改进的BP神经网络在港口物流量预测中的应用  

Application of Improved BP Neural Network in the Prediction of Port Logistics Data

作　　者：林连[1] 林桦[1] 

机构地区：[1]武汉理工大学交通学院,武汉430063

出　　处：《交通信息与安全》2009年第5期161-165,共5页Journal of Transport Information and Safety

摘　　要：针对传统BP神经网络模型收敛速度慢的缺陷,对其进行改进,以提高收敛速度。经运用厦门港物流出口量的历史数据进行检验分析,给出BP神经网络仿真计算方法,其仿真结果与实际结果比较,具有较高的可信度。证明了改进后的模型加快了收敛速度,提高预测结果的准确性。Because of the limitation of traditional BP neural network models （NNM）, their convergence speed is generally slow. This paper is to modify and improve the BP NNMs, in order to enhance their convergence speed. The improved BP NNM is used to predict the historical export volumes of Xiamen Port and it is found that it has a very good accuracy through comparison of the simulation results and actual results. The prediction results show that the proposed model reduces the time required for the convergence and enhances the prediction accuracy.

关 键 词：港口物流量 BP神经网络 预测模型 

分 类 号：TP389.1[自动化与计算机技术—计算机系统结构]