Empirical comparison of methods of fuzzy relational identification (Q3986447)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 28919
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Empirical comparison of methods of fuzzy relational identification |
scientific article; zbMATH DE number 28919 |
Statements
Empirical comparison of methods of fuzzy relational identification (English)
0 references
27 June 1992
0 references
relational model
0 references
discrete-time fuzzy relational models
0 references
reference fuzzy sets
0 references
0.8472802639007568
0 references
0.8000809550285339
0 references
0.7973376512527466
0 references
0.7958738207817078
0 references
Discrete-time fuzzy relational models describe relationships between reference fuzzy sets of system's variables defined in the corresponding universes of discourse. The model discussed in the paper takes on the following form, \[ Y=U_ 1\circ U_ 2\circ\cdots\circ U_ p\circ R, \] where \(U_ 1,U_ 2,\dots,U_ p\) are the grades of membership of the reference fuzzy sets expressed in \(U\) while \(Y\) stands for a vector of membership values of the corresponding reference sets defined in \(X\). These vectors are related via composition operator ``\(\circ\)'' (in general, this operator can be viewed as the max-\(t\) convolution with ``\(t\)'' denoting a triangular norm; in particular one can refer to the max-min and max-product composition). The paper is concerned with an empirical comparison of several methods of identification of the above model, namely \textit{E. Czogala} and the reviewer, Fuzzy Sets Syst. 6, 73- 83 (1981; Zbl 0473.93070); the reviewer, ibid. 13, 153-167 (1984; Zbl 0554.93070); \textit{C. Lu} and \textit{Y. Xu}, IEEE Trans. Syst. Man Cybern. SMC-17, 683-689 (1987; Zbl 0643.93018). The numerical experiments deal with a relational model of a fed-batch fermenter and involve several parametric studies including those pertaining to noise immunity of the methods. Among these three algorithms, the method proposed by the reviewer produced the best overall performance.
0 references