Cascading failure simulation method, device and storage medium based on weighted dependency network

By using a weighted dependent network method in cascade failure simulation, setting the initial load and load redistribution, the problem of poor accuracy of cascade failure simulation in the prior art is solved, and the simulation effect is significantly improved.

CN115016939BActive Publication Date: 2024-11-12NAT UNIV OF DEFENSE TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210653783.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-06-10
Publication Date
2024-11-12
Estimated Expiration
2042-06-10

AI Technical Summary

Technical Problem

Due to poor robustness, the existing cascade failure simulation methods have low accuracy in cascade failure simulation, especially in multi-layer weighted networks, which affects the simulation effect.

Method used

The cascade failure simulation method based on weighted dependent network is adopted. The initial load is set by obtaining the node strength and neighbor node strength of the multi-layer network, and load redistribution is performed according to the nonlinear relationship between node capacity and load, and the cascade failure process is simulated. , improve the robustness of the model.

Benefits of technology

The robustness of the cascade failure model is improved, thereby improving the accuracy and effectiveness of the cascade failure simulation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115016939B_ABST
    Figure CN115016939B_ABST
Patent Text Reader

Abstract

The present invention provides a cascading failure simulation method, device and storage medium based on a weighted dependent network, wherein the cascading failure simulation method includes: obtaining a cascading failure model based on a weighted dependent network, the cascading failure model includes multiple network layers, each network layer has multiple nodes; setting the initial load of the node according to the node strength of the node and the node strength of its neighboring nodes, and setting the node capacity of the node according to the nonlinear relationship between the node capacity of the node and the initial load; using the cascading failure model to simulate the cascading failure process, the cascading failure process includes inter-layer cascading failure and intra-layer cascading failure, and in the simulation process, load redistribution is performed according to the remaining capacity and importance of the neighboring nodes of the failed node. The technical solution of the present invention can improve the robustness of the cascading failure model, thereby improving the simulation effect of cascading failure.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of cascading failure simulation, and in particular relates to a cascading failure simulation method based on a weighted dependency network, a computer device and a computer storage medium. Background Art

[0002] With the development of network science, complex network theory provides an effective research approach for understanding the properties, component structure and functional structure of complex systems. After continuous development and improvement, complex network theory has been widely used in complex systems such as biological systems, power systems, and social systems. Although the research framework of complex networks plays an important role in promoting related research, simply abstracting complex systems into a single network often ignores the multi-type relationships that may exist between multiple types of nodes, while multi-layer networks break through the limitations of homogeneity of nodes and edges in single-layer networks. In addition, complex systems characterized by the interaction relationship of several networks are ubiquitous, such as different social relationships between individuals in social relationship networks, different means of transportation between cities in transportation networks, different collaborative relationships between equipment in combat systems, and interdependence between power networks and communication networks.

[0003] In reality, there are many complex systems with interdependent relationships. Such huge networks are prone to failure. For example, a network virus attack can cause the Internet to fail, leading to the collapse of a series of infrastructure systems such as communication networks, power networks, transportation networks, etc. Due to the mutually coupled structural and functional relationships in complex networks, when a failure occurs, it may trigger a series of chain failure reactions, and even expand to the entire network, thereby producing an effect similar to an "avalanche".

[0004] The cascading failure model is a model used to simulate the above network and perform fault simulation. The existing classic models commonly used in cascading failure research mainly include the sandpile model, CASCADE model, OPA model and load-capacity model. In the cascading failure model, it is proposed that the main factors affecting the cascading effect are initial load, capacity and load redistribution strategy. Many scholars also mainly propose improvements and innovations around these three parts when conducting research related to cascading failure. Despite this, the existing research on cascading failure in complex networks pays less attention to multi-layer weighted networks. On the one hand, the design of the cascading failure model rarely considers the impact of network weights on node capacity and load redistribution strategy. On the other hand, the cascading failure process in the multi-layer weighted network is rarely considered, resulting in poor robustness of the cascading failure model, which in turn makes the simulation of cascading failures poor.

