FUNDAMENTAL ASPECTS OF METROLOGICAL SUPPORT IN IoT

Abstract

The application of intelligent sensors, network technologies, and machine learning in IoT and industry is increas- ingly widespread as a part of the development and implementation of Industry 4.0, Industry 5.0, and Smart City. It is necessary to review the fundamental principles of metrological support for production. This includes calibration, estimation of measurement uncertainty, traceability, and processing of large data sets to reproduce and compare the results of measurements of physical quan- tities remotely. Modern smart sensors are cost-effective, which makes traditional sensor calibration methods increasingly uneco- nomical. The utilization of advanced networking technologies, along with machine learning, complicates the pre-processing of measured values. Therefore, new solutions are required when it comes to implementing digital metrology. In this article, a metrological framework for the full life cycle of measured data in IoT is presented. It ensures transparency, comparability, consistent quality and reliability of measured data, processing methods and results. The OPC-UA digital data com- munication standard is considered, which provides a single interface for exchanging digital data with devices from different manu- facturers or via different protocols. The syntax of a machine-readable representation of SI units and derived quantities as well as the structure of the sensor network metadata model are also described. Special emphasis is placed on dynamic calibration of sen- sors, determining measurement uncertainty in sensor networks, and implementing digital calibration certificates in IoT and industry.