Fault Warning Method of Guidance, Navigation & Control System based on Adaptive Correlation Network

As one of the key components of the spacecraft, it is crucial to ensure the safe operation of Guidance, Navigation & Control (GNC) system. Early detection of abnormalities plays a pivotal role in enhancing mission success rates and reducing operational costs. In this paper, a fault warning method of GNC based on association analysis, complex network and adaptive algorithm was proposed. First of all, a new correlation identification method was defined to construct a time-varying correlation network. Then, an improved particle swarm optimization algorithm was used to choose hyperparameters. Meanwhile, a comprehensive index reflecting the network topology was constructed to assess the abnormality of the GNC system. Finally, the real monitoring data of a certain case was used to verify the effectiveness of the proposed method.