Research on EMU cable fault diagnosis based on deep belief network and signal domain conversion

Abstract

In order to realize the intelligent and accurate evaluation of EMU cable health, a cable fault diagnosis method based on signal domain transformation and deep belief network is proposed. Firstly, the collected partial discharge signal is transformed into signal domain after de-noising to realize the visualization of the collected signal and increase the amount of information contained in the signal; then the visualized signal is imported into the deep belief network for evaluation. The test data show that the proposed method can maintain high recognition rate for three typical cable faults, and is superior to other traditional fault diagnosis methods in recognition rate and time-consuming. The diagnosis method based on the combination of deep belief network and partial discharge signal domain conversion has a good engineering application prospect.