A matching method, apparatus, electronic device, storage medium, and program product.

Abstract

This disclosure provides a matching method, apparatus, electronic device, storage medium, and program product. The disclosure determines the state of each node; when a faulty node exists, it determines a set of candidate nodes based on node type; and it determines the target node based on the candidate node set and dependencies. The dependencies are determined based on the fusion dependency strength between candidate nodes and faulty nodes. Thus, compared to existing technologies, this disclosure considers not only the differentiated characteristics of node types but also the dynamically changing dependencies between candidate nodes and faulty nodes, allowing for comprehensive consideration and dynamic capture of node changes. In summary, the technical solution provided by this disclosure overcomes the limitations of traditional static matching logic, achieving accurate selection of target nodes, improving the system's response efficiency to extreme disasters and other emergencies, and adapting to various application scenarios.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to a matching method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] Currently, all existing nodes have interactive relationships. When faced with sudden situations or extreme disasters causing node failures, most systems directly determine the target node to replace the failed node based on their fixed static interactive relationships. However, once a failed node appears, at least one of its interactive relationships or dependency strengths changes dynamically. Fixed static matching logic cannot capture these real-time changes, easily leading to a mismatch between the actual correlation between the target node and the failed node, resulting in decreased matching performance and difficulty in adapting to various application scenarios. Summary of the Invention

[0003] This disclosure provides a matching method, apparatus, electronic device, storage medium, and program product to address, to some extent, the problem that existing fixed static matching logic cannot capture such real-time changes, easily leading to a mismatch between the actual correlation between the target node and the faulty node, resulting in decreased matching performance and difficulty in adapting to various application scenarios.

[0004] According to one aspect of this disclosure, a matching method is provided, the method comprising: determining the state of each node; when a faulty node exists, determining a set of candidate nodes based on node type; determining a target node based on the set of candidate nodes and dependencies; the dependencies being determined based on the fusion dependency strength between the candidate node and the faulty node.

[0005] Furthermore, according to one aspect of the method of this disclosure, determining the state of each node includes: for any given node, determining the node's load and capacity based on the node's betweenness centrality; and determining whether the node is a faulty node based on the relationship between the load and capacity.

[0006] Furthermore, according to one aspect of the method of this disclosure, the node types include: social function type and infrastructure type; when the node type is social function type, when there is a faulty node, a candidate node set is determined based on the node type, including: determining the activity correlation between the faulty node and the set of adjacent nodes; the activity correlation is determined based on activity volume and spatial distance; determining the dynamic traffic impedance between the faulty node and the set of adjacent nodes; the dynamic traffic impedance is determined based on traffic engineering BPR; and selecting and determining the candidate node set based on the activity correlation and the dynamic traffic impedance.

[0007] Furthermore, according to one aspect of the method of this disclosure, a candidate node set is selected based on the correlation between activity volume and dynamic traffic impedance, including: when the correlation between activity volume and dynamic traffic impedance is positive and the traffic impedance is less than a first threshold, determining the matching degree between the correlation between activity volume and dynamic traffic impedance; and when the matching degree meets a second threshold, determining the node as a candidate node.

[0008] Furthermore, according to one aspect of the method of this disclosure, when the node type is infrastructure type, when there is a faulty node, a candidate node set is determined based on the node type, including: determining the redundancy capacity for any adjacent node set; and determining a node as a candidate node when the redundancy capacity satisfies the faulty node.

[0009] Furthermore, according to one aspect of the method of this disclosure, determining a target node based on a set of candidate nodes and dependencies includes: selecting candidate nodes from the set of candidate nodes; determining the dependency value between the candidate nodes and the faulty node based on at least one of geographical constraint dependency strength, topological constraint dependency strength, and operational constraint dependency strength; and determining the target node based on the dependency value, a third threshold, and a fourth threshold.

[0010] Furthermore, according to one aspect of the method of this disclosure, candidate nodes are selected from a set of candidate nodes, and the dependency value between the candidate nodes and the faulty node is determined based on at least one of geographical constraint dependency strength, topological constraint dependency strength, and operational constraint strength, including: selecting candidate nodes based on the distance between the candidate nodes and the faulty node and the failure propagation range; determining the reciprocal of the physical distance between the candidate nodes and the faulty node as the geographical constraint dependency strength; determining the ratio of the network average degree of the candidate nodes and the faulty node as the topological constraint dependency strength; determining the influence distance between the candidate nodes and the faulty node as the operational constraint dependency strength; and obtaining the dependency value by weighted extreme values ​​based on the geographical constraint dependency strength, topological constraint dependency strength, and operational constraint strength; the weighted calculation includes at least one of the following: weighted average and weighted summation.

[0011] Furthermore, according to one aspect of the method of this disclosure, a target node is determined based on a dependency value, a third threshold, and a fourth threshold, including: when the dependency value is greater than or equal to the third threshold, a candidate node is determined as the target node; when the dependency value is less than the third threshold but greater than or equal to the fourth threshold, the candidate node with the largest dependency value is determined as the target node.

[0012] Furthermore, according to one aspect of the method disclosed, the method further includes: evaluating the performance of the target node based on evaluation metrics; the evaluation metrics include at least one of the following: total failure ratio, step failure ratio, and cross-network failure ratio.

[0013] According to another aspect of this disclosure, a matching apparatus is provided, comprising: a first determining unit for determining the state of each node; a second determining unit for determining a set of candidate nodes based on node type when a faulty node exists; and a third determining unit for determining a target node based on the set of candidate nodes and dependencies; wherein the dependencies are determined based on the fusion dependency strength between the candidate nodes and the faulty nodes.

[0014] According to another aspect of this disclosure, an electronic device is provided, comprising: a memory for storing computer-readable instructions; and a processor for executing the computer-readable instructions, causing the electronic device to perform the method as described in any embodiment of one aspect.

[0015] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided for storing computer-readable instructions that, when executed by a processor, cause the processor to perform the method as described in any embodiment of one aspect.

[0016] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method as described in any embodiment of one aspect.

