Machine learning assisted digital twin for event identification in electrical power system

The challenges of stable operation in the electrical power system are increasing with the infrastructure shifting of the power grid from the centralized energy supply with fossil fuels towards sustainable energy generation. The predominantly RES plants, due to the non-linear electronic switch, have brought harmonic oscillations into the power grid. These changes lead to difficulties in stable operation, reduction of outages and management of variations in electric power systems. The emergence of the Digital Twin in the power system brings the opportunity to overcome these challenges. Digital Twin is a digital information model that accurately represents the state of every asset in a physical system. It can be used not only to monitor the operation states with actionable insights of physical components to drive optimized operation but also to generate abundant data by simulation according to the guidance on design limits of physical systems. The work addresses the topic of the origin of the Digital Twin concept and how it can be utilized in the optimization of power grid operation.