基于BP神经网络的厦门港集装箱吞吐量预测研究

厦门港是我国东南沿海的集装运输基本港口.集装箱吞吐量的预测是港口制定发展规划的重要依据.在研究了常用的输入数据归一化方法之后,提出了新的归一化方法.该方法能加快BP神经网络的训练速度并提高精度.运用BP神经网络建立了厦门港集装箱吞吐量预测模型,并计算出2020至2024的集装箱吞吐量预测值.无论从拟合值,还是预测值检验来看,该方法都具有很高精度.

Research on the Container Throughput Prediction of Xiamen Port Based on BP Neural Network

The Xiamen Port is the main trade port of container transport in the coastal area of southeast China.Prediction of container throughput is the important basis for making port development plan.The common methods of input data normalization in BP neural network are Researched,and a new method of normalization is presented.This new method can quicken up the learning of BP neural network,and improve the precision.A model for predicting container throughput is developed for the Xiamen Port based on BP neural network,and the container throughputs from 2020 to2024 are calculated.This method has a high precision in terms of either simulation,or the result of the prediction.