[0017] This disclosure provides a matching method, apparatus, electronic device, storage medium, and program product. The disclosure determines the state of each node; when a faulty node exists, it determines a set of candidate nodes based on node type; and it determines the target node based on the candidate node set and dependencies. The dependencies are determined based on the fusion dependency strength between the candidate node and the faulty node. Thus, compared to existing technologies, this disclosure considers not only the differentiated characteristics of node types but also the dynamically changing dependencies between candidate nodes and faulty nodes, allowing for a comprehensive and dynamic capture of node changes. In summary, the technical solution provided by this disclosure overcomes the limitations of traditional static matching logic, achieving accurate selection of target nodes, improving the system's response efficiency to extreme disasters and other emergencies, and adapting to various application scenarios.

[0018] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further illustration of the claimed technology. Attached Figure Description

[0019] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0020] Figure 1 A flowchart illustrating a matching method provided in an embodiment of this disclosure;

[0021] Figure 2 A schematic diagram of the code for a complete matching method provided in the embodiments of this disclosure;

[0022] Figure 3 A structural block diagram of another matching method provided in this disclosure embodiment;

[0023] Figures 4(a) and 4(b) are schematic diagrams comparing RCF under different parameters provided in the embodiments of this disclosure;

[0024] Figures 5(a) and 5(b) are schematic diagrams comparing RICF under different parameters provided in this disclosure;

[0025] Figures 6(a) and 6(b) are schematic diagrams comparing RTCF under different parameters provided in the embodiments of this disclosure;

[0026] Figures 7(a) and 7(b) are schematic diagrams showing the changes in network dependency under different parameters in the embodiments of this disclosure;

[0027] Figure 8 A structural block diagram of a matching device provided in an embodiment of this disclosure;

[0028] Figure 9 A hardware block diagram of an electronic device provided in an embodiment of this disclosure;

[0029] Figure 10 This is a schematic diagram of a computer-readable storage medium provided in an embodiment of this disclosure. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this disclosure more apparent, exemplary embodiments according to this disclosure will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this disclosure, and not all embodiments of this disclosure. It should be understood that this disclosure is not limited to the exemplary embodiments described herein.

[0031] Currently, all existing nodes have interactive relationships. When faced with sudden situations or extreme disasters causing node failures, most systems directly determine the target node to replace the failed node based on their fixed static interactive relationships. However, once a failed node appears, at least one of its interactive relationships or dependency strengths changes dynamically. Fixed static matching logic cannot capture these real-time changes, easily leading to a mismatch between the actual correlation between the target node and the failed node, resulting in decreased matching performance and difficulty in adapting to various application scenarios.

[0032] Therefore, to address the aforementioned problems, this disclosure provides a matching method, apparatus, electronic device, storage medium, and program product. These methods consider not only the differentiated characteristics of node types but also the dynamic dependencies between candidate and faulty nodes, allowing for comprehensive consideration and dynamic capture of node changes. This overcomes the limitations of traditional static matching logic, achieving accurate selection of target nodes, improving system response efficiency to extreme disasters and other emergencies, and adapting to various application scenarios.

[0033] This disclosure provides a matching method. Please refer to... Figure 1 , Figure 1 This is a flowchart illustrating a matching method provided in an embodiment of this disclosure. Figure 1 As shown, the method includes:

[0034] In step S101, the status of each node is determined;

[0035] In step S102, when a faulty node exists, a set of candidate nodes is determined based on the node type;

[0036] In step S103, the target node is determined based on the candidate node set and the dependency relationship; the dependency relationship is determined based on the fusion dependency strength between the candidate node and the faulty node.

[0037] In this disclosure, node status can be understood as a comprehensive representation of a node's current operational capabilities, resource usage, and functional availability, and is the core basis for judging whether a node is working properly.

[0038] In this disclosure, dependency can be understood as the adaptive association between the task undertaken by the faulty node in the system and at least one of the functions, resources, topological locations, etc. of the candidate nodes. The fusion dependency strength can be understood as the comprehensive fit score obtained by weighting the matching degree between the candidate node and the faulty node and the global impact on the system through a quantification model, which is used to select the optimal target node from multiple candidate nodes.

[0039] Specifically, during the matching process, the following steps can be executed:

[0040] First, the status of each node can be determined through a combination of real-time monitoring and periodic inspections, including but not limited to functional availability (such as whether it can respond to service requests normally). When a faulty node is detected, its type is first identified, and then nodes of the same type and currently in normal status are selected from the system to form a candidate node set. For each node in the candidate node set, its fusion dependency strength with the faulty node is calculated: on the one hand, the matching scores of the candidate node and the faulty node in dimensions such as functional adaptability, resource matching degree, and topology compatibility are quantitatively evaluated; on the other hand, the impact of the candidate node replacing the faulty node on the overall system (such as whether it will cause a surge in load on related nodes or whether a large number of links need to be reconstructed) is evaluated, and corresponding weights are assigned. The scores of the above dimensions are weighted and summed to obtain the fusion dependency strength of each candidate node. Finally, the candidate nodes are sorted according to the fusion dependency strength, and the node with the highest score is selected as the target node. If there are cases with the same score, a secondary screening can be performed by combining secondary indicators such as the current load redundancy of the node to ensure that the target node can efficiently replace the faulty node and minimize the impact on the overall operation of the system.

[0041] The following will explain in detail how this disclosure determines the state of each node, including:

[0042] For any given node, determine its load and capacity based on its betweenness centrality.

[0043] Based on the relationship between load and capacity, determine whether a node is a faulty node.

[0044] In this disclosure, betweenness centrality can be understood as a quantitative indicator of the importance of a node as a transit point in the system topology, specifically manifested as the proportion of the node's frequency of occurrence in all shortest paths. The higher the betweenness centrality, the more important the transit role the node plays in the system's information flow, and it is a key parameter for measuring the topological position of a node.

[0045] In this disclosure, load can be understood as the sum of the actual amount of tasks currently undertaken by a node and the pressure of data flow. Its size is directly related to the betweenness centrality of the node. Nodes with higher betweenness centrality usually have a larger load.

[0046] In this disclosure, capacity can be understood as the maximum amount of tasks and data flow pressure threshold that a node can stably bear, which is the upper limit of the node's ability to work stably.

