关键词: 短时交通流, 预测模型, 模糊神经网络, 粒子群算法, 蚁群算法

Abstract: Real-time and accurate traffic flow prediction is very important to the intelligent traffic guidance and control.According to the characteristics of short-time traffic flow，a fuzzy neural network model has been proposed to solve short-time traffic flow prediction.The paper combines Particle Swarm Optimization（PSO） algorithm with ant algorithm for training the fuzzy neural network.The algorithm is formulated in a form of hierarchical structure.The master level is ant algorithm and slave level is PSO.The global search is performed at the master level，while the local search is carried out at the slave level.The simulation results demonstrate the proposed model can improve prediction accuracy，compared with BP based training techniques.

Key words: short-time traffic flow, prediction model, fuzzy neural network, Particle Swarm Optimization algorithm, ant algorithm