Skip links

ECLIPSE DIGITAL at EPEi 2026: Exploring AI-Based Solutions for More Resilient Power Grids

ECLIPSE DIGITAL participated in the 14th IEEE International Conference and Exposition on Electrical and Power Engineering (EPEi 2026), held in Iași, Romania, from 21 to 23 October 2026. The event, organised by the Gheorghe Asachi Technical University of Iași, brought together around 500 participants from the scientific community, industry and other sectors to exchange knowledge and discuss the latest developments in electrical and power engineering.

Representing the project, UPB contributed to the dissemination of scientific knowledge related to ECLIPSE DIGITAL through the presentation of a research paper focused on the application of Artificial Intelligence (AI) and Graph Neural Networks (GNNs) to electrical distribution networks.

Improving Distribution Grid Monitoring with Graph Neural Networks

As electricity distribution networks evolve towards more active and digitalised systems, reliable monitoring and control become increasingly important. Distribution System State Estimation (DSSE) plays a key role in this process, providing an overview of grid conditions even when measurements are incomplete or unavailable.

The paper presented by UPB investigated how Graph Neural Networks can support DSSE under degraded observability conditions. The research evaluated the performance of GNN-based state estimation using two standard test networks: the IEEE 33-bus and IEEE 37-bus systems.

The results highlighted the robustness of the proposed approach when the availability of input measurements is reduced, demonstrating the potential of GNNs to support reliable grid monitoring in challenging conditions.

Supporting Grid Stability and Resilience

The discussions at EPEi 2026 highlighted the growing importance of Artificial Intelligence in the evolution of electrical networks. The presentation also provided an opportunity to explore how data-driven solutions can contribute to improving network monitoring, situational awareness and operational resilience.

These research findings are particularly relevant to the objectives of the Romanian demonstrator within ECLIPSE DIGITAL, where reliable monitoring and a better understanding of grid conditions are important aspects of supporting a more resilient energy system.

Participation in EPEi 2026 enabled ECLIPSE DIGITAL to share its research with the international scientific and technical community, while contributing to the wider exchange of knowledge on AI-based solutions for the digitalisation of energy systems.

The full paper is available online for those interested in exploring the research in more detail.

EPEI_Paper_IEEE33_37_Preprint.pdf (44 downloads )

Leave a comment

Light