混沌粒子群优化神经网络在铁路客运量预测中的应用  

APPLICATION OF CPS OPTIMISED NEURAL NETWORK IN RAILWAY PASSENGER TRAFFIC PREDICTION

作　　者：赵清艳[1] 熊茂华[2] 

机构地区：[1]中山职业技术学院广东中山528404 [2]广州番禺职业技术学院广东广州511483

出　　处：《计算机应用与软件》2013年第6期225-227,242,共4页Computer Applications and Software

摘　　要：针对常规RBF神经网络在铁路客运量预测中存在的收敛速度慢、易陷入局部极小等缺陷,提出一种基于混沌粒子群优化RBF神经网络算法,实现对RBF神经网络参数进行优化,并对我国1985年-2008年铁路客运量数据进行仿真实验。仿真结果表明,该算法很好地解决常规RBF神经网络参数优化问题,提高了铁路客运量预测精度,预测结果对铁路企业的决策有更加实用的参考价值。Conventional RBF neural network has the defects of slow convergence, easy to fall into local minimum and so on in railway passenger traffic prediction. In light of this, we propose to optimise the parameters of the RBF neural networks based on chaotic particle swarm （CPS）, and carry out the simulation experiment on the railway passenger traffic of China during 1985-2008. Simulation results show that the proposed algorithm is a good solution for parameter optimisation problem of conventional RBF neural networks, and improves the prediction accuracy of railway passenger traffic. The prediction results has more practical reference value for the decision-making of railway enterprises.

关 键 词：铁路客运量 神经网络 预测 混沌粒子群算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]