Urban Road Traffic Flow Forecasting

On the back ground of the ITS(Intelligent Transportation System),how to make full use of the traffic data collected to the RTMS to forecast the traffic condition in the urban road,is meaningful and valuable in theory and practice to improve the efficiency of transportation,to save the traveler's time and to reduce traffic jam and traffic accident.The short-time traffic flow prediction is one of the chief problems needing to be solved in the field of traffic controlling,vehicle guidance and so on.It is important for us to study the method and theory on predicting the traffic condition in the future 5 minute or shorter.Well social prediction is valuable for alleviating traffic jam in the city and avoiding the resource wasting.Considering traffic flow in the intersection in the past certain times and in the adjacent intersection,this paper develops a mix predicted model which based on time series model for the first step prediction and neutral network for the second step prediction to predict the traffic flow in the two adjacent intersections of an urban street network.By analysis of the characteristics of the samples from a street corner transportation flow,the conclusion can be drawn that because the data of short time transportation flow is non-linear, time-variant and uncertain,using one estimate method can't meet the requirement of prediction accuracy.Accordingly,this thesis decomposes the transportation flow data in frequency domain. Firstly wavelet transformation issued to filter transportation flow,and then according to frequency the compositions of the transportation sudden change that results from uncertainty are decomposed to four frequency segments.After this procedure the signal is decomposed into one basic signal serial and four signal serials on different frequency,which are all steady signal serials.Then using the AR model on the data of different frequency makes the prediction.Then,all the predict results are added up for the first step prediction.At last,using neural network model on the data of the first step prediction and the adjacent intersection makes the second step prediction.By comparison it shows that the method obtains the high prediction result.