[0005] In summary, it can be seen that the cascading failure simulation method in the prior art has the problem of poor accuracy in cascading failure simulation due to poor robustness of the cascading failure model. Summary of the invention

[0006] The object of the present invention is to provide a cascading failure simulation method, device and storage medium based on a weighted dependency network, which is used to at least solve the above-mentioned problem of poor accuracy of cascading failure simulation due to poor robustness of the cascading failure model during cascading failure simulation.

[0007] In order to at least solve the above-mentioned technical problems, in a first aspect, the present invention provides a cascading failure simulation method based on a weighted dependent network, comprising: obtaining a cascading failure model based on a weighted dependent network, the cascading failure model comprising multiple network layers, each network layer having multiple nodes; setting the initial load of the node according to the node strength of the node and the node strength of its neighboring nodes, and setting the node capacity of the node according to the nonlinear relationship between the node capacity of the node and the initial load; using the cascading failure model to simulate the cascading failure process, the cascading failure process comprising inter-layer cascading failure and intra-layer cascading failure, and in the simulation process, performing load redistribution according to the remaining capacity and importance of the neighboring nodes of the failed node.

[0008] According to an embodiment of the present invention, setting the initial load of a node according to the node strength of the node and the node strength of its neighboring nodes includes: calculating the initial load of each node using the following formula:

[0009]

[0010] in is the initial load of the ith node, s i is the strength of the i-th node in the cascading failure model, Γ i is the set of neighbor nodes of the ith node, s′ m For the set Γ i The node strength of the mth node in im is the i-th node and the set Γ i is the edge weight between the mth nodes in , and α is the parameter for adjusting the initial load intensity.

[0011] According to an embodiment of the present invention, setting the node capacity of the node according to the nonlinear relationship between the node capacity and the initial load of the node includes: calculating the node capacity of each node using the following formula:

[0012]

[0013] Among them C i is the capacity of the i-th node in the cascading failure model, β and θ are the first capacity parameter and the second capacity parameter, respectively. is the initial load of the i-th node in the cascading failure model.

[0014] According to an embodiment of the present invention, the load redistribution includes: performing load redistribution using the following formula:

[0015]

[0016] Where L j (t+1) and L j (t) are the actual loads of the jth node at time step t+1 and time step t, respectively. i is the capacity of the i-th node, ΔL ji is the additional load of the j-th node, which is obtained according to the remaining capacity and importance of the j-th node, and the j-th node is the neighbor node of the ith node.

[0017] According to an embodiment of the present invention, when obtaining the additional load of the j-th node according to the remaining capacity and importance thereof, the following calculation formula is used:

[0018]

[0019] Where L j (t) is the actual load of the ith node at time step t, C j is the capacity of the jth node in the cascading failure model, w j is the importance of the jth node.

[0020] According to one embodiment of the present invention, the simulation of intra-layer cascading failure includes: obtaining neighboring nodes of the failed node that participate in redistribution in its network layer; calculating the remaining load and node importance of the neighboring nodes of the failed node; calculating the load allocated by the failed node to its neighboring nodes, and redistributing the load according to the calculation result; updating the real-time load of all nodes participating in load redistribution, and determining whether the real-time load of the node exceeds its own capacity, and if so, treating the node as a failed node.

[0021] According to one embodiment of the present invention, the cascading failure model is provided with an adjacency matrix, and the simulation of inter-layer cascading failure includes: obtaining a failed node, and obtaining nodes that depend on the failed node according to the adjacency matrix; blocking the dependency relationship between the failed node and the nodes that depend on the failed node, and treating the nodes that depend on the failed node as failed nodes.

[0022] According to an embodiment of the present invention, the cascading failure model is further provided with a failure node list, and the cascading failure simulation process includes updating the failure node list, and stopping the cascading failure simulation when there is no failure node in the failure node list.