[0047] Specifically, when determining the state of each node, the process can be executed step by step according to the following logic:

[0048] First, a connection graph between system nodes can be constructed using algorithms such as topology traversal. Then, the shortest paths between all pairs of nodes in the graph are calculated, and the frequency of each node's appearance on these paths is counted. This frequency is then compared to the total number of shortest paths in the system to obtain the betweenness centrality value (range 0-1). Next, using betweenness centrality as the base weight and combined with real-time monitoring data, node load can be calculated and determined. For example, the load value can be obtained by weighting the betweenness centrality with the number of task requests processed by the node per unit time. Then, the basic capacity is determined based on node configuration or limitations and betweenness centrality. Finally, the load factor (the ratio of load value to capacity) is calculated, and the status can also be determined by considering the duration: if the load factor exceeds 100% within a preset time, the node can be considered a faulty node.

[0049] For example, in this disclosure, the load and capacity of a node can be determined by the following formula:

[0050] The calculation formula is as follows:

[0051]

[0052]

[0053] in, and These represent the load and capacity of node i, respectively; λ represents the betweenness centrality of node i, indicating the proportion of shortest paths passing through that node; the parameter λ is a tolerance factor used to control capacity redundancy. This refers to redundant capacity that can be used to absorb external disturbances.

[0054] The following section will elaborate on the types of nodes and how to determine the candidate node set for different types of nodes:

[0055] Node types include: social function type and infrastructure type.

[0056] In this disclosure, the social function category can be understood as a node type that directly undertakes social services, meets specific public needs, or realizes humanistic interaction functions. Examples include: transportation systems, community service center nodes, medical information registration nodes, educational resource allocation nodes, and public cultural activity scheduling nodes.

[0057] In this disclosure, infrastructure nodes can be understood as the underlying architecture node types that support the operation of social function nodes and provide basic resources and technical support. Examples include network communication base station nodes, power supply and distribution nodes, and data storage server nodes.

[0058] The following will explain how to determine the candidate node set when the node type is social function, including the following methods:

[0059] Determine the activity correlation between the faulty node and the set of neighboring nodes; the activity correlation is determined based on activity and spatial distance.

[0060] Determine the dynamic traffic impedance between the faulty node and its adjacent node set; the dynamic traffic impedance is determined based on traffic engineering business dynamics (BPR).

[0061] Based on the correlation of activity volume and dynamic traffic impedance, a set of candidate nodes is selected and determined.

[0062] In this disclosure, activity volume correlation can be understood as the degree of association between the scale and type of public service activities undertaken by a faulty social function node and its neighboring nodes within a unit of time. It reflects the substitution potential or attractiveness of neighboring nodes to the service scope of the faulty node. It can be calculated by multiplying activity volume similarity and spatial decay coefficient. Activity volume similarity requires statistical analysis of the overlap in the number of people served by the nodes, the frequency of activities held, etc., while the spatial decay coefficient decreases as the distance between the two nodes increases, to reflect the impact of geography on service substitution.

[0063] In this disclosure, dynamic traffic impedance can be understood as the travel status from a socially functional failure node to an adjacent node (e.g., congestion level), and is an indicator of the ease with which residents / users can transfer from the service area of ​​the failure node to an adjacent node. Traffic engineering (BPR) can be understood as the classic delay function model proposed by the U.S. Public Roads Authority, whose core function is to calculate the travel time of road segments under different traffic flow conditions. This model incorporates dynamic factors such as traffic flow and road segment capacity into the impedance calculation, enabling real-time quantification of travel costs. Dynamic traffic impedance can be quantitatively represented based on the actual travel time calculated by the traffic engineering BPR model.

[0064] Specifically, when the node is a social function node, the following steps can be performed:

[0065] First, taking the faulty social function node as the center, the search radius is defined based on the geographical space range. All social function nodes of the same or compatible type within this range are included in the initial neighbor node set, and nodes that are already in a faulty or sub-healthy state are excluded.

[0066] Next, the correlation between activity volume and dynamic traffic impedance is calculated: the activity volume indicators of the fault node and its adjacent nodes, including the core business processing volume, are statistically analyzed. The activity volume similarity can be calculated using the cosine similarity formula, and then the spatial attenuation coefficient is calculated based on the straight-line distance D (unit: kilometers) between the two nodes. Finally, the activity volume correlation = activity volume similarity × spatial attenuation coefficient.

[0067] Dynamic traffic impedance calculation can be based on the BPR function of traffic engineering calculations. arrive Travel time as dynamic traffic impedance: obtaining connectivity and Key traffic segment parameters, including free-flow travel time Actual traffic volume Q and road segment capacity C are used, employing the classic BPR formula: Actual travel time t = × [1 + α×(Q / C)^β] (default parameters α=0.15, β=4.0); if there are multiple travel paths, the minimum actual travel time of each path is taken as the final dynamic traffic impedance.

[0068] Finally, the correlation between activity volume and dynamic traffic impedance are standardized and then weighted and summed to obtain the matching degree, which is then compared with a scoring threshold (e.g., 0.6). Neighboring nodes with scores higher than the threshold are included in the candidate node set; if the scores are the same, nodes with lower dynamic traffic impedance are prioritized.

[0069] The following section will elaborate on how to determine the candidate node set based on activity correlation and dynamic traffic impedance, including the following methods:

[0070] When the activity level correlation is positive and the traffic impedance is less than the first threshold, the matching degree between the activity level correlation and the dynamic traffic impedance is determined.

[0071] When the matching degree meets the second threshold, the node is determined as a candidate node.

[0072] Specifically, the process can be executed in a progressive manner: "positive correlation determination → traffic impedance screening → matching degree calculation → candidate node confirmation". The detailed logic of each step is as follows:

