| 摘要 |
A rail transportation fault diagnosis method and system based on time series analysis. The method comprises: 1) collecting the historical and real-time monitoring data of a rail transportation signal device; 2) on the basis of the historical data, generating an initial parametric model of a time series curve to obtain fault determining parameters; 3) on the basis of the real-time data, re-adjusting the parameters of the parametric model of the time series curve to obtain fault determining parameters in current environmental conditions; 4) by means of the fault determining parameters obtained in step 3), implementing fault early warning for the time series curve data in the current environmental conditions; 5) training a fault classifier using the fault determining parameters obtained in step 2), and inputting into said fault classifier the extracted curve model parameters of a fault curve in the early warning results of step 4) to obtain fault diagnosis results. The present method can effectively solve the problems of large workload, low efficiency, and high-level risk during manual diagnosis of railway signal system faults. |
