In this paper, we propose a new approach to spacecraft fault analysis based on an appropriate combination of data mining techniques. More precisely, we first transform the time series data representing the spacecraft parameters into points in a 9-dimension feature space, then we apply the K-means clustering algorithm so as to identify the typical parameter patterns. Finally, we adopt the Apriori algorithm to derive interesting association rules between typical patterns and the occurrence of spacecraft anomalies. We show the successful application of the found association rules to a benchmarking set of house-keeping telemetries (HKTMs).
Spacecraft fault analysis using data mining techniques
LAZZERINI, BEATRICE;
2005-01-01
Abstract
In this paper, we propose a new approach to spacecraft fault analysis based on an appropriate combination of data mining techniques. More precisely, we first transform the time series data representing the spacecraft parameters into points in a 9-dimension feature space, then we apply the K-means clustering algorithm so as to identify the typical parameter patterns. Finally, we adopt the Apriori algorithm to derive interesting association rules between typical patterns and the occurrence of spacecraft anomalies. We show the successful application of the found association rules to a benchmarking set of house-keeping telemetries (HKTMs).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
