Dynamic Time Warping and Spectral Clustering Based Fault Detection and Diagnosis of Railway Point Machines

Point machine plays a very important role in the metro operation. Among these infrastructure failures in urban rail transit systems, the vast majority of them are triggered by railway point machines. Thus, fault detection and diagnosis should be well concerned to ensure traffic safety. In this paper, we propose to employ dynamic time warping and spectral clustering to handle this problem. Firstly, the dynamic time warping method is used to compare unequal sequences with phase-shifted shape. Secondly, the spectral clustering method is applied to deal with the classification problem without training steps. At last, simulation results demonstrate well performance of the proposed scheme.