A fault diagnosis approach for railway track circuits trimming capacitors using EMD and Teager Energy Operator

The trimming capacitor is an important component of a track circuit used for train detection in the railway train control system. In order to ensure the required dependability and availability levels, their working conditions shall be monitored in a timely and efficient manner. This paper presents a diagnosis approach of trimming capacitors based on Empirical Mode Decomposition (EMD) and Teager Energy Operator (TEO) theory. The EMD is used to decompose the short-circuit current signal into two parts. One is the sum of a set of Intrinsic Mode Functions (IMFs), which feature the fault information of the defective trimming capacitors; the other is the residual element relating to the signal variation trend. Then the TEO of IMFs is performed and the instantaneous frequencies are derived, so as to highlight the fault features. Experiments with simulated data show that multiple defects can be detected effectively using this approach. Compared with the existing methods which are only able to detect one failed capacitor, the integrated method represents a rather improved performance.