Intersection Importance Assessment And Regional Control Algorithms Of Urban Road Network

With the development of urban economy and the increase of car ownership,the increasingly serious regional traffic congestion brings a lot of inconvenience to people's life.In order to solve the above problems,this paper focuses on optimizing the signal timing of urban traffic network to alleviate traffic pressure.This thesis selects the Wangjing area with more complete functions as the experimental area,and combines the floating car data to simulate,which increase the accuracy and universality of research results.First,through the MapInfo and Hidden Markov methods for map matching of floating vehicle data,the information of the traffic flow,average speed and travel path for each road segment are obtained.The above data is used to construct the PTV-VISSIM simulation platform,and the state of each time period in the urban traffic road network is reproduced by the simulation platform.Based on the simulation platform built by PTV-VISSIM,more traffic flow data such as queue length,delay and occupancy rate can be obtained.The macroscopic fundamental diagram is fitted by traffic flow data,which is used to supplement and improve the cluster partitioning algorithm.Therefore,a partitioning method with faster speed and obvious partition boundary is obtained.In this thesis,a new scheme of regional boundary control is proposed,which includes the macroscopic fundamental diagram,fuzzy adaptive control and the concept of pinning control.Firstly,the fuzzy adaptive control algorithm,the traditional timing control and the PI control are used for the traffic signal control,and the number of vehicles in the current road network and number of output vehicles of the road network can be obtained,which proved the effectiveness of the fuzzy adaptive control algorithm.Secondly,according to the idea that changes all the signal of regional boundary intersections is highly complex and difficult to implement,this paper proposes a comprehensive evaluation method of intersection importance based on neural network.It combines the characteristics of social network,system scientific analysis and traffic flow of the transportation network.By use this method,the important intersections on the boundary can be gotten.By using the concept of pinning control,the fuzzy adaptive control algorithm to control all intersections and important intersections of regional boundaries are simulated and analyzed in detail.Finally,the simulation results show that the implementation of the new regional boundary control scheme can effectively improve the operation level of the road network and reduce vehicle congestion.There are 64 figures,17 tables and 61 references in this thesis.