New energy automobile supply chain network risk node identification method, system and device

Abstract

The invention provides a new energy automobile supply chain network risk node identification method, system and device, and the method comprises the steps: taking a new energy automobile supply chain enterprise as a node, constructing an adjacent matrix, and further constructing a new energy automobile supply chain network; calculating a network index of each node in the network; calculating an immune propagation evaluation index of each node in the network, wherein the immune propagation evaluation index is used for reflecting the influence intensity of infection propagation caused by a single node; simulating attacks by traversing nodes, and calculating survivability indexes of the nodes; and sorting the nodes based on the network index, the immune propagation evaluation index and the survivability index, and determining key nodes in the nodes according to a sorting result. According to the scheme, through comprehensive and multi-dimensional fusion risk propagation key node identification, potential hidden key nodes in the new energy automobile supply chain network can be found, and risk prevention of the new energy automobile supply chain network is optimized.

Description

Technical Field

[0001] The present invention relates to the fields of new energy vehicles, automobile manufacturing and supply chain data processing, and in particular to a method, system and device for identifying risk nodes in a new energy vehicle supply chain network. Background Art

[0002] Today, the automotive industry is moving towards electrification, networking, intelligence and sharing, which will lead to huge changes in the supply chain and supplier structure. The continuous development of the concept of new energy connected vehicles and the manufacturing industry has accelerated the iteration of new technologies. In the context of fierce market competition, automobile manufacturers are more urgent than ever in their demand for the stability of complex supply chain networks. In the development of the new energy vehicle industry, it is particularly important to establish a safe, efficient and sustainable supply chain.

[0003] The automotive supply chain network presents typical complex network characteristics due to the large number of entities involved. As the number of uncontrollable factors that parts supply nodes may suffer in the increasingly complex automotive supply chain network continues to increase, the automotive supply chain becomes more vulnerable. For complex automotive supply chain networks, network structure characteristic indicators are usually introduced to analyze the impact of network nodes on the entire network, provide the basis for the importance of nodes in the entire network, and then analyze the significance of risk control. From the perspective of complex automotive supply chain networks, studying the cascading failure mechanism and corresponding management measures that have a huge impact on the global network during the generation and propagation of risk factors in the automotive supply chain network has important practical significance for improving the risk resistance and stability of the automotive supply chain.

[0004] The complex automotive supply chain network has the characteristics of a long supply chain, many participating enterprises, and complex relationships between enterprises. Therefore, when a single node or node group in the network fails, the network is prone to cascading failures, resulting in reduced efficiency of the automotive supply chain and failure to complete the expected supply chain functions. In the existing technology, the risk identification and processing of the new energy vehicle supply chain usually focuses on the analysis of a single risk factor, which may result in the risk control not being optimal. Summary of the invention

[0005] In view of this, the embodiments of the present invention provide a method, system and device for identifying risk nodes in a new energy vehicle supply chain network, so as to at least partially solve the deficiencies in the prior art. Specifically, the present invention provides the following technical solutions:

[0006] In one aspect, the present invention provides a method for identifying risk nodes in a new energy vehicle supply chain network, the method comprising:

[0007] S1. Take new energy vehicle supply chain enterprises as nodes and construct an adjacency matrix;

[0008] S2. Construct a new energy vehicle supply chain network based on the adjacency matrix;

[0009] S3, calculating the network index of each node in the new energy vehicle supply chain network, where the network index is one or more centrality indexes;

[0010] S4. Calculate the immune propagation evaluation index of each node in the new energy vehicle supply chain network, where the immune propagation evaluation index is used to reflect the impact intensity of infection propagation caused by a single node;

[0011] S5. Perform a simulated attack by traversing the nodes and calculating the node's invulnerability index, where the node's invulnerability index is determined based on the ratio of the number of network nodes before the node fails to the number of valid network nodes after the node fails;

[0012] S6. Sort the nodes based on the network index, the immune propagation evaluation index and the indestructibility index, and determine the key nodes among the nodes according to the sorting results.

