基于LSTM网络的铁路货运量预测Railway Freight Volume Prediction Based on LSTM Network
程肇兰,张小强,梁越
摘要(Abstract)：
准确预测铁路货运量对铁路货运组织工作的开展极为重要,特别是短期(月、日)货运量数据直接关系到铁路各项运输计划的编制。人工神经网络模型因其强大的学习能力而被广泛运用于各领域的预测,其中的LSTM网络适合处理和预测铁路货运量这类间隔和延迟相对较长的时间序列。考虑不同时期货运数据的特点分别建立基于月货运量数据的LSTM多变量预测模型和基于日货运量数据的LSTM时间序列模型。基于广铁2010—2017年的货运量数据,运用所建模型预测各月和每日的货运发送量,并与ARIMA模型预测方法和BP神经网络方法的预测结果相比较。结果表明,LSTM网络预测效果更佳。
关键词(KeyWords)： 铁路运输;货运量预测;LSTM网络;时间序列;深度学习
基金项目(Foundation): 国家铁路局科技开发项目(KF2019-101-B);; 成都市科技局软科学项目(2017-RK00-00050-ZF);; 广州铁路(集团)科技开发项目(2017K020)
作者(Author): 程肇兰,张小强,梁越
参考文献(References)：
- [1] 汤银英,李龙.基于Holt-Winters模型的铁路月度货运量预测研究[J].交通运输工程与信息学报,2017,15(2):1-5,13.TANG Yinying,LI Long.Prediction Research about Monthly Traffic Volume Based on Holt-Winters Model[J].Journal of Transportation Engineering and Information,2017,15(2):1-5,13.
- [2] KAMINSKI W.Comparison of Forecasting Methods for Freight Transportation by Polish Railways[J].Pollack Periodica,2019,14(2):51-60.
- [3] 安永娥,鲍学英,王起才.基于无偏灰色Verhulst模型的铁路货运量预测研究[J].铁道科学与工程学报,2016,13(1):181-186.AN Yonge,BAO Xueying,WANG Qicai.Railway Freight Volume Forecasting Based Onunbiased Grey Verhulst Model[J].Journal of Railway Science and Engineering,2016,13(1):181-186.
- [4] 张蕾,孙德山,张文政,等.基于灰色关联分析的支持向量机的铁路货运量预测研究[J].经济数学,2018,35(2):58-61.ZHANG Lei,SUN Deshan,ZHANG Wenzheng,et al.Research on the Rail Freight Volumes Forecast by Support Vector Machine and Gray Relational Analysis[J].Mathematics in Economics,2018,35(2):58-61.
- [5] 王巍,陈崇双,周宏宇,等.基于KS检验的铁路车站日常货运量预测[J].综合运输,2018,40(12):93-96,108.WANG Wei,CHEN Chongshuang,ZHOU Hongyu,et al.Daily Freight Volume Prediction of Railway Station Based on KS Test[J].Comprehensive Transportation,2018,40(12):93-96,108.
- [6] BENGIO Y,DUCHARME R,VINCENT P,et al.Neural Probabilistic Language Models[J].Journal of Machine Learning Research,2006,3:137-186.
- [7] HOCHREITER S,SCHMIDHUBER J.Long Short-term Memory[J].Neural Computation,1997,9(8):1735-80.
- [8] 伊恩·古德费洛,约书亚·本吉奥,亚伦·库维尔.深度学习[M].赵申剑,译.北京:人民邮电出版社,2017.
- [9] OLAH C.Understanding LSTM Networks[EB/OL].(2015-08-27) [2019-03-17].http://colah.github.io/posts/2015-08-Understanding-LSTMs/.
- [10] PEDREGOSA F,VAROQUAUX G,GRAMFORT A,et al.Scikit-learn:Machine Learning in Python[J].Journal of Machine Learning Research,2011,12:2825-2830.
- [11] 李高盛,彭玲,李祥,等.基于LSTM的城市公交车站短时客流量预测研究[J].公路交通科技,2019,36(2):128-135.LI Gaosheng,PENG Ling,LI Xiang,et al.Study on Short-term Traffic Forecast of Urban Bus Stations Based on LSTM[J].Journal of Highway and Transportation Research and Development,2019,36(2):128-135.
- [12] BRINK H,RICHARDS J,FETHEROLF M.Real-world Machine Learning:Model Evaluation and Optimization[M].Greenwich:Manning Publications,2016.
- [13] 张岄.铁路货运量预测及影响因素研究[D].北京:北京交通大学,2016.
- [14] 国家统计局.分省月度数据[EB/OL].(2018-09-09) [2019-04-20].http://www.stats.gov.cn/.
- [15] LO W W L,WAN Y L,ZHANG A M.Empirical Estimation of Price and Income Elasticities of Air Cargo Demand:the Case of Hong Kong[J].Transportation Research Part A:Policy and Practice,2015,78:309-324.
- [16] BROWNLEE J.Deep Learning for Time Series Forecasting[M].(2017-03-11) [2019-05-10].https://download.csdn.net/download/l1505624/9777540.
- [17] KINGMA D,BA J.Adam:a Method for Stochastic Optimization [EB/OL].(2017-01-30)[2018-09-23].https://arxiv.org/abs/1412.6980.
- [18] 汤银英,朱星龙,李龙.基于SARIMA模型的铁路月度客运量预测 [J].交通运输工程与信息学报,2019,17(1):25-32.TANG Yinying,ZHU XingLong,LI Long.Monthly Railway Passenger Traffic Volume Forecasting Based on SARIMA Model [J].Journal of Transportation Engineering and Information,2019,17(1):25-32.
- [19] 鲜敏.基于灰色神经网络的铁路客运量预测方法研究 [D].成都:西南交通大学,2016.