Traffic network traffic node importance identification method

Abstract

The invention provides a traffic network traffic node importance identification method, which considers node paths and state disturbance, and specifically comprises the following steps: firstly, establishing a complex traffic network, then calculating ks values of all traffic nodes in the traffic complex network, then calculating information entropies of the traffic nodes, and calculating the importance of the traffic nodes; and then deleting the traffic node with the minimum information entropy from the traffic network, and finally selecting the traffic node with the maximum information entropy from the traffic node set under the current iteration frequency, and orderly adding the traffic node with the maximum information entropy into a traffic node importance sorting set. And all the traffic nodes in the traffic network are put into the importance traffic node sorting set. According to the method, node paths and state disturbance are comprehensively considered, so that the key nodes in the traffic network can be widely and uniformly identified, and the robustness and the operation efficiency of the network are remarkably improved.

Description

Technical Field

[0001] The invention belongs to the technical field of complex traffic networks, and in particular relates to a method for identifying the importance of traffic nodes in a traffic network. Background Art

[0002] The transportation system is the context of urban development and the guarantee of residents' travel. By modeling the urban transportation system as a complex network, we can gain an in-depth understanding of the structure, characteristics and behavior of the transportation network from both macro and micro levels. At the macro level, complex network models can help understand the topological structure of urban transportation networks, such as the connectivity, agglomeration and centrality of the network. These characteristics are of great significance for evaluating the efficiency and robustness of transportation networks, as well as predicting traffic flow and congestion. At the micro level, complex network models can reveal local interactions and dynamic behaviors in transportation networks, such as traffic flow at intersections, vehicle speeds and driving paths, etc. This information is of great value for optimizing traffic signal control, improving road utilization and improving the quality of transportation services.

[0003] However, existing technologies have limitations when dealing with the complexity of traffic networks. The identification and analysis of key traffic nodes requires the integration of road segment capacity, traffic demand and control strategies, but traditional algorithms, such as K-shell decomposition and betweenness centrality measurement, often fail to meet the requirements due to the neglect of the dynamic and nonlinear characteristics of traffic networks. In addition, the failure to effectively integrate road segment functional attributes and real-time traffic data has limited the accuracy of traffic node importance assessment.

[0004] To overcome the above challenges, traffic network analysis urgently needs technological innovation. First, the algorithm should be enhanced to capture the temporal changes in traffic demand, such as introducing a time dimension to reflect the dynamic characteristics of traffic flow. Second, traffic node attribute information needs to be further enriched, integrating road section carrying capacity, traffic control parameters and real-time traffic data to more accurately identify traffic nodes that have a significant impact on traffic flow, travel time and safety.

[0005] In summary, through algorithm innovation and data integration, traffic network analysis is expected to achieve more accurate identification and evaluation of key traffic nodes, thereby optimizing traffic control strategies, improving the overall efficiency and robustness of the traffic network, and providing strong support for future traffic needs and urban sustainable development planning. This direction not only promotes the progress of traffic management and planning, but also provides a theoretical basis and practical guide for the continuous optimization and intelligence of urban transportation systems. Summary of the invention

[0006] The purpose of the embodiment of the present invention is to provide a method for identifying the importance of traffic nodes in a transportation network, so as to accurately identify key traffic nodes in a complex transportation network, so as to optimize traffic flow, reduce congestion, improve the stability of the transportation system, and reduce the risk of traffic accidents, so as to further improve the performance of the transportation system.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is a method for identifying the importance of traffic nodes in a traffic network, comprising the following steps:

[0008] S1. Establish a complex transportation network G, G = (V, E), and determine the adjacency matrix A of each node in the transportation network, where G represents the complex transportation network, V represents the set of nodes in the network, and E represents the set of edges in the network;

[0009] S2, calculating the ks value of all traffic nodes in the traffic complex network, wherein the ks value is the shell level where the traffic node is located when it is removed;

[0010] S3, record the number of iterations of the traffic node with the smallest degree in the current network as IT, and consider the path and state disturbance information of the traffic nodes to calculate the information entropy of these traffic nodes;

[0011] S4, delete the traffic node with the smallest information entropy from the complex traffic network G;

[0012] S5. If the number of traffic nodes in the complex traffic network is 0, record the current iteration number as IT-1 and proceed to S6; otherwise, record the iteration number as IT+1 and return to S3;

[0013] S6. Under the current number of iterations, select the traffic node with the largest information entropy from the traffic node set, and add it to the traffic node importance ranking set in order; when there are multiple traffic nodes with equal information entropy values, add all the traffic nodes with equal information entropy to the importance ranking set in descending order of the traffic node numbers;

[0014] S7. When the current number of iterations is 0, proceed to S8; otherwise, record the current number of iterations as IT-1 and proceed to S6;

[0015] S8. When all the traffic nodes in the traffic network have been placed in the ordered set of important traffic nodes, the iteration ends; otherwise, return to S6.

