The Method Of Forecasting And Controlling The Ramp Flow Of Urban Expressway

With the construction of a large-scale expressway system among cities,expressway congestion has already become one of the main problems of urban traffic.In order to alleviate traffic jams and reduce a series of problems caused by queuing vehicles,it is necessary to deal with the traffic flow of expressway.Ramp control is an effective way to solve the problem of expressway congestion.By controlling the ramp traffic flow at a reasonable level,it could affect the interval density of the vehicles on the expressway.To control ramp traffic,accurate prediction of the traffic flow is an effective way,as clearing traffic congestion requires a certain degree of foresight.However,because the prediction process involves complex non-linear data models,it becomes difficult to accurately predict the traffic flow of expressway.This paper proposes a traffic flow prediction model SSVRCSA,which combines seasonal support vector regression model（SSVR）,chaos theory（C）and simulated annealing algorithm（SA）to predict expressway traffic flow.In addition,in order to solve the missing data problem caused by sensor failure or communication failure,that could affect the prediction accuracy,this paper uses a neural network-based traffic flow missing value filling algorithm to improve the accuracy of the prediction.In order to solve the problems of the traditional ALINEA ramp control algorithm,a new expressway singlepoint ramp control method based on ramp traffic prediction is proposed.The main contributions of this article are as follows:（1）In terms of the prediction of expressway ramp flow,a model combining support vector regression（SVR）and chaotic simulated annealing（CSA）is proposed.The SVRCSA model can describe the randomness in the ramp flow data.In addition,by adding the seasonal mechanism to deal with the problem of seasonal trend time series.By using a neural network-based traffic data missing value completion algorithm to fill missing data caused by sensor failures or communication failures,thus the traffic prediction accuracy is further improved.（2）A ramp control method based on the ALINEA algorithm is proposed for the single-point control of the expressway ramp.This method establishes a queuing by considering the determination of the ALINEA control threshold that can be inserted into the gap and improving the ALINEA hierarchical control method of queuing,so as to control the model to alleviate the problem of congestion on expressway.（3）Through comparison with existing methods,prediction and simulation experiments both have achieved better results.