[0013] Preferably, in S3, the network index includes one of degree centrality, eigenvector centrality, betweenness centrality and closeness centrality, or any combination thereof.

[0014] Preferably, the degree centrality of a node is determined by the ratio of the degree of the node to the number of remaining network nodes excluding the node;

[0015] The eigenvector centrality of a node is determined at least in part by the sum of the eigenvector centralities of the node's neighbor nodes;

[0016] The betweenness centrality of a node is calculated as: Among them, gjk represents the number of shortest paths between node j and node k, gj k (i) represents the number of shortest paths between node j and node k that pass through node i, and n represents the total number of network nodes;

[0017] The proximity centrality of a node is determined at least in part based on the distance of the node from the rest of the nodes in the network, as represented by the shortest path length between the nodes.

[0018] Preferably, in S4, the immune propagation evaluation index of the node is determined as follows:

[0019] Take this node as the initial infected node, and divide the sum of the number of infected nodes and the number of recovered nodes after propagation by the total number of network nodes.

[0020] Preferably, S4 further includes constructing an immune propagation model, wherein the propagation mode of the immune propagation model is: Among them, α is the infection rate, β is the recovery rate, and t represents the time variable in the model, which is used to describe the evolution of the transmission process over time.

[0021] Preferably, the immune propagation evaluation index of the node takes the average value after multiple calculations as the final result.

[0022] Preferably, S5 further includes: the simulated attack is implemented by simulating the attack on each node in the new energy vehicle supply chain network through a cascading failure model, and the network load is rebalanced by distributing the load of the failed node to the node with large load capacity among the adjacent nodes;

[0023] The adjacent nodes include all nodes connected to the failed node.

[0024] Preferably, the load distribution method is: determining an initial load of a node, and calculating a load capacity of the node based on the initial load;

[0025] Calculate the load percentage assigned to each adjacent node;

[0026] Based on the load percentage and the load of the failed node, the load allocated to each adjacent node is calculated.

[0027] Preferably, if the accumulated load of the adjacent node exceeds the load capacity of the adjacent node after the adjacent node receives the allocated load, the adjacent node fails.

[0028] Preferably, in S1, the new energy vehicle industry enterprise is taken as the central node, and the enterprises cooperating therewith are taken as other nodes; the cooperative relationship between enterprises is taken as the edge; and an adjacency matrix is ​​constructed based on the edges and nodes.

[0029] Preferably, the degree centrality calculation method is: Among them, K(i) represents the degree of node i, and n represents the total number of nodes.

[0030] Preferably, the eigenvector centrality calculation method is: Among them, λ is a constant, e(i) is the set of neighbor nodes of node i, j represents the neighbor nodes of node i, E j Represents the eigenvector centrality of node j.

[0031] Preferably, the closeness centrality calculation method is: Among them, d(i, j) represents the distance between node i and node j in the network.

[0032] Preferably, the initial load is calculated as follows: Among them, q≥0, α≥0 are parameters used to control the initial load distribution. represents the betweenness of node i.

[0033] Preferably, the load capacity is calculated as follows: i (0) = (1 + β) L i (0), where β>0 represents the tolerance coefficient of the node.

[0034] Preferably, the load percentage is calculated as follows: Among them, C e represents the capacity of the adjacent node e, represents the sum of the capacities of all adjacent nodes j of node i, F i represents the set of adjacent nodes of node i.

[0035] On the other hand, the present invention also provides a new energy vehicle supply chain network risk node identification system, the system comprising:

[0036] The network building module is used to input the information of new energy vehicle supply chain enterprises, build an adjacency matrix with the new energy vehicle supply chain enterprises as nodes, and build a new energy vehicle supply chain network based on the adjacency matrix;

[0037] A network index calculation module, used to calculate the network index of each node in the new energy vehicle supply chain network, wherein the network index is one or more centrality indexes;

[0038] An immune propagation evaluation index module, used to calculate the immune propagation evaluation index of each node in the new energy vehicle supply chain network, wherein the immune propagation evaluation index is used to reflect the impact intensity of infection propagation caused by a single node;

[0039] An invulnerability index calculation module, used to simulate attacks by traversing nodes and calculate the invulnerability index of the node, wherein the invulnerability index of the node is determined based on the ratio of the number of network nodes before the node fails to the number of valid network nodes after the node fails;

[0040] The key node determination module sorts the nodes based on the network index, the immune propagation evaluation index and the invulnerability index, and determines the key nodes among the nodes according to the sorting results.

