A unified parameterized formulation of reasoning in fuzzy modeling and control (Q1808918): Difference between revisions

The authors of this paper construct a general and unified framework for the reasoning process in fuzzy modeling and control. They start with the basic elements of reasoning, the connective operators (AND, ALSO, and IF-THEN) and select a suitable parameterized family of triangular functions which are necessary for the purpsoe of this work. Next, the authors extend the binary operation to \(n\)-ary operation and show the validity of De Morgan law for \(n\)-ary operation, and thus they parameterize the reasoning formulation for multi-input single-output systems. In the case of crisp input variables it is shown that the two methods of inference from a rule set, first-aggregate-then-infer (FATI) and first-infer-then-aggregete (FITA), give the same fuzzy output. Also, the authors introduce four reasoning parameters, and thus the reasoning operation varies continuously among the extreme cases in each step of the inference. In order to obtain a simple computation, a fast algorithm for the calculation of a parameterized family of triangular functions is proposed. On the other hand, using an approximation for this family, the defuzzified output can be calculated directly from the individual consequent fuzzy sets of the rule set, by a simplified parameterized reasoning formulation. Several examples complete this paper.