基于改进粒子群优化算法的灰色神经网络的铁路货运量预测  

Railway freight volume prediction based on grey neural network with improved particle swarm optimization

作　　者：雷斌[1] 陶海龙[1] 徐晓光[1] 

机构地区：[1]兰州交通大学机电技术研究所,兰州730070

出　　处：《计算机应用》2012年第10期2948-2951,2962,共5页journal of Computer Applications

基　　金：甘肃省自然科学基金资助项目(1014RJZA042)

摘　　要：针对现有铁路货运量预测方法的不足,提出基于改进粒子群优化算法的灰色神经网络(IPSO-GNN)的铁路货运量预测方法,通过IPSO对常规灰色神经网络(GNN)的白化参数进行优化,改善了GNN的不足,保证了预测精度;同时利用灰色关联分析法,计算了铁路货运量和影响因素间的关联度,以最主要的6个关联因素,建立了基于IPSO-GNN的铁路货运量预测模型。仿真实验结果表明,在铁路货运量预测中此模型预测精度优于常规GNN及其他预测方法,说明此预测方法有效可行。Concerning the shortcomings of the methods which forecast railway freight volume, the paper proposed Grey Neural Network （GNN） based on the Improved Particle Swarm Optimization algorithm （ IPSO-GNN）. To make up for the shortfall of the conventional GNN and guarantee the prediction accuracy, it optimized the GNN whitening parameters through the IPSO. And it computed the railway freight volume and the correlation degree of influential factors. It built a railway freight volume model based on IPSO-GNN with six relating factors. The simulation results show that the prediction method is effective and feasible. The prediction precision of the given model in the railway freight volume forecast is better than those of the conventional GNN prediction method and other prediction methods.

关 键 词：铁路货运量预测 粒子群优化算法 灰色神经网络 灰色关联分析 BP神经网络 ELMAN神经网络 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]