Real-time detection using wavelet transform and neural network of short-circuit faults within a train in DC transit systems

Abstract

A method is proposed for the real-time detection of DC-link short-circuit faults in DC transit systems. The discrete wavelet transform is implemented to detect any surges in the DC third-rail current waveform. In the event of a surge the wavelet transform extracts a feature vector from the current waveform and feeds it to a self-organising neural network. The neural network determines whether the feature vector belongs to a normal or a fault current surge.