Robust frequency domain fault detection/diagnosis
1991 (English)In: Proceedings of the 11th Triennial World Congress of the International Federation of Automatic Control / [ed] Jaaksoo U., Utkin V.I., 1991, Vol. 4, no Elmsford, NY, United States, 373-378 p.Conference paper (Refereed)
Undermodelling can lead to poor behavior of fault detection (diagnosis) algorithms based on estimated parametric models. The aim of this paper is to study how to reduce this problem by working in the frequency domain and using flexible models. Simple and practically appealing frequency domain fault detection/diagnosis schemes are derived using recent results of Ljung (1987) on a high-order black-box identification theory.
Place, publisher, year, edition, pages
1991. Vol. 4, no Elmsford, NY, United States, 373-378 p.
Computer Programming--Algorithms, Mathematical Techniques--Frequency Domain Analysis, Black-box identification theory, Fault analysis, Fault detection, Control Systems
IdentifiersURN: urn:nbn:se:kth:diva-55440OAI: oai:DiVA.org:kth-55440DiVA: diva2:471583
the 11th Triennial World Congress of the International Federation of Automatic Control. Tallinn, USSR. 13 August 1990 - 17 August 1990
QC 20120104. Sponsors: IFAC; Int Assoc for Mathematics & Computer Simulation; IFIP; Int Federation of Operational Research Soc; Int Measurement Confederation2012-01-022012-01-022013-09-05Bibliographically approved