Improved Particle Swarm Optimization for Fuzzy Neural Network Traning

Traffic flow prediction is one of the important components of ITS. However, satisfactory results of prediction cannot be obtained by classic mathematical methods. Fuzzy neural networks are extensively used in the prediction of traffic flow. In a fuzzy neural network, the optimization of each neuron and the connection weights between the layers is very critical. In this paper, an improved particle swarm optimization method is used to optimize the fuzzy neural network parameters. Simulation results show that this method can improve the efficiency of fuzzy neural network training and has good potential of generalization.