| タイトル | Causal modeling and event-driven simulation for monitoring of continuous systems |
| 著者（英） | Wyatt, Jay; Robertson, Charles; Charest, Leonard, Jr.; Doyle, Richard; Rouquette, Nicolas |
| 著者所属（英） | Jet Propulsion Lab., California Inst. of Tech. |
| 発行日 | 1993-01-01
1993 |
| 言語 | eng |
| 内容記述 | We describe extensions to the traditional techniques for anomaly detection, as well as new anomaly detection techniques based on alternate models of what distinguishes normal from abnormal behavior. Some of these techniques are designed to capture anomalies at individual sensors; some detect anomalies across collections of sensors. To assist in reasoning about complex global behaviors, we construct and simulate a causal model of the physical system being monitored. These techniques have been tested on data from the ECLSS of Space Station Freedom (SSF) and are being applied in advanced monitoring prototypes for the SSF External Active Thermal Control System of SSF and the Environmental Emergency and Consumables Management subsystem of the Space Shuttle. |
| NASA分類 | COMPUTER PROGRAMMING AND SOFTWARE |
| レポートNO | 94A11449
AIAA PAPER 93-4525 |
| 権利 | Copyright |
|
