R²D² Workshop in Serbia Highlights Innovation and Future Opportunities

R²D² Workshop in Serbia Highlights Innovation and Future Opportunities

At the end of May 2025, the mountain resort of Kopaonik became a meeting point for experts, researchers, and industry leaders in the energy sector. As part of the 37th Session of CIGRE Serbia, the R²D² project held a workshop showcasing the progress and achievements of its Serbian Pilot Site, titled “R²D², Reliability, Resilience and Defense technology for the grid – Achievement of the Serbian Pilot Site”. The workshop was organised by the R²D² partners EMS Services (EMSS),  Institute Mihajlo Pupin (IMP), and Security Coordination Centre SCC Ltd. Belgrade.

The event drew around 50 participants from across Serbia and neighboring countries, eager to learn how R²D² is shaping the future of grid reliability, resilience, and defense.

Bringing Innovation Closer to Practice

During the session, three tools developed within the Serbian Pilot Site were in the spotlight:

  • The KSI Tool (PRECOG product): Provides an additional layer of trust for electricity grid operations. It uses blockchain-based signatures to secure data exchanges between Transmission System Operators (TSOs) and Regional Coordination Centres (RCCs), ensuring integrity, preventing manipulation, and strengthening cybersecurity in grid and IoT data processes.

  • The Remedial Action Automation Tool (IRIS product): Automates the determination and execution of remedial actions, which are essential for maintaining operational security in power grids. By enabling fast and accurate responses to unexpected system changes, it supports dispatchers in emergencies, reduces manual workload, and strengthens grid resilience at critical moments.
  • The Outage Planning (OP) Tool (EMMA product): Designed to improve outage planning coordination at national and regional levels. It streamlines a process that often relies on human negotiation and experience, providing faster security analysis, scenario testing, and better visualization (such as automated Gantt charts) for Transmission System Operators and RCCs.

Each represents a step forward in tackling complex challenges in energy system management. While still at the stage of concept validation, they have already proven their value in real environments — a rare achievement in such an early phase of development.

The discussion revealed strong interest from the audience, particularly in the RA automation tool, which supports operators in decision-making during emergencies. Rather than replacing human expertise, the tool is designed to help dispatchers handle the overwhelming flow of data that modern power systems generate.

Achievements and Challenges

One of the major takeaways from the workshop was the recognition that these tools are not only practical but also carry academic and scientific potential, with applications in research and education. Their proactive approach,  anticipating problems and offering solutions in advance,  was praised as essential for building a more resilient energy system.

At the same time, participants noted the road ahead: to reach full market readiness (TRL9), tools will need to become more user-friendly, more responsive, and equipped with stronger cybersecurity features. Some solutions may also deliver their greatest impact at the Pan-European level, requiring alignment with organizations such as ENTSO-E or ACER.

Looking Forward

Perhaps the most exciting news came during the discussion of next steps. The Serbian partners are already exploring commercialisation opportunities, with trilateral talks under way and the first agreements expected before the end of 2025.

What started as pilot-site experimentation is now on track to become part of the region’s operational toolbox — a clear sign that R²D² is not only advancing research but also creating tangible pathways to strengthen Europe’s energy resilience.

 

 

Replay the session conducted in Serbian here:

Further information:

info@r2d2project.eu

This project has received funding from the  European Union’s Horizon Europe research and innovation programme under grant agreement No 101075714.


EMMA-SURVEILLANCE: Enhancing Substation Security with AI-Powered Visual Detection

EMMA-SURVEILLANCE: Enhancing Substation Security with AI-Powered Visual Detection

Within the context of the R2D2 project, an innovative tool known as EMMA-SURVEILLANCE is currently in development to bolster the security of critical facilities located in electrical substation transformer centers. This solution integrates an artificial vision algorithm that has been honed by retraining the well-known YOLO (You Only Look Once) model. This algorithm is equipped with the ability to efficiently detect fires, smoke, and the presence of animals in the vicinity of the substation. The purpose of identifying fires and smoke is to promptly alert personnel to potential emergencies, ensuring a swift and effective response. Furthermore, recognizing animals is of utmost importance, as many of them tend to come into contact with the substation structures, posing the risk of electrocution and causing significant disruptions to the electrical system. This model will be deployed in a stationary camera situated within the corresponding pilot substation.

In this initial phase, the algorithm has exhibited remarkable precision, achieving an F1 score of 0.84. This achievement is particularly noteworthy, especially when considering the project’s early stages. As the next steps, the plan involves expanding the dataset, with the objective of collecting more images of fires and smoke to enhance the model’s accuracy. Concurrently, the load testing phase will be initiated, assessing the model’s inference capacity when operating in a real camera and continuously processing real-time video streams. This process is critical to ensure that the algorithm can perform effectively without significant delays, thereby guaranteeing its practicality for real-time monitoring scenarios.

Some examples of recorded images of the AI-powered visual detection.

Further information:

Ugo Stecchi (Project coordinator)

This project has received funding from the  European Union’s Horizon Europe research and innovation programme under grant agreement No 101075714.


Privacy Preference Center