[0023] In a second aspect, the present invention further provides a computer device, comprising a processor and a memory, wherein the memory stores computer program instructions for execution on the processor, and when the processor executes the computer program instructions, the cascading failure simulation method based on a weighted dependency network described in any one of the above embodiments is implemented.

[0024] In a third aspect, the present invention further provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the cascading failure simulation method based on a weighted dependency network described in any one of the above embodiments.

[0025] The technical solution provided by the present invention first obtains a cascading failure model based on a weighted dependent network, sets the initial load of the node according to the node strength of the node and the node strength of its neighboring nodes, sets the node capacity of the node according to the nonlinear relationship between the node capacity of the node and the initial load, and in the process of simulating cascading failures, redistributes the load according to the remaining capacity and importance of the neighboring nodes of the failed node. Therefore, the technical solution of the present invention can improve the robustness of the cascading failure model, thereby improving the simulation effect of cascading failures. In summary, it can be seen that the technical solution of the present invention can improve the accuracy of cascading failure simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0027] Figure 1 is a flow chart of a cascading failure simulation method based on a weighted dependency network according to an embodiment of the present invention;

[0028] Figure 2 is a schematic diagram of a cascading failure model according to an embodiment of the present invention;

[0029] Figure 3 A schematic diagram of a simulation of intra-layer cascading failure according to an embodiment of the present invention;

[0030] Figure 4 A schematic diagram of an inter-layer cascading failure simulation according to an embodiment of the present invention;

[0031] Figure 5 A schematic diagram of a further inter-layer cascading failure simulation according to an embodiment of the present invention;

[0032] Figure 6 A schematic diagram of the relationship between model parameters and the robustness of a weighted dependency network according to an embodiment of the present invention;

[0033] Figure 7 It is a schematic diagram of the effect of a BA-BA-BA network under intentional attack and under random attack according to an embodiment of the present invention;

[0034] Figure 8 It is a schematic diagram of the effect of a WS-WS-WS network under a deliberate attack and a random attack according to an embodiment of the present invention;

[0035] Fig. 9 Schematic diagram of the effect of an ER-ER-ER network under intentional attack and under random attack according to an embodiment of the present invention;

[0036] Fig.10 It is a schematic diagram of the effect of a BA-WS-ER network under a deliberate attack and a random attack according to an embodiment of the present invention;

[0037] Fig.11 A schematic diagram of network robustness under the RCNI-LC model according to an embodiment of the present invention;

[0038] Fig.12 A schematic diagram of the structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Those skilled in the art should know that the embodiments described below are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0040] See also Figure 1 , Figure 1 The flowchart of the cascading failure simulation method based on the weighted dependency network of the present invention is shown. The method simulates the cascading failure process by establishing a cascading failure model based on the weighted dependency network to improve the reliability of the cascading failure simulation. Figure 1 The process shown is a detailed introduction to the cascading failure simulation method based on weighted dependency network of the present invention.

[0041] like Figure 1 As shown, the cascading failure simulation method based on a weighted dependency network provided by the present invention includes:

[0042] In step S1, a cascading failure model based on a weighted dependency network is obtained. In this embodiment, the cascading failure model has multiple network layers, each network layer has multiple nodes, nodes with relationships in each network layer are connected by corresponding edges, and nodes with dependencies between each network layer are connected by corresponding edges, that is, in the cascading failure model, the edges within the network layer represent the relationship between the corresponding nodes, and the edges between the network layers represent the mutual dependence between the corresponding nodes.

