基于改进灰色神经网络模型的三峡枢纽过坝货运量预测  

Prediction of freight volume through Three Gorges Dam based on improved grey neural network model

作　　者：柯桥 邓萍 KE Qiao;DENG Ping(School of Economics and Management, Chongqing Jiaotong University, Chongqing 400074, China)

机构地区：[1]重庆交通大学经济与管理学院,重庆400074

出　　处：《上海海事大学学报》2021年第1期82-87,共6页Journal of Shanghai Maritime University

基　　金：重庆市教育委员会人文社会科学研究项目(16SKJD17)。

摘　　要：针对三峡枢纽过坝货运量预测受多种因素影响及其具有的非线性特点,提出一种基于改进灰色模型和神经网络的组合预测模型。针对传统组合预测模型在赋权上的局限性,提出基于诱导有序加权几何平均(induced ordered weighted geometric averaging,IOWGA)算子的赋权方法。计算结果比较:组合预测模型的均方误差和均方百分比误差都比各单一预测模型的小。利用组合预测模型对2019—2022年三峡枢纽过坝货运量进行了预测,可为相关决策者提供参考。In view of the fact that the prediction of freight volume through Three Gorges Dam is influenced by many factors and is of nonlinear characteristic,a prediction method based on the combination of the improved grey model and neural network is proposed.Aiming at the limitation of the traditional combination prediction model in weighting,a weighting method based on the induced ordered weighted geome-tric averaging(IOWGA)operator is proposed.The results indicate that the mean square error and the mean square percentage error of the combination prediction model are smaller than those of each single prediction model.The freight volume through Three Gorges Dam of 2019-2022 is predicted by the combination prediction model,which can provide reference for relevant decision makers.

关 键 词：三峡枢纽 货运量 诱导有序加权几何平均(IOWGA)算子 组合预测 

分 类 号：U641.7+3[交通运输工程—船舶及航道工程]