Study on Short-Time Traffic Flow Forecasting Methods

This thesis introduces the forecasting methods of domestic and foreign road traffic flow, analyzes the advantages and shortcomings of all sorts of traffic flow forecasting methods and the actual forecasting effects. For the complexity of the urban traffic, the precision of some current traffic flow forecasting methods is not high. With respect to these questions, this thesis applies the chaotic neural network to establish the chaotic neural network forecasting model of traffic flow of urban intersection exit. Compared with the forecasting results obtained by the traditional BP neural network and exponential smoothing method, it is showed that such model has highly good effect.