Design and analysis of diagnosis systems using structural by Mattias Krysander.

By Mattias Krysander.

Show description

Read Online or Download Design and analysis of diagnosis systems using structural methods PDF

Best analysis books

System Analysis and Modeling: About Models: 6th International Workshop, SAM 2010, Oslo, Norway, October 4-5, 2010, Revised Selected Papers

This ebook constitutes the throughly refereed post-proceedings of the sixth foreign Workshop on platforms research and Modeling, SAM 2010, held in collocation with types 2010 in Oslo, Norway in October 2010. The 15 revised complete papers awarded went via rounds of reviewing and development. The papers are geared up in topical sections on modularity, composition, choreography, program of SDL and UML; SDL language profiles; code new release and version ameliorations; verification and research; and consumer specifications notification.

Additional info for Design and analysis of diagnosis systems using structural methods

Sample text

34), and we are particularly interested in sets with small cardinality, because the number of models in ω is equal to the number of tests in the diagnosis system based on ω. It is desirable to use few tests to reduce the online computational cost of running the diagnosis system. 11. 11 is not sound. However, if we add a test δ3 such that M3 = {e2 , e3 , e4 , e5 }, Φ3 = assump(M3 ) = {NF}, T3 = y1 − y2 , and R3 = R \ {0}, then a sound and complete diagnosis system is obtained. 2 that the minimum number of models that must be tested to obtain a sound diagnosis system is three.

The first starts from a given set of null hypotheses Hi0 , and the second from the diagnosis model M of the system to be diagnosed. 2 Starting From Null Hypotheses One way of starting the design of a diagnosis system is simply to decide which null hypotheses to test, and then construct a suitable test quantity and rejection region for each hypothesis test. One straightforward strategy is for example to have one diagnosis test for each of the system behavioral-modes. This is especially attractive when only single faults are considered.

One alternative is to eliminate all unknown variables in Mi to derive equations containing only known variables. 19) and is said to be a consistency relation, also called analytical redundancy relation or parity relation. The consistency relation a(z) = 0 can then used in the construction of a test quantity. Some works on the consistency relation based approach are (Staroswiecki and Comtet-Varga, 2001; Frisk and Åslund, 2005; Yu and Shields, 1997; Basseville and Nikiforov, 1993). , 1989) or to make an identification based design (Isermann, 1993).

Download PDF sample

Rated 4.84 of 5 – based on 48 votes