[0016] Furthermore, the specific steps of S2 are as follows:

[0017] S201, initializing the traffic network: calculating the degrees of all traffic nodes;

[0018] S202, deleting traffic nodes: deleting the traffic nodes with the smallest degree and the edges connected to them from the network, and recording the ks values ​​of these traffic nodes as 1;

[0019] S203, updating the degree: recalculating the degree of the remaining traffic nodes;

[0020] S204, repeating steps S202 and S203: continue to delete the traffic node with the smallest current degree, and record its ks value as the current iteration number, until all traffic nodes are deleted.

[0021] Furthermore, the traffic node information entropy in S3 is specifically:

[0022]

[0023] Where, e(i) represents the information entropy of traffic node i; Γ i is the set of all neighboring traffic nodes of traffic node i, I j represents the set of neighboring traffic nodes j of traffic node i, where:

[0024]

[0025] In the formula, k i is the degree of transportation node i; k j is the degree of transportation node j; I i represents the ratio of the degree of traffic node i to the sum of the degrees of all its neighboring traffic nodes; N represents the number of traffic nodes in the entire complex network;

[0026]

[0027] Among them, a ij Represents the connection relationship between traffic nodes i and j;

[0028]

[0029] Among them, e η (i) represents the information entropy of traffic node i considering the betweenness centrality and the degree of disturbance of traffic node status; BC(i) represents the betweenness centrality of traffic node i, P(i) represents the average impact of the failure of traffic node i on the load status of its surrounding neighboring traffic nodes; ε represents the weight parameter.

[0030] Furthermore, the method for calculating the betweenness centrality of the traffic node i is specifically as follows:

[0031]

[0032] Among them, s and t are two different traffic nodes in the traffic node set of the complex traffic network; σ strepresents the number of shortest paths from traffic node s to traffic node t; σ st (i) represents the number of shortest paths from traffic node s to traffic node t that pass through traffic node i.

[0033] Furthermore, when traffic node i fails, the evaluation of the load status of its neighboring traffic nodes includes evaluating the remaining load capacity and the load remaining capacity of the neighboring traffic nodes, specifically:

[0034] Calculate the remaining load capacity:

[0035]

[0036]

[0037] C i =(1+β)F i

[0038] Among them, Y i is the remaining load capacity of traffic node i; C i is the capacity of traffic node i; F i represents the real-time load of traffic node i, where the degree of traffic node i is k; a and α are adjustment parameters, a is used to adjust the initial load size, and α is used to adjust the size of the load parameter; β represents the traffic load tolerance parameter;

[0039] Then determine the load remaining capacity of the neighboring traffic nodes:

[0040]

[0041] Among them, S i is the state expression of traffic node i; Y i represents the remaining load capacity of traffic node i; Y j represents the remaining load capacity of traffic node j; Γ i is the set of all neighboring traffic nodes of traffic node i; γ is the weight parameter, which is used to allocate the state proportion of each neighboring traffic node in traffic node failure and load redistribution. Represents the empty set.

[0042] Furthermore, the load and load redistribution of each neighboring traffic node are as follows:

[0043] F i '=F i +ΔF i ,i∈Γ f

[0044]

[0045] Among them, Ff represents the load borne by the traffic node f before failure; F i and F i ' are the loads before and after the traffic node is updated; ΔF i is the load increment; S i is the state of traffic node i; S j is the state of neighboring traffic node j of traffic node i; Γ f It is the aggregation of neighboring traffic nodes of the failed traffic node.

[0046] Furthermore, during the load redistribution process, the load increment ΔF i <βF i , β represents the traffic load tolerance parameter;

[0047] The average load fluctuation P(f) caused by the failure of traffic node f on all its neighboring traffic nodes is:

[0048]

[0049] Among them, |Γ f | represents the number of neighboring traffic nodes of the faulty traffic node f.

