中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2015, Vol. 34 ›› Issue (5): 135-138.DOI: 10.3969/j.issn.1674-0696.2015.05.27

• 交通运输及管理工程 • 上一篇    下一篇

基于灰色神经网络的港口集装箱吞吐量预测模型研究

张树奎1,2,肖英杰2,鲁子爱3   

  1. 1.江苏海事职业技术学院 航海技术系,江苏 南京 211170;2.上海海事大学 商船学院,上海 201306;3.河海大学 港口、海岸与近海工程学院,江苏 南京 210098
  • 收稿日期:2014-04-24 修回日期:2014-07-08 出版日期:2015-11-04 发布日期:2015-11-04
  • 作者简介:张树奎(1973—),男,安徽阜阳人,副教授,博士,主要从事港口、海岸与近海工程及航海技术方面的研究。E-mail: zhangshkfy@163.com。
  • 基金资助:
    2013年度江苏省教育教学研究课题项目(ZYB210);2013年交通运输职业教育科研立项项目(2013A03)

Prediction Model of Port Container Throughput Based on Grey Neural Network

Zhang Shukui1,2, Xiao Yingjie2, Lu Zi’ai3   

  1. 1. Department of Navigation, Jiangsu Maritime Institute, Nanjing 211170, Jiangsu, China; 2. College of Merchant, Shanghai Maritime University, Shanghai 201306, China;3. College of Harbor, Coastal & Offshore, Hohai University, Nanjing 210098, Jiangsu, China
  • Received:2014-04-24 Revised:2014-07-08 Online:2015-11-04 Published:2015-11-04

摘要: 为降低港口集装箱吞吐量的预测误差,提高预测精度,在分析传统的灰色预测模型和BP神经网络预测模型的优缺点的基础上,构建了灰色神经网络港口集装箱吞吐量预测模型,该模型充分发挥了灰色模型所需初始数据少和BP神经网络非线性拟合能力强的特点。以实际数值作为初始数据,各种灰色模型的预测值为神经网络的输入值,神经网络的输出值为组合预测结果。通过实例分析,结果表明:灰色神经网络预测模型提高了预测精度,预测结果比较理想,优于单一预测模型,因此,该模型用于港口集装箱吞吐量预测是可行的、有效的。

关键词: 交通运输工程, 吞吐量, 预测, 灰色模型, 灰色神经网络

Abstract: In order to reduce prediction error of port container throughput and improve its prediction accuracy, a Grey neural network model of port container throughput was constructed after analyzing the advantages and disadvantages of the conventional Grey model and BP neural network model. The new model made full use of the characteristic of low data demand of Grey model and strong nonlinear fitting ability of BP neural network. The actually measured values were used as the initial data, and various prediction values of Grey model were used as input data of neural network and final output data of neural network was used as combination prediction result. A case study shows that the Grey neural network model can offer improved prediction accuracy and ideal prediction results, which is better than single forecasting model. Therefore, it is feasible and effective to predict port container throughput by Grey neural network model.

Key words: transportation engineering, throughput, prediction, grey model, grey neural network

中图分类号: