Research On Cascading Failure Behaviors Of Networks Based On Improved Threshold Model

The structure of various real systems in the real world is more and more complex,and the connection between systems is more and more close.The failure of some nodes may cause cascading collapse in the whole system or even damage the associated system.Therefore,it is of great practical significance to study the cascading failure behavior of network systems deeply.Threshold model is one of the classical models used to describe the cascade failure process,but the traditional threshold model ignores the node difference and only considers the influence of the direct neighbor node.In this paper,an improved threshold model considering individual loyalty and multi-order neighborhood nodes is proposed to study the evolution of cascading failure behavior in complex networks.The research results can further enrich the understanding of the complexity of cascading failure behavior and provide theoretical support for the effective prevention and treatment of intermediate cascading failure in complex network systems.The primary content and innovation results of this paper are as follows:(1)Study on Cascading Failure Behaviors based on Improved threshold model considering individual loyaltyIn order to improve the traditional threshold model,the concept of individual loyalty is introduced,the standby target node group is set,and the characteristics of structural connectivity change during failure process are analyzed.By means of numerical simulation,the dynamic law of intermediate joint failure behavior of three random network and real network systems is analyzed.The results show that the cascade failure effect is weakest when individual loyalty and degree are positively correlated.When the degree value is larger,the failure speed of nodes can be accelerated significantly.Node failure will destroy the overall network connectivity to a large extent,and when individual loyalty is negatively correlated with degree value,the network is more vulnerable to damage.Increasing the network average and strengthening the protection of key nodes can effectively restrain the occurrence of cascading failure behavior.(2)Study on Cascading Failure Behaviors based on Improved threshold model considering the influence of multi-hop neighborsIn order to solve the problem that the existing threshold model only considers the effect of first-order neighborhood nodes,the role of multi-order neighborhood nodes is introduced into the study of cascading failure behavior.Based on the improved model,the evolution characteristics of intermediate joint failure behaviors of random network and real network systems are analyzed by numerical simulation.The results show that the second-order neighborhood nodes can significantly accelerate the cascade failure speed,and the higher the weight of second-order neighborhood nodes in the model,the faster the cascade failure speed,but the marginal effect of acceleration will be weakened.The influence of second-order neighborhood can be eliminated to some extent by changing the network structure.The influence of third-order neighborhood is related to the network structure.The increase of the second-order neighborhood weight will weaken the influence of the propagation threshold.