Research On Iterative Learning Control Method Of Region Perimeter Control Based On MFD

With the rapid increase in the number of vehicles,the problem of traffic congestion is getting worse.Maximizing the potential of the road under the existing road conditions and tapping the capacity of the road is an effective way to alleviate the increasingly serious problem of traffic congestion.With the development of the Macroscopic Fundamental Diagram(MFD)theory,more and more scholars have begun to study the methods of perimeter control and guidance collaboration of regional traffic flow,in order to propose a method of balancing traffic conditions at the macro level.Therefore,this paper is based on the Macroscopic Fundamental Diagram theory to discuss the method of u sub-area division and the strategy of perimeter control between sub-regions.First,using Baidu traffic index data to analyze the road network and discussing the feasibility of the perimeter control method based on MFD.Compare and analyze the advantages and disadvantages of the traditional methods of the sub-area division based on the MFD.Improving the K-means clustering algorithm,considering the time-space characteristics of urban road network density,and using license plate data to discuss perimeter control.The method of sub-area division is programmed using MATLAB software.The average density uniformity is introduced as the criterion for judging the results of sub-area division,and the actual road network is divided by the continuous dichotomy.Analysis and comparison of the results show that the improved algorithm can divide the road sections with high compactness and similar density into the same Sub-area.Second,apply iterative learning control theory to perimeter control.Constructing an iterative learning control model aiming at consisting road network traffic load,and using MATLAB software to program.Then taking the traffic data of Qingshan District,Baotou City as an example to analyze convergence,verifying iterative learing control model and providing parameters for perimeter control.Finally,method for dynamic identification of controlled boundary intersections is proposed,which optimizes signal timing based on the consideration of road capacity and minimum pedestrian crossing time.Taking the data of Qingshan District of Baotou City as research object,carrying out simulation parameter correction and simulation evalution of methodology in the paper by PTV Vissim professional traffic software.The simulation results illustrate that after perimeter control the average delay and the number of stops in selected road network decreased by 11.7% and 10.8%,respectively,and the average travel speed of the network increased by 15.5%.After perimeter control,the average delay and the number of stops in subzone-1 decreased by 4.1% and 5.8%,and the average travel speed of the subzone-1increased by 6.7%.The average delay and the number of stops in subzone-2 after perimeter control decreased by 21.2% and 17.0%,respectively,the average travel speed of the subzone-2increased by 16.3%.The resultes of simulation evalution indicate that the perimeter control method can effectively improve the traffic operation status of the entire road network.