[0050] The beneficial effects of the present invention are:

[0051] The present invention introduces node importance indicators constructed by path and state disturbances, and the method of the present invention can more accurately identify the importance of each node. This is not only reflected in the accurate assessment of the importance of a single node, but also in significant improvements at the overall network level. Traditional methods tend to identify a large number of key nodes in certain specific areas. This phenomenon is called the "rich club", that is, some nodes are overvalued while other nodes are ignored. The method of the present invention overcomes this limitation, making the distribution of the identified important nodes more extensive and even throughout the network, improving the robustness and efficiency of the network.

[0052] Specifically, the method of the present invention first uses the K-shell algorithm to stratify the network, and combines the betweenness centrality to evaluate the global importance of the node. Then, by considering the path information and state disturbance of the node, the evaluation criteria of the node importance are further refined. This method not only takes into account the attributes of the node itself, but also comprehensively considers the position of the node in the traffic network and its impact on the surrounding nodes, so that the importance of the node can be evaluated more comprehensively. In addition, the present invention also considers the load redistribution mechanism between nodes, which is crucial for evaluating the impact of node failure on the overall stability of the network. The experimental results show that the method of the present invention can distinguish the importance of nodes more finely, and the distribution of the identified nodes in the entire network is more reasonable. This means that through the effective management and optimization of these key nodes, the overall operation efficiency and reliability of the traffic network can be significantly improved, traffic congestion and accident risks can be reduced, and then strong support can be provided for the sustainable development of the city. In summary, the present invention can bring substantial improvements to traffic management and planning, and help promote the continuous optimization and intelligent development of urban transportation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 is a flow chart of a method for implementing the present invention;

[0055] Figure 2 It is the Nguyen-Dupuis complex network structure diagram;

[0056] Figure 3 Schematic diagram of output results of an implementation method of the present invention, (a) is the output diagram of the method result, and (b) is the output diagram of the IKS algorithm. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] A complex network is an abstract model for understanding complex systems in the real world. It is a network structure composed of a huge number of traffic nodes and the intricate relationships between them. It abstracts the entities in a complex system into traffic nodes and the relationships between entities into lines.

[0059] Information entropy is used to quantify the uncertainty of information and can be understood as a measure of the complexity of a system. If a system is more complex and has more different situations, its information entropy is larger. On the contrary, if a system is simpler and has fewer situations, its information entropy is smaller.

[0060] In complex networks, the K-shell algorithm is a coarse-grained decomposition method based on the global structure of the network, which is used to determine the importance of traffic nodes. The K-shell method divides the network into layers, obtains the K-shell value of each traffic node, and measures the impact of the global structure of the complex network through the K-shell value. For traffic nodes with the same K-shell value, the traffic node with a larger comprehensiveness is more important.

[0061] like Figure 1 ,This embodiment provides a method for identifying the importance of traffic nodes in a traffic network, taking into account the path and state disturbance of the traffic nodes, and specifically includes the following steps:

[0062] S1. Before executing the algorithm, a complex transportation network G(V,E) is established and the adjacency matrix A is determined; where G represents the complex transportation network, V represents the set of nodes in the network, and E represents the set of edges in the network; where the adjacency matrix is ​​used to represent the connection relationship between each node in the transportation network;

[0063] S2. In the first iteration of the algorithm, the K-shell algorithm is used to calculate the ks values ​​of all traffic nodes in the traffic complex network. The specific steps of the K-shell algorithm are as follows:

[0064] S21. Initialization: Calculate the degrees of all traffic nodes (ie, the number of edges connected to the traffic node).

[0065] S22. Delete traffic nodes: Delete the traffic nodes with the smallest degree and their connected edges from the network, and record the K-shell values ​​(i.e., ks values, ks refers to the shell level when the traffic node is removed) of these traffic nodes as 1.

[0066] S23. Update degree: recalculate the degrees of the remaining traffic nodes.

[0067] S24. Repeat steps S22 and S23: continue to delete the traffic node with the smallest degree, and record its k-shell value as the current iteration number until all traffic nodes are deleted.

[0068] S3. The iterative step of the traffic node with the smallest degree in the current network is recorded as IT, and the information entropy of these traffic nodes (i.e., the traffic nodes with the smallest degree) is calculated by considering the path and state disturbance information of the traffic nodes:

[0069] In the Improved K-shell Algorithm (IKS), the traffic node information entropy is defined as:

[0070]

[0071] Where, e(i) represents the information entropy of traffic node i; j∈Γ i , Γ i Indicates that traffic node j is a neighboring traffic node of traffic node i, I j represents the set of neighboring traffic nodes of traffic node i, where:

[0072]

[0073] In the formula, k i is the degree of transportation node i, I i represents the ratio of the degree of traffic node i to the sum of the degrees of all its neighboring traffic nodes; N represents the number of traffic nodes in the entire complex network; a ij represents the connection relationship between traffic nodes i and j, a ij =1 means that traffic nodes i and j are connected, a ij =0 means that traffic nodes i and j are not connected.

