River-sea combined transportation network reliability analysis method and enhancement method based on network survivability

Abstract

The invention relates to a river-sea combined transportation network reliability analysis method and enhancement method based on network survivability, and the analysis method comprises the steps: S1, obtaining a river-sea combined transportation network to be detected, and constructing a non-cascade port failure model and a cascade port failure model according to the river-sea combined transportation network; and S2, attacking the non-cascade port failure model and the cascade failure model by adopting a random attack mode and a deliberate attack mode respectively, and calculating a relative size performance parameter and a network global efficiency performance parameter of the attacked maximum connected subgraph. Compared with the prior art, the method has the advantages that more accurate global influence can be obtained after a single node fails, and the like.

Description

Technical Field

[0001] The present invention relates to the field of transportation, and in particular to a reliability analysis method and enhancement method for a river-sea combined transport network based on network survivability. Background Art

[0002] The river-sea transport network model is a model used to evaluate the actual situation of river-sea transport. In this network model, ports are used as nodes and routes between ports are used as edges. The reliability of river-sea shipping is related to the safety, efficiency, and traffic planning of port shipping logistics. The analysis of reliability is an integral part of the study of river-sea transport.

[0003] There are currently very few studies and analyses on the overall network's anti-destruction capabilities in the market, which makes it difficult for people to globally predict the risk resistance of the entire port intermodal transport operation. It is unclear what degree of damage will be caused to the entire port shipping when some ports fail to operate under the influence of force majeure. Summary of the invention

[0004] The purpose of the present invention is to provide a reliability analysis method and enhancement method for river-sea transport network based on network survivability.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A reliability analysis method for river-sea transport network based on network survivability includes:

[0007] Step S1: obtaining a river-sea transport network to be tested, and constructing a non-cascade port failure model and a cascade port failure model according to the river-sea transport network;

[0008] Step S2: Use random attack and deliberate attack to attack the non-cascading port failure model and the cascading failure model respectively, and calculate the relative size performance parameter of the maximum connected subgraph and the global efficiency performance parameter of the network after the attack.

[0009] In the cascading failure model, when any node fails, the following steps are performed:

[0010] Load distribution step: distribute the load of the failed node to the neighboring nodes connected to the failed node;

[0011] Capacity determination step: After executing the load distribution step, determine one by one whether the load of each neighboring node is greater than its own capacity. If so, the node with a load greater than its own capacity is regarded as a failed node, and the load distribution step is executed.

[0012] In the load distribution step, the newly added load distributed to each neighboring node is:

[0013]

[0014] Where: △L m The additional load allocated to node m, P Fn For the distribution ratio, is the load of the failed node, O m is the capacity of node m, F n is the set of neighbor nodes of the failed node, L m Load before distribution to node m.

[0015] The process of the random attack includes:

[0016] Step S2-A-1: Initialize the failure model of the attacked port and set the number of attack rounds;

[0017] Step S2-A-2: randomly select a node as a node to be failed;

[0018] Step S2-A-3: removing the node to be failed and the edges connected to the node to be failed from the port failure model;

[0019] Step S2-A-4: Calculate the maximum connected subgraph relative size performance parameter and the network global efficiency performance parameter;

[0020] Step S2-A-5: Determine whether the set number of attack rounds has been reached. If yes, end the random attack. Otherwise, execute step S2-A-6;

[0021] Step S2-A-6: Determine whether there are any nodes that have not failed. If yes, return to step S2-A-2. Otherwise, end the random attack.

[0022] The process of the deliberate attack includes:

[0023] Step S2-B-1: Initialize the failure model of the attacked port and set the number of attack rounds;

[0024] Step S2-B-2: Sort all nodes that have not failed according to the importance characteristic value of each node, and select the node with the largest importance characteristic value as the node to be failed;

[0025] Step S2-B-3: remove the node to be failed and the edges connected to the node to be failed from the port failure model;

[0026] Step S2-B-4: Calculate the relative size performance parameter of the largest connected subgraph and the global efficiency performance parameter of the network;

[0027] Step S2-B-5: Determine whether the set number of attack rounds has been reached. If yes, end the random attack. Otherwise, execute step S2-B-6;

[0028] Step S2-B-6: Determine whether there are any nodes that have not failed. If yes, return to step S2-A-2. Otherwise, end the random attack.

