The neural network technology applied on fault detection of locomotive converter

The paper analyzes the fault principle of the SS7E locomotive converter, and classifies the fault sorts of it. The paper describes a method of adopting wavelet analysis to extract energy eigenvector from the output voltage wave of the converter, and proposes a fault diagnosis method of the converter by using energy eigenvector and neural-network technology. The hardware frame and software design of the fault diagnosis system are discussed and the application scheme to diagnose the specific fault position of converter online by increasing a few monitoring points is proposed. The validity of the method is proved by computer simulation.