IoT-Based Energy Management Assistant Architecture Design

The Internet of Things (IoT) provides opportunities to control interconnected smart devices via pre-designed scenarios with littleor no human involvement. Due to the need for systematic improvement of industrial energy efficiency, the relevance of IoT-based energy management systems (EMS) is constantly increasing. Industrial IoT (IIoT)-enhanced EMS are created to support the digital transformation of enterprises. They increase the transparency of energy consumption statistics, enhance the personnel awareness of energy losses, provide predictive analytics tools for forecasting potential industrial accidents and future energy demand. This paper hence provides a system architecture of the energy management assistant that can be used at enterprises to archive the aforementioned aims. Authors identify the relevance of EMS, list the conditions at modern enterprises that define the requirements for the energy management assistant, demonstrate the identified requirements in the UML diagrams. Specification of energy planning and energy monitoring stages demonstrate the possibilities of the designed system architecture. For demonstration purposes, authors present examples of linear and nonlinear regression models implemented to specify energy consumption target functions based on real data. Finally, future research directions and open energy management problems are presented. The analysis has been carried out within the priority area of scientific development established in the National Research University Higher School of Economics - "Research on control methods in Cyber-Physical Systems".