This thesis presents an approach for qualitative model based fault diagnosis via parameter estimation. Estimating parameters qualitatively in this research is based on the relationships among involved qualitative quantities. To maintain the consistency of ordinal relationships an inequality reasoner is developed. Existing inequality reasoners (or commonly in the AI community referred to as quantity lattices) infer relationships between quantities involved in a system (qualitative or quantitative) based on the real values associated with them. The novel feature of our inequality reasoner is that it infers and maintains relationships among pure qualitative quantities. Inferring ordinal relationships between sign based quantities removes the traditional drawback of qualitative domain, the lack of discriminating power.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:493495 |
Date | January 2008 |
Creators | Al-Ballaa, Hind Rasheed |
Publisher | University of Aberdeen |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Page generated in 0.0022 seconds