stringent condition of a 15% disruption rate, the RASR strategy reduced the path risk value by 7.5%, 8.0%, and 8.5% compared to the shortest path strategy, the minimum delay path strategy, and the maximum bandwidth path strategy, respectively.

[0099] Figure 8 The results of a network throughput comparison experiment are shown in the figure. Figure 8 As shown, the RASR strategy also demonstrates excellent performance in terms of network throughput, maintaining a high level of network throughput under various node failure rates.

[0100] Figure 9 The results of the path reliability comparison experiment are shown in the figure. (See figure below.) Figure 9 As shown, in terms of path reliability, the RASR strategy significantly improves the overall path reliability by actively avoiding high-risk nodes. Experimental results demonstrate that the RASR strategy maintains the highest path reliability index under various node failure rates, and can maintain a stable reliability level even under high failure rate environments.

[0101] Experimental results verify the comprehensive advantages of the reliable routing scheme for power communication networks based on path risk quantification proposed in this invention across multiple performance dimensions.

[0102] In summary, when routing power communication networks, the controller constructs a power communication network topology diagram, which includes several power nodes and edges connecting them. The controller then obtains the historical failure rate and load factor of each power node. The load factor is the ratio of the node's degree to the bandwidth of the connection link between the node and other nodes; in other words, the load factor characterizes the potential load pressure on a power node. Therefore, based on the historical failure rate and load pressure of each power node, a node risk value can be defined for each node. The failure probability of each power node is then predicted based on its risk value. After obtaining the aforementioned data for each node, path planning can be used to determine the target routing path of the power network, aiming to minimize the path cost. The target routing path obtained by this scheme considers the node length, the delay of traversed edges, the failure probability of traversed nodes, and the node risk value. It simultaneously ensures transmission efficiency, security, and stability when selecting the target routing path, improving the reliability and security of data transmission.

[0103] Figure 10 This document illustrates a structural block diagram of a routing device for a power communication network based on path risk quantification, as shown in an embodiment of this application. Figure 10 As shown, the device includes: The topology graph construction module 1001 is used to construct a power communication network topology graph based on the network structure of the power communication network; the power communication network topology graph includes several power nodes and edges connecting the power nodes; The risk acquisition module 1002 is used to calculate the node risk value of each power node based on the historical failure rate and traffic load factor of each power node; the traffic load factor is used to characterize the ratio between the degree of the power node and the link bandwidth of the power node. The fault probability acquisition module 1003 is used to determine the fault probability of each power node based on the node risk value. The routing path determination module 1004 is used to determine the target routing path of the power network with the goal of minimizing the path cost of the routing path; the path cost is used to indicate the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

[0104] In summary, when routing power communication networks, the controller constructs a power communication network topology diagram, which includes several power nodes and edges connecting them. The controller then obtains the historical failure rate and load factor of each power node. The load factor is the ratio of the node's degree to the bandwidth of the connection link between the node and other nodes; in other words, the load factor characterizes the potential load pressure on a power node. Therefore, based on the historical failure rate and load pressure of each power node, a node risk value can be defined for each node. The failure probability of each power node is then predicted based on its risk value. After obtaining the aforementioned data for each node, path planning can be used to determine the target routing path of the power network, aiming to minimize the path cost. The target routing path obtained by this scheme considers the node length, the delay of traversed edges, the failure probability of traversed nodes, and the node risk value. It simultaneously ensures transmission efficiency, security, and stability when selecting the target routing path, improving the reliability and security of data transmission.

[0105] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0106] The system in this embodiment is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0107] Please see Figure 11 , Figure 11 This is a schematic diagram of the structure of an electronic device provided in an optional embodiment of the present invention. The electronic device can be a computer device used to achieve, for example... Figure 10 The device shown includes one or more processors 10, a memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise as required. The processor can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface).

[0108] The processor 10 may further include a hardware chip. This hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0109] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0110] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the use of the electronic device based on the display of a mini-program landing page. Furthermore, the memory 20 may include high-speed random access memory (RAM), and may also include non-transient memory, such as at least one disk storage device, flash memory device, or other non-transient solid-state storage device. The memory 20 may include volatile memory, such as RAM; the memory may also include non-volatile memory, such as flash memory, hard disk, or solid-state drive; the memory 20 may also include combinations of the above types of memory.

