A novel approach of health monitoring and anomaly detection applied to spacecraft telemetry based on PLSDA multivariate latent technique

For any space mission operations, safety and reliability are the most important issues. Sophisticated and accurate fault detection and diagnosis of monitoring processes can minimize downtime, increase safety of space and ground operations, and reduce costs. To tackle this problem, system operations telemetry data was studied and analyzed to automatically characterize normal system behavior and anomaly detection and fault diagnosis methods for spacecraft systems based on multivariate latent techniques. In these methods, the knowledge or model which is necessary for monitoring a spacecraft system is acquired from the spacecraft telemetry data. In this paper, overview the anomaly detection - diagnosis problem in the spacecraft systems and modern techniques was discussed. Then explanation the concept of multivariate latent based approach was introduced. Furthermore, the results of applying dimensionality reduction algorithm to spacecraft telemetry using a novel technique called projection to latent structure discriminant analysis PLSDA were explained. Moreover, it compared with another multivariate technique principal component analysis PCA to provide robust information about the system's condition of the attitude determination and control subsystem ADCS of actual artificial satellite.