基于组合模型的铁路客运量预测

Railway Passenger Volume Prediction Based on Hybrid Model

为了提高铁路客运量的预测精度,针对单一铁路客运量预测模型以及传统组合预测模型的缺陷,设计了基于组合模型(ARIMA-LSSVM)的铁路客运量预测方法,采用ARIMA对铁路客运量的周期性变化特点进行建模预测,从整体上把握铁路客运量的变化特点,采用LSSVM对铁路客运量的随机性变化特点进行预测,采用具体铁路客运量预测实例对性能进行测试和分析。结果表明,ARIMA-LSSVM可以准确、全面描述铁路客运量的变化特点,提高了铁路客运量的预测准确性,预测结果可以为铁路管理者提供有价值信息。

According to the defects of passenger traffic volume of railway single prediction model and traditional combination forecasting model,in order to improve the prediction accuracy of railway passenger volume,the combined model（ARIMALSSVM）,aprediction method for railway passenger volume is designed.The railway passenger volume change periodic characteristics are predicted by ARIMA to grasp the whole railway passenger traffic volume,then the random variation characteristics are predicted by LSSVM for railway passenger volume forecast,at last passenger railway concrete is verified by test and analysis.The prediction results of the model show that the ARIMA-LSSVM model can accurate and comprehensive describe the railway passenger volume.The model improves the prediction accuracy of railway passenger volume,forecast results can help to provide valuable information to the railway management.