[0041] On the other hand, the present invention also provides a new energy vehicle supply chain network risk node identification device, which device includes at least a processor and a memory, and the processor calls the instructions stored in the memory to execute the new energy vehicle supply chain network risk node identification method as described above, or the new energy vehicle supply chain network risk node identification system as described above is set in the device.

[0042] Compared with the prior art, the present invention can discover potential hidden key risk control nodes in the new energy vehicle supply chain network through comprehensive and multi-dimensional integrated risk propagation key node identification, thereby optimizing the risk prevention of the new energy vehicle supply chain network. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] 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.

[0044] Figure 1 It is a schematic diagram of a flow chart of an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a system construction framework of an embodiment of the present invention;

[0046] Figure 3 A simplified architecture diagram of the SIR model embedded in the immune propagation model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. It should be clear that 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.

[0048] Those skilled in the art should know that the following specific embodiments or specific implementations are a series of optimized settings listed by the present invention to further explain the specific content of the invention, and these settings can be combined or used in association with each other, unless the present invention clearly states that some or a specific embodiment or implementation cannot be associated or used together with other embodiments or implementations. At the same time, the following specific embodiments or implementations are only used as the most optimized settings, and are not to be understood as limiting the scope of protection of the present invention.

[0049] The purpose of this invention is to propose an identification modeling technology for key risk nodes in the new energy vehicle supply chain network, which can identify nodes with greater potential risks in the new energy vehicle supply chain network and reduce supply chain risks.

[0050] Figure 1 This is a flowchart of the steps of a key risk node identification modeling technology for a new energy vehicle supply chain network of the present invention. Figure 2The detailed system architecture design method of this embodiment is demonstrated. Figure 3 The method of embedding the SIR model into the immune propagation model in this embodiment is illustrated. In one embodiment, the implementation method of this solution includes steps S1-S6, which are as follows:

[0051] S1: Data input and processing. Various types of enterprises cooperating with new energy vehicle industry enterprises are inputted, and the input constitutes an adjacency matrix. In this embodiment, when constructing a supply chain network, the network is constructed with new energy vehicle manufacturing enterprises as the core, and the nodes of the network are enterprise names. Based on the cooperative relationship between the enterprises, the edges between the nodes in the network are determined. Based on the relationship between the nodes and the edges, the composition of the adjacency matrix is ​​determined.

[0052] Furthermore, here we take a certain production enterprise in the new energy vehicle industry as the center to construct the corresponding adjacency matrix as an example. Suppose the core enterprise is "Enterprise A", and the enterprises cooperating with it include "Enterprise B", "Enterprise C" and "Enterprise D". The cooperative relationship between these enterprises constitutes the adjacency matrix. Define the node set: {Enterprise A, Enterprise B, Enterprise C, Enterprise D}, assuming that Enterprise A has a cooperative relationship with Enterprise B and Enterprise C, but not with Enterprise D; Enterprise B has a cooperative relationship with Enterprise C, but not with Enterprise D; Enterprise C has a cooperative relationship with Enterprise D. From this, a 4×4 matrix can be determined, with rows and columns corresponding to the enterprises in the node set, then the adjacency matrix can be expressed as:

[0053]

[0054] The indexes of the rows and columns are respectively Enterprise A, B, C, and D. For example, "1" in the second column of the first row indicates that there is a cooperative relationship between Enterprise A and Enterprise B, while "0" in the fourth column of the first row indicates that there is no cooperative relationship between Enterprise A and Enterprise D. The meanings of the remaining elements are similar. It should be noted that only 1 is used here to represent the cooperation between enterprises. The values ​​of these elements can be adjusted based on the needs of the user. For example, different element values ​​can be set according to the importance or closeness of the cooperation between enterprises, or with reference to more factors (such as the overlap of business between enterprises, distance, etc.). The element value in this example is 1, which is only used as an example for the convenience of explanation.

