摘要
针对铁路货运量的主要13项内部影响因素,以2004~2014年间11年的铁路运输指标统计数据为训练样本,采用BP神经网络建立这些因素与货运量的映射关系,再根据该映射以权积法求解货运量对各因素的敏感度系数,从而定量计算出各项因素对铁路货运量的影响程度。研究结果表明:国铁正式营业里程、货车保有量、货运密度和货车机车平均牵引总重这4项因素对货运量的影响要显著大于其他因素,若需在货源充足的情况下提高货运量,则以上4项因素是需要着重优先考虑的因素。
Taking 13 internal factors of railway system as main factor for freight volume, railway transportation index statistical data in the year 2002 to 2012 as training sample, then the mapping relationship between these factors and freight volume can be established by BP neural network. Then, the sensitivity coefficient of freight traffic to can be calculated, this coefficient can be used to measure the influence of various factors on railway freight volume. The results show that the 4 factors, freight mileage, freight car ownership, freight density and average traction gross, are more important than other factors. If we need to increase the volume of freight in sufficient supply, the above 4 factors need to be given priority.
作者
颜保凡
郭垂江
李夏苗
YAN Baofan;GUO Chuijiang;LI Xiamiao(School of Transportation Management, Hunan Vocational College of Railway Technology, Zhuzhou 412000, China;School of Traffic＆ Transportation, Central South University, Changsha 410075, China)
出处
《铁道科学与工程学报》
CAS
CSCD
北大核心
2018年第5期1341-1346,共6页
Journal of Railway Science and Engineering
基金
湖南省教育科学"十三五"规划课题资助项目(XKJ17BZY040)
湖南省高校科研资助项目(15C0908)
关键词
铁路货运量
内部影响因素
BP神经网络
敏感度分析
railway freight volume
internal influencing factors
BP neural network
sensitivity analysis
