Fault diagnosis method of launch vehicle based on sequential neural network

Traditional fault diagnosis technology or a single defect diagnostic approach cannot match the actual needs of specific and sophisticated large-scale systems such as launch rockets. The wide application of artificial intelligence technology provides a new means for fault diagnosis of complex large-scale systems. Combining the test data and fault diagnosis experience accumulated in the aerospace range with artificial intelligence technology has important significance and practical value for the optimization of the efficiency and accuracy of carrier rocket fault diagnosis. According to the requirements of fault diagnosis in the launch vehicle test and launch process, based on the analysis of the existing fault diagnosis methods in the launch vehicle test and launch process, a fault diagnosis method based on sequential neural network is proposed in this paper. The diagnosis approach is validated by using real-world measured data, demonstrating the validity of the fault diagnosis system’s architecture.