新型组合模型在铁路客运量预测中的应用

客运量是用来测算交通运输业所承担的工作量,反映了运输业为国民经济和人民生活服务的数量指标,准确的客运量预测直接影响到铁路项目的经济效益评价及铁路交通组织安排。根据客流量数据的特点,提出新的组合预测方法,构建线性时间序列灰色GM(1,1)模型和考虑客流量影响因素的非线性遗传算法优化BP神经网络模型。最后结合新建兰州至中川机场铁路项目及调查数据进行客流量的预测研究,并将组合模型预测结果和单一模型相比,得出新型线性和非线性组合模型预测精度更高,取得了满意的效果,为客流量的预测提供了一种新的工具。

A New Combination Model for Forecasting Railway Passenger Volume

Passenger traffic is used to measure transportation workload,reflecting the transportation service level for the national economy and people's living index. Accurate forecast of passenger traffic directly affects the evaluation of economic benefits of railway project and railway traffic organization arrangement. According to the characteristics of the traffic data, this paper puts forward a new combination forecast method,the establishment of the linear time series grey GM( 1,1) model and the nonlinear genetic algorithm to optimize the BP neural network model that considers the influence factor of traffic. Finally,the combination model is illustrated by Lanzhou to Zhongchuan airport railway new project and the survey data. The results are compared with the single model,concluding that the new model of the linear and nonlinear combination forecasting is higher in accuracy with satisfactory results and ideal for predicting passenger traffic.