Short-Term Traffic Congestion Prediction Model Based On Deep Learning

With the progress of the urbanization process,a series of traffic problems have occurred because of the growing urban population and the far lower growth rate of roads than the growth rate of car.One of the most prominent problem is the traffic congestion problem.The prediction of traffic congestion is the key to alleviate traffic congestion.For the sake of guaranteeing the real-time performance and accuracy of the traffic congestion prediction,this paper proposed a short-term traffic congestion prediction model based on deep learning.By processing a mess of urban taxi transportation data,this paper extracts traffic flow and average speed which are the most important parameters in the assessment of traffic congestion.After analyzing the temporal and spatial distribution characteristics of the traffic flow and average speed,this paper have proposed the short-term traffic flow prediction model based on a deep learning method Stacked Auto Encoder(SAE)and the short-term average speed prediction model based on SAE.By comparing the other traffic flow forecasting methods and average speed forecasting methods,the methods proposed by this paper have improved the accuracy rate.For traffic congestion recognition,this paper used a model that based on three parameters(average speed,traffic flow and density)which uses standard function method to standardize the parameters and calculate the congestion comprehensive threshold to determine the congestion level by the range of value.Combining these two SAE prediction models mentioned above,a traffic congestion prediction method based on deep learning was proposed.At last,this paper implement the simulation verification of the proposed traffic congestion prediction method.According to the experimental result,the prediction model based on traffic volume and average speed proposed in this paper got higher accuracy than other prediction model.In the aspects of traffic congestion prediction,this method can acquire satisfied accuracy rate and real-time performance.