[0029] The importance characteristic value is determined by the node degree value, node betweenness and node PR value.

[0030] The relative size of the largest connected subgraph is the ratio of the number of nodes of the largest connected subgraph to the number of nodes of the port failure model.

[0031] The global efficiency of the network is specifically:

[0032]

[0033] Where: E global is the global efficiency of the network, N is the total number of nodes, and d(i,j) is the distance between node i and node j.

[0034] A method for enhancing the reliability of a river-sea transport network based on the above analysis method comprises:

[0035] In the load distribution step, multiple target nodes are selected from all non-neighboring nodes of the failed node, and load distribution is performed together with the neighboring nodes of the failed node.

[0036] The target node includes selecting a node with the highest degree value and selecting a node with the highest PR value.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. A non-cascading failure model and a cascading failure model for river-sea transport were constructed. In the non-cascading failure model, the network model was analyzed for failure through random attacks or intentional attacks. In the cascading failure model, a node was made to fail through random attacks or intentional attacks, and the load of the failed node was distributed to the adjacent nodes. The adjacent nodes were judged to exceed their own capacity after receiving the load of the failed node, and whether the adjacent nodes continued to fail was determined by repeating the cycle to analyze the indestructibility of the network model. In the analysis of indestructibility, the two performance parameters of the relative size of the maximum connected subgraph and the global efficiency of the network were mainly used for analysis and verification. Through this analysis method, the impact of the failure of each port node in the actual river-sea transport on the indestructibility of the entire port transport can be predicted.

[0039] 2. By adding edges to port nodes, the anti-destruction capability of the entire network can be improved, providing improvement measures for increasing the risk of port intermodal transport. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of the random attack implementation method for this application;

[0041] Figure 2 Flowchart of the method for implementing the deliberate attack for this application;

[0042] Figure 3 This is a schematic diagram of load distribution for cascading failures in this application;

[0043] Figure 4 This is the flow chart of cascading failure operation for this application;

[0044] Figure 5 This is a schematic diagram of the relative size change of the largest connected subgraph of the network under non-cascading failure of this application;

[0045] Figure 6 This is a schematic diagram of the change of global network efficiency under non-cascading failure in this application;

[0046] Figure 7 This is a schematic diagram of the relative size change of the largest connected subgraph of the network under cascading failure of this application;

[0047] Figure 8 This is a schematic diagram of the change in global network efficiency under cascading failure in this application;

[0048] Fig. 9 This is a schematic diagram of the relative size change of the largest connected subgraph of the network after random edge addition in this application;

[0049] Fig.10 This is a schematic diagram of the change in global network efficiency after random edge addition in this application;

[0050] Fig.11 This is an operational flow chart for this application that uses the edge-adding method to improve network reliability;

[0051] Fig.12 for Fig.11 Schematic diagram of the change in global network efficiency after adding edges;

[0052] Fig.13 It is a schematic diagram of the main steps of the present invention. DETAILED DESCRIPTION

[0053] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0054] A reliability analysis method for river-sea transport network based on network survivability, such as Fig.13 As shown, including:

[0055] Step S1: Obtain the river-sea transport network to be tested, and construct a non-cascade port failure model and a cascade port failure model according to the river-sea transport network. Each node in the river-sea transport network represents a port, and the edges between nodes represent the existence of direct routes between ports.

[0056] Step S2: Use random attack and deliberate attack to attack the non-cascading port failure model and the cascading failure model respectively, and calculate the relative size performance parameter of the maximum connected subgraph and the global efficiency performance parameter of the network after the attack.

[0057] Among them, the non-cascading failure model refers to the use of random attack or deliberate attack mode to attack the network nodes in the river-sea transport network model and observe the reliability of the river-sea transport network; the cascading failure model refers to the use of random attack or deliberate attack mode to attack a certain network node or edge in the river-sea transport network model to make it fail. The load on the node will redistribute its own load to the nodes adjacent to the node according to the corresponding distribution rules. After the adjacent node receives the load from the failed node and compares it with its own capacity, if it is greater than, the adjacent node fails, and so on, resulting in a chain reaction of network node failure; if it is less than, the adjacent node does not fail.

[0058] When this application attacks the model, the following environmental conditions are met:

[0059] (1) Each attack can completely disable the node and its surrounding edges.

