Research On The Method For Dynamic Subdivision Of Regional Traffic Signal Control Based On Multi-source Data

The distribution of traffic flow in various regions of the urban road network is uneven.Local congestion occurs frequently in large and medium cities.Local congestion can easily spread and cause regional congestion.Consequently,the development of urban traffic signal control systems requires specifically designing a signal coordinated control plan from the perspective of the overall strategy of the road network.The division of urban road network traffic control sub-areas is the basis for the coordinated control of regional road signals.A reasonable sub-division method can maximize the complementary advantages of intersections and make the traffic flow between the intersections relatively balanced.Further,it will be more conducive to select and implement the signal control scheme of the sub-area,and thus improving the operating efficiency of the entire road network significantly.This paper proposes a method for dynamic regional sub-division of regional traffic signal control based on multi-source data through integrating multiple types of data,analyzing the traffic flow characteristics of the road network.According to the changes in the traffic flow and paths in the road network,the sub-areas are divided,the signal timing is optimized and the path is coordinated,so that the traffic status of the road network is improved.First of all,this paper reviews the domestic and foreign status of the relevance degree model and the division of traffic control sub-areas at home and abroad.Secondly,the paper uses Baidu traffic index data,ultrasonic testing data,license plate data,and Didi turning data as the main data sources in order to tap the traffic flow characteristics of the road network,vehicle driving trajectory,path flow,turning ratio,etc.The data basis for subarea division is derived from analyzing and integrating these different data sources.Thirdly,this paper selects a local road network and the incidence degree is calculated by using Whitson model,Whitson improved model,route incidence degree model.Sub-area division is based on cluster analysis,simultaneously,carrying out signal timing optimization and route coordination.It is concluded that the use of the route incidence model can optimize the sub-area division through simulation evaluation conducted by PTV Vistro and PTV Vissim professional traffic software.Finally,according to the principles and methods of sub-area division,this paper uses a hierarchical clustering method to establish a dynamic sub-area division method based on multisource data,and takes the urban road network of Qingshan District of Baotou as an example.This paper uses incidence degree model based on route to divide the sub-area,then,analyzes the characteristics of each control sub-area and boundary of the sub-area of the road network,and optimizes the control parameters of each signal control intersection so as todetermine the coordination phase and mode of the paths in each control sub-area of the road network.Further,this paper coordinates the control sub-area with its adjacent sub-area or a single-point control intersection in order to achieve a smooth transition between control subareas.Using PTV Vistro and PTV Vissim professional traffic software to simulate the actual traffic situation of the road network,the methods proposed in the paper are evaluated.The simulation results illustrate that the average delay of the road network is reduced by 28.93%,the number of stops is reduced by 17.81%,and the average travel speed is increased by 20.26%.The traffic operation status of the road network has been significantly improved.