Research On Risk Prevention And Control Of Complex Network System With Reinforced Nodes And Dependence

Networks in the real world are often not isolated networks.There are various connections between them,such as physical dependence,logical dependence,energy or information exchange.The situation of interdependent and cooperative networks is more common in modern society.phenomenon.Although this interdependent relationship can greatly improve our efficiency,it also brings greater potential danger to the entire system.In recent years,many scholars have abstracted a variety of network models with different special characteristics according to the real-world network conditions.In these network models,cascading fault propagation does not fully follow the theory of percolation,and the attributes of nodes in the network also change.There are big differences.Xin et al.Proposed a network model with both reinforced nodes and dependencies in December 2016,which attracted some attention.In this model,there are some nodes.These nodes are pre-reinforced.They can withstand the failure caused by the disconnection of the maximum connected graph during the cascade,but they will still be affected by the dependencies.In addition,if the node can survive,It can also guarantee the survival of small nodes in the vicinity.Based on the model,this paper mainly has the following two contributions:(1)A new reinforcement node selection algorithm is proposed,which makes the robustness of the entire system after reinforcement stronger than other existing reinforcement algorithms;(2)A dynamic recovery algorithm adapted to the model is proposed to reduce the impact of the system when it encounters a fault.First of all,we analyze the properties of the nodes and the structure of the network system to clarify the experimental objectives.To make the entire system more robust,we must choose the nodes that contribute to the stability of the system to strengthen and quantify the contribution of each node.In this paper,a reinforcement weighting algorithm is proposed.This algorithm combines the attributes of connected edges and dependent edges of nodes,which can fully reflect the contribution value of a single node to the system’s robustness.We sort and select the nodes that should be strengthened according to the weight of all nodes.With the same proportion of reinforced nodes,the reinforced nodes selected by this algorithm can greatly improve the robustness of the entire dependent network model.Therefore,this algorithm can provide a theoretical basis for the network construction method of this type of network system.Secondly,for the network system of the strengthened nodes that has been determined,we can use the attributes of the strengthened nodes to design a clever dynamic recovery algorithm.By adding a failed coupled node and a strengthened node without a coupled node in its network,Connected edges in the network can ensure that the failed reinforced nodes and their coupled nodes resume their functions in this round of cascading,and support a small range of connected graphs near themselves to resist the negative impact of cascade diffusion.We reflect this through the side of the NOI value Our dynamic recovery algorithm is effective and applicable to this systemThe dependent network model with reinforced nodes is a model with realistic basis.Therefore,our analysis of node attributes,reinforcement algorithms and dynamic recovery algorithms in this model can be used for real-world network creation,risk prevention and control,and disaster relief.And other aspects to provide effective theoretical support.