Vehicle Commuting Characteristic Analysis And Mining Of Key Travel Path Based On Electronic Police Data

With the explosive growth of urban population and motor vehicle ownership in China,problems such as traffic accidents,traffic congestion,and environmental pollution are becoming increasingly serious,especially during morning and evening rush hours.This puts forward higher requirements for intelligent traffic control technology and systems.Peak hour traffic congestion is mainly caused by the excessive concentration of commuting vehicles in both time and space dimensions.Traditional information collection technologies cannot extract the spatiotemporal features of large-scale travel vehicles,making it difficult to support traffic control strategies and plan development at the road network level.In recent years,high-definition electric police systems have been widely used to capture illegal behavior of vehicles,providing new ideas for the research of intelligent traffic control technology at the road network level.To this end,this article conducts in-depth research on the analysis of commuting vehicle travel characteristics and the mining of key road sections in the urban road network to investigate the inherent correlation between electrical police data and traffic flow characteristics parameters.The main content is as follows:(1)Preprocessing and quality assessment of electronic police data.Based on the actual situation,this document integrated the actual situation to manage duplicate data,abnormal data,missing data and invalid data in the electric police data,and conducted a qualitative assessment of data integrity and accuracy,which ensured technical feasibility.accuracy of subsequent research content and research results.(2)Vehicle travel trajectory extraction and analysis of commuter vehicle characteristic Variables.According to the characteristics and principles of traffic flowed characteristic parameters,this paper extracted the travel time,average travel speed and vehicle travel chain of the road section on the basis of the electric police data.By combining the previously extracted travel time and speed,threshold parameters were set to achieve the extraction of travel trajectories.Then,combined with the road network topology,the TOPSIS(Technique for Order Preference by Similarity to Ideal Solution)method was used to complete the extraction of the travel chain,completing the complete extraction of the travel chain.By reflecting vehicle travel information through travel trajectories,the spatiotemporal characteristics of commuting vehicles were further obtained,and three feature variables were extracted and analyzed.(3)Characteristic clustering analysis of urban commuting vehicles.In this paper,by extracting three features in the spatiotemporal direction of commuter vehicles,K-Means Clustering Algorithm and Gaussian Mixed Model were used to cluster and analyze vehicle data with commuting characteristics in the electric police vehicle data.The results showed that both methods could effectively acquire vehicle information with commuting characteristics in the data source but with the same characteristics results in clustering of the GMM algorithm were more stable than the K-Means algorithms.(4)Excavation of key travel paths of the urban road network.Firstly,the FP-Growth algorithm was improved by optimizing the process of constructing FP-Tree(Frequent Pattern Tree).Using the improved FP-Growth algorithm,the key road segments of weekday peak commuters and weekday all day traffic were identified and corresponding association rules were analysed.Ultimately,on the basis of the results of the excavations,appropriate proposals were put forward for the organization and control of the movement.