[0074] The calculation of traffic node information entropy mainly depends on the information of the surrounding adjacent traffic nodes. Since the Ks value obtained by relying solely on the information entropy of local information of traffic nodes is not refined enough, traffic nodes on the periphery of a complex network have the same Ks value. However, these traffic nodes have different influences on the network path. In order to further improve the importance of traffic nodes in the K-shell algorithm, this implementation proposes a method for calculating the information entropy of traffic nodes from the perspective of traffic node paths and traffic network functional characteristics, combined with betweenness centrality and the degree of disturbance of traffic node states:

[0075]

[0076] Among them, e η(i) represents the information entropy of traffic node i considering the betweenness centrality and the degree of disturbance of traffic node status; BC(i) represents the betweenness centrality (BC) of traffic node i; P(i) represents the average impact of the failure of traffic node i on the load status of its surrounding neighboring traffic nodes; ε represents the weight parameter, and ε∈[0,1].

[0077] The betweenness centrality in formula (3) is an indicator used to measure the importance of traffic nodes in the network, reflecting the global properties of traffic nodes. It measures the degree to which a traffic node acts as a bridge or connection point in the network, that is, the traffic node plays a key role in the transmission of information and resources in the network. Its definition is as follows:

[0078]

[0079] Among them, s and t are two different traffic nodes in the traffic node set of the complex traffic network; σ st represents the number of shortest paths from traffic node s to traffic node t; σ st (i) represents the number of shortest paths from traffic node s to traffic node t that pass through traffic node i.

[0080] The evaluation parameter P(i) in formula (3) is calculated from the load status of the target traffic node and the neighboring traffic nodes. The status of a traffic node is not only related to its own load capacity, but also to the load capacity of its neighboring traffic nodes. Therefore, the real-time evaluation of the status of a traffic node is divided into two stages. In the first stage, the remaining load capacity is as follows:

[0081]

[0082] Among them, Y i is the remaining load capacity of traffic node i; C i is the capacity of traffic node i; F i represents the real-time load of traffic node i, where the degree of traffic node i is k, F i is defined as (a and α are adjustment parameters used to adjust the size of the initial load and load parameters); Assume that the capacity C of traffic node i i With its initial load F i Then the traffic node capacity model is C i =(1+β)F i , the constant β represents the maximum tolerance of the network to the additional load. From formula (5), it can be seen that when the load is too large (exceeding the capacity of the traffic node), the difference between the capacity of the traffic node with load surplus capacity and the real-time load is 0. The load surplus capacity reflects the ability of the traffic node to accommodate the load under the initial load conditions, which is related to Yi The values ​​are positively correlated.

[0083] In the second stage, the residual load capacity of neighboring traffic nodes should be considered, and a traffic node status model that can more accurately measure this value should be proposed. The model comprehensively considers the residual carrying capacity of the evaluated traffic node and its neighboring traffic nodes, as shown in the following formula:

[0084]

[0085] Among them, S i is the state expression of traffic node i; Y i represents the remaining load capacity of traffic node i; Γ i is the set of all neighboring traffic nodes of traffic node i; γ is a weight parameter, which is adjusted to distribute the state proportion of each neighboring traffic node in traffic node failure and load redistribution.

[0086] Load redistribution refers to the rule that when a traffic node fails for some reason, the load it originally borne is distributed to other neighboring traffic nodes. After the update, the load of the neighboring traffic nodes and the load distribution model are shown in equations (7) and (8):

[0087] F i '=F i +ΔF i , i∈Γ f (7)

[0088]

[0089] Among them, F f represents the load borne by the traffic node f before failure; F i and F i ' are the loads before and after the traffic node is updated; ΔF i is the load increment; S i is the state of traffic node i; S j is the state of neighboring traffic node j of traffic node i; Γ f It is the aggregation of neighboring traffic nodes of the failed traffic node.