[0060] (2) Only one port node is attacked each time. The network input for the next attack is the network that was attacked last time. That is, the number of port nodes in the network each time is one less than the previous one.

[0061] like Figure 1 As shown in the figure, during the random attack process, a port node is randomly selected from the network port nodes to attack each time, and then the relative maximum connected subgraph of the network after the attack and the global efficiency of the network are calculated, and then the above operations are repeated until the reliability index of the network drops to 0, which indicates that the river-sea transport network has completely failed.

[0062] Specifically, the process of random attack includes:

[0063] Step S2-A-1: Initialize the failure model of the attacked port, including the weights of all port nodes and edges, and set the number of attack rounds and the total number of port nodes to be actively removed. Since each attack only attacks one node, the total number of port nodes to be actively removed is the same as the total number of attacks.

[0064] Step S2-A-2: randomly select a node as the node to be invalidated. Specifically, in each round of attack, randomly select a node from the port nodes that are still in the network. The selection can be achieved by generating a random number, and the range of the random number corresponds to the number of nodes in the current network.

[0065] Step S2-A-3: removing the node to be failed and the edges connected to the node to be failed from the port failure model;

[0066] Step S2-A-4: Calculate the maximum connected subgraph relative size performance parameter and the network global efficiency performance parameter;

[0067] Step S2-A-5: Determine whether the set number of attack rounds has been reached. If yes, end the random attack. Otherwise, execute step S2-A-6;

[0068] Step S2-A-6: Determine whether there are any nodes that have not failed. If yes, return to step S2-A-2. Otherwise, end the random attack.

[0069] like Figure 2 As shown in the figure, a deliberate attack is to first attack the important port nodes in the river-sea transport network. Usually, the failure of these nodes has a greater impact on the reliability of the network. Specifically, the process of a deliberate attack includes:

[0070] Step S2-B-1: Initialize the failure model of the attacked port and set the number of attack rounds;

[0071] Step S2-B-2: Sort all nodes that have not failed according to the importance characteristic value of each node, and select the node with the largest importance characteristic value as the node to be failed, where the importance characteristic value is determined by the node degree value, node betweenness and node PR value;

[0072] Step S2-B-3: remove the node to be failed and the edges connected to the node to be failed from the port failure model;

[0073] Step S2-B-4: Calculate the relative size performance parameter of the largest connected subgraph and the global efficiency performance parameter of the network;

[0074] Step S2-B-5: Determine whether the set number of attack rounds has been reached. If yes, end the random attack. Otherwise, execute step S2-B-6;

[0075] Step S2-B-6: Determine whether there are any nodes that have not failed. If yes, return to step S2-A-2. Otherwise, end the random attack.

[0076] like Figure 4 As shown in the cascading failure model, when any node fails, the following steps are performed:

[0077] Load distribution step: distribute the load of the failed node to the neighboring nodes connected to the failed node;

[0078] Capacity determination step: After executing the load distribution step, determine one by one whether the load of each neighboring node is greater than its own capacity. If so, the node with a load greater than its own capacity is regarded as a failed node, and the load distribution step is executed.

[0079] Among them, Figure 3 As shown, in this embodiment, in the load distribution step, the newly added load distributed to each neighboring node is:

[0080]

[0081] Where: △L m The additional load allocated to node m, P Fn For the distribution ratio, is the load of the failed node, O m is the capacity of node m, F n is the set of neighbor nodes of the failed node, L m Load before distribution to node m.

[0082] In this embodiment, the capacity of the node is specifically:

[0083]

[0084] Where β is the tolerance factor, generally 0.1, 0.2, 0.3, 0.4 or 0.5. For node v i initial load.

[0085] After redistribution, the load of node m after receiving the load from the failed node is:

[0086] L′ m =L m +ΔL m

[0087] If L' m >O m , then node m fails, and a new round of load distribution is performed on the neighboring nodes of node m;

[0088] If L' m ≤O m , indicating the port node m Not saturated, not failed, and no new round of cascading failure occurs.

