摘要
为提高铁路货运量的预测准确性,运用灰色关联分析法,计算分析了与铁路货运量相关的主要社会指标,确定铁路货运量的影响因子分别为铁路运营里程、铁路电气化里程、铁路复线比重、公路运营里程、固定资产投资总额和钢材产量。将所确定的因子作为铁路货运量的预测指标,建立基于BP神经网络的铁路货运量预测模型,并对模型进行了应用测试。结果表明:BP神经网络模型具有较高的精度,最大相对误差为3.7%,平均相对误差为2.3%。该方法具有较快的收敛速度和较高的预测精度,可为我国铁路货运量的预测研究提供方法支撑。
In order to improve the forecast ability of railway freight volume,a gray correlation method is used. The predictors are railway operating mileage,railway electrification mileage,the proportion of double-track railway,highway operating mileage,total fixed asset investment and steel production. The prediction model of railway freight volume is establish based on the BP neural network,and then is verified with tests. The results show that railway freight volume can be predicted accurately by the model based on BP neural network. The maximum relative error is 3. 7% and the average relative error is 2. 3%. In addition,the proposed forecast method provides a better convergence rate and higher predicting accuracy and the predictive model can provide a method for railway freight volume.
出处
《江南大学学报(自然科学版)》
CAS
2015年第1期80-84,共5页
Joural of Jiangnan University (Natural Science Edition)
基金
陕西省教育科学"十二五"规划项目(SGH140790)
西安航空学院科研基金项目(2014KY1212)
关键词
铁路货运量预测
灰色关联分析
BP神经网络
railway freight volume
prediction grey relational analysis
BP neural network