[0055] S2: Based on the adjacency matrix, construct a new energy vehicle supply chain network;

[0056] S3: Analyze the network index of the new energy vehicle supply chain network. The network index can be set as a centrality index, for example, calculate the centrality index of each node in the network, and sort the nodes from high to low according to the centrality index value according to the centrality index value;

[0057] In this embodiment, the network indicators include four types of centrality indicators: degree centrality, eigenvector centrality, betweenness centrality and closeness centrality.

[0058] In a further preferred embodiment, the above four centrality indicators can be set as follows:

[0059] The degree centrality of the node is Where K(i) represents the degree of node i, and n represents the total number of network nodes. K(i) can be determined by the number of edges associated with the node.

[0060] The eigenvector centrality is Where λ is a constant, e(i) is the set of neighbor nodes of node i, j represents the neighbor nodes of node i, E j represents the eigenvector centrality of node j. In the calculation of eigenvector centrality, the initial eigenvector centrality of all nodes is usually set to the same value, and a common choice is 1. As the iterations proceed, these values ​​will be gradually updated.

[0061] The betweenness centrality is Where gjk represents the number of shortest paths between nodes j and k, g j k(i) represents the number of shortest paths between node j and node k that pass through node i.

[0062] The closeness centrality is Where d(i, j) represents the distance between node i and node j in the network, and the distance is represented by the shortest path length between the nodes.

[0063] The four centrality indicators are combined to obtain the comprehensive network indicator of a single node. For example, the weighted average method can be used to set a weight for each centrality indicator and calculate the weighted average according to its importance: C total (i) = ω1 × C D (i)+ω2×E i +ω3×C B (i)+ω4×C C (i). Among them, ω1, ω2, ω3, ω4 are the weights of each centrality index, and satisfy ω1+ω2+ω3+ω4=1. Finally, the sorting results of the four centrality indexes can be integrated by sorting according to the weighted average. It should be noted that the weighted average method is only used as a preferred implementation method. Those skilled in the art can appropriately adjust the centrality index fusion calculation method based on the detection requirements to facilitate the sorting of nodes.

[0064] S4: In addition, in this embodiment, the supply chain network is also analyzed by immune propagation model, which analyzes each node in the new energy vehicle supply chain network and obtains the propagation model evaluation index of each node. In a more preferred real-time method, the average value of the propagation model evaluation index calculated by multiple experimental results can be taken, and the average value of the propagation model evaluation index can be sorted from high to low, thereby realizing the sorting of nodes. Figure 3 As shown, in this embodiment, the overall architecture of the immune propagation model is divided into several parts: a node set, which represents each enterprise or node in the supply chain network; an edge set, which represents the cooperative relationship between nodes to form an adjacency matrix; a SIR model, which is used to simulate the propagation process of information or infection in the network; and a propagation model evaluation index, which measures the propagation influence of each node.

[0065] In this embodiment, the above-mentioned immune propagation model includes the SIR model, which analyzes each node in the network as an initial infected node, and uses the value of the number of infected nodes plus the number of recovered nodes divided by the total number of nodes as an evaluation indicator for measuring the propagation caused by each node, that is, the propagation model evaluation indicator.

[0066] Further preferably, the SIR model defines S as a susceptible node, I as an infected node, and R as an immune node that recovers after infection. The model propagation follows Where α is the infection rate, β is the recovery rate, and t represents the time variable in the model, which is used to describe the evolution of the propagation process over time. The propagation caused by the node is Among them I i is the number of nodes infected after propagation with node i as the initial node, R i is the number of nodes recovered after propagation with node i as the initial node, and n is the total number of nodes. In a preferred embodiment, we can set α = 0.3, β = 0.1, the number of experimental steps for each node is 10, the propagation caused by each node is experimented 100 times, and the average value is taken as the final result. The simplified architecture diagram of the SI R model embedded in the immune propagation model is shown in Figure 3 shown.

