Traffic Flow Forecasting Model of Signalized Intersections Based on Particle Swarm Optimization

This paper's research object is two typical adjacent intersections of city road network. A fuzzy neural network model based on particle swarm optimization to forecast the traffic flow is developed, aiming at the characteristic of the city road intersection having signal lamp control. The model comprises a fuzzy cluster module and several feed-forward neural network modules with a particle swarm optimization algorithm to optimize the weights of the NN and parameters of its activation functions. The simulation results show that the model is effective. Compared with BP NN, the model is more practical.