Expressway Traffic Flow Prediction And Traffic Status Determination Based On Travel Plan

With the rapid development of the economy and the expansion of urban population,the demand for transportation has increased sharply,and the spatial structure of the city has also undergone great changes.In this context,the problem of traffic congestion is becoming more and more serious,especially the expressway traffic which is the main artery of the city.At the same time,the continuous development of science and technology has gradually matured the big data,cloud computing,artificial intelligence and 5G networks required for smart transportation,which provides infinite possibilities for smart transportation solutions.For the sharing of big data,the government has actively issued relevant policy documents,which provides the possibility for the acquisition of massive travel plan data and corresponding research.This article is based on the National Natural Science Foundation of the mentor team on the idea of travel planning.In summary of the current situation of short-term traffic flow forecasting and analyzing the characteristics of traffic flow on urban expressways,a short-term traffic flow forecasting method considering travel planning is studied.The model algorithm adopts two neural networks optimized by the Sparrow Algorithm(SSA),and then proposes a traffic congestion discrimination method based on basic traffic parameters.The main contents are as follows:(1)Based on a full understanding of domestic and foreign literature on short-term traffic flow prediction,traffic operation status discrimination and other research on user travel planning data,the current research status is summarized and analyzed.(2)This paper studies the characteristics of traffic flow based on the urban expressway,mainly including the analysis of the relationship between traffic flow parameters and time correlation analysis,spatial correlation analysis,focusing on the urban expressway traffic speed-density fitting model and the time series autocorrelation of traffic flow data.(3)Acquisition and processing of traffic data.The data involved in this paper include travel plan data and road traffic speed historical data.Considering the multi-source and heterogeneous characteristics of travel plan data,the model algorithm of the paper is cross-validated by instance data and traffic simulation.The collected data is processed by outliers and data noise reduction,which can be used in the later traffic forecasting research.(4)Build a neural network prediction model based on travel planning.On the basis of considering the temporal and spatial correlation of traffic flow,adding the travel plan as an input variable,constructing a neural network,and using the sparrow algorithm for optimization,to realize the traffic prediction of a fast road section and improve the accuracy of the prediction.(5)In order to better characterize the urban expressway traffic operation state,this paper uses the fuzzy C-means clustering algorithm,and on this basis adds a kernel function to improve it,divides the urban expressway traffic state,and then evaluates the division results.To sum up,this paper proposes two models based on travel planning data,which can effectively improve the timeliness and accuracy of traffic flow prediction,and determine the traffic state based on the predicted data.For travel users,travel routes can be reasonably selected to save time;for traffic managers,effective traffic control and inducement measures can be formulated to avoid possible traffic congestion.