Stability And Optimization Decision Analysis Of Complex Networks

With the continuous development of human society and the deepening exploration of nature,many entities are connected through different corresponding relationships,forming a huge complex network system.As an important research tools of complex system,complex network theory explores the synchronization,consistency and stability in the process of network dynamic evolution,explores the network behavior,excavates the properties and characteristics of different networks,and reveals the knowledge and laws contained in the network.In recent years,a series of complex networks,such as traffic network system,financial network system and brain neural network system,have become the focus of complex network research.In particular,finance and transportation have played a cornerstone role in social development.At the same time,they also have a huge network scale and complex connections.Maintaining the stability of financial network and transportation network is of great significance to ensure the well-being of people’s livelihood and maintain social stability.This dissertation focuses on the financial network and traffic network,analyzes the similarities,differences and connections between the two types of networks in structure and nature,and deeply studies the problems of financial price stability,data security,traffic flow control and consistency.Aiming at the problem of financial price regulation in financial network,this dissertation gives a new control strategy and idea;Aiming at the problem of congestion and waiting in urban traffic network,this dissertation puts forward a new stability scheme in optimizing the peak traffic flow.The main research results of this dissertation are as follows:(1)As for the fluctuation of financial prices affected by institutional decisions and policies,this dissertation studies the use of Boolean network to model the financial system,which effectively realizes the stability of the financial network.In the research,the matrix semi tensor product method is used to analyze the cyclic Boolean network based on financial network,and a single node containment control strategy using model reduction method is designed.Therefore,this strategy can find and control the core nodes in the network and use the least control resources to achieve the stability and oscillation of the financial network.Compared with the traditional control methods,the proposed scheme effectively reduces the computational complexity by reducing the number of control nodes,and can better achieve the stability of the financial network.At the same time,the good performance of the scheme is verified by numerical simulation.(2)For the characteristics that the supervision power and openness of data of financial institutions are difficult to coexist,this dissertation proposes a weak central data management scheme based on blockchain technology,which provides a new way for data security supervision and the security and stability of financial system.At present,financial institutions represented by banks have not formed a perfect public supervision system in order to maintain their main right of data control.Therefore,this dissertation analyzes the security and stability strategy of bank data network,gives the three-tier definition of bank data security,and constructs a weakly centralized bank data management scheme based on blockchain technology.On the basis of not changing the central position of bank data management,using block bookkeeping,encryption algorithm and total score cross double chain structure,this dissertation realizes the supervision transparency and operation security of network in data.At the same time,the scheme enables the participants of financial institutions to exercise their regulatory power in real time,effectively avoids the tampering of data by strong nodes in the network,and contributes to the stable operation of the financial network.(3)For the command and control of traffic flow at the core intersection of urban traffic roads,this dissertation uses the actual finite time stability definition to give the intersection flow stability strategy based on sliding mode control method,which ensures the traffic efficiency of key intersections.Due to the uneven traffic flow at complex intersections such as overpasses in different periods,the time of each traffic signal light needs to be adjusted according to the flow to ensure the traffic efficiency and stability of the intersection.This dissertation takes different intersections as network nodes and the flow trajectory of vehicles between intersections as the edge to model the traffic intersection.Through the control scheme design and numerical test based on sliding mode control,it is found that the control scheme used in this dissertation can effectively alleviate the congestion problem at the intersection in peak hours.(4)For the uneven flow of urban road network,this dissertation uses multiagent to model the urban traffic network,and uses iterative learning method to optimize the existing control scheme.In the modeling part,this dissertation creatively regards different roads in the city as multi-agent,and models the connection relationship of roads as edges,so as to establish the continuous time dynamic model of traffic system.Secondly,combined with the characteristics that the iterative learning method does not strictly rely on the mathematical model and is easy to design the control strategy,based on the terminal iterative learning control method,a control algorithm meeting the state constraints is designed to optimize the existing methods.Finally,through strict mathematical derivation and proof,the sufficient conditions for realizing system consistency tracking are given.The numerical simulation process verifies the effectiveness of the proposed control method to maintain the stable traffic and alleviate the traffic pressure in peak hours.Finally,according to the research conclusions on the stability and optimization decision-making of financial network and transportation network,this dissertation gives targeted policies and suggestions,which provides a theoretical basis for promoting the stable development of society.At the same time,this dissertation also summarizes the above research results and shortcomings,and gives the future research plan and prospect according to the research conclusions of financial network and transportation network at this stage.