基于BP神经网络的新建支线机场货邮量综合预测

BP Neural Network Helps to Comprehensively Predicting Cargo Volumes of the Newly Constructed Airports

新建支线机场的货邮量预测对合理规划其航空物流体系具有重要意义。首先分析了支线机场货邮量的影响因素,然后用因子分析法对BP神经网络输入层进行影响因素(指标)进行约简,以减少BP人工神经网络的复杂度,在此基础建立了拟合模型对新建支线机场历史航空货邮量进行拟合,最后建立BP三层神经网络模型对新建机场航空货邮量量进行预测。以江苏省淮安市新建支线机场以例进行实证研究,结果表明该模型有较高精度。

For reasonably regulating the air logistics systems of the newly constructed regional airports, it is of great importance to predict their respective cargo volumes. This essay flint clarifies what factors decide the cargo volume of a regional airport, then applies factor analysis to simplifying the influential elements of input parameter of BP neural network so that the artificial BP neural network is simplified. This model can be adopted to build up a similar model that cargo volume of a newly constructed regional airport bears. At last, the article shows the predicted result, which is exemplified by the ease of Huai＇an. Jiangsu. and the result indicates that model are of high exactitude.