[0090] In the load redistribution model based on the evolution of traffic node status, the importance evaluation of traffic nodes has important application value and significance. In the model, the more important the traffic node is, the greater the impact of its failure or failure on the entire network, and the more protection the traffic node should have. The importance evaluation of traffic nodes can be measured by the average failure degree of these neighboring traffic nodes caused by transferring the load of the faulty traffic node to the surrounding neighboring traffic nodes. Due to the limitation of traffic node capacity, further cascading failures should be avoided, as shown in formula (9):

[0091]

[0092] According to formula (9), we can use represents the load oscillation degree of the neighboring traffic nodes of the faulty traffic node f. β represents the tolerance parameter, which represents the additional traffic load that the traffic node can bear. Since the failure of a traffic node will cause load fluctuations of all its neighboring traffic nodes, the importance of traffic node f can be represented by the average load fluctuation of all its neighboring traffic nodes, as shown in formula (10):

[0093]

[0094] Where P(f) represents the average load fluctuation caused by the failure of traffic node f to all its neighboring traffic nodes, that is, the importance of traffic node f; |Γ f | represents the number of neighboring traffic nodes of the faulty traffic node f.

[0095] S4. Delete these traffic nodes with the smallest information entropy from the network.

[0096] S5. If the number of traffic nodes in the network is 0, record IT max = previous iteration step IT, jump to S6; otherwise, IT increases by 1 and jump to S3.

[0097] S6. In the current iteration step IT, the traffic node with the largest information entropy is selected from the traffic node set, and it is added to the traffic node importance ranking set in order. If there are multiple traffic nodes with equal information entropy values, then all traffic nodes with equal information entropy are added to the importance ranking set in descending order of the traffic node numbers.

[0098] S7. If IT=0, jump to step 8; otherwise, IT decreases by 1 and jumps to S6.

[0099] S8: If all the traffic nodes in the network have been put into the important traffic node sorting set, the algorithm ends. Otherwise, set the previous iteration step IT to IT(max), then jump to S6 and continue to perform the same operation.

[0100] This implementation uses the Nguyen-Dupuis network to verify the feasibility and effectiveness of the proposed method. Figure 2 ,The Nguyen-Dupuis example network consists of 11 traffic nodes and 17 edges. Tables 1 and 2 show the ranking results of the IKS algorithm and the traffic node importance identification algorithm considering the path and disturbance state, respectively.

[0101] Table 1 Ranking results of existing (IKS) algorithm

[0102]

[0103]

[0104] Table 2 Ranking results of traffic node importance identification algorithm considering path and disturbance state

[0105]

[0106]

[0107] The comparison between the traffic node importance recognition results studied in this paper and the IKS recognition results is shown in Table 3.

[0108] Table 3 Comparison of recognition results

[0109] Ranking IKS Methods of the present invention 1 2、4、5 2 2 3 4 3 7、9、10 1 4 6 3 5 8 10 6 1、11 7 7 8 8 11 9 5 10 6 11 9

[0110] Compared with the original algorithm, the traffic node importance index constructed by the method of the present invention through path and state disturbance can more accurately identify the importance of each traffic node. Figure 3 (a) The method of the present invention not only makes the importance difference of each traffic node more significant, but also makes the distribution of the identified important traffic nodes more extensive in the entire network, thus overcoming the "rich man's club" phenomenon commonly seen in traditional methods. Figure 3 (b) Traditional algorithms tend to focus on identifying a few traffic nodes as key traffic nodes, causing these traffic nodes to cluster in the network, forming a structure similar to a "rich man's club", and ignoring other potential important traffic nodes. The method of the present invention makes the distribution of important traffic nodes in the network more uniform and diversified through more detailed path and state disturbance analysis, thereby improving the robustness and efficiency of the entire network.

[0111] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A method for identifying the importance of traffic nodes in a traffic network, characterized in that: The following steps are involved: S1. Establish a complex transportation network G, G = (V, E), and determine the adjacency matrix A of each node in the transportation network, where G represents the complex transportation network, V represents the set of nodes in the network, and E represents the set of edges in the network; S2, calculating the ks value of all traffic nodes in the traffic complex network, wherein the ks value is the shell level where the traffic node is located when it is removed; S3, record the number of iterations of the traffic node with the smallest degree in the current network as IT, and consider the path and state disturbance information of the traffic nodes to calculate the information entropy of these traffic nodes; S4, delete the traffic node with the smallest information entropy from the complex traffic network G; S5. If the number of traffic nodes in the complex traffic network is 0, record the current iteration number as IT-1 and proceed to S6; otherwise, record the iteration number as IT+1 and return to S3; S6. Under the current number of iterations, select the traffic node with the largest information entropy from the traffic node set, and add it to the traffic node importance ranking set in order; when there are multiple traffic nodes with equal information entropy values, add all the traffic nodes with equal information entropy to the importance ranking set in descending order of the traffic node numbers; S7. When the current number of iterations is 0, proceed to S8; otherwise, record the current number of iterations as IT-1 and proceed to S6; S8. When all the traffic nodes in the traffic network have been placed in the ordered set of important traffic nodes, the iteration ends; otherwise, return to S6.

