基于协整理论的接触网故障强度回归分析

接触网作为一种沿铁路线上空架设的户外供电装置,其结构复杂,工作环境恶劣,且为无备用系统,极易出现故障,影响列车的运行安全。通过对接触网故障进行回归分析可以掌握其故障发生的规律以预测未来趋势。本文基于协整理论的多元时间序列回归模型(ARIMAX模型)对接触网的故障数据进行预测分析,通过对输入和输出的时间序列数据进行协整检验,避免了虚假回归,得到了较好的预测结果。预测结果表明,ARIMAX模型对比ARIMA模型和SARIMA模型的预测效果更好,且该模型对短期内的预测效果较好,时间步长越长,预测效果越差。预测结果可为实际线路的接触网故障检修工作提供指导。

As an outdoor power supply device,installed along the railway,without the backup system,the overhead contact line(OCL)is liable to be faulty due to its complex structure and severe working environment,which will impact the train’s safety running.The regression analysis of OCL faults may grasp the regularities of occurrence of the faults so as to predict the future tendency.The paper illustrates that the multiple time series regression model(ARIMAX model)based on the cointegration theory is used to predict and analyze the fault data of OCL,and the cointegration test of input and output time series data may avoid the pseudo return and the better prediction results may be obtained.The prediction results show that the prediction results of the ARIMAX model is better than that of the ARIMA model,especially within the short-term,the longer the time step is,the worse the prediction effect will be.The prediction results may provide instructions to the inspection,repair and maintenance of the faults on the OCL of the actual railway lines.