| With the rapid development of economy and the acceleration of urbanization,the scale of urban road network is increasingly large,meanwhile,the number of cars also increases sharply.Urban traffic congestions,traffic safety and other road traffic problems are getting more and more serious.At the same time,the low road area rate of China,serious mixed traffic flows and other factors have made the urban traffic problems more difficult to solve.How to solve the problems of urban traffic congestions has always been a hot and difficult issue in geographic information system(GIS),traffic management and operations researches.In recent years,the vigorous development of complex network theory has expanded the research ideas,provided new perspectives and methods for in-depth exploration of the structure and functional characteristics of realistic complex systems,which has been successfully applied to the study of urban traffic system complexity and dynamic traffic inductions.Therefore,this paper focuses on the transmission mechanism of traffic flow on dynamic traffic networks by combining GIS network analysis methods and complex network theory researches,and explores efficient dynamic path selection methods to control traffic congestions,provide network traffic carrying capacity and balance network traffic load.In this paper,a simple dynamic path selection algorithm named time-optimization is proposed by modeling the Time series of dynamic traffic processes.Specifically,the main research work and achievements of this paper include the following aspects:(1)It studies the network traffic dynamics characteristics of different routing strategies,and focuses on the analysis of network throughput and distribution of mesocentric.Based on the routing simulation of BA scale-free network under shortest distance routing algorithm(S-P),medium optimal routing algorithm B-O),gravitational field dynamic routing algorithm(G-F)and path confidence routing algorithm(P-C),the performance of different routing algorithms is studied bycalculating the network throughput and traffic congestion distribution.(2)It establishes a measurement method of the time required for traffic flow to transfer along any path on the dynamic traffic networks.This method considers the traffic flow analysis on the whole network from the global optimal point of view by using the time series analysis theory to express and model the time sequence of the traffic flow transfer process,has established a mathematical function to accurately describe the time required for the individual traffic flow to transfer along any path.The results show that this method can describe the interaction mechanism of traffic flow travel time on dynamic traffic network.(3)It proposed a dynamic path selection algorithm based on time optimization.Based on the measurement model of travel time in dynamic traffic network,the algorithm takes the real-time global time optimization of traffic flow individual as the objective constraint to dynamically plan all traffic flow transmission paths,and then realizes the balanced distribution of network traffic load and the control of traffic congestion.(4)a variety of static and dynamic routing algorithms are selected for routing simulation to verify the effectiveness of the method.The network throughput and mesocentric distribution of BA scale-free network,ER random network and actual road network under different routing algorithms are calculated respectively.The results show that compared with the s-p,g-f,b-o and p-c routing algorithms,the time-optimal routing strategy significantly improves the network throughput and effectively balances the distribution of mesocents. |
PDF Full Text Request