Study Of Traffic State Correlation Identification And Multiple Ramps Cooperative Control Strategy For Urban Expressway

The on-ramps and off-ramps on the expressway are an important part of the expressway network,and the vehicle merging behavior of the entrance ramps can greatly affect the traffic operation status of the mainline of the expressway,which leads to the reduction of the overall traffic efficiency of the expressway.Cooperative control of multiple ramps can efficiently utilize the limited space of the mainline and each ramp,thus achieving the purpose of traffic congestion relief and conflict reduction.However,centralized control of large-scale expressway networks has bottlenecks of difficult optimization search and high computational complexity,so how to make reasonable control range division based on traffic state correlation to reduce control complexity is a difficult problem to be solved for multiple ramp cooperative control.The specific research of this paper is as follows:(1)Traffic state spatio-temporal correlation identification and delineation of cooperative control zones.The factors affecting traffic state correlation are analyzed,and the spatio-temporal delay is used to qualitatively describe the relationship between each road segment and each time period and construct a spatio-temporal weight matrix.Based on the spatio-temporal weight matrix,the global spatio-temporal correlation degree and local spatio-temporal correlation degree are calculated by combining Moran index and spatio-temporal object speed.With the spatio-temporal correlation degree as the weight,the Newman algorithm is improved to realize the division of the collaborative control area.(2)MPC-based hierarchical multi-ramp cooperative control model.Design hierarchical multi-ramp cooperative control architecture and control flow.The MPC-based hierarchical multi-ramp cooperative control model is constructed,and a dynamic critical occupancy estimation model is proposed to reflect the best operating conditions of the mainline.The cooperative control model takes into account both the mainline operation status and the ramp queuing time to determine the control rate of each sub-area entrance ramp and to coordinate and optimize the traffic status of the whole expressway network.Simulation experiments are designed to validate the model in comparison with the Bottleneck control algorithm,and the validity of the model is verified in terms of ramp delays and mainline capacity.(3)Example analysis of control strategy.The multi-ramp section of the northern expressway of Changchun City is selected as the research object,and the division of cooperative control area is carried out based on the spatio-temporal correlation degree of each road section.The expressway traffic flow simulation scenario is built based on SUMO and simulation parameters are set.The traffic flow density,flow rate,speed and delay are selected for evaluation,and the traffic flow conditions under no control,centralized control and zonal control are compared.The simulation results show that the vehicle density of high-density road sections decreases significantly under zoning control,while the vehicle density of low-density road sections increases,and the density difference of each road section decreases,which effectively utilizes the space capacity of each road section and relieves traffic congestion;meanwhile,zoning control can improve the traffic flow and average speed and reduce the average delay,which improves the traffic flow efficiency of the expressway.