[0073] First, clarify the criteria for determining the positive correlation of activity volume. Using the activity volume correlation values ​​calculated earlier (ranging from 0 to 1), set a positive correlation threshold (usually 0.3, adjustable based on node service type). When the activity volume correlation between an adjacent node and a faulty node is greater than or equal to the positive correlation threshold, a positive correlation is established, meaning the service scale and business type of the adjacent node overlap to some extent with the faulty node, providing a basis for service substitution. If the correlation is less than the positive correlation threshold, the node is directly excluded and does not need further screening. Next, combining the service radius of social function nodes with user travel, set a first threshold (i.e., the upper limit of traffic impedance). For adjacent nodes already determined to be positively correlated, if their dynamic traffic impedance is less than the first threshold, it indicates that the ease of user transfer from the faulty node's service area to this node meets the requirements, and the matching degree calculation begins; if it is greater than or equal to the first threshold, it is excluded due to excessive travel. Then, the standardized values ​​of activity volume correlation and dynamic traffic impedance are fused and calculated according to preset weights. Set a second threshold (i.e., the matching degree qualification line), which needs to be determined based on the system's expectation of the number of candidate nodes and service quality requirements, typically between 0.6 and 0.7. For nodes that have completed the matching degree calculation, if the matching degree is greater than or equal to the second threshold, it is identified as a candidate node; if the matching degree is less than the second threshold, it is excluded. If multiple nodes meet the matching degree requirement, they can be sorted from high to low to form an ordered set of candidate nodes, providing a priority basis for subsequent target node selection. In this way, through the above process, candidate nodes with "high functional matching and good user accessibility" can be accurately screened, ensuring effective replacement of services from faulty nodes while minimizing the user's migration costs, thus meeting the core service needs of social functional nodes.

[0074] For example, when a node is a social function node, the candidate node set can be determined by the following formula:

[0075] The amount of load redistributed is proportional to the social attraction between the failed node and its neighbors, as defined below:

[0076]

[0077]

[0078] in, The load redistributed by the failed node i to its neighboring node j at time step t; Let i be the set of neighbors of node i; Let α be the social attraction between nodes i and j; parameter α is the global scaling constant; parameter β is the distance decay coefficient.

[0079] Here This represents the dynamic traffic impedance between nodes i and j, not just the physical distance. Its calculation method is based on a modified Bureau of Public Roads (BPR) function and incorporates topological features to reflect the congestion effect when the load approaches capacity, thus more realistically quantifying the difficulty of interaction between nodes. The specific definition is as follows:

[0080]

[0081]

[0082]

[0083] in, Let the edge flow between nodes i and j at time step t be determined by the normalized betweenness centrality of the edges. Perform calculations; Let be the edge betweenness centrality between nodes i and j, and let represent the proportion of the shortest paths through that edge. Let the edge capacity between nodes i and j be given by... The minimum-maximum normalization result is obtained.

[0084] The following will elaborate on how to determine the candidate node set when the node type is infrastructure, including the following methods:

[0085] For any set of adjacent nodes, determine the redundancy capacity;

[0086] When the redundancy capacity meets the requirements of the faulty node, it is identified as a candidate node.

[0087] In this disclosure, redundancy capacity can be understood as the additional resource capacity that an infrastructure-type node possesses, which can temporarily take over the business needs of other faulty nodes, on the premise that it is normally carrying its own established business load.

[0088] Specifically, in this embodiment, the available redundancy capacity of a single node needs to be calculated first for the target infrastructure node type. For example, if a computing node has a total CPU core capacity of 100 cores, a security threshold of 70%, and currently uses 50 cores (50% utilization), then the available redundancy resources of this node's CPU are 100 cores × (70% - 50%) = 20 cores. If the node also needs to consider memory resources (total memory 128GB, security threshold 75%, currently using 64GB), then the available redundancy resources for memory are 128GB × (75% - 50%) = 32GB. Next, it is determined whether the redundancy capacity meets the requirements: the service load requirements of the faulty node are obtained and compared item by item with the available redundancy resources of adjacent nodes. Only when the available redundancy resources of adjacent nodes are all greater than or equal to the service load requirements of the corresponding dimension of the faulty node can it be determined that the redundancy capacity of the adjacent node meets the faulty node takeover requirements, and then it is included in the candidate node set.

[0089] For example, when the node type is infrastructure, it can satisfy the following calculation formula:

[0090]

[0091] in, Let i be the set of neighbors of node i.

[0092] The following, as one of the core points of this disclosure, will specifically explain how to determine the target node, including the following methods:

[0093] Select candidate nodes from the candidate node set, and determine the dependency value between the candidate nodes and the faulty nodes based on at least one of the geographical constraint dependency strength, topological constraint dependency strength, and operational constraint dependency strength.

[0094] The target node is determined based on the dependency value, the third threshold, and the fourth threshold.

[0095] In this disclosure, a candidate node can be understood as a candidate node selected from the subnetwork that is within the failure propagation range of the faulty node and may form a dependent edge with the faulty node.

[0096] In this disclosure, the geographical constraint dependency strength can be understood as a quantitative indicator characterizing the impact of the physical spatial proximity between the faulty node i and the candidate node j on the dependency relationship.

[0097] In this disclosure, the topological constraint dependency strength can be understood as a quantitative indicator reflecting the impact of the structural importance of candidate node j in its sub-network on the dependency relationship.

[0098] In the present disclosure, the running constraint dependence strength can be understood as a binary quantization index for identifying whether the alternative node j is within the actual influence boundary of the faulty node i.

[0099] In the present disclosure, the dependence value can be understood as a normalized comprehensive score for the tightness of the dependence relationship between the faulty node i and the alternative node j by integrating the geographical constraint dependence strength, the topological constraint dependence strength, and the running constraint dependence strength. It can directly reflect the potential probability of forming a dependence edge and the functional influence strength, and is the core judgment basis for subsequent distinguishing strong / weak dependence and determining the target node.

[0100] Specifically, when determining the target node, first screen out the alternative nodes that meet the failure propagation range condition from the candidate node set, then obtain the dependence value by calculating or selecting at least one of the geographical, topological, and running constraint dependence strengths, and finally compare the dependence value with the preset third threshold (T3) and fourth threshold (T4, T3>T4): if the dependence value ≥ T3, the alternative node is a strongly dependent target node and will immediately fail and trigger load reallocation; if T4 < dependence value < T3, it is a weakly dependent target node and its capacity is reduced proportionally; if the dependence value ≤ T4, it does not constitute a target node.

[0101] Next, how to determine the dependence value according to the three dependence strengths will be specifically elaborated. The method includes:

[0102] Select alternative nodes based on the distance between the candidate node and the faulty node and the failure propagation range;

[0103] Determine the reciprocal of the physical distance between the alternative node and the faulty node as the geographical constraint dependence strength;

[0104] Determine the ratio of the network average degree between the alternative node and the faulty node as the topological constraint dependence strength;

[0105] Determine the influence distance between the alternative node and the faulty node as the running constraint dependence strength;

[0106] Based on the geographical constraint dependence strength, the topological constraint dependence strength, and the running constraint strength, obtain the dependence value by weighted extreme; the weighted calculation includes at least one of the following: weighted average, weighted sum.

