Research On Freeway Traffic Flow Parameter Prediction And Active Traffic Congestion Prevention And Control

As a tightly-connected inter-city channel,the freeway occupies a leading position in the external traffic system of the city by virtue of its characteristics of fast speed,high efficiency and strong flexibility.In recent years,the increasingly diversified traffic demand and the limited road capacity increasingly sharp contradiction.How to use information to gather traffic elements,drive the transformation of management methods,transform passive processing into active prevention,and ensure the smooth and efficient operation of freeways is the trend of current traffic management.Therefore,this paper takes the prevention of traffic congestion on freeways as the research,combined with toll data to predict traffic conditions in advance,predict the time and place of congestion,and control congestion within a controllable range,so as to formulate scientific and reasonable management and control measures to limit and induce traffic passive treatment translates into active prevention.The main features of this paper are to predict traffic flow parameters by combining toll data and to use variance analysis to judge the rationality of clustering algorithm.Firstly,based on the traffic flow theory,mining the regularity of OD distribution characteristics,flow characteristics and speed running characteristics.Secondly,a travel time prediction model of K-means clustering algorithm based on analysis of variance is established,and the significance level of each cluster is tested using analysis of variance to determine the effectiveness of clustering,output the clustering center and calculate the distance between each cluster center and the feature vector,and give the cluster center corresponding weight according to the distance,and use the weighted strategy method to predict the travel time for the cluster center.And then use the on ramp as input to estimate the traffic flow of the road section.Finally,starting from the diagnostic freeway congestion reasons,analyzed at multiple levels from excessive traffic,traffic accidents,bad weather,driver characteristics,and inadequate management plans.Select variable speed limit and ramp metering control methods as the active traffic control method in this paper,respectively introduce the principles,functions and classic algorithms of these two control methods,then formulate the control strategy of this paper and select TTT,TTD and TWT is used as a control evaluation index.The paper takes the bottleneck section of the freeway as an example and uses matlab programming software to predict the travel time of holidays and working days.The results show that the prediction errors are 4.22% and 4.74% respectively,and the prediction accuracy is within the acceptable range.The results show that the model has a high prediction accuracy;the VISSIM is used to simulate the RM control strategy,VSL control strategy,RM & VSL cooperative control strategy respectively.The results show that the three proposed ATM control strategies can reduce the total travel time and total delay to different degrees and improve the road utilization rate,which has important application for preventing freeway congestion and improving traffic conditions.