Research On Automatic Division Method Of Traffic Control Subarea In Urban Road Network

The sub-control-area division is the basement of performing regional signal coordination control on a large scale urban road network.At present,dividing the traffic control subarea uses traversal search methods or cluster analysis methods to divide the intersections with similar characteristics into the same subarea based on the correlation between adjacent intersections.However,the characteristics and applicability of the different methods are not yet clear.Therefore,this paper compares and analyzes the existing division methods and explores the characteristics and applicability of the different methods.All of these can provide a basis for choosing the sub-control-area division method,and lay a foundation of improving the efficiency of traffic control.First of all,this paper researches the correlation between adjacent intersections of urban road network.Analyzing the main factors that influencing the correlation between intersections through the simulation experiments.And then,from this point of view,the four most commonly used correlation models are comparably analyzed,including Whitson model,Whitson improvement model,coherence degree model and coordinability benefit model.In addition,the characteristics and applicability of four modelss are explored.The Whitson model is suitable when the intersection saturation is low,and the Whitson improved model is suitable when the intersection is in a high saturation state.The coordination benefit model is mostly used in trunk line coordination control.In contrast,the convergence degree model has the widest application range.Secondly,this paper studies the different partitioning algorithms of urban road network.Fousing on the deficiencies of the present clustering analysis methods,the network clustering analysis algorithm in complex network theory is applied to the traffic sub-control-area division of urban road network in this paper.The advantages and disadvantages of ergodic search partitioning algorithm and cluster analysis partitioning algorithm are compared and analyzed,and some suggestions for improvement are put forward.Before executing the algorithm,the area that needs coordinated control is obtained by calculating the degree of congestion of the road segment to simplify the road network and improve the efficiency of the algorithm.The paper further builds the urban traffic network model,and studies the AE secondary development function and visual c# programming language,developed a software for automatically dividing the traffic sub-control-area of the urban road network.Finally,taking the main area of Nanjing as an example,this paper applies the development program to verify and analyze the ergodic search partitioning method and cluster analysis partitioning method,and then get the road network partition schemes.Based on the partition schemes,the regional network is carried out by phase difference optimization coordination control with Synchro software,and access to optimization before and after the road network performance.At the same time,as a comparison of the partition schemes,Synchro is used to divide the sub-control-area of the same parameters to verify the feasibility of the ergodic search partitioning method and the clustering analysis partition method,as well as the practicability of the program.In the rush hour of urban traffic congestion,the results show that when the degree of intersection is calculated by the convergence degree model,the control subdivision scheme obtained by using the network clustering analysis method based on the feature nodes is the best for the coordinated control.