Virtualization topologies of IEDs with the IEC61850 protocol and their application in edge devices

The application domain of this work lies within smart grids, with a specific focus on primary substations. These critical infrastructures serve as the backbone of power distribution networks, where efficient management of data exchange and computational resources is essential for reliable and scalable operation. Our work explores the reliability and scalability of containerized IEC 61850 logical nodes and the use of LSTM models for data network load forecasting. The deployment demonstrated satisfactory performance at SV rates of 4000 and 4800 packets per second, confirming the feasibility of hosting multiple containerized nodes without compromising reliability. Additionally, LSTM models were applied to forecast data network loads, focusing on peak demand prediction. The results highlight the potential of integrating predictive analytics with containerized logical nodes to optimize computational resources and enhance network performance in IEC 61850 environments.