您目前所在的位置:首页 - 期刊简介 - 详细页面

铁道科学与工程学报

JOURNAL OF RAILWAY SCIENCE AND ENGINEERING

第12卷    第5期    总第68期    2015年10月

[PDF全文下载]    [Flash在线阅读]

    

文章编号:1672-7029(2015)05-1227-05
基于灰色和神经网络的铁路客运量预测研究
冯冰玉,鲍学英,王起才

(兰州交通大学土木工程学院,甘肃兰州 730070)

摘 要: 准确的客流量预测在国家交通规划与管理中具有重要意义,预测方法的选择直接影响到预测的精度。客运量的预测具有小样本和非线性的特点。结合灰色理论和RBF神经网络的特点形成灰色-RBF神经网络模型,并采用客流运量分担率的方式对拟建铁路客流量进行预测。通过灰色理论对原始数据进行生成处理,将无规律的原始数据变为较有规律的生成数列,再利用RBF神经网络的超强适应能力和学习能力,大大加快学习速度并避免出现局部极小问题对生成数列进行预测,再将模型运用到客运量的预测中。最后结合新建兰州至中川机场铁路项目及调查数据进行客流量的预测研究,得出灰色-RBF神经网络模型对客流量具有很好的预测性。

 

关键字: 客运量;预测;灰色理论;RBF神经网络

Research of railway passenger volume forecast based on grey and neural network
FENG Bingyu, BAO Xueying, WANG Qicai

School of Civil Engineering , Lanzhou Jiaotong University, Lanzhou 730070, China

Abstract:Accurate forecast for passenger volume plays an important role in the national traffic planning and management. The accuracy of prediction is directly influenced by its prediction technique. There are two features of the prediction of traffic volume: small samples and nonlinear. According to the characteristics of the grey theory and RBF neural network, the grey RBF neural network model was formed in this paper. Moreover, the passenger volume was forecasted by virtue of the traffic share rate. The initial data was first generated and processed within the framework of grey theory to turn the erratic than the raw data into a regular sequence generation. By using the high adaptability and learning ability of RBF neural network, which can greatly accelerate the learning speed and avoid the local minimum problem to predict the generated sequence. At last, the high prediction of the grey-RBF neural network model for passenger traffic volume will be illustrated by Lanzhou to Zhongchuan airport railway new project and the survey data.

 

Key words: passenger traffic volume; prediction; grey theory; RBF neural network

ISSN 1672-7029
CN 43-1423/U

主管:中华人民共和国教育部 主办:中南大学 中国铁道学会 承办:中南大学
湘ICP备09001153号 版权所有:《铁道科学与工程学报》编辑部
------------------------------------------------------------------------------------------
地 址:湖南省长沙市韶山南路22号 邮编:410075
电 话:0731-82655133,82656174   传真:0731-82655133   电子邮箱:jrse@mail.csu.edu.cn