Short-term Traffic Flow Forecasting Model of Elman Neural Network Based on Dissimilation Particle Swarm Optimization

Typical main multi- intersection of urban road is researched in this paper. Since traffic flow has the property of periodicity and randomicity, a dynamic recursion network, which called Elman neutral network model, is presented. Compared with other static neural network model, the model has the ability to adapt the time-varying and can approximate the dynamic system more dramatically and directly. Dissimilation particle swarm optimization (DPSO) algorithm is used to determine the parameters of the model respectively while it has solved the defects such as prematurity of traditional PSO. In particular, our experiments show that the method can both enhance training speed and mapping accurate than other algorithms. The simulation results of traffic flow collected from Chinese national urban road show that the model has greater efficiency and better performance.