基于组合预测的港口货物吞吐量分析——以长三角港口群为例
Analysis of Port Cargo Throughput Based on Combination Forecasting:A Case Study of the Yangtze River Delta Port Group
邹荣妹 1兰国辉 1杨霞2
作者信息
- 1. 安徽理工大学,安徽 淮南 232001
- 2. 安徽理工大学,安徽 淮南 232001;淮南师范学院,安徽 淮南 232038
- 折叠
摘要
科学预判长三角港口群的发展趋势对加快建设交通强国、海洋强国的意义重大,而港口货物吞吐量的精准预测是促进港口发展的重要一点.为了提高预测的精准度,采取BP神经网络算法、灰色预测GM(1,1)模型和三次指数平滑法对2009-2021年长三角港口群的宁波舟山港、上海港、苏州港三大港口货物吞吐量进行分析,建立组合预测模型.结果显示,该组合预测模型降低了预测误差,提高了预测精准度,对长三角港口群发展具有一定的参考价值.
Abstract
Scientific prediction of the development trend of the Yangtze River Delta port group is of great significance for accelerating the construction of a transportation and maritime power,and the accurate and scientific prediction of port cargo throughput is an important factor in promoting port development.In order to improve the accuracy of prediction,this paper adopts BP neural network algorithm,grey prediction GM(1,1)model and cubic exponential smoothing method to analyze the cargo throughput of Port of Ningbo-Zhoushan,Port of Shanghai and Port of Suzhou in the Yangtze River Delta port group from 2009 to 2021,and establishes a combined prediction model.The results show that the combined prediction model re-duces prediction errors and improves prediction accuracy,which can provide certain reference value for the development of the Yangtze River Delta port group.
关键词
港口货物吞吐量/BP神经网络算法/灰色预测GM(1,1)/三次指数平滑法/组合预测Key words
port cargo throughput/BP neural network algorithm/grey prediction GM(1,1)/cubic exponential smoothing method/combination prediction引用本文复制引用
基金项目
安徽省教学研究重点项目(2021jyxm0351)
淮南市科技研发项目(2021A244)
安徽省高等学校人文社会科学研究项目(2021)(SK2021A0543)
出版年
2024
NETL
NSTL
万方数据