Study On Traffic Flow And Conflict In Underground Interchange Diversion Area Of Urban Roads

Underground roads have the advantage of connecting key areas of the city,break the "bottleneck" of node,release urban space,improve the urban environment,but as it is built into a network,a large number of interchanges will be formed,as the key nodes of the road network,the underground interchange includs the diversion and other complex sections.The diversion area often becomes a common accident area due to limited visibility,large speed difference between vehicles and conflict points.However,there are few studies on traffic flow and traffic conflicts in underground interchange diversion areas.Based on this,this paper explores the traffic flow and traffic conflict of typical urban roads underground interchange diversion areas based on the basic route of "data collection-phenomenon analysis-modeling and mechanism analysis".Taking Xiamen Wanshi-mountain underground interchange C-ramp diversion area as the research object,firstly,traffic flow and conflict data are obtained through "surveillance video + video recognition ".Secondly,the traffic operation characteristics of the underground interchange diversion area are revealed through statistical analysis,including traffic volume,speed distribution,following behavior distribution and lane changing behavior distribution.Then,based on the structural characteristics of the underground interchange and the survey data,the conflict classification work is carried out,and the existing conflict indicators are compared to select suitable indicators for conflict measurement,and the threshold values of different conflict types are determined by the "cumulative frequency curve analysis method",and then the causes of conflict formation are analyzed,and the influencing factors of traffic conflicts are initially analyzed,and The simulation software is used to correct the driving behavior model to supplement the specific conflict data.On this basis,a traffic conflict model is established based on Bayesian network,and the accuracy of the model is verified to be high.The main influencing factors of traffic conflict in the underground interchange diversion area are determined through inverse reasoning and sensitivity analysis,and the influence relationship between each main influencing factor and the outcome variable is analyzed through a posteriori probability analysis.The results show that traffic volume,lane change execution position and distance and speed difference with the back of the target lane have significant influence on the severity of conflict with sensitive values of: 0.124,0.05,0.119,0.049,while traffic volume,lane change execution position,ramp design speed,diversion rate,and lane change duration have significant influence on the degree of influence on the longitudinal location of conflict occurrence with sensitive values of.0.089,0.061,0.08,0.08,0.015,and the key link affecting the severity of the traffic conflict is "traffic volume-distance to the back of the target lane-severity of the conflict",and the key link affecting the longitudinal position of the traffic conflict is "traffic volume-vehicle lane change execution position-longitudinal position of conflict occurrence".