[0089] The details are as follows:

[0090] (1) At a certain moment, the network is in normal operation;

[0091] (2) At the next moment, node P fails due to some force majeure factors, and the edges between node P and its neighboring nodes M, K, and N also fail. Then node P distributes the load to its neighboring nodes M, K, and N according to the load distribution rule;

[0092] (3) When the neighbor node's load is greater than its own capacity after receiving the load from node P, the neighbor node becomes a saturated node and then fails, and then continues with process (2) until the load of all nodes in the network is less than their own capacity.

[0093] The relative size of the largest connected subgraph is an intuitive indicator of network connectivity. In the river-sea transport network, connectivity ensures that goods, information and resources can flow between different parts of the network. A large connected subgraph means that most nodes in the network are reachable to each other, which can effectively support transportation and logistics activities. The size of the largest connected subgraph can also reflect the robustness of the network in the face of node failures. If a network can still maintain a large maximum connected subgraph after some nodes fail, it means that the network has relatively strong robustness and can resist failures and maintain to a certain extent. Specifically, the relative size of the largest connected subgraph is the ratio of the number of nodes in the largest connected subgraph to the number of nodes in the port failure model.

[0094] The global efficiency of the network measures the average transmission efficiency between any two nodes in the network. High global efficiency means high transmission efficiency between nodes in the network. The failure of key port nodes may significantly reduce the global efficiency of the network. If the network can still maintain a relatively high global efficiency and maintain normal operation after the node fails, it means that the network is stable. Specifically, the global efficiency of the network is:

[0095]

[0096] Where: E global is the global efficiency of the network, N is the total number of nodes, and d(i,j) is the distance between node i and node j.

[0097] In this embodiment, for a certain actual case, the network reliability analysis results under non-cascading failure are as follows:

[0098] like Figure 5 The following are the simulation results of the river-sea transport network under several different attack modes, and the changing trend of the relative size of the network's largest connected subgraph. The horizontal axis represents the number of attacks, that is, the number of port node failures, and the vertical axis represents the relative size of the largest connected subgraph. Figure 5 It can be seen that:

[0099] When the random attack strategy is adopted, the relative size of the largest connected subgraph of the river-sea transport network decreases the most slowly compared with the other four attack methods (node ​​PR value attack, node betweenness attack, node degree attack, and node strength attack). This is because in this river-sea transport network, most of the port nodes are closely connected. After randomly attacking some nodes, the remaining port nodes still have a relatively strong connection. However, it can be seen from the curve of the random attack that there are some points that decrease very quickly between two adjacent attacks, which means that the attack happened to hit the nodes with relatively high importance in the network. Overall, after 20 random attacks, the relative size of the largest connected subgraph of the network dropped to 0, indicating that the network is basically disconnected.

[0100] like Figure 6 As shown, it is a schematic diagram of the change of network global efficiency under non-cascading failure; Figure 6 It can be seen that the initial global efficiency of the network is 0.1714. Under random attacks, the decline of the global efficiency of the network is relatively slow. After 20 attacks, the global efficiency of the network drops by about 90%, and the network efficiency fluctuates around 0.01, indicating that as the scope of the attack expands, the network can basically not operate normally.

[0101] In this embodiment, for a certain actual case, the results of network reliability analysis under cascading failure are as follows:

[0102] like Figure 7 The following are the simulation results of the river-sea transport network under several different attack modes, and the changing trend of the relative size of the network's largest connected subgraph. The horizontal axis represents the number of attacks, that is, the number of port node failures, and the vertical axis represents the relative size of the largest connected subgraph. Figure 7 It can be seen that:

[0103] After 10 consecutive attacks, all nodes in the network failed. Compared with non-cascading failure, cascading failure is more destructive to the network. Random attacks have the least impact on the river-sea transport network, and the maximum connected subgraph decreases at a slower rate. This is because random attacks are unlikely to immediately select key nodes in the network. The connectivity of the network decreases slowly, indicating that the network has good robustness to random failures. Node degree attacks and node PR value attacks are the most damaging to the network. Especially after the first three attacks, the network connectivity drops sharply to below 0.05, indicating that removing high-degree nodes or high-PR value nodes will quickly destroy the connectivity of the network. Compared with node strength attacks, node betweenness also significantly affects the connectivity of the network after the ninth attack, but the degree of damage is slightly lower than that of node degree attacks and node PR value attacks. This is because some key port nodes that act as bridges in the network are removed.

