Study On Road Traffic State Identification And Traffic State Trend Deduction Method In Major Activity Area

During the major events,there will be strong gathering and high impact traffic demand,which will put forward a severe test to the current urban transportation system,which focuses on solving the daily traffic of commuting,general school,business and so on.With the development of domestic economy and the improvement of urban quality,some major events with international influence are held frequently in many cities in China,which can not only guarantee the smooth progress of major events,but also reduce the impact of major events on daily traffic,which is an urgent problem to be solved.Mastering the regional traffic characteristics during major events,accurately identifying the regional traffic operation status,and inferring the future traffic situation are helpful for traffic managers to formulate and issue corresponding control strategies,which is of great significance for ensuring the smooth development of major events.Based on alleviating regional traffic congestion during major events,this paper takes the regional traffic status during typical major events as the research object.Firstly,traffic characteristics and influencing factors of traffic state in major activity areas are analyzed.Secondly,the change law of road congestion was analyzed based on wind roses,and a quantitative model of congestion degree was established based on Pearson correlation coefficient.A set of real-time traffic state identification method for major activity areas was established.Finally,based on the Gaussian smoke plume model,the limit influence distance of the major event area was established,and the BP neural network was used to predict the future road congestion degree.Considering the change of traffic state after the launch of the travel guidance scheme,the traffic situation of the major event area was deduced.The specific research contents of this paper are as follows:(1)Taxi GPS data analysis and pretreatment.This paper expounds the detection principle and data advantages of GPS,and puts forward the detailed steps and process of raw data preprocessing for GPS data redundancy and error problems.Finally,the technical process of GPS data association matching is given to obtain effective and complete visual data.(2)Research and analysis of transportation elements in major activity areas.Major activity definition and classification,based on traffic demand,traffic flow,traffic state three characteristics,analyzed the major areas of traffic characteristics,master the basic techniques of major events regional traffic status changes,finally,in combination with characteristics of big event attributes,traffic conditions,traffic flow parameters,and other factors,to analyze its impact on the major activities of running status of regional transportation.(3)Research on road traffic status identification methods in major activity areas.The spatiotemporal characteristic index to describe the change of traffic running state is put forward.Based on the wind rose diagram,the congestion changes in major activity areas are represented from the congestion distance,congestion diffusion intensity and congestion diffusion speed,and a complete congestion distance determination algorithm is given.Pearson correlation coefficient is introduced to quantify the congestion degree of major activity area,and a method to express the traffic state by the approximate degree of the current path to the congestion state in its limit state is proposed.(4)Research on traffic situation inference method in major activity area.Using the Gaussian plume model,the traffic congestion diffusion is compared to the gas diffusion in the atmosphere,and the traffic congestion limit distance in the major event area is predicted.Based on the BP neural network algorithm,the historical road delay index is studied and trained to predict the future road congestion degree.Finally,considering the release of the travel guidance scheme and changing the traffic demand,the method of predicting the path traffic running state is established based on the flow speed fitting.(5)Empirical research.This chapter selects Chongqing Yubei District Yuelai International Museology Center and its influence area to carry out empirical research.The traffic congestion distance,congestion intensity and congestion diffusion speed in Yuelai area are calculated based on the wind rose diagram idea.The congestion degree in Yuelai area is calculated based on Pearson correlation coefficient,and the proximity degree between real-time congestion and the limit value is calculated.The Gaussian plume model is used to predict the congestion distance of each path.The prediction accuracy considering the congestion diffusion coefficient reaches 87.51%,and the prediction accuracy considering the limit influence of each path reaches 89.30%.The prediction accuracy of the two methods is high,and the prediction accuracy of the latter is more accurate.BP neural network in Matlab was used to predict the road delay index,and the average prediction error was 0.1116.The traffic situation after the launch of the travel guidance scheme was calculated.