Self-Calibration of Intelligent Sensors Based on Virtual Analysis

Abstract

The paper considers the methods of calibration of measuring instruments using virtual analyzers. The advantages of using a virtual quality analyzer to determine the metrological condition of the measuring instrument are shown, where the main attention is paid to the accuracy of measurement. To obtain accurate calibration data, large-volume stored data is used, which is processed by a neural network model of a virtual analyzer trained with a large volume of samples. The developed self-calibration method based on the processed data of the virtual analyzer evaluates the metrological condition of the measuring instrument and generates calibration data with which the intelligent sensor performs self-calibration. The article also analyzes the trends in the development of intelligent measuring devices, in particular the functions of intelligent sensors within the framework of "Industry 4.0".