[0107] In the present disclosure, the failure propagation range can be understood as the average Euclidean distance between the faulty node i and its neighbor nodes, which is a key parameter for defining the spatial boundary of the influence of disasters or emergencies. It comprehensively considers the geographical distribution density and topological connection characteristics of the network where the nodes are located, and is used to judge whether the candidate node is within the direct influence range of the faulty node, and is the core spatial constraint basis for screening alternative nodes.

[0108] Specifically, when determining the dependence value, the following steps can be taken: First step, screening alternative nodes: Calculate the Euclidean distance between the candidate node and the fault node i. When the Euclidean distance ≤ the failure propagation range, the candidate node is determined as the alternative node j. Second step, calculating the single-constraint dependence strength: The geographical constraint dependence strength takes the reciprocal of the Euclidean distance and is normalized (to make the result fall within the range of [0, 1]). The topological constraint dependence strength is the ratio of the degree of the alternative node j to the average degree of the sub-network it belongs to. The operation constraint dependence strength is a binary variable. Third step, weighted fusion to obtain the dependence value: Set weights ω1, ω2, ω3 (satisfying ω1 + ω2 + ω3 = 1, which can be adjusted through empirical calibration). If weighted summation is used, the dependence value = ω1 × geographical constraint dependence strength + ω2 × topological constraint dependence strength + ω3 × operation constraint dependence strength. If weighted averaging is used, it is divided by the sum of the weights on the basis of weighted summation (since the sum of the weights is 1, the result is the same as weighted summation). Finally, a normalized comprehensive score reflecting the tightness of the dependence relationship is obtained, which is the dependence value.

[0109] Next, how to determine the target node according to the dependence value and the preset threshold will be specifically elaborated, including:

[0110] When the dependence value is greater than or equal to the third threshold, the alternative node is determined as the target node;

[0111] When the dependence value is less than the third threshold and greater than or equal to the fourth threshold, the alternative node with the largest dependence value is determined as the target node.

[0112] In this disclosure, the third threshold can be understood as the critical scoring value (denoted as T3) for determining that the alternative node and the fault node form a strong dependence relationship. This threshold needs to be set in combination with the actual coupling strength of the interdependent network and the requirements of the disaster scenario, and usually takes a value within the range of [0.6, 0.8]. Its core role is to screen out the key nodes that are closely dependent on the fault node and will trigger direct failure once a dependence edge is formed, and it is the determination benchmark for strong dependence target nodes.

[0113] In this disclosure, the fourth threshold can be understood as the critical scoring value (denoted as T4) for defining whether the alternative node is within the weak dependence influence range of the fault node, and satisfies T4 < T3, and generally takes a value within the range of [0.3, 0.5]. This threshold distinguishes "nodes with the potential for weak dependence influence" from "nodes without dependence influence", and it is the lower limit standard for screening weak dependence candidate nodes.

[0114] Specifically, the steps for determining the target node based on the dependency value and the preset threshold are as follows: The first step is to traverse the dependency values of all alternative nodes and compare each value with the third threshold T3 and the fourth threshold T4 respectively; the second step is to directly determine all the nodes that meet this condition as the target nodes (strong dependency target nodes) if there are alternative nodes with a dependency value ≥ T3; the third step is to screen out the alternative nodes with dependency values within the interval [T4, T3) if there are no alternative nodes with a dependency value ≥ T3, calculate and select the node with the largest dependency value among them as the target node (weak dependency target node); the fourth step is that if the dependency values of all alternative nodes are < T4, there are no eligible target nodes in this time. This method, through the determination logic of strong dependency first and then weak dependency, not only ensures the timeliness of the failure response of strongly associated nodes preferentially, but also can accurately locate the node with the greatest impact within the weak dependency range, improving the scientificity and pertinence of the determination of the target node.

[0115] Exemplarily, when the present disclosure determines the target node, the following formula can be satisfied:

[0116] First, select candidate nodes j in the sub-network that may be affected by the failed node i within the feasible interaction range. The rules are as follows:

[0117]

[0118] Among them, dist(i, j) is the Euclidean distance between nodes i and j, is the failure propagation range, calculated by the average distance between node i and its neighbors. The random variable r ~ U(0, 1) is used to control the switching of the two selection mechanisms. When r > 0.5, node j is randomly selected within the local range, reflecting the random propagation of disaster risks; if r ≤ 0.5, the node j closest to i is selected.

[0119] Subsequently, the present disclosure assigns a strength score to each potential dependency edge, indicating the probability of forming the dependency edge. Specifically, the probability of generating a dependency edge between the failed node i in the current sub-network and the candidate node j in another sub-network is represented by an adjustable function as follows:

[0120]

[0121]

[0122] Among them, represents the interaction score between nodes i and j, considering three key constraints: spatial proximity (geographical constraint ), structural importance (topological constraint ), and adjacency to the failure boundary (operational constraint ). is the reciprocal of the physical distance between nodes i and j; It is measured by the ratio of the degree of node j to the average degree of the network; For a binary variable, the influence distance of node j from node i. The value is 1 if the condition is met, and 0 otherwise. Parameters ω1, ω2, and ω3 are weights that can adjust the importance of each factor; parameter γ controls the steepness of the curve; and θ is the activation threshold. A larger γ indicates that the system is more sensitive to the intensity of interactions; a smaller θ means that dependency edges are easier to generate, and the system is more coupled.

[0123] Based on the calculated probability Pij(t), this disclosure further defines the dependency effect of a failed node i in one subnetwork on a candidate node j in another subnetwork. Iij(t):

[0124]

[0125] That and It is an adjustable threshold (0~1) used to distinguish between strong dependencies, weak dependencies, and no dependencies. If If it's a strong dependency, node j will immediately fail and its load will be redistributed to its neighboring nodes; if If it is a weak dependency, then the capacity of node j will be determined by probability. Reduced proportionally, as defined below:

[0126]

[0127] Where ρ is an adjustable parameter used to control the strength of the influence exerted by weakly dependent edges.

