Analysis On Merging Behaviors For Expressway And Optimization Of Traffic Organization For Confluence Area

The bottleneck of expressway refers to the phenomenon of vehicles queuing in a certain section of expressway,which is caused by insufficient capacity,emergencies or traffic flow conflicts.Due to the dive in speed and the surge in density in bottleneck area,expressway congestion occurs frequently.As the key area where the bottleneck of expressway occurs,there are two traffic flows,the on-ramp flow and the flow on the outer lane of main road,directly affect confluence area capacities.The main reasons for the bottleneck of expressway are the merging behavior of on-ramp traffic flow and the chain reaction of the traffic flow on the outside lane of the main road.Therefore,for revealing the complex interaction mechanism of the two traffic flows,it is of vital importance to study the characteristics of the entrance behavior from the perspective of on-ramp vehicles and vehicles in the shoulder lane of the main road.In this paper,by means of actual survey data analysis and video processing technology,the spatial and temporal information of the vehicles on the ramp and the vehicles on the main road in the whole confluence area,at the south entrance of nanhu square of the western expressway in changchun city,were obtained.The merging behavior of the expressway was studied as follows:(1)Focusing on on-ramp merging vehicles,on-ramp gap was studied and a gap selection model was proposed firstly.Then the vehicle merging behaviors of parallel acceleration lane on expressway are classified through clustering analysis.On this basis,distribution model of merging position is proposed in terms of various categories,and distribution characteristics of merging position under different densities and merging speeds are analyzed.(2)Focusing on main road vehicles,this paper analyzes the influence of the merging maneuver on the main road by taking the following stability,speed and lane change rate as indicators.Based on “maneuvers’ ’ and “lanes”,the influence of different numbers of merging vehicles on the main road and the influence of merging vehicles on different lanes of the main road were studied respectively,and the lane change prediction model of the shoulder lane of the main road was proposed.Finally,the accuracy of the model is verified by the measured data.(3)Based on research of merging maneuver characteristics and its impact influence,two optimization strategies for designing confluence area are introduced:(1)Optimizing parallel acceleration lane in peak period.(2)Guiding the shoulder lane of mainline traffic flow changing behaviors.Employing micro-simulation analysis,the application conditions of two optimized design methods are studied,and obtain the threshold values of flow in two methods.In general,this paper based on massive measured data,proposes the gap selection model,the merging position distribution model and the prediction model of lane change on the shoulder lane of the main road.The significant characteristic indexes,which affect gap selection of on-ramp vehicles,the merging position distribution of on-ramp vehicles and the lane change of main road vehicles,are clarified and provide theoretical basis and technical support for further studies on more accurate and more practical micro model of confluence area.The proposed two optimization strategies supply data support for traffic management and control of expressway confluence area.Finally,on the basis of summarizing the main research results of this paper,the existing problems and the next research plan are put forward.