基于GM(1,1)-SVM模型的货运量预测研究

Research on Freight Volume Prediction Based on GM(1,1)-SVM Model

在线阅读下载PDF

导出

摘要针对目前铁路运输货运量预测在精度方面的不足,考虑铁路货运量的非线性特征,基于灰色模型(Grey Model(1,1),GM(1,1))、支持向量机模型(Support Vector Machine,SVM),建立灰色支持向量机模型(GM(1,1)-SVM)。以2021—2023年西部陆海新通道各个季度铁海联运的货运量作为原始数据进行灰色预测,将GM(1,1)模型预测结果作为SVM模型的输入变量,通过数值模拟对比研究GM(1,1)模型、SVM模型、GM(1,1)-SVM模型预测结果,通过MAE、MBE、RMSE指标检验模型精度。结果表明:GM(1,1)的误差最大,分别为3755.600、11188.000、50063.749,其次是SVM模型,其误差为1933.386、-680.045、1934.501,GM(1,1)-SVM模型的误差最小,分别为867.693、82.456、867.922,提高了模型的预测精度。 Aiming at the shortcomings of the current railroad transport freight volume prediction in terms of accuracy,this paper considers the nonlinear characteristics of the railroad freight volume,and establishes a gray support vector machine model based on the GM(1,1)model and the support vector machine model.Based on the freight volume of rail-sea intermodal transport in each quarter of New International Land-Sea Trade Corridor from 2021 to 2023 as the original data for gray prediction,the prediction results of the GM(1,1)model were used as the input variables of the support vector machine model,and the prediction results of the GM(1,1)model,the support vector machine model,and the gray support vector machine model were studied comparatively through numerical simulation.The model accuracy was examined by MAE,MBE,and RMSE indexes.The results showed that GM(1,1)had the largest errors of 3755.600,11188.000,and 50063.749,followed by the support vector machine model with errors of 1933.386,-680.045,and 1934.501,and the gray support vector machine model had the smallest errors of 864.693,82.456,respectively,867.922,which improves the prediction accuracy of the model.

作者周小琬赵影朱江洪杨强 ZHOU Xiao-wan;ZHAO Ying;ZHU Jiang-hong;YANG Qiang(School of Traffic and Transportation,Chongqing Jiaotong University,Chongqing,400074,China;School of Economic and Management,Chongqing Jiaotong University,Chongqing,400074,China)

机构地区重庆交通大学交通运输学院重庆交通大学经济与管理学院

出处《广州航海学院学报》 2024年第4期53-58,共6页 Journal of Guangzhou Maritime University

基金重庆市自然科学基金面上项目(2022NSCQ-MSX2158) 重庆交通大学科研启动项目(F1210016)。

关键词货运量预测 GM(1 1)模型支持向量机组合模型西部陆海新通道 freight volume forecasting GM(1,1)model support vector machine combinatorial model new international land-sea trade corridor

分类号 U2-9 [交通运输工程—道路与铁道工程]

作者简介第一作者:周小琬(2000-),女,四川德阳人,硕士生,主要从事集装箱货运量预测研究,E-mail:622220950139@mails.cqjtu.edu.cn;通信作者:朱江洪(1986-),男,重庆人,工学博士,讲师,主要从事水上交通安全与环境研究,E-mail:jhzhu@cqjtu.edu,cn。