Fault Diagnosis Method of Launch Vehicle Based on Deep Neural Network

This paper presents a fault diagnosis method of launch vehicle based deep neural network and long-term and short-term memory artificial neural network. This approach extracts fault features from each operation data of launch vehicle for fault diagnosis, which has the characteristics of high accuracy and real-time computing performance. Based on the DNN and LSTM, the prediction model of the characteristic signal of the launch vehicle is established, and the steps of fault diagnosis using the model are given. The effectiveness of the model is simulated and analyzed by using the actual faults of launch vehicle data. The simulation test results show that the accuracy of the model for fault diagnosis can reach 90%.