[0043] Take one implementation as an example, Figure 2 As shown, the weighted dependency network has three network layers, Layer1, Layer2 and Layer3, with 5 nodes, 5 nodes and 4 nodes respectively, where ω1 to ω 15 is the edge between the corresponding nodes. In this embodiment, the weighted dependency network used by the cascading failure model is assumed to be M = (G, C), and the weighted dependency network always has M network layers, for example Figure 2 The value of M is 3; the above G={G a ; a∈{1,...,M}}, G is the network G of each layer in M a =(X a ,E a ), X a represents the node set in the a-th network layer, E a Represents the edge in the a-th network layer, and its corresponding adjacency matrix is ​​A a ; C=E ab (b∈{1,...,M},a≠b) represents the edge between the corresponding nodes of the a-th layer network and the b-th layer network, and the corresponding adjacency matrix is ​​A ab . Let the projection network of the weighted dependency network M be proj(M)=(X M ,E M ),in:

[0044]

[0045]

[0046] represents the set of nodes in the a-th network layer, N a represents the number of nodes in the a-th network layer, is the mth node in the ath network layer. Let the super-adjacency matrix of the weighted dependency network M be but It can be expressed as:

[0047]

[0048]

[0049]

[0050] in represents the weight of the edge between the i-th node and the j-th node in the a-th network layer, Represents the dependency between the i-th node in the a-th network layer and the j-th node in the b-th network layer. Indicates the mutual dependence between the corresponding two nodes. Indicates that the corresponding nodes are independent of each other.

[0051] In step S2, the initial load of each node in the cascading failure model is set according to the strength of the node and the neighbor degree of its neighbor nodes, and the capacity of each node in the cascading failure model is set according to the nonlinear relationship between the node capacity and the initial load. In order to improve the robustness of the cascading failure model, in this step S2, the initial load of each node in the cascading failure model is set according to the node strength of the node and the node strength of its neighbor nodes, and the node capacity of each node in the cascading failure model is set according to the nonlinear relationship between the node capacity and the node initial load.

[0052] In step S3, according to the above cascading failure model, cascading failure is simulated. In this step S3, the simulation of cascading failure includes the failure simulation of inter-layer nodes and the failure simulation of intra-layer nodes, that is, the failure simulation of nodes in each network layer and the failure simulation of nodes between each network layer, and in the simulation process, load redistribution is performed according to the remaining capacity and importance of the neighboring nodes of the failed node to improve the robustness of the cascading failure model.

[0053] In summary, the technical solution provided by the present invention sets the initial load of the node according to the node strength of the node and the node strength of its neighboring nodes, sets the node capacity of the node according to the nonlinear relationship between the node capacity of the node and the initial load, and in the process of simulating cascading failures, redistributes the load according to the remaining capacity and importance of the neighboring nodes of the failed node. Therefore, compared with the prior art, the cascading failure model in the technical solution of the present invention is more robust, thereby improving the simulation effect of cascading failures.

[0054] The above is an overall introduction to the technical solution of the present application. The following is a detailed description of the method for setting the initial load of the node in the cascading failure model in combination with a specific application scenario.

[0055] In one embodiment, when setting the initial load of each node in the cascading failure model, first obtain the neighbor nodes of each node, then obtain the node strength of each node and the node strength of its neighbor nodes, and finally obtain the initial load of each node according to the node strength of the node, the node strength of its neighbor nodes and the edge weights between the node and its neighbor nodes. Taking the i-th node in the cascading failure model as an example, let its neighbor node set be Γ i , whose node strength is s i , the mth node is the neighbor node of the ith node, and its node strength is s m , and the weight of the edge between the i-th node is w im , then the initial load of the i-th node in the cascading failure model is It is obtained by the following calculation formula:

[0056]

[0057] The node strength s of the i-th node i for α is a parameter for adjusting the initial load intensity and is a positive number.

[0058] The setting method of this embodiment not only takes into account the node strength and the strength of neighboring nodes when obtaining the initial load of a node, but also takes into account the influence of the edge weights between nodes, so the obtained node initial load information is more comprehensive.

[0059] The above article has introduced in detail the method of calculating the initial load of the node in the cascading failure model. The following article will introduce in detail the method of setting the node capacity in the cascading failure model in combination with specific application scenarios.

