基于GA-BP模型的铁路货运量预测

针对于目前已有铁路货运量预测方法的缺陷与不足,提出基于遗传算法和神经网络的混合预测模型对铁路货运量的预测方法进行改进优化,目的保证其预测精度.首先引用灰色关联分析法,以此来确定全国铁路货运量与其主要影响因子之间的关联度,然后按照其关联度在标准值之上的关联因子,建立GA-BP神经网络预测模型.最后通过实例分析表明,此模型预测精度及推广能力均优于传统的预测方法,从而证明该方法的可行性和有效性.

Railway Freight Volume Forecasting Method Based on GA-BP Model

Because of deficiencies and shortcomings of existing rail freight prediction methods,an improved hybrid prediction model based on genetic algorithm and neural network is presented for predicting the railway freight volume,aiming to ensure the accuracy of the forecast.Firstly,the gray correlation analysis is introduced to determine the correlation of the national rail freight and its main impact factors,then GA-BP neural network prediction model is established according to their correlation factors associated with above standard values.Finally,the case analysis shows that the prediction accuracy and generalization ability of this model is superior to the traditional forecasting method,and the feasibility and effectiveness of this method are proved.