Estimation Of Macroscopic Fundamental Diagram And Perimeter Flow Dynamic Perimeter Flow Control Strategy In The Multi-region Network

Resently,the frequent traffic congestion at the peak of urban network has seriously affected the level of urban transportation network systems.When the number of accumulated vehicles in the network is close to the critical value,after local traffic congestion occurs inside the network,like the road section or network node,the traffic flow of the adjacent link is close to saturation,and the traffic diversion strategy can not quickly reduce the network traffic to a reasonable level.Therefore,in order to avoid local traffic congestion spreading to the entire network to form a large-scale regional traffic congestion,it is necessary to adopt a network boundary flow control method to maintain the overall flow of the network within a reasonable range,and avoid accidental traffic congestion resulting in the road network system.The Macroscopic Fundamental Diagram(MFD)is a research tool for urban traffic network flow control because it presents a low-discrete and stable unimodal function relationship between the average network flow and density.It measures the optimal state of the number of vehicles accumulated inside the network.The primeter control of macro network traffic is actually to adjust the degree of traffic transfer and transfer time between regional networks at the boundary of regional road network.The primeter control can coordinate and control the traffic flow of each regional network,so that they all carry out the road network system within a reasonable traffic range and operation to maximize network system performance.Based on the floating car data(FCD),this paper proposes research methods such as acquiring the MFD and dynamic control of boundary flow,including the following three aspects:(1)This paper proposes an analytical method for traffic flow estimation in urban arterial corridor based on FCD trajectories.Different from the existing models,the proposed model considers traffic states and the traffic signal coordination among adjacent intersections,therefore,can capture the delay and queuing dynamics in arterial corridors.The queue spillback phenomenon is explicitly considered by applying the shockwave theory.The analysis of the example shows that the proposed model can accurately predict the trend of traffic flow data and obtain high traffic flow estimation accuracy.Then,based on the traffic estimation results,we can effectively derive the macro basic map information;(2)A multi-region network primeter flow control method based on dynamic route choice is proposed.Different from the conventional primeter flow control method,the influence of route choice behavior factors on regional network MFD information is considered in the model.At the same time of primeter flow control,the dynamic user equilibrium model is used to reasonably adjust the traffic flow distribution in the regional network.The network state maintains an approximate dynamic user equilibrium state as it evolves toward a steady state.The results of the example show that the approximate dynamic user equilibrium state is obtained when the model converges.In addition,the model proves that the model has obvious operational efficiency improvement compared with the conventional DUE model from theoretical analysis and case verification.(3)This paper proposes a dynamic control method for multi-region network primeter flow based on variable structure control.The method fully considers the dynamic change of the traffic flow queue at the regional network boundary in the actual traffic environment,that is,when the model changes the length of the known traffic flow queue,the constraint on the signal timing type of the boundary node can be relaxed.Based on the idea of variable structure control,a sliding mode control structure is designed to adjust the primeter flow control range with reasonable approach rate in the process of network system evolution from initial state to network performance maximization state.The results of the example show that the model can reasonably give the dynamic control scheme of regional boundary traffic under different scenarios and the dynamic control scheme of the corresponding boundary node traffic.In summary,the research results of this paper can obtain the MFD information of the multi-region network in real time according to the floating car data and the network OD matrix in any research period.The traffic transfer scheme between the regional networks is adjusted in real time to coordinate and control the traffic of each regional network to maximize the performance of the network system.