[0060] In one embodiment, when setting the node capacity of a node in a cascading failure model, the calculation formula used is:

[0061]

[0062] Among them C i is the capacity of the i-th node in the cascading failure model, β and θ are the first capacity parameter and the second capacity parameter, respectively, and their values ​​are both greater than 0. It can be seen from the above formula (7) that the node capacity obtained by the setting method of this embodiment is in a nonlinear relationship with the initial load, and as the initial load increases, the node capacity gradually tends to the initial load, that is, the remaining capacity gradually decreases. Therefore, the node capacity obtained by the setting method of this embodiment can more accurately simulate the actual state of the node.

[0063] The above text explains in detail how to set the node capacity in the cascading failure model. The following text introduces the node load redistribution strategy in detail in combination with specific application scenarios. It can be understood that the introduction of the load redistribution strategy below is exemplary rather than restrictive. The introduction of load redistribution above is also applicable to the description of load redistribution below.

[0064] In one embodiment, when a node fails, its load will be distributed to its neighboring nodes during the cascading failure process. When the load of the neighboring nodes of a node exceeds its node capacity, the neighboring nodes will trigger the next round of cascading failure. In this embodiment, the i-th node in the cascading failure model is an example of a failed node, and the j-th node is its neighboring node. After t rounds of cascading failures, the actual load of the j-th node is L j (t), then after t+1 rounds of cascading failures, the actual load of the jth node is:

[0065]

[0066] Where ΔL ji is the additional load that the jth node should receive from the ith node, C i is the capacity of the i-th node.

[0067] In the simulation process of cascading failure, the larger the remaining capacity of the neighboring node of the failed node, the more load the failed node allocates to it. When the neighboring node of the failed node is more important, in order to avoid the important node being destroyed and causing greater losses, the neighboring node needs to allocate less load. Based on the above theory, in one embodiment, when obtaining the additional load allocated by the j-th node to the i-th node, first calculate the frequency division ratio Π of the load allocated by the i-th node to the j-th node according to the real-time remaining capacity and unit weight of the j-th node. j , the formula used in the calculation is:

[0068]

[0069] Where L n (t) is the real-time load of the nth node after t rounds of cascading, and the nth node is the neighbor node of the ith node, C n is the capacity of the nth node, w j is the importance of the jth node and is calculated by the following formula:

[0070]

[0071] Then the load of the i-th node is divided into the above distribution ratio Π j Combined with the above, the additional load that the j-th node should receive from the i-th node is obtained by the following calculation formula:

[0072]

[0073] The cascading failure model is introduced in detail above. The following is a detailed introduction to the intra-layer cascading failure simulation method and the inter-layer cascading failure simulation method based on specific application scenarios.

[0074] In one embodiment, the method for simulating intra-layer cascading failures includes: first obtaining the neighboring nodes of each failed node that participate in the redistribution in its network layer, then obtaining the remaining load and node importance of the neighboring nodes of each failed node, then calculating the load allocated to each failed node by its neighboring nodes, and redistributing the load according to the calculation result, and finally updating the real-time load of all nodes participating in the load redistribution, and determining whether the real-time load of each node exceeds its own capacity. If it exceeds, the node is regarded as a failed node; after updating the real-time load of the node, the super-adjacency matrix of the cascading failure model is updated. Figure 3 As shown, assuming that the second layer network node in the cascading failure model If attacked, all edges connected to it in layer 2 will fail. Assuming that the load in the layer is redistributed, the nodes in the layer 2 network layer Because the cascading effect within the layer fails, the node As a failed node.

[0075] In one embodiment, the method for simulating inter-layer cascading failures includes: first obtaining dependent nodes of the failed node according to the adjacency matrix, then blocking the dependency relationship between the failed node and its dependent nodes, and treating the nodes that depend on the failed node as failed nodes. Due to the node dependencies between layers, nodes in the first network layer node Nodes in the Layer 3 network layer The corresponding failure leads to the failure of the first network layer and the node node The connected edges fail and the total nodes at the layer 3 network layer The connected edges fail, such as Figure 4 As shown; further, all nodes in the first network layer fail. Due to the cascading failure relationship between layers, nodes in the second network layer fail. and nodes Also failed, the third layer network layer total node and nodes Also invalid, such as Figure 5 shown.

