Transient Fault Diagnosis Algorithm based on Feature Classification and Its Application in High-Power Laser facility

High-power laser facility plays an import role both in scientific research and military application, and the occurred faults are usually transient. In this paper, a transient fault diagnosis algorithm based on feature classification is proposed, and the algorithm has been applied in the high-power laser facility. The method extracts the features of input data with an autoencoder network. And based on the extracted features, a multi-layer full connection neural network is adopted for fault diagnosis. By comparing the results with those obtained by traditional fault diagnosis methods, the performance of the algorithm has been verified.