Research On Characteristic Parameter Prediction Of Urban Road Traffic Flow Based On RFID Electronic License Plate Data

Accurate and comprehensive understanding of traffic flow characteristics is the basis for effective control of urban road networks.Due to the limitation of traffic information sensing means,the analysis and prediction of traffic flow characteristic parameters in the past are mostly based on single parameters and single road segments.It is impossible to obtain a comprehensive traffic state description from multiple sides.They also ignored the impact of the road network related sections on the traffic flow characteristic parameters of the target section.RFID electronic license plates can collect single vehicle information and cover a wide range of vehicles,which makes up for the limitations of the above-mentioned collection methods,and brings new opportunities for more comprehensive analysis and prediction of traffic flow characteristic parameters.Under the new data conditions,it is of great significance to explore accurate and reliable traffic flow characteristic parameter prediction methods to improve urban service level,alleviate traffic congestion and ensure traffic safety.Based on the data of RFID electronic license plate data,this paper establishes a method for predicting the characteristic parameters of urban road traffic flow based on RFID electronic license plate data based on the factors affecting the accuracy of urban road traffic flow characteristic parameters.This paper focuses on the judgment analysis of slow-moving vehicles,the spatio-temporal correlation analysis of traffic flow characteristic parameters and the prediction of traffic flow characteristic parameters.It mainly includes the following contents:Based on the data of RFID electronic license plate data,this paper establishes a method for predicting the characteristic parameters of urban road traffic flow based on RFID electronic license plate data based on the factors affecting the accuracy of urban road traffic flow characteristic parameters and the characteristics of RFID electronic license plate data.The key points are to realize the judgment analysis of the slow-moving vehicle,the spatio-temporal correlation analysis of the traffic flow characteristic parameters and the prediction of the traffic flow characteristic parameters.It mainly includes the following contents:(1)Extraction of traffic flow characteristic parameters.Road traffic conditions and traffic efficiency are greatly affected by the proportion of slow-moving vehicles,and the greater the proportion of slow-moving vehicles,the greater the impact.In order to obtain a clear definition of slow-moving vehicles,this paper focuses on the analysis method of judging the slow-moving vehicles.Firstly,this paper extracts the travel time data of all vehicles in the fixed section,and determines the observation value of the judgment analysis according to the statistical characteristic parameters of the travel time for different vehicle types.Then the improved TOPSIS model is used to analyze the extracted observations.The Lagrangian method is used to obtain the optimal weight of the “evaluation analysis”,and the type of the slow-moving vehicle is obtained.(2)Temporal and spatial correlation analysis of traffic flow characteristic parameters.At the spatial level,considering the correlation analysis of single parameters can not fully reflect the correlation between multiple parameters,this paper establishes a comprehensive correlation analysis model and obtains typical correlation variables.The correlation between the two groups of variables is obtained from the correlation coefficient between typical variables,which proves the rationality of using the relevant traffic parameters of the road segment to predict the traffic flow characteristic parameters of the target road segment.At the time level,this paper aims at maximizing the correlation coefficient between two pairs of typical variables,and establishes the target planning model.Finally,the maximum correlation time delay of the two traffic flow characteristic parameters is obtained.(3)Traffic flow characteristic parameter prediction method.Many studies do not classify discussions based on actual traffic characteristics when predicting traffic flow characteristic parameters,and only pursue improvements in algorithm models.In order to avoid this situation,this paper first establishes an adaptive fading Kalman filter model for steady state traffic.A wavelet neural network model is established for unsteady traffic,and then the two models are combined to establish the characteristic parameters forecast model of urban road traffic flow adapted to various traffic characteristics by using the combination of 0-1 planning and average linear programming.Finally,we prove the superiority of the model by comparing the experimental results.