[0076] In one embodiment, the cascading failure model is also provided with a corresponding failure node list. During the cascading failure simulation process, the failure node list is updated until there are no failure nodes in the failure list. For example, when performing intra-layer cascading failure simulation, if the load redistribution of a failed node is completed, the failed node can be deleted from the failure node list, and if the neighbor node of the failed node becomes a failed node, the neighbor node is added to the failure node list; when performing inter-layer cascading failure, after blocking the dependency relationship between the failed node and its dependent nodes, the nodes that depend on the failed node are treated as failed nodes and added to the failure list; when there are no failed nodes in the failure node list, it can be determined that the cascading failure simulation is completed.

[0077] The technical effect of the technical solution of the present application is verified below in conjunction with specific verification experiments.

[0078] In order to make the verification experiment more intuitive and representative, this experiment selects BA (Barabási-Albertmodel) scale-free network, WS (Watts-Strogatz) small-world network and ER ( The simulation experiment was carried out by using a weighted dependent network composed of a random network (Erdos-Rayleigh) network. The network types involved in the experiment include BA-BA-BA network, WS-WS-WS network, ER-ER-ER network and BA-WS-ER network. The cascading failure model provided in this application is called RCNI-LC (RCNI-Load-Capacity, cascading failure load capacity) model. In order to avoid the impact of different scales of different network layers, it is stipulated that each layer of the network has 100 nodes. The BA network follows: when a new node joins the BA network, three of the nodes are randomly connected; the WS network follows: each node in the network has an edge with any other five nodes, and the probability of generating an edge between two nodes is 0.5; the ER network follows: the probability of generating an edge between nodes in the network is 0.2. The rule for generating network dependency relationships between layers is a randomly generated directed dependency. All experiments include 15 rounds of network collapse, and the surviving nodes in the network are attacked in each round of collapse. The attack methods include random attacks and deliberate attacks against the highest load. All experimental results are the average values ​​after 20 experiments. The robustness metric of the network is the relative size of the largest connected subgraph Where N' is the maximum connected subgraph size after the network attack, and N is the maximum connected subgraph size of the initial network.

[0079] The node capacity in the network is determined by the parameters α, β and θ, where the parameter α adjusts the initial load of the network nodes. The relationship between the model parameters and the robustness of the weighted dependent network is as follows: Figure 6As shown in Figure 2, the results are the average of 20 experiments on the BA-WS-ER network. Figure 6 It can be seen that no matter what level α is at, the robustness of the network increases with the increase of β and θ. By comparing the robustness of the network under three different levels of parameter α, it can be seen from the folding degree and slope of the surface that as α increases, the robustness of the network decreases, and the robustness is more sensitive to the changes in parameters β and θ. The blue curve in the figure is the relationship between parameter β and network robustness when θ = 0.8. For a given θ, the robustness increases with the increase of β, but as α increases, its robustness is at a low level for a longer period of time. When β = 1.2, the relationship between parameter θ and network robustness. For a given β, the robustness shows a similar relationship, and similar trends are presented at different α levels. It can also be seen from the trend of the curve that when β and θ are at a high level, the network shows strong robustness, and as the values ​​of β and θ gradually decrease to the critical value (θ ≤ 0.9 or β ≤ 1.2), the robustness of the network drops sharply.

[0080] In order to verify the effectiveness of the proposed model, three cascading failure models were selected as baselines for comparative experiments. The first cascading model is based on residual capacity (RC-LC model for short), in which the load of the faulty node is redistributed according to the residual load of its neighboring nodes; the second cascading failure model is based on capacity (C-LC model for short), which redistributes the load of the faulty node considering the capacity of the neighboring nodes; the third cascading failure model is a model based on node strength (NS-LC model for short).