[0128] The matching method disclosed herein will be further elaborated below, including:

[0129] The performance of the target node is evaluated based on the evaluation metrics. The evaluation metrics include at least one of the following: total failure ratio, step failure ratio, and cross-network failure ratio.

[0130] In this disclosure, the evaluation metrics can be understood as a measure of the effectiveness of the target node determination method in controlling cascading failures in interdependent networks, and can intuitively reflect the accuracy and rationality of target node identification. Specifically, the total failure ratio can be understood as the ratio of the total number of failed nodes in the entire interdependent network (including all sub-networks) to the total number of nodes in the network during disaster impact and cascading failure, with a value ranging from [0,1]. The closer the value is to 0, the better the method's control effect on failure propagation. The step-by-step failure ratio can be understood as the ratio of the number of failed nodes at each time step in the cascading failure process to the number of surviving nodes in the network at the corresponding time step. The cross-network failure ratio can be understood as the ratio of the number of failed nodes triggered by the target node through cross-network dependency edges to the total number of failed nodes in the entire interdependent network.

[0131] Specifically, the performance of the target node can be evaluated using evaluation indicators according to the following steps: First, set the initial failure node and disaster propagation parameters (such as failure propagation range). The steps are as follows: 1) Identify target nodes based on the target node determination method disclosed herein, and simulate the cascading failure process, recording the failure node information (including the sub-network to which the failure node belongs, failure time, failure type, etc.) in real time; 2) Calculate the evaluation indicators: Total failure ratio = (total number of failure nodes in all sub-networks) / (total number of network nodes); Step failure ratio = (number of failure nodes in each time step) / (number of network surviving nodes in the corresponding time step), used to analyze the rate characteristics of failure propagation; Cross-network failure ratio = (number of failure nodes triggered by the target node across networks) / (total number of failure nodes), used to evaluate the propagation contribution of cross-network dependent edges; 3) Set a performance benchmark value according to the scenario requirements, and compare the calculated evaluation indicators with the benchmark value. If the total failure ratio is lower than the benchmark value, the peak of the step failure ratio is flat, and the cross-network failure ratio is controllable, it indicates that the target node determination method meets the performance standards; otherwise, the threshold, weights, and other parameters need to be adjusted for optimization.

[0132] For example, the evaluation indicators of this disclosure can satisfy the following formula:

[0133]

[0134] Where RCF represents the total failure ratio; N represents the total number of nodes. This represents the total number of nodes that eventually fail.

[0135]

[0136] in, Indicates the step failure ratio; This represents the number of nodes that fail at the l=n time step.

[0137]

[0138] Wherein, RICF represents the cross-network failure ratio; This indicates the number of nodes that failed due to cross-network dependencies.

[0139] For example, Figure 2 A schematic diagram of the code for a complete matching method provided in the embodiments of this disclosure. For example... Figure 2 As shown, it includes:

[0140] Step 1 (lines 1–5): Combine the network generation model with the layout algorithm to generate two sub-networks with explicit spatial coordinates, and calculate the load and capacity for each node.

[0141] Step 2 (lines 6–12): Introduce an extreme disaster impact to determine the initial set of failed nodes F in the two sub-networks; evaluate each node in F and calculate the generation probability of its cross-network dependency edges (direction includes from sub-network A to B, and from B to A); determine the dependency strength: if it is a strong dependency, node j will fail directly; if it is a weak dependency, the capacity of node j will be reduced proportionally.

[0142] Step 3 (lines 13–19): Traverse each node i in the new set of failed nodes and identify its neighboring nodes in the same subnetwork; subnetworks A and B perform load redistribution based on the social gravity model and infrastructure redundancy mechanism, respectively; update the set of failed nodes and use it as the initial set of failed nodes F for the next iteration; at the same time, calculate the local evaluation index RTCF.

[0143] Step 4 (line 21): The dependency determination and load redistribution process is executed iteratively until cascading failure terminates; at the end of the simulation, the global evaluation metrics RCF and RICF are calculated.

[0144] For example, this disclosure also provides a complete simulation verification process for applying the matching method. Figure 3 Figure 4(a) and Figure 4(b) are schematic diagrams of RCF comparison under different parameters provided in the embodiments of this disclosure. Figure 5(a) and Figure 5(b) are schematic diagrams of RICF comparison under different parameters provided in the embodiments of this disclosure. Figure 6(a) and Figure 6(b) are schematic diagrams of RTCF comparison under different parameters provided in the embodiments of this disclosure. Figure 7(a) and Figure 7(b) are schematic diagrams of network dependency changes under different parameters in the embodiments of this disclosure.

[0145] As shown in Figure 3, interdependent and independent networks exhibit significant differences in the phase transition characteristics of cascade failure. The RCF of the interdependent sub-network A changes dramatically with increasing λ, showing a first-order phase transition characteristic at λ=0.5; while the RCF of the independent sub-network A shows a gradual degradation trend with increasing λ, exhibiting a second-order phase transition characteristic. Figures 4(a) and 4(b) show a critical phenomenon in the RICF changes of sub-network A in Figure 4(a) and sub-network B in Figure 4(b), which is particularly evident at λ=0.5. In sub-network A, the strong dependency RICF increases with increasing λ when λ<0.5, but the opposite occurs when λ>0.5; in sub-network B, both strong and weak dependency RICFs increase significantly when λ>0.5, with the growth of dependency edges becoming the key driving force for the amplification of cascade failure. As shown in Figures 5(a) and 5(b), the local variation (RTCF) of cascaded failures follows a similar pattern to the global variation, with a significant abrupt change at λ=0.5. At λ=0.5, subnetworks A in Figure 5(a) and B in Figure 5(b) experience the maximum failure impact at earlier time steps, after which the fluctuations tend to stabilize. When λ>0.5, the propagation time is prolonged, the degree of fluctuation disorder increases, and with the increase of λ, the fluctuation range narrows, and failure propagation is suppressed. As shown in Figures 6(a) and 6(b), the parameters γ and θ significantly affect the degree of network dependency. Increasing γ shifts the Pij distribution towards higher values, increasing the proportion of strong dependencies; increasing θ shifts the overall Pij distribution towards lower values, increasing the proportion of no dependencies. The method can adapt to various interdependent patterns and has strong generalization ability. As can be seen from Figures 7(a) and 7(b) (based on the analysis logic of network dependency in the previous text), the distribution and proportion of network dependency under different parameter settings show corresponding changes, further verifying the conclusion that the degree of network dependency can be controlled by adjusting γ and θ to adapt to different interdependence patterns.

