Anomaly detection refers to the problem of finding patterns in data that do not conform to expected behavior. Development of advanced anomaly detection and failure diagnosis technologies for satellite launch vehicle (SLV) is a quite significant issue in the aerospace industry, because the space environment is harsh, distant and uncertain. Current SLV health monitoring and fault diagnosis practices involve around-the-clock limit-checking or simple trend analysis using text or graphical displays on large amount of telemetry data. This procedure, which requires large numbers of human experts, is of course cumbersome and time-consuming. Furthermore, humans are not always able to recognize anomalous situations. In this paper, a novel knowledge-free anomaly detection method is proposed and developed for SLV based on three algorithms, K-NN, CART, and FCM, which constructs a system behavior model from the past normal telemetry data and monitors the current system status by checking incoming data with the model.

Login Required

Please login to access the full PDF of this article.

🔑 Author Login