A COMPARATIVE ANALYSIS OF FUZZY T-NORM APPROACHES TO THE MEASUREMENT UNCERTAINTY EVALUATION

Claudio De Capua, Emilia Romeo

Abstract

The demands of statistical investigations in measurements inspired the remarkable development of probabilistic methods. However, the probability theory didn’t prove to be fully adequate for all types of uncertainty. Probability theory is excellent if the ambiguity is to be modelled, but its attempts to describe vagueness is quite inconsistent with common sense. Fuzzy theorists have often argued that a major motive behind the theory of fuzzy sets has been the treatment of uncertainty. In particular way, it’s well accepted that a measurement result (no matter what kind of instruments we are using in our process) is just a number which is only known to lie within an interval, and this is the reason for which fuzzy sets can be successfully applied. To consider both systematic and random effect of measurement operation, in agreement with [4], we have chosen to use Random Fuzzy Variables, proposing to describe the correlation or interaction of repeated measurements by triangular norm based arithmetics.

Download

IMEKO-TC4-2004-018.pdf

DOI

-

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings