Statistical fault isolation with PCA
(2001) Fault Detection, Supervision and Safety for Technical Processes - SAFEPROCESS 2000 — Location: Budapest, Hungary (14.June.2000)
Files
No attached file found for this publication.
Details
- Authors
- Abstract
- The paper considers a novel approach to isolating instrumentation faults. The key ideas involved are the existing connections between PCA and parity equations, the extension of PCA-like techniques to nonlinear cases and, finally, the use of the parametric statistical approach as a framework to detect and isolate faults. The methodology presented is based on a combination of PCA-based modeling techniques with the parametric statistical approach for residual generation and evaluation.
- Affiliations
Citations
Gomez, G., & Lendasse, A. (2001). Statistical fault isolation with PCA. In Edelmayer, A.M.; Banyasz, C.; (ed.), Fault Detection, Supervision and Safety for Technical Processes 2000(SAFEPROCESS 2000). Proceedings volume from the 4th IFAC Symposium (p. Vol. 3, p. 1029-34). Elsevier sci. https://hdl.handle.net/2078.5/230676