Research On A Short-Trem Traffic Flow Prediction Method Based On Deep Learning

Today,the traffic jams in the city center are quite serious,which makes it more difficult for people to travel.Therefore,the short-term traffic flow prediction has received more and more attention and research from scholars.However,most of the traditional short-term traffic flow prediction researches predict single intersections without considering the correlation between the intersections.To solve this problem,a short-term Traffic Flow Prediction Method based on Spatial-Temporal Correlation(TFPM-STC)is proposed in this subject,and the research content of the subject are as follows:(1)The mean value processing method and threshold value processing method are used to deal with the missing values and outliers in the original traffic flow data,which has achieved the purpose of data cleaning.(2)Principal component analysis(Principal Component Analysis,PCA)is used to analyze the correlation of traffic flow between each intersection in the road network,and several intersections with greater correlation with the predicted intersection are selected for short-term traffic flow prediction.(3)Convolution-Gated Recurrent Unit(Conv-GRU)and Bidirectional Gated Recurrent Unit(Bi-GRU)are used to extract the spatial-temporal and periodic features of traffic flow,the spatial-temporal features and periodic features are fully integrated,and finally the traffic flow prediction results are obtained.(4)It can be found from the experimental results that the prediction results of the method proposed in the subject has smaller error,faster model convergence,and can adapt to the characteristics of rapid changes in traffic flow compared with othermethods.The data used in this subject comes from the real traffic data of Qingdao,rather than the existing data set on the Internet,which makes our experiment more authentic and objective.Moreover,the method proposed in this subject can fully adapt to different traffic scenarios and has certain portability.It can not only provide important basis for the timing of signal lights at city intersections and people's travel,but also play a vital role in the construction of smart cities.