A big data framework for spacecraft prognostics and health monitoring

Abstract

The capability of prognostics and health monitoring of spacecraft such as satellites and spaceships is of key importance for their correct operation and the success of their mission. Before launching, various experiments are usually carried out on the spacecraft to thoroughly examine its performance and reliability, resulting in a large amount of experiment/test data. During its operation in orbit, the spacecraft will periodically generate operation data which can be collected by ground stations. The analysis and management of the spacecraft experiment and operation big data are essential for the monitoring of the system health and the prediction of potential failures. In this paper, we propose a complete framework for the collection, cleansing, storage, analysis and visualization of the spacecraft experiment and operation big data. All components of this framework such as the database, webserver and middleware are built on mature open source software frameworks. This big data framework has been deployed on a number of Linux servers and operates well in production.