A Combination Predicted Model of Short Term Traffic Flow

This paper proposes a combination forecasting model for short-term traffic flow based on a wavelet neural network. The model has three stages: the relevant forecasting variable to the traffic flow is selected by use data mining technology (such as the genetic algorithm); a training pattern of wavelet neural network that is similar to the forecast term is carried out by using data mining technology; and the wavelet neural network is then used to forecast the traffic flow. The authors used the traffic flow at Xinhua Street in Huhehot (China) to demonstrate that this model has a higher precision and reliability than the grey model and the BP artificial neural network model. The authors conclude that this wavelet neural network-based model provides a new, reliable, and effective strategy to forecast short-term traffic flow of nodes in urban road networks.

Language

  • English

Media Info

  • Media Type: Print
  • Features: References;
  • Pagination: pp 2075-2080

Subject/Index Terms

Filing Info

  • Accession Number: 01054245
  • Record Type: Publication
  • ISBN: 7560323553
  • Files: TRIS
  • Created Date: Jul 24 2007 1:37PM