8th International Conference on Advanced Power System Automation and Protection (APAP2019)

1044 / 2019-05-19 10:55:26

Fault Identification of Hydroelectric Sets Based on Time-frequency Diagram and Convolutional Neural Network

hydroelectric sets , fault diagnosis , time-frequency transform , time-frequency diagram , convolutional neural network

Aiming at the poor generalization ability of traditional hydropower unit fault diagnosis methods, a fault diagnosis method for hydroelectric sets based on time-frequency diagram and convolutional neural network(CNN) is proposed. First, the hydroelectric sets vibration signal is time-frequency transformed to construct a time-frequency diagram. Then, combined with the convolutional neural network, the fault state identification of the hydropower unit is realized. The method realizes the automatic extraction of the texture features of the time-frequency diagram, avoids manual identification, and can quickly and accurately identify the state of the hydropower unit. The results show that the method can effectively identify the type of fault.