Research on fault diagnosis method of data-driven launch vehicle control system

The launch vehicle control system is one of the key subsystems responsible for the state acquisition and the transmission and execution of control instructions during the rocket flight. Its reliability is of great significance to the success of the launch mission. At present, the traditional diagnosis method is mainly used for the fault diagnosis system of the control system, such as the expert system, which requires a large number of professional staff to perform manual interpretation. Whether it is from the needs of industrial development or the needs of national security, a more reliable, faster and smarter fault diagnosis system is the focus of the next development in the field of rocket testing. This paper conceives a fault diagnosis method for launch vehicle control system that combines deep learning and information fusion. It combines some time-consuming and laborious work in traditional fault diagnosis with artificial intelligence methods and historical data, using various data processing tools and feature extraction. Analysis of tools and integrated diagnosis at the feature level and decision-making level have improved the diagnostic efficiency and system reliability. The research results will effectively promote the automation and intelligent development of China's launch vehicle fault diagnosis.