FROM INFORMATION-MEASURING TO CYBER-PHYSICAL SYSTEMS

UDC 681.2: 621.337.12
DOI: 10.36871/2618-9976.2023.01.007

Authors

Tatyana V. Lazovskaya,
Senior Lecturer, Peter the Great St. Petersburg Polytechnic University, Russia, Saint-Petersburg, Russia
Galina F. Malykhina,
Doctor of Technical Sciences, Professor, Peter the Great St. Petersburg Polytechnic University, Saint-Petersburg, Russia
Dmitry A. Tarkhov,
Doctor of Technical Sciences, Professor, Peter the Great St. Petersburg Polytechnic University, Saint-Petersburg, Russia

Abstract

The expansion of the capabilities of informationmeasuring systems in the framework of the fourth industrial revolution can be associated with the use of artificial intelligence methods and algorithms for measurement tasks. Approaches based on intelligent technologies are methodically most developed within the framework of the concept of cyberphysical systems, therefore it is advisable to analyze the possibilities of using these technologies in informationmeasuring systems. It is expedient to use measuring systems for obtaining digital twins of controlled physical objects, forecasting and prediction, diagnosing and classification. The authors propose, along with wellknown artificial intelligence technologies using classical and deep neural networks, to use physically informed neural networks, which reveal the black box model to a certain extent. Physically informed artificial neural networks are more suitable for measurement tasks, since they combine knowledge of physical laws and learning processes based on a constantly updated set of measurements and a digital twin of a physical object.

Keywords

Information-measuring systems, Cyber-physical systems, Digital twin, Machine learning, Physically informed neural networks serial systems