[0081] In order to better demonstrate the cascading failure process and perform robustness analysis, the parameters of the RCNI-LC model in this experiment are set to α = 1.0, β = 1.2, and θ = 0.8. The comparison results of the model effects are shown in Figure 2. Figure 7 , Figure 8 , Fig. 9 and Fig.10 As shown, Figure 7 This is a schematic diagram of the effect of the BA-BA-BA network under deliberate attack and random attack. Figure 8 The following is a diagram of the effect of a WS-WS-WS network under deliberate and random attacks. Fig. 9 Schematic diagram of the effect of ER-ER-ER network under deliberate attack and random attack. Fig.10Schematic diagram of the effect of BA-WS-ER network under deliberate attack and random attack. In general, the robustness of weighted interdependent network is better under the RCNI-LC model proposed in this paper, whether under random attack or deliberate attack. Among them, the robustness is better in WS-WS-WS and ER-ER-ER networks, which means that the proposed model has stronger network protection ability compared with other cascading failure models. The robustness of BA-BA-BA and BA-WS-ER networks is relatively common, but the robustness of BA-WS-ER network is stronger in comparison. From the perspective of attack type, due to the randomness of random attack in node selection, the performance of the proposed model is slightly different from other models, but it is better than other models overall.

[0082] Furthermore, this experiment analyzes the effect of the model under different network and attack types. The results are as follows: Fig.11 As shown in the figure, the ER-ER-ER network is the most robust under network attacks, while the BA-BA-BA network is more vulnerable. This is the opposite of the case of single-layer networks. When the heterogeneity of degree distribution increases, the vulnerability of dependent networks increases. At the same time, due to the existence of BA networks, the robustness of composite networks is also reduced.

[0083] Since the cascading failure phenomenon is widely present in real complex systems, this experiment abstractly models the complex system as a weighted dependent network and studies the network robustness under cascading failure. This application proposes a new cascading failure model based on a weighted dependent network, in which the initial load is measured by the node strength, and the load redistribution strategy considers the combined effect of the remaining load and the node importance. In addition, considering the time-varying nature of the load and the inter-layer dependency of the nodes, a cascading failure algorithm is designed. Finally, the effectiveness of the model is verified through simulation experiments. Despite this, there are still many problems in this study that need further study, such as the design of network weights, the generation of inter-layer dependency rules, and the criticality of the network under different conditions.

[0084] According to a second aspect of the present invention, the present invention also provides a computer device, such as Fig.12 As shown, the device includes a processor, a memory, a communication interface and a communication bus, and the processor, the memory and the communication interface communicate with each other through the communication bus. The above-mentioned processor is used to provide computing and control capabilities, and the above-mentioned memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer program instructions, and the internal memory provides an environment for the operation of the operating system and computer program instructions in the non-volatile storage medium. The communication interface of the above-mentioned device is used to communicate with an external terminal in a wired or wireless manner. The wireless method can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The above-mentioned memory is used to store computer program instructions, and the computer program instructions enable the processor to execute the above-mentioned present invention as Figure 1 The cascading failure simulation method based on weighted dependency network is shown.

[0085] According to the third aspect of the present invention, the present invention further provides a computer-readable storage medium. A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through computer program instructions, and the computer program instructions can be stored in a non-volatile computer-readable storage medium. When the computer program instructions are executed, the above-mentioned present invention can be implemented as follows: Figure 1 The cascading failure simulation method based on the weighted dependency network shown. Any reference to memory, storage, database or other media used in the embodiments provided by the present invention may include non-volatile and / or volatile memory, wherein the non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory, and the volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0086] The terms "first" or "second" and the like used in this specification to refer to numbers or ordinals are used for descriptive purposes only and are not to be understood as explicitly or implicitly indicating relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of this specification, "plurality" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0087] Although this specification has shown and described multiple embodiments of the present invention, it is obvious to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, modifications and alternatives without departing from the thought and spirit of the present invention. It should be understood that in the process of practicing the present invention, various alternatives to the embodiments of the present invention described herein may be adopted. The claims appended hereto are intended to define the scope of protection of the present invention, and therefore cover the modular composition, equivalent or alternatives within the scope of protection of these claims.

