摘要
准确的区域铁路货运量预测在区域物流顶层规划、运输资源合理配置及其他物流辅助活动中起着重要的参考作用。针对LSSVM模型参数选择敏感和选择随意,且多输入条件下模型过程计算复杂的问题,提出一种融合PCA方法、WOA算法和LSSVM模型的区域铁路货运量预测新方法。采用PCA方法提取样本数据的主成分作为模型的输入,利用WOA算法全局搜索能力强、寻优效率高的优点对LSSVM模型的参数组合(λ,δ)进行寻优,得到基于改进LSSVM的区域铁路货运量预测模型。以陕西省2001—2019年与铁路货运量相关的18个指标数据作为样本,通过实际算例验证模型的预测性能。结果表明,所建模型的最大相对误差绝对值达到2.724%,相较于传统LSSVM模型和WOA-LSSVM模型降低了7.748%和3.589%,且模型的泛化能力和稳定性都得到了提升。
Accurate forecasting of regional railway freight volume plays a vital role in regional logistics planning,reasonable allocation of transportation resources,and other logistics auxiliary activities.Given the sensitive and arbitrary selection of parameters for the least square support vector machine(LSSVM)model and the complicated model calculation under multi-input conditions,a new method for forecasting regional railway freight volume was proposed,which combined the principal component analysis(PCA),whale optimization algorithm(WOA),and LSSVM model.PCA was utilized to extract the principal components of the sample data as the input of the model,and WOA with strong global search capability and high optimization efficiency was used to optimize the parameters(λ,δ)of the LSSVM model,thereby obtaining a forecasting model of regional railway freight volume based on improved LSSVM model.With the 18 indicators related to railway freight volume in Shaanxi Province from 2001 to 2019 as samples,the forecasting performance of the built model was verified.The results show that the absolute value of the maximum relative error of the built model reaches 2.724%,which is 7.748%and 3.589%lower than the traditional LSSVM model and the WOA-LSSVM model.The generalization ability and stability of the built model have also been improved.
作者
陈鹏芳
孟建军
李德仓
胥如迅
CHEN Pengfang;MENG Jianjun;LI Decang;XU Ruxun(Mechatronics T&R Institute,Lanzhou Jiaotong University,Lanzhou 730070,Gansu,China;Gansu Provincial Engineering Technology Center for Informatization of Logistics&Transport Equipment,Lanzhou 730070,Gansu,China;Gansu Province Logistics and Transportation Equipment Industry Technology Center,Lanzhou 730070,Gansu,China)
出处
《铁道运输与经济》
北大核心
2022年第2期59-65,共7页
Railway Transport and Economy
基金
国家自然科学基金项目(62063013,72061021)
国家铁路局科研课题(TYFH201931,TYSH202022)。
