Premium
Qualitative/quantitative simulation of process temporal behavior using clustered fuzzy digraphs
Author(s) -
Li R. F.,
Wang X. Z.
Publication year - 2001
Publication title -
aiche journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.958
H-Index - 167
eISSN - 1547-5905
pISSN - 0001-1541
DOI - 10.1002/aic.690470413
Subject(s) - digraph , process (computing) , computer science , fuzzy logic , artificial intelligence , cluster analysis , data mining , variable (mathematics) , machine learning , mathematics , mathematical analysis , combinatorics , operating system
A methodology is presented for qualitative modeling and simulation of the temporal behavior of individual variables and the composite process. The temporal behavior of a variable in a windowed time scale is captured categorically using principal‐component analysis, while the process is modeled as a causal digraph with interacting and recycle nodes. Rigorous, rather than ad hoc, reasoning rules can be devised for the causal digraph using a learning mechanism. Quantitative information can also be incorporated into the method with the introduction of fuzzy c‐means clustering without compromising the cognitive level of information embedded in the nodes and the digraph.