[0067] S5: Construct a cascading failure model for the new energy vehicle supply chain network, and perform node analysis based on the cascading failure model. In this embodiment, the cascading failure model attacks each node and compares the proportion of valid nodes that still exist in the network after attacking different nodes. The failed node is used to simulate the attack behavior of each node; the adjacent nodes include all network nodes connected to the failed node; the load distribution includes the load capacity of the node and the node distributes the load to the adjacent nodes in a certain proportion when the node fails.

[0068] The cascading failure model achieves the goal of rebalancing the network load by distributing the load of the failed node to the node with large load capacity among its adjacent nodes.

[0069] In a preferred embodiment, the initial load of the node is Where q ≥ 0, α ≥ 0 are parameters used to control the initial load distribution, is the betweenness of node i. More preferably, we can set q=0.5 and α=0.6 to select nodes that still have a greater impact on the network when they fail under the condition of more stable network anti-destruction performance. It should be noted here that the initial load of the node can also be determined in different ways, such as by using a random number method, by calculating the node's degree and weight coefficient, etc. The method given in this embodiment is only described as a preferred method, and should not be understood as the only way to implement the function here.

[0070] More preferably, the node load capacity C i (0) Determined by its initial load C i (0) = (1 + β) L i (0), where β>0 represents the tolerance coefficient of the node. The tolerance coefficient can be set based on actual detection requirements or through experiments. For example, in this embodiment, the tolerance coefficient is set to β=0.5 as a constant parameter.

[0071] When performing load distribution, preferably, the distribution strategy can be set as follows: when node i fails, its load is distributed to the node with larger capacity among the adjacent nodes. At this time, the load percentage allocated to each adjacent node is The load assigned to the neighboring node e of the failed node is ΔL ei =∏ e L i , where L i represents the load of failed node i, C e represents the capacity of the adjacent node e, represents the sum of the capacities of all adjacent nodes of node i, F i represents the set of neighboring nodes of node i. If the accumulated load of neighboring node e exceeds its capacity after receiving the allocated load, that is, L e (t) = L e (t-1)+ΔL ei ≥C e, it will cause the adjacent node to fail. At this time, the failed adjacent node will continue to distribute the load again based on the load distribution strategy, but the redistribution at this time will no longer consider the failed node at the previous level. After redistribution, mark the failed node and continue to observe the load of the adjacent node to ensure that there is no failure again. It should be noted here that the load percentage can also be calculated in different ways, such as calculating it based on the load capacity of the node, or using the initial load value to calculate it, or calculating it based on a random weight method, etc. The method given in this embodiment is only described as a preferred method, and should not be understood as the only way to achieve the function here.

[0072] In another specific embodiment, the cascading failure model also includes a node indestructibility analysis; the node indestructibility is analyzed based on the ratio of the number of network nodes before the node failure and the number of valid nodes after the failure, and the higher the ratio, the higher the indestructibility.

[0073] In this implementation, the network's invulnerability index when attacking node i is Where N′ i represents the valid nodes that still exist in the network after attacking node i, and n is the total number of nodes in the network. In the cascading failure model, the invulnerability index S finally calculated for each node is usually used. i As the final calculation indicator of the influence of each node. i The smaller the value, the greater the impact on the network when it fails, and the higher its importance.

[0074] It should be noted here that the above steps S3, S4, and S5 in this embodiment can be connected in series or in parallel, and the order of the above three steps can also be adjusted, that is, the indicators of the three models (i.e., the centrality model SIR model and the cascading failure model described in this embodiment) can be calculated simultaneously, or the indicators of the three models can be calculated in parallel, or the indicators of the three models can be calculated in any order. The numbers S3, S4, and S5 here are only for the convenience of description, and should not be understood as a necessary fixed order between the above three steps. Therefore, any change in the order of the above three steps should be regarded as falling within the scope of protection of the present invention.

[0075] S6: After the nodes based on the above three models are sorted, the sorting of the above multiple models is integrated to obtain the key nodes.

