Research On Intersection Pinning Control Under Coupling Relationship

Regional signal control is the most commonly used traffic control method at present,but if all nodes in the network are under control,the cost will increase significantly,and the regional road network will be less robust.Pinning control is a complex network control method,whose essence is to control a small number of key nodes to make the network achieving the desired effect,so it can make up for the shortcomings of traditional signal control methods.This paper studies the coupling relationship between adjacent intersections and establishes a complete control scheme based on the importance evaluation of regional road network intersections.This paper analyzes the small world and scale-free characteristics of road network from the perspective of complex network,determines the complex network characteristics of road network and puts forward the unidirectional coupling model of adjacent intersections and expounds the coupling state analysis process of adjacent intersections.Firstly,the coupling synchronization process of unified chaotic system under the linear coupled synchronization method is described.Taking Chen system as an example,the unidirectional inflow and outflow chaotic model of intersection is established by using the actual evolution data of traffic flow.Next,the unidirectional outflow and inflow chaotic models of intersections in the same direction of the same road section are analyzed,and the unidirectional coupling model of adjacent intersections is formed by combining the two models.Finally,the coupling degree between adjacent intersections is defined based on the error evolution process of the combined model.In order to select the nodes(intersections)that need to be controlled in regional traffic network,based on the weighted PageRank algorithm,an intersections importance evaluation model is proposed.The weight value,initial value and dynamic problem in the original PageRank algorithm were analyzed,and the coupling degree between adjacent intersections was taken as the weight value,and the importance evaluation model of intersections was established based on the improvement of other parameters.The importance of intersections is ranked by this model to select key intersections.The superiority is obtained by comparing this algorithm with the original PageRank method and the basic parameter method of complex network.Based on the determination of key intersections,a road network containment control model is constructed,which makes the closer to key nodes(intersections),the smaller the intersections’ periodic delay is.The ideal delay matrix of road network intersection is calculated and combined with the actual delay matrix during the evolution of road network,the delay state equation of intersection with traffic signal information is constructed,and the Gram matrix is used to judge the controllability of the equation of state.Then reduce the difference between the actual error of the intersection and the ideal error matrix in the evolution process and calculate the green ratio matrix under the containment control scheme,obtain the traffic signal scheme under the containment control scheme.In the example,six intersections near Huangshan road in Hefei city,Anhui province are ranked in importance,and the control scheme is set up.The control results show that this strategy can greatly optimize the key intersections and their adjacent intersections,proving that the control method is feasible in practical traffic control applications.In this paper,the coupling relationship between adjacent intersections is microdisplayed by mathematical model,and the result is embodied as the importance evaluation index of intersections by PageRank algorithm.From the perspective of containment control,the actual control scheme of intersections is set up to prove the rationality of the analysis of abstract coupling relationship between adjacent intersections.This study improves the theory of coupling between adjacent intersections and provides the basis for the application of complex network theory in traffic control.