Sensor fault detection and isolation. A blind approach
Résumé
This paper considers an application of blind identification to fault detection and isolation. The key point is to use estimates of the unknown input of a process in order to detect the presence of fault. By using blind identification i.e. without knowing the structure of the model of the process and its parameters, it is shown how to use the data collected on the process to perform fault detection and isolation. As it is well known in the field of signal processing, the main assumption of blind identification relies on the statistical independence of the input signal of the process under consideration. Example is provided to illustrate the proposed approach even in the case where the previous hypothesis is not fully satisfied.
Domaines
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |