Modelization of reliability of functions of the network infrastructures of the cyberphysical systems

Abstract

Within the framework of this research, there has been developed a structural model of an industrial cyberphysical system architecture using cloud computing as a base platform has been developed. The proposed cyber-physical system integrates all elements in four levels of abstraction, based on a multi-agency approach. The proposed approach based on the Intelligent Data Analysis System of the Monitoring System allows searching and identifying vulnerable elements of the cyber-physical network infrastructure based on the cloud platform. The research has developed a model for securing the operation of network infrastructure facilities presented in the form of a measured multigraph, which forms a plan for collecting, analyzing and verifying data from the monitoring system to provide a consolidated assessment of the current state of the network elements. At the same time, the parameters for securing the infrastructure of the cyber-physical system and the cloud platform were selected as the top of the column. As arcs are given the relationships between the fixed dimensions of reliability, which reflect the relationship between the operation and the parameters of the connected nodes of the cyber-physical system, taking into account the current parameters of the column data streams. This allows us to identify system segments that will reduce the overhead needed to make changes. At the same time, the neural network approach is used to predict the continuous infrastructure of the cyber-physical system. The use of the proposed hybrid approach allowed us to predict the behavior of the infrastructure over time and to warn of possible failures in the operation of individual components and critical nodes.