[0104] like Figure 8As shown in the figure, the changes in the global efficiency of the network under different attack strategies such as random attack, node PR value attack, node betweenness attack, node degree attack, and node strength attack are shown. All types of attacks have caused a significant decline in network efficiency. Under random attacks, the global efficiency of the network has declined more slowly. This is because random attacks do not continuously remove multiple key nodes, so the network can maintain a relatively high efficiency. However, when removing key nodes, the curve of random attacks will also fluctuate greatly, such as from the 4th to the 5th attack. This also shows that the river-sea transport network has a certain stability and robustness against random attacks. After the first three attacks, as the number of attacks increases, the impact of these strategies on network efficiency decreases. This is because the network is basically paralyzed, and there are only weak connections between individual nodes.

[0105] Finally, the present application also provides a method for enhancing the reliability of a river-sea transport network based on the above analysis method, including:

[0106] In the load distribution step, multiple target nodes are selected from all non-neighboring nodes of the failed node, and load distribution is performed together with the neighboring nodes of the failed node.

[0107] The target nodes include selecting the node with the highest degree value and selecting the node with the highest PR value.

[0108] Specifically, in some embodiments, nodes in the river-sea transport network are randomly connected to achieve network edge addition. Edge addition means increasing the mutual connection between nodes in the river-sea transport network. In order to determine whether the edge addition strategy can improve the reliability index of the network, a random edge addition method is adopted to verify whether edge addition can improve the reliability of the network. The main steps are as follows: two nodes are randomly selected from the original network nodes to form an edge, and the edge is added 20 times continuously. After each edge addition, the global efficiency of the network and the relative size of the largest connected subgraph are calculated.

[0109] like Fig. 9 and Fig.10 As shown in the figure, the relative size change of the largest connected subgraph after adding edges and the change of the global efficiency of the network. It can be seen from the figure that adding edges will enhance the reliability of the network and improve the global efficiency of the network. In addition, the first ten times of adding edges will greatly improve the performance of the network, and the subsequent additions will have less impact on the overall performance of the network, which shows that there is also a threshold for the number of added edges.

[0110] like Fig.11 As shown in the figure, the operation process of increasing the edge to improve network reliability is as follows:

[0111] Step M01: Initialization: input the initial state of the network, including the weights of all port nodes and edges, and set the capacity and load of the port nodes;

[0112] Step M02: randomly select a node as a failed node, and extract a list of non-neighboring nodes of the selected failed node;

[0113] Step M03: randomly select a node from the non-neighbor node list, select a node with the highest degree value, and select a node with the highest PR value to form an edge with the failed node. Then these selected nodes are the new neighbor nodes of the failed node.

[0114] Step M04: after each edge is added, the selected failed nodes are removed to redistribute the load of the failed nodes;

[0115] Step M05: Verify whether the load of the neighboring node exceeds its capacity after the failed node load is redistributed. If so, proceed to step M04 until no node fails in the network. If not, proceed to step M06.

[0116] Step M06: Calculate the global efficiency index of the network.

[0117] According to the above steps M01-M06, a node is randomly selected as the failed node. After each edge addition, the network removes the selected port node. The change in the global efficiency of the network is as follows: Fig.12 shown.

[0118] from Fig.12 It can be seen that even after a port node is removed from the network, the global efficiency of the river-sea transport network is gradually increasing with the increase in the number of added edges, indicating that adding edges can effectively improve the network's resistance to cascading failures and reliability. Under the three strategies of degree value addition, PR value edge addition, and random edge addition, priority is given to establishing connections with nodes with higher PR values ​​among the non-adjacent nodes of the failed node, and the improvement of network efficiency is more obvious. Therefore, under limited costs, priority is given to increasing routes with port nodes with high PR values.

[0119] This application deeply analyzes the changes in the reliability of the river-sea transport network when facing different attack strategies, and simulates the failure of port nodes through random attacks and deliberate attacks. The analysis results show that: (1) Under non-cascading failures, although the river-sea transport network has good resistance to random node failures, targeted and deliberate attacks on core port nodes will greatly reduce the connectivity and efficiency of the network. (2) Under cascading failures, it is found that removing hub nodes will cause serious network cascading failures. (3) It is further proposed to improve the overall reliability of the network by increasing edges. Through the changes in the global efficiency of the network, it is shown that the network efficiency is more significantly improved by preferentially establishing connections with nodes with higher PR values ​​among the non-adjacent nodes of the failed node.

