SUBSUMPTION AND IMAGES RECOGNITION ALGORITHMS OF THE ANALYSED MATERIALS IN THE INTELLECTUAL DATAMEASURING SYSTEM

D.Yu. Muromtsev, Z.M. Selivanova, D.S. Kurenkov, V.I. Pavlov, P.S. Belyaev, O.V. Trapeznikova
DOI: 10.25699/SSSB.2020.30.62752 Download PDF
Abstract: Providing that the commercially produced thermal insulation materials quality depends on their thermophysical properties as well as a number of influencing factors, an actual problem is both accurate and operational non-destructive testing provision of the thermophysical properties parameters at the measurement uncertainty conditions. The problem solution based on the intelligent data-measuring system application aimed to insulation materials thermal properties determination is proposed. The study is targeted to both the accuracy and efficiency testing increasing of the thermophysical materials properties at their production and operation in order to preserve the thermal insulation properties under the destabilizing factors influence. In the research context the artificial intelligence methods were applied at the decisions making in a fuzzy environment. Aimed to thermophysical measurements adaptation to the materials class as well as the next non-destructive testing method selecting both the problems and objects classifying algorithms at the thermal conductivity analysis were solved. The pattern recognition algorithm was developed in accordance with the category of received measurement information (fuzzy or reliable) at the thermophysical measurements. The developed algorithm accuracy and efficiency is experimentally proved.
Index terms: objects classification, pattern recognition, fuzzy sets, intelligent data-measuring system, thermophysical properties

Contacts

Russia, 659305, Altai region, Biysk,
Trofimova Street, 27, room 404B
Tel. + 7-923-162-93-27
(executive secretary -
Golykh Roman Nikolayevich)
e-mail: info@s-sibsb.ru

The certificate