Claims

1. A cascading failure simulation method based on a weighted dependency network, characterized in that: include: Acquire a cascading failure model based on a weighted dependency network, wherein the cascading failure model includes a plurality of network layers, each of which has a plurality of nodes; The initial load of the node is set according to the node strength of the node and the node strength of its neighboring nodes, and the node capacity of the node is set according to the nonlinear relationship between the node capacity of the node and the initial load; The cascading failure model is used to simulate the cascading failure process, wherein the cascading failure process includes inter-layer cascading failure and intra-layer cascading failure, and during the simulation process, load redistribution is performed according to the remaining capacity and importance of neighboring nodes of the failed node; The load redistribution includes: using the following formula to perform load redistribution: Where L j (t+1) and L j (t) are the actual loads of the jth node at time step t+1 and time step t, respectively. i is the capacity of the i-th node, ΔL ji is the additional load of the jth node, which is obtained according to the remaining capacity and importance of the jth node, and the jth node is the neighbor node of the ith node; When obtaining the additional load of the j-th node according to the remaining capacity and importance of the node, the following calculation formula is used: Where L j (t) is the actual load of the ith node at time step t, C j is the capacity of the j-th node in the cascading failure model, and wj is the importance of the j-th node.

2. The cascading failure simulation method based on weighted dependency network according to claim 1, characterized in that: The step of setting the initial load of a node according to the node strength of the node and the node strength of its neighboring nodes includes: calculating the initial load of each node using the following formula: in is the initial load of the ith node, s i is the strength of the i-th node in the cascading failure model, Γ i is the set of neighbor nodes of the ith node, s′ m For the set Γ i The node strength of the mth node in im is the i-th node and the set Γ i is the edge weight between the mth nodes in , and α is the parameter for adjusting the initial load intensity.

3. The cascading failure simulation method based on weighted dependency network according to claim 1, characterized in that: The step of setting the node capacity of the node according to the nonlinear relationship between the node capacity and the initial load of the node includes: calculating the node capacity of each node using the following formula: Among them C i is the capacity of the i-th node in the cascading failure model, β and θ are the first capacity parameter and the second capacity parameter, respectively. is the initial load of the i-th node in the cascading failure model.

4. The cascading failure simulation method based on weighted dependency network according to claim 1, characterized in that: The simulation layer cascading failure includes: Obtain the neighbor nodes of the failed node that participate in the redistribution in its network layer; Calculate the remaining load and node importance of the neighboring nodes of the failed node; Calculate the load that the failed node allocates to its neighboring nodes, and redistribute the load based on the calculation results; Update the real-time load of all nodes participating in load redistribution, and determine whether the real-time load of the node exceeds its own capacity. If so, the node will be regarded as a failed node.

5. The cascading failure simulation method based on weighted dependency network according to claim 1, characterized in that: The cascading failure model is provided with an adjacency matrix, and the simulation of inter-layer cascading failure includes: Obtaining a failed node, and obtaining nodes that depend on the failed node according to the adjacency matrix; The dependency relationship between the failed node and the nodes that depend on the failed node is blocked, and the nodes that depend on the failed node are treated as failed nodes.

6. The cascading failure method based on weighted dependency network according to claim 4 or 5, characterized in that: The cascading failure model is also provided with a failure node list, and the cascading failure simulation process includes updating the failure node list, and when there is no failure node in the failure node list, stopping the cascading failure simulation.

7. A computer device, characterized in that: It comprises a processor and a memory, wherein the memory stores computer program instructions for execution on the processor, and when the processor executes the computer program instructions, the cascading failure simulation method based on a weighted dependency network as described in any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium, characterized in that: It stores computer instructions, and when the computer instructions are executed by a processor, the cascading failure simulation method based on a weighted dependency network according to any one of claims 1 to 6 is implemented.