Fuzzy Identification of Technological Objects

In practice, information on statistical characteristics of series mode interference is available only in certain cases. Moreover, for this or that reason input variables are measured inexactly, and their values can be specified with some uncertainty. The identification task is significantly complica...

Full description

Saved in:
Bibliographic Details
Main Authors: Mykhailo Horbiychuk, Taras Humeniuk, Dmytro Povarchuk
Format: Article
Language:English
Published: Lviv Polytechnic National University 2015-08-01
Series:Energy Engineering and Control Systems
Subjects:
Online Access:https://science.lpnu.ua/jeecs/all-volumes-and-issues/volume-1-number-1-2015/fuzzy-identification-technological-objects
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In practice, information on statistical characteristics of series mode interference is available only in certain cases. Moreover, for this or that reason input variables are measured inexactly, and their values can be specified with some uncertainty. The identification task is significantly complicated when the measuring signal passes through a natural channel with unknown statistical characteristics. This is, for instance, the case with well drilling when axial weight on the drilling bit and drilling bit rotations per minute are gauged by above-ground devices. The study resulted in developing a method of building polynomial empirical models for a case when input factors are fuzzy variables with known Gaussian membership functions. The output variable of the model was shown to be also a fuzzy variable in such conditions; the corresponding membership function on the basis of which the identification task is formulated was obtained. To determine the parameters of the model with fuzzy input factors, the least square method was applied, which allowed obtaining a respective formula incorporating the information on input factors fuzziness.
ISSN:2411-8028
2415-7287