[0111] The electronic device also includes a communication interface 30 for communicating with other devices or communication networks.

[0112] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0113] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0114] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A routing method for power communication networks based on path risk quantification, characterized in that, The method includes: Based on the network structure of the power communication network, a power communication network topology diagram is constructed; the power communication network topology diagram includes several power nodes and edges connecting the power nodes; the power communication network topology diagram also stores the network parameters of each power node and each edge; Based on the historical failure rate and traffic load factor of each power node, the node risk value of each power node is calculated; the traffic load factor is used to characterize the ratio between the degree of the power node and the link bandwidth of the power node. Based on the node risk value, determine the failure probability of each power node; The target routing path of the power network is determined with the goal of minimizing the path cost of the routing path; the path cost is used to indicate the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

2. The method according to claim 1, characterized in that, The path cost is a weighted sum of the original path cost, the maximum risk cost, the total risk cost, and the failure probability cost; The original path cost is used to characterize the delay and number of nodes traversed by the routing path; The maximum risk cost is used to characterize the maximum node risk value on the routing path; The total risk cost is used to characterize the sum of the node risk values ​​of the power nodes on the routing path; The failure probability cost is the sum of the failure probabilities of the power nodes on the routing path.

3. The method according to claim 2, characterized in that, The original path cost is a weighted sum of the edge delay cost of each edge in the routing path and the number of nodes; The edge delay cost for each edge is a weighted sum of the edge's original transmission delay, the predicted failure probability of the target node toward which the edge is directed, and the node risk value of the target node.

4. The method according to any one of claims 1 to 3, characterized in that, The step of calculating the node risk value of each power node based on its historical failure rate and traffic load factor in the power communication network topology includes: For any power node in the power communication network topology, obtain the neighboring nodes of the power node; the distance between the neighboring nodes and the power node is less than k hops; The fault impact radius of the power node is obtained based on the importance score of the power node in the power communication network topology, the path quality weight between each neighboring node and the power node, and the node distance. The node risk value of the power node is calculated based on the fault impact radius of the power node, the load flow factor of the power node, and the historical failure rate of the power node.

5. The method according to claim 4, characterized in that, The node risk value of the power node is positively correlated with the fault impact radius of the power node, the degree of the power node, and the fault probability of the power node.

6. The method according to claim 4, characterized in that, Before obtaining the fault impact radius of the power node based on its importance score in the power communication network topology, the path quality weights between each neighboring node and the power node, and the node distance, the method further includes: The power communication network topology is input into a graph neural network to obtain the importance score of each power node in the power communication network topology.

7. The method according to any one of claims 1 to 3, characterized in that, The step of determining the failure probability of each power node based on the node risk value includes: For each power node, a candidate probability is generated according to the node risk value of the power node in a first proportion; When the node risk value is less than the first threshold, the minimum value between the candidate probability and the first probability is determined as the failure probability of the power node. When the node risk value of the power node is greater than or equal to the first threshold and less than or equal to the second threshold, the minimum value between the candidate probability and the second probability is determined as the failure probability of the power node. When the node risk value of the power node is greater than the second threshold, the minimum value of the candidate probability and the third probability shall be determined as the failure probability of the power node.

8. A routing device for a power communication network based on path risk quantification, characterized in that, The device includes: The topology graph construction module is used to construct a power communication network topology graph based on the network structure of the power communication network. The power communication network topology graph includes several power nodes and edges connecting the power nodes. The power communication network topology graph also stores the network parameters of each power node and each edge. The risk acquisition module is used to calculate the node risk value of each power node based on the historical failure rate and traffic load factor of each power node; the traffic load factor is used to characterize the ratio between the degree of the power node and the link bandwidth of the power node. The fault probability acquisition module is used to determine the fault probability of each power node based on the node risk value. The routing path determination module is used to determine the target routing path of the power network with the goal of minimizing the path cost of the routing path; the path cost is used to indicate the node length of the routing path, the delay of each edge, the failure probability of each node, and the node risk value of each node.

9. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the routing method for a power communication network based on path risk quantification as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the routing method for a power communication network based on path risk quantification as described in any one of claims 1 to 7.

Images