Research On Traffic Congestion Judgment Based On Short-term Traffic Flow Prediction

With the rapid development of the economy,the road construction measures are far from being able to meet the growing needs of travel.Therefore,traffic congestion and large area paralysis often occur in many large cities.Traffic congestion status determination,as the core content of intelligent traffic research,is a reliable basis for preventing traffic congestion,reducing traffic pollution and so on.It is also a necessary means and method for traffic information service and traffic science management.Urban road traffic system is a complex system which is changing at all times.Its operation behavior is difficult to predict.Therefore,it is necessary to use the historical traffic data effectively and predict the traffic data in real time to get better application of the intelligent transportation system.This paper mainly studies the prediction model of the urban short time traffic flow and the decision model of urban traffic congestion,and applies the model to the example of the traffic flow.It is expected that these forecasting methods can be combined with the judgment of road congestion to apply to the actual traffic flow guidance and control system.The specific contents are as follows:(1)from the background and significance of the research,the importance of the short-term traffic flow prediction in the field of intelligent transportation is deeply analyzed,and the development and present situation of the traffic flow prediction method are elaborated in detail.(2)the types and types of data error are described.Aiming at the processing of fault data,this paper uses the method of threshold and traffic flow mechanism to deal with the data to meet the accuracy requirements of the next traffic flow prediction.(3)it introduces Gauss process regression and neural network related knowledge,and uses Gauss process regression and BP neural network model to predict the traffic flow of Dalian East fast road.(4)the road congestion is judged by example,and the traffic data collected by radar and card mouth are determined by fuzzy comprehensive evaluation method.From the traffic flow forecast,the thesis has completed four links from the theoretical analysis,the data processing,the model construction to the model simulation.At the same time,in view of the research on the short-term traffic flow prediction,the paper introduces the fuzzy comprehensive determination method to judge the urban traffic condition,in order to provide the urban traffic management.Reliable results of traffic situation assessment.