An Approach to Predictive Digital Twin Modelling of Complex Multmode Objects

The development of modelling tools is a key factor in the creation and implementation of intelligent (predictive) digital twins. As a rule, many complex objects (CO) are multi-mode, i.e. a priori they have the property of non-monotony. Changes in the content of the goals and tasks required from the object, as well as the destructive effects of the external environment, lead to uncertainty in the CO functioning, which is associated with the intensity and nature of their various modes of operation. Purpose: the stated above situation requires a new approach to the study of multimode non-monotonic systems in conditions of significant uncertainty. Results: the object operation modes in the form of vertices of functional integrity schemes of the general logical-probabilistic method (GLPM) of calculus shows that the functional structure of a multi-mode object becomes non-monotonic. This situation, as well as lack of knowledge about the intensity and nature of these modes’ use, required the authors to develop a model-algorithmic superstructure over the GLPM. It is based on the concept of a parametric genome of functional structures of multi-mode CO. Practical significance: the proposed approach made it possible to evaluate the indicators of structural and functional reliability and sustainability of such objects in the absence of knowledge about their operating mode cyclograms. Based on the approach outlined in the article, it is necessary to develop new methods that allow monitoring and management of CO operating modes.