Neural Network Fault Diagnosis for Open Circuit Fault of Synchronous Generator Rectifier Element

Polyphase rectifier synchronous generator has been widely used in marine power system and marine electric propulsion because of its excellent power supply reliability and power quality. It is of practical significance to study its fault diagnosis technology. This paper takes the open-circuit fault of rectifier components of polyphase rectifier synchronous generator as an example. According to the fault mechanism, a fault diagnosis model is designed by combining the results of fault feature wavelet packet decomposition with SOM neural network, and compared with PNN neural network. The extracted fault features are subdivided by wavelet packet decomposition, and the detail spectrum features of the subdivided fault features are used in fault diagnosis and fault feature extraction based on intelligent algorithm, which improves the reliability and accuracy of fault diagnosis.