[0076] Preferably, the final ranking obtained by integrating multiple models includes the ranking results of the network obtained by the centrality model, the SIR model, and the cascading failure model to obtain the final overall ranking result, and the key nodes include the nodes with the highest ranking in the overall ranking result. Further, in this embodiment, the overall ranking result can be determined in the following manner: according to the importance of each model, a weight is assigned to each model, for example, the centrality model weight ω ′ 1. SIR model weight ω ′ 2. Cascading failure model weight ω ′ 3. For each node i, the comprehensive score is calculated according to the weights assigned by each model, and all nodes are sorted based on this score to obtain the final sorting result.

[0077] In another embodiment, the present solution can also be implemented in a system manner, again in combination with Figure 2 , the system can be set up as follows:

[0078] The network building module is used to input the information of new energy vehicle supply chain enterprises, build an adjacency matrix with the new energy vehicle supply chain enterprises as nodes, and build a new energy vehicle supply chain network based on the adjacency matrix;

[0079] A network index calculation module, used to calculate the network index of each node in the new energy vehicle supply chain network, wherein the network index is one or more centrality indexes;

[0080] An immune propagation evaluation index module, used to calculate the immune propagation evaluation index of each node in the new energy vehicle supply chain network, wherein the immune propagation evaluation index is used to reflect the impact intensity of infection propagation caused by a single node;

[0081] An invulnerability index calculation module, used to simulate attacks by traversing nodes and calculate the invulnerability index of the node, wherein the invulnerability index of the node is determined based on the ratio of the number of network nodes before the node fails to the number of valid network nodes after the node fails;

[0082] The key node determination module sorts the nodes based on the network index, the immune propagation evaluation index and the invulnerability index, and determines the key nodes among the nodes according to the sorting results.

[0083] In a more specific implementation, one or more units may be provided in the network index calculation module to calculate the network centrality index. Figure 2 As shown, one of degree centrality unit, eigenvector centrality unit, betweenness centrality unit, closeness centrality unit or any combination thereof can be set. In this way, a single or multiple centrality indexes can be calculated, and then a comprehensive ranking is performed to obtain the node-based network index in the module and the ranking result.

[0084] In the immune propagation evaluation index module, after calculating the immune propagation evaluation index of each node, the nodes are sorted based on the index and the sorting result of this module is obtained.

[0085] In the invulnerability index calculation module, after calculating the invulnerability index of each node, the nodes are sorted based on the index to obtain the sorting result of this module.

[0086] It should be noted here that the calculation method or model of each module can be implemented using the specific algorithms and methods listed in the above embodiments.

[0087] In another embodiment, the present solution can be implemented by means of a device, which may include a corresponding module for performing each or several steps in each of the above-mentioned embodiments. Therefore, each step or several steps of each of the above-mentioned embodiments may be performed by a corresponding module, and the device may include one or more of these modules. The module may be one or more hardware modules specifically configured to perform the corresponding steps, or implemented by a processor configured to perform the corresponding steps, or stored in a computer-readable medium for implementation by a processor, or implemented by some combination.

[0088] Any process or method description in the flowchart or otherwise described herein can be understood as a module, fragment or portion of a code representing one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiment of the present solution includes other implementations, in which the functions may not be performed in the order shown or discussed, including performing the functions in a substantially simultaneous manner or in a reverse order according to the functions involved, which should be understood by a person skilled in the art of the present solution. The processor performs the various methods and processes described above. For example, the method implementation in the present solution can be implemented as a software program, which is tangibly contained in a machine-readable medium, such as a memory. In some embodiments, part or all of the software program can be loaded and / or installed via a memory and / or a communication interface. When the software program is loaded into the memory and executed by the processor, one or more steps in the method described above can be performed. Alternatively, in other embodiments, the processor can be configured to perform one of the above methods in any other appropriate manner (e.g., by means of firmware).