2. A method for identifying the importance of traffic nodes in a traffic network according to claim 1, characterized in that: The specific steps of S2 are as follows: S201, initializing the traffic network: calculating the degrees of all traffic nodes; S202, deleting traffic nodes: deleting the traffic nodes with the smallest degree and the edges connected to them from the network, and recording the ks values ​​of these traffic nodes as 1; S203, updating the degree: recalculating the degree of the remaining traffic nodes; S204, repeating steps S202 and S203: continue to delete the traffic node with the smallest current degree, and record its ks value as the current iteration number, until all traffic nodes are deleted.

3. A method for identifying the importance of traffic nodes in a traffic network according to claim 1, characterized in that: The traffic node information entropy described in S3 is specifically: Where, e(i) represents the information entropy of traffic node i; Γ i is the set of all neighboring traffic nodes of traffic node i, I j represents the set of neighboring traffic nodes j of traffic node i, where: In the formula, k i is the degree of transportation node i; k j is the degree of transportation node j; I i represents the ratio of the degree of traffic node i to the sum of the degrees of all its neighboring traffic nodes; N represents the number of traffic nodes in the entire complex network; Among them, a ij Represents the connection relationship between traffic nodes i and j; Among them, e η (i) represents the information entropy of traffic node i considering the betweenness centrality and the degree of disturbance of traffic node status; BC(i) represents the betweenness centrality of traffic node i, P(i) represents the average impact of the failure of traffic node i on the load status of its surrounding neighboring traffic nodes; ε represents the weight parameter.

4. A method for identifying the importance of traffic nodes in a traffic network according to claim 3, characterized in that: The specific method for calculating the betweenness centrality of the traffic node i is: Among them, s and t are two different traffic nodes in the traffic node set of the complex traffic network; σ st represents the number of shortest paths from traffic node s to traffic node t; σ st (i) represents the number of shortest paths from traffic node s to traffic node t that pass through traffic node i.

5. A method for identifying the importance of traffic nodes in a traffic network according to claim 3, characterized in that: When traffic node i fails, the evaluation of the load status of its neighboring traffic nodes includes evaluating the remaining load capacity and the load remaining capacity of the neighboring traffic nodes, specifically: Calculate the remaining load capacity: C i =(1+β)F i Among them, Y i is the remaining load capacity of traffic node i; C i is the capacity of traffic node i; F i represents the real-time load of traffic node i, where the degree of traffic node i is k; a and α are adjustment parameters, a is used to adjust the initial load size, and α is used to adjust the size of the load parameter; β represents the traffic load tolerance parameter; Then determine the load remaining capacity of the neighboring traffic nodes: Among them, S i is the state expression of traffic node i; Y i represents the remaining load capacity of traffic node i; Y j represents the remaining load capacity of traffic node j; Γ i is the set of all neighboring traffic nodes of traffic node i; γ is the weight parameter, which is used to allocate the state proportion of each neighboring traffic node in traffic node failure and load redistribution. Represents the empty set.

6. A method for identifying the importance of traffic nodes in a traffic network according to claim 5, characterized in that: The load and load redistribution of each neighboring traffic node are as follows: Among them, F f represents the load borne by the traffic node f before failure; F i and F i ' are the loads before and after the traffic node is updated; ΔF i is the load increment; S i is the state of traffic node i; S j is the state of neighboring traffic node j of traffic node i; Γ f It is the aggregation of neighboring traffic nodes of the failed traffic node.

7. A method for identifying the importance of traffic nodes in a traffic network according to claim 6, characterized in that: During load redistribution, the load increment ΔF i <βF i , β represents the traffic load tolerance parameter; The average load fluctuation P(f) caused by the failure of traffic node f on all its neighboring traffic nodes is: Among them, |Γ f | represents the number of neighboring traffic nodes of the faulty traffic node f.

Images