Monitoring Systems for Critical Infrastructure

This chapter presents research related to real-time monitoring of Key Performance Indicators (KPIs) for critical infrastructure (CI) technological processes with the aim of identifying deviations in technological processes and preventing attacks (failures) by analyzing abnormal equipment behavior, changes in load modes, resource consumption, etc. In this context, cloud technologies open up new opportunities for scalable, reliable, and secure performance monitoring with remote access and centralized management. The KPI cortege model has been further developed, which allows you to systematize monitoring parameters in the Information and Communication Systems of CI objects, formalize automatic data processing, support integration with analytical cloud platforms, and identify deviations (anomalies), cyber incidents, degradation, overload, or sabotage. As a result, it enables the generation of aggregated KPIs for daily monitoring by operators and IT services. An example of the implementation of the developed model for the electric arc treatment (EAT) technological process in unmanned aerial vehicle (UAV) engineering, which is part of CI in the industrial and defense-industrial production sector, is considered. In addition, a KPI monitoring scheme for the EAT technological process has been developed; an example of the model implementation infrastructure in the Azure cloud and a data model in JSON format is provided, as well as a general scheme of the EAT technological process in UAV engineering based on cloud technologies. Verification of the model based on current, voltage, and temperature data proved its functional correctness, resistance to changing conditions, and high applied value for monitoring technological processes in CI. The model can become the basis for the development of a digital twin of the production system, ensuring reliability, predictability, and security in the context of digitalization and cyber risks. Integration with AI/ML algorithms such as LSTM opens up prospects for advanced analytics, adaptive process control, and automatic decision-making in real time.