[0089] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for identifying risk nodes in the new energy vehicle supply chain network, characterized in that: The method comprises: S1. Take new energy vehicle supply chain enterprises as nodes and construct an adjacency matrix; S2. Construct a new energy vehicle supply chain network based on the adjacency matrix; S3, calculating the network index of each node in the new energy vehicle supply chain network, where the network index is one or more centrality indexes; S4. Calculate the immune propagation evaluation index of each node in the new energy vehicle supply chain network, where the immune propagation evaluation index is used to reflect the impact intensity of infection propagation caused by a single node; S5. Perform a simulated attack by traversing the nodes and calculating the node's invulnerability index, where the node's invulnerability index is determined based on the ratio of the number of network nodes before the node fails to the number of valid network nodes after the node fails; S6. Sort the nodes based on the network index, the immune propagation evaluation index and the indestructibility index, and determine the key nodes among the nodes according to the sorting results.

2. The method according to claim 1, characterized in that In S3, the network index includes one of degree centrality, eigenvector centrality, betweenness centrality and closeness centrality, or any combination thereof.

3. The method according to claim 2, characterized in that The degree centrality of a node is determined by the ratio of the degree of the node to the number of remaining network nodes except the node; The eigenvector centrality of a node is determined at least in part by the sum of the eigenvector centralities of the node's neighbor nodes; The betweenness centrality of a node is calculated as: Among them, g jk represents the number of shortest paths between node j and node k, g jk (i) represents the number of shortest paths between node j and node k that pass through node i, and n represents the total number of network nodes; The proximity centrality of a node is determined at least in part based on the distance of the node from the rest of the nodes in the network, as represented by the shortest path length between the nodes.

4. The method according to claim 1, characterized in that: In S4, the immune propagation evaluation index of the node is determined as follows: Take this node as the initial infected node, and divide the sum of the number of infected nodes and the number of recovered nodes after propagation by the total number of network nodes.

5. The method according to claim 1, characterized in that The S5 further includes: the simulated attack is implemented by simulating the attack on each node in the new energy vehicle supply chain network through a cascading failure model, and the network load is rebalanced by distributing the load of the failed node to the node with large load capacity among the adjacent nodes; The adjacent nodes include all nodes connected to the failed node.

6. The method according to claim 5, characterized in that The load distribution method is: determining the initial load of the node, and calculating the load capacity of the node based on the initial load; Calculate the load percentage assigned to each adjacent node; Based on the load percentage and the load of the failed node, the load allocated to each adjacent node is calculated.

7. The method according to claim 6, characterized in that If the accumulated load of an adjacent node exceeds the load capacity of the adjacent node after the adjacent node receives the allocated load, the adjacent node fails.

8. The method according to claim 1, characterized in that: In S1, the new energy vehicle industry enterprise is taken as the central node, and the enterprises cooperating with it are taken as other nodes; the cooperative relationship between enterprises is taken as the edge; and the adjacency matrix is ​​constructed based on the edges and nodes.

9. New energy vehicle supply chain network risk node identification system, characterized by: The system comprises: The network building module is used to input the information of new energy vehicle supply chain enterprises, build an adjacency matrix with the new energy vehicle supply chain enterprises as nodes, and build a new energy vehicle supply chain network based on the adjacency matrix; A network index calculation module, used to calculate the network index of each node in the new energy vehicle supply chain network, wherein the network index is one or more centrality indexes; An immune propagation evaluation index module, used to calculate the immune propagation evaluation index of each node in the new energy vehicle supply chain network, wherein the immune propagation evaluation index is used to reflect the impact intensity of infection propagation caused by a single node; An invulnerability index calculation module, used to simulate attacks by traversing nodes and calculate the invulnerability index of the node, wherein the invulnerability index of the node is determined based on the ratio of the number of network nodes before the node fails to the number of valid network nodes after the node fails; The key node determination module sorts the nodes based on the network index, the immune propagation evaluation index and the invulnerability index, and determines the key nodes among the nodes according to the sorting results.

10. New energy vehicle supply chain network risk node identification device, characterized by: The device includes at least a processor and a memory, and the processor calls instructions stored in the memory to execute the new energy vehicle supply chain network risk node identification method as described in any one of claims 1-8.

Images