[0146] In summary, the above matching method effectively reflects the differences in cascading failure characteristics between interdependent and independent networks, and exhibits good sensitivity to parameters such as network capacity (represented by λ) and network dependency (represented by γ and θ). It not only demonstrates the significant first-order phase transition cascading failure characteristics of interdependent networks at specific critical values ​​(e.g., λ=0.5), but also reflects the differentiated impact of different dependency types (strong and weak dependencies) on cascading failure propagation. Furthermore, it verifies that adjusting relevant parameters can adapt to various interdependence modes, providing an effective tool for resilience analysis and risk management of interdependent networks such as urban infrastructure. Returning to this disclosure, the evaluation index system constructed based on the total failure ratio (RCF), step failure ratio (RTCF), and cross-network failure ratio (RICF) can systematically evaluate the performance of the target node determination method from three core dimensions: global, time dynamics, and cross-network propagation. From a global perspective, RCF can accurately measure the overall failure scale of interdependent networks under the control of the target node. For example, the sudden change in RCF of the interdependent network at λ=0.5 in the simulation can intuitively reflect the control effect of the target node on the failure propagation under critical conditions. From a time dynamic perspective, RTCF can capture the changing characteristics of the failure propagation rate at each time step. For example, the pattern of extended propagation time and narrowed fluctuation range when λ>0.5 can provide a basis for optimizing the timing of target node intervention. From the perspective of cross-network propagation, RICF can quantify the contribution of the target node to cross-network failures caused by dependent edges. For example, the phenomenon of a significant increase in RICF of subnetwork B when λ>0.5 can clarify the role of the target node in suppressing cross-network failure chains.

[0147] This disclosure also provides a matching device. Figure 8 A structural block diagram of a matching device provided in an embodiment of this disclosure, such as... Figure 8 As shown, the matching device 800 includes:

[0148] The first determining unit 801 is used to determine the state of each node;

[0149] The second determining unit 802 is used to determine a set of candidate nodes based on the node type when a faulty node exists.

[0150] The third determining unit 803 is used to determine the target node based on the candidate node set and the dependency relationship; the dependency relationship is determined based on the fusion dependency strength between the candidate node and the faulty node.

[0151] In one exemplary embodiment, the first determining unit 801 is specifically used to: for any given node, determine the load and capacity of the node based on the betweenness centrality of the node; and determine whether the node is a faulty node based on the relationship between the load and the capacity.

[0152] In one exemplary embodiment, the second determining unit 802 is specifically used for: node types including social function type and infrastructure type; when the node type is social function type, when there is a faulty node, determining a candidate node set based on the node type, including: determining the activity correlation between the faulty node and the set of adjacent nodes; the activity correlation is determined based on activity and spatial distance; determining the dynamic traffic impedance between the faulty node and the set of adjacent nodes; the dynamic traffic impedance is determined based on traffic engineering BPR; and selecting and determining the candidate node set based on the activity correlation and the dynamic traffic impedance.

[0153] In one exemplary embodiment, the second determining unit 802 is specifically used to: select and determine a set of candidate nodes based on the correlation between activity volume and dynamic traffic impedance, including: when the correlation between activity volume and dynamic traffic impedance is positive and the traffic impedance is less than a first threshold, determining the matching degree between the correlation between activity volume and dynamic traffic impedance; when the matching degree meets a second threshold, determining the node as a candidate node.

[0154] In one exemplary embodiment, the second determining unit 802 is specifically used to: when the node type is infrastructure type, when there is a faulty node, determine a set of candidate nodes based on the node type, including: determining the redundancy capacity for any adjacent set of nodes; when the redundancy capacity meets the requirements of the faulty node, it is determined as a candidate node.

[0155] In one exemplary embodiment, the third determining unit 803 is specifically used to: determine the target node based on the candidate node set and the dependency relationship, including: selecting candidate nodes in the candidate node set; determining the dependency value between the candidate node and the faulty node based on at least one of geographical constraint dependency strength, topological constraint dependency strength and operational constraint dependency strength; and determining the target node based on the dependency value, a third threshold and a fourth threshold.

[0156] In one exemplary embodiment, the third determining unit 803 is specifically configured to: select candidate nodes from the candidate node set; and determine the dependency value between the candidate nodes and the faulty node based on at least one of geographical constraint dependency strength, topological constraint dependency strength, and operational constraint strength, including: selecting candidate nodes based on the distance between the candidate nodes and the faulty node and the failure propagation range; determining the reciprocal of the physical distance between the candidate nodes and the faulty node as the geographical constraint dependency strength; determining the ratio of the network average degree of the candidate nodes and the faulty node as the topological constraint dependency strength; determining the influence distance between the candidate nodes and the faulty node as the operational constraint dependency strength; and obtaining the dependency value by weighted extreme values ​​based on the geographical constraint dependency strength, topological constraint dependency strength, and operational constraint strength; the weighted calculation includes at least one of the following: weighted average and weighted summation.

[0157] In one exemplary embodiment, the third determining unit 803 is specifically used to: determine a target node based on the dependency value, a third threshold, and a fourth threshold, including: when the dependency value is greater than or equal to the third threshold, determining a candidate node as the target node; when the dependency value is less than the third threshold but greater than or equal to the fourth threshold, determining the candidate node with the largest dependency value as the target node.

[0158] In one exemplary embodiment, the third determining unit 803 is further configured to: evaluate the performance of the target node based on evaluation metrics; the evaluation metrics include at least one of the following: total failure ratio, step failure ratio, and cross-network failure ratio.

[0159] Figure 9 This is a hardware block diagram of an electronic device provided according to an embodiment of the present disclosure. The electronic device 900 according to an embodiment of the present disclosure includes at least a processor; and a memory for storing computer-readable instructions. When the computer-readable instructions are loaded and executed by the processor, the processor performs the matching method described in any of the preceding embodiments of the present disclosure.

