Development of Digital Twins to Support the Functioning of Cyber-physical Systems

Abstract

The peculiarities of developing a digital twin for cyberphysical systems using an analytical model of the investigated process are considered. The proposed approach takes into account the conceptual uncertainty of the analytical model parameters with the subsequent passive identification by adapting the analytical model to the dynamic characteristics of the real physical process. In the article, the digital twin development is carried out on the analytical model air heating process example with the help of an electric calorifier. The analytical model uncertain parameters of the electric heater are analyzed and an integral quality index is proposed to evaluate the dynamic model adequacy in the process of air heating. For electric heater model adaptation, the passive identification algorithm of uncertain parameters is developed, in which deviations of the mathematical model uncertain coefficients are minimized. A numerical study of the considered approach has been carried out. It is shown that uncertain parameters’ passive identification of the model belongs to the problem of single-extremal optimization. The considered modeling examples confirmed the effectiveness of the proposed approach to digital twins’ development for cyber-physical systems.