Research on Automatic Fault Diagnosis of Environmental Control and Life Support Systems

摘要

To provide a lasting suitable cabin environment for the astronauts and payload specialists, it is necessary to achieve a long-term well running of environmental control and life support systems(ECLSS) of the space station. To resolve this issue, an effective means is to construct a model to carry out a real-time online diagnosis and decision support for fault. Based on the design for fuzzy automata to signal processing system, neural networks and expert systems principle, this paper designs a basic structure of the ECLSS's automatic fault diagnosis system (EAFDS), and gives a wavelet function to carry out an early diagnosis for weak fault. Moreover, this paper achieves an online automatic fault diagnosis to ECLSS by using the designed model. In the two-men-two-days simulated Shenzhou manned space flight test, the software of automatic fault diagnosis system works well, which has been assessed and verified comprehensively, and has provided a good technical support and security for the consummation of this flight. This designed automatic fault diagnosis system can judge data and fault diagnosis exactly and rapidly.