Study On Optimal Timing Of Intersection Based On Short-term Traffic Flow Forecasting

In recent years,with the rapid development of city traffic construction,intelligent transportation technology has attracted more and more attention,and it is considered a powerful tool to solve the city Atraffic problems.The traffic flow prediction is the key to implement intelligent traffic management,efficient operation of city traffic signal control system is based on accurate traffic prediction.In this dissertation,the optimal timing of intersection is studied based on the short-term traffic flow forecasting.Firstly,the characteristics of urban road traffic flow are introduced,and the correlation between traffic flow sequence of each section is analyzed;Due to the spatial distance between the detection sections,there must be a certain time difference between the sections.By calculating the correlation of time delay based on the traffic flow time series of each section,the traffic flow series with stronger correlation of traffic flow are selected as the training sample set of the prediction model.The BP neural network is used to establish the prediction model,and the actual road network data are analyzed to evaluate the accuracy of the model prediction.Then,this dissertation analyzes the existing HCM timing method,TRRL timing method and ARRB timing method,and the performance evaluation index of the intersection control scheme,a multi-objective optimization model is established.And the optimal allocation scheme is selected by the dimensionless coefficient and the weighted coefficient of variation of the total flow rate at the intersection.In order to obtain better timing scheme,multi-objective optimization are introduced in the original Dragonfly hybrid mutation operator algorithm,and adopts the dynamic external archive maintenance strategies to improve,finally using benchmark functions are used to test the improved algorithm proves that the improved method is effective and the performance of the algorithm improvement.Finally,the traffic simulation experiment is carried out based on the accurate short-term traffic flow forecasting.In this dissertation,the multi-objective timing model is analyzed about the average delay of vehicles,the average number of stops and the intersection of the total capacity of three indicators,under different flow rates.The validity of the timing model is verified by comparing with the Webster method.