[0120] The present application provides a feasible method for reliability analysis of river-sea transport network. Through the present application, the reliability analysis of the river-sea transport network model composed of port nodes formed by each port and edges formed by each ship shipping can be carried out, providing a prediction method for actual river-sea transport reliability issues.

[0121] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

Claims

1. A reliability analysis method for river-sea transport network based on network survivability, characterized in that: include: Step S1: obtaining a river-sea transport network to be tested, and constructing a non-cascade port failure model and a cascade port failure model according to the river-sea transport network; Step S2: Use random attack and deliberate attack to attack the non-cascading port failure model and the cascading failure model respectively, and calculate the relative size performance parameter of the maximum connected subgraph and the global efficiency performance parameter of the network after the attack.

2. The reliability analysis method of river-sea transport network based on network survivability according to claim 1 is characterized in that: In the cascading failure model, when any node fails, the following steps are performed: Load distribution step: distribute the load of the failed node to the neighboring nodes connected to the failed node; Capacity determination step: After executing the load distribution step, determine one by one whether the load of each neighboring node is greater than its own capacity. If so, the node with a load greater than its own capacity is regarded as a failed node, and the load distribution step is executed.

3. The reliability analysis method of river-sea transport network based on network survivability according to claim 2 is characterized in that: In the load distribution step, the newly added load distributed to each neighboring node is: Where: △L m The additional load allocated to node m, P Fn For the distribution ratio, is the load of the failed node, O m is the capacity of node m, F n is the set of neighbor nodes of the failed node, L m Load before distribution to node m.

4. The reliability analysis method of river-sea transport network based on network survivability according to claim 1 is characterized in that: The process of the random attack includes: Step S2-A-1: Initialize the failure model of the attacked port and set the number of attack rounds; Step S2-A-2: randomly select a node as a node to be failed; Step S2-A-3: removing the node to be failed and the edges connected to the node to be failed from the port failure model; Step S2-A-4: Calculate the maximum connected subgraph relative size performance parameter and the network global efficiency performance parameter; Step S2-A-5: Determine whether the set number of attack rounds has been reached. If yes, end the random attack. Otherwise, execute step S2-A-6; Step S2-A-6: Determine whether there are any nodes that have not failed. If yes, return to step S2-A-2. Otherwise, end the random attack.

5. A reliability analysis method for river-sea transport network based on network survivability according to claim 1, characterized in that: The process of the deliberate attack includes: Step S2-B-1: Initialize the failure model of the attacked port and set the number of attack rounds; Step S2-B-2: Sort all nodes that have not failed according to the importance characteristic value of each node, and select the node with the largest importance characteristic value as the node to be failed; Step S2-B-3: remove the node to be failed and the edges connected to the node to be failed from the port failure model; Step S2-B-4: Calculate the relative size performance parameter of the largest connected subgraph and the global efficiency performance parameter of the network; Step S2-B-5: Determine whether the set number of attack rounds has been reached. If yes, end the random attack. Otherwise, execute step S2-B-6; Step S2-B-6: Determine whether there are still nodes that have not failed. If yes, return to step S2-A-2. Otherwise, end the random attack.

6. The reliability analysis method of river-sea transport network based on network survivability according to claim 5 is characterized in that: The importance characteristic value is determined by the node degree value, node betweenness and node PR value.

7. The reliability analysis method of river-sea transport network based on network survivability according to claim 1 is characterized in that: The relative size of the largest connected subgraph is the ratio of the number of nodes of the largest connected subgraph to the number of nodes of the port failure model.

8. The reliability analysis method of river-sea transport network based on network survivability according to claim 1 is characterized in that: The global efficiency of the network is specifically: Where: E global is the global efficiency of the network, N is the total number of nodes, and d(i,j) is the distance between node i and node j.

9. A method for enhancing the reliability of a river-sea transport network based on the analysis method of claim 2 or 3, characterized in that: include: In the load distribution step, multiple target nodes are selected from all non-neighboring nodes of the failed node, and load distribution is performed together with the neighboring nodes of the failed node.

10. The enhancement method according to claim 9, characterized in that: The target node includes selecting a node with the highest degree value and selecting a node with the highest PR value.

Images