State-of-health analysis applied to spacecraft telemetry based on a new projection to latent structure discriminant analysis algorithm

The potential for space mission operations and the supporting ground infrastructure is growing dramatically, fueled by new technologies, but with that growth comes increased complexity, and daunting reliability and security challenges. And like most complex endeavors, space operations are being asked to do more with less. In order to deliver cost-effective space operations services, researchers must explore novel ways to build and operate systems under study. Innovation is the engine that drives progress in today´s high-tech global economy. Statistical multivariate latent techniques are one of the vital learning tools that are used to tackle the aforementioned problem coherently. There has been a tremendous increase in the volume of telemetry data over the last decade from contemporary spacecrafts. All these datasets need to be analyzed for finding interesting patterns or for searching for both moderate and significant outliers. Information extraction from such rich data sources using advanced statistical methodologies is a challenging task due to the massive volume of data. To solve this problem, in this paper, we present a novel supervised learning algorithm based on projection to latent structure discriminant analysis technique (PLS-DA). The algorithm is particularly uses to model, analyze, classify telemetry data and identify key contributors to anomalous events while simultaneously measuring several predictors and response variables. The performance of the algorithm using the telemetry acquired from of attitude determination and control system (ADCS) of actual remote sensing spacecraft was presented. In addition, a critical compression between the analysis results obtained by the algorithm and the multivariate statistical analysis software Simca-P developed by Umetrics was presented. Finally, the algorithm provides competent information in modelling, classifying, diagnosis and prediction as well as adding more insight and physical interpretation to the ADCS s- ate of health (SOH).