[0160] Figure 9 The illustrated electronic device 900 specifically includes a central processing unit (CPU) 901, a graphics processing unit (GPU) 902, and a memory 903. These units are interconnected via a bus 904. The CPU 901 and / or GPU 902 can function as the aforementioned processor, and the memory 903 can function as the aforementioned memory storing computer-readable instructions. Furthermore, the electronic device 900 may also include a communication unit 905, a storage unit 906, an output unit 907, an input unit 908, and an external device 909, all of which are also connected to the bus 904.

[0161] Figure 10 This is a schematic diagram of a computer-readable storage medium provided in an embodiment of this disclosure. (As shown...) Figure 10 As shown, a computer-readable storage medium 1000 according to an embodiment of the present disclosure stores computer-readable instructions 1001 thereon. When the computer-readable instructions 1001 are executed by a processor, the matching method described above with reference to any embodiment of the present disclosure is performed. The computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0162] This disclosure further provides a computer program product, including a computer program that, when executed by a processor, implements the matching method described in any of the preceding embodiments of this disclosure.

[0163] In summary, this disclosure provides a matching method, apparatus, electronic device, storage medium, and program product. This disclosure determines the state of each node; when a faulty node exists, it determines a set of candidate nodes based on node type; and it determines the target node based on the candidate node set and dependencies; the dependencies are determined based on the fusion dependency strength between candidate nodes and faulty nodes. Thus, compared to existing technologies, this disclosure considers not only the differentiated characteristics of node types but also the dynamically changing dependencies between candidate nodes and faulty nodes, allowing for a comprehensive and dynamic capture of node changes. In conclusion, the technical solution provided by this disclosure overcomes the limitations of traditional static matching logic, achieving accurate selection of target nodes, improving the system's response efficiency to extreme disasters and other emergencies, and adapting to various application scenarios.

[0164] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0165] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0166] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0167] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0168] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0169] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0170] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0171] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A matching method, characterized by, The method comprises: determining the state of each node; when there is a fault node, determining a candidate node set based on the node type; determining a target node based on the candidate node set and the dependency relationship; the dependency relationship is determined based on the fusion dependency strength of the candidate node and the fault node; wherein, determining a target node based on the candidate node set and the dependency relationship comprises: selecting a candidate node in the candidate node set, determining the dependency value of the candidate node and the fault node based on at least one of the geographic constraint dependency strength, the topological constraint dependency strength and the operation constraint dependency strength; determining the target node based on the dependency value, the third threshold value and the fourth threshold value; determining the target node based on the dependency value, the third threshold value and the fourth threshold value comprises: when the dependency value is greater than or equal to the third threshold value, determining the candidate node as the target node; when the dependency value is less than the third threshold value and greater than or equal to the fourth threshold value, determining the candidate node with the maximum dependency value as the target node.

2. The method of claim 1, wherein, The determination of the state of each node comprises: for any one node, determining the load and capacity of the node based on the betweenness centrality of the node; determining whether the node is a fault node based on the relationship between the load and the capacity.

3. The method of claim 1, wherein, The node type includes: social function class and infrastructure class; when the node type is the social function class, the determination of the candidate node set based on the node type when there is a fault node comprises: determining the activity volume correlation between the fault node and the adjacent node set; the activity volume correlation is determined based on the activity volume and the spatial distance; determining the dynamic traffic impedance between the fault node and the adjacent node set; the dynamic traffic impedance is determined based on the traffic engineering BPR; selecting the candidate node set based on the activity volume correlation and the dynamic traffic impedance.

4. The method of claim 3, wherein, The selection of the candidate node set based on the activity volume correlation and the dynamic traffic impedance comprises: when the activity volume correlation is positively correlated and the traffic impedance is less than the first threshold value, determining the matching degree of the activity volume correlation and the dynamic traffic impedance; when the matching degree meets the second threshold value, determining the node as the candidate node.

5. The method of claim 3, wherein, When the node type is the infrastructure class, the determination of the candidate node set based on the node type when there is a fault node comprises: determining the redundant capacity for any one adjacent node set; when the redundant capacity meets the fault node, determining as the candidate node.

6. The method of claim 1, wherein, The determination of the dependency value of the candidate node and the fault node based on at least one of the geographic constraint dependency strength, the topological constraint dependency strength and the operation constraint strength comprises: selecting the candidate node based on the distance and failure propagation range between the candidate node and the fault node; determining the reciprocal of the distance between the candidate node and the fault node as the geographic constraint dependency strength. determining a ratio of the alternative node and a network average degree of the failure node as the topology constraint dependency strength; determining an influence distance of the alternative node and the failure node as the operation constraint dependency strength; calculating the dependency value based on the geographic constraint dependency strength, the topology constraint dependency strength and the operation constraint strength by weighted calculation; the weighted calculation includes at least one of weighted average and weighted summation.

7. The method of claim 1, wherein, The method further includes: evaluating performance of the target node based on evaluation indexes; the evaluation indexes include at least one of total failure ratio, step failure ratio and cross-network failure ratio.

8. A matching device, characterized by The apparatus includes: a first determining unit configured to determine states of nodes; a second determining unit configured to determine a candidate node set based on node types when there is a failure node; a third determining unit configured to determine a target node based on the candidate node set and dependency relationships; the dependency relationships are determined based on fusion dependency strengths of the candidate nodes and the failure node; wherein the determining of the target node based on the candidate node set and the dependency relationships includes: selecting an alternative node from the candidate node set, determining a dependency value of the alternative node and the failure node based on at least one of geographic constraint dependency strength, topology constraint dependency strength and operation constraint dependency strength, and determining the target node based on the dependency value, a third threshold value and a fourth threshold value; the determining of the target node based on the dependency value, the third threshold value and the fourth threshold value includes: when the dependency value is greater than or equal to the third threshold value, determining the alternative node as the target node; and when the dependency value is less than the third threshold value and greater than or equal to the fourth threshold value, determining the alternative node with the largest dependency value as the target node.

9. An electronic device, comprising: including: a memory configured to store computer readable instructions; and a processor configured to run the computer readable instructions, so that the electronic device performs the method according to any one of claims 1-7.

10. A non-transitory computer-readable storage medium, comprising: computer readable instructions for storing, when executed by a processor, cause the processor to perform the method according to any one of claims 1-7.

11. A computer program product, characterised in that, a computer program, when executed by a processor, implements the method according to any one of claims 1-7.

Images