Font Size: a A A

Research On Network Modeling Method Based On Local Information

Posted on:2014-07-13Degree:MasterType:Thesis
Country:ChinaCandidate:Q FuFull Text:PDF
GTID:2250330401482695Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
Network science is the most concerned direction of the complexity research in recent years, its research contents are to use the network to describe the actual system, then study the dynamics of complex systems, it provides the help for people to understand the law of the variety of natural and artificial complex system.This paper combines the characteristics of the local information in the real network systems, which use network science theory to establish two network models related to local information, as well as research the impact of the local characteristics in the evolution of the whole network structure and the network characteristics; we established a bus network model based on the merging of the local stations, and used its path search method in the Bus Transfer System of Hangzhou. It combines the theoretical methods and practical applications, which has an important practical significance. The main work and results are as follows:First, many practical system’s scale is so large that it is difficult to obtain its global information and the interaction in a local scope is more frequently than the global, a evolutional network model based on the scale of local world and local preferential attachment mechanism is proposed here, the simulation result shows the property of this model can reappear the real network’s topological property, especially the distribution of the node’s degree, node’s strength is scale-free.Then, considering the merging and regeneration process of nodes in networks is a common phenomenon in nature and human societies, a self-organized evolutionary model of network has been proposed to study such phenomena. During the evolution process of the network, the network always holds its total number of nodes and single edges constantly. The simulation result shows that, a network with scale-free degree distribution, scale-free node strength distribution and scale-free edge weight distribution can be generated.Finally, a bus network model based on the merging of the local stations is established using network science’s method, a new bus transfer arithmetic has been proposed based on this model. Using the new arithmetic, we design and realize the Hangzhou Intellegence Bus Transfer System, it’s worth noting that this system can provide different transfer scheme for users’different requirements, and it is a visual system which is convenient for users.
Keywords/Search Tags:network science, local information, preferential attachment, non-growth, multiple-edges, scale-free
PDF Full Text Request
Related items