Webinar - Applying Foundation Models to Power Grids and Emerging AI Paradigms
Dear Members,
We are pleased to announce the webinar from Dr. Ricardo Bessa from INESC TEC.
This webinar explores the potential and challenges of applying foundation models to power grid operation. It introduces the core concepts behind foundation models, with a particular focus on transformer-based architectures, and discusses how to identify and select promising use cases. Key challenges related to data availability, sharing, and governance are also examined, as these are critical to unlocking the full potential of such models. In addition, the talk highlights emerging AI paradigms that take a complementary approach: rather than relying solely on large-scale general knowledge, they integrate domain expertise with data. In this context, neuro-symbolic learning architectures are presented as a promising direction for combining expert knowledge with data-driven methods in power system applications.
Ricardo Bessa (b. 1983, Viseu), IEEE Fellow, earned his 5-years in Electrical and Computer Engineering (2006), M.Sc. in Data Analysis (2008), and Ph.D. in Sustainable Energy Systems (2013) from the University of Porto. He coordinates the Center for Power and Energy Systems at INESC TEC. His research spans energy forecasting, computational intelligence, and smart grids. He led projects like AI4REALNET and contributed to FP6 ANEMOS.plus, H2020 Smart4RES and H2020 InteGrid. Received the ESIG (Energy Systems Integration Group) Excellence Award (2022) for its contributions to renewable energy forecasting.
The webinar will be held virtually on the 13th of May 2026, from 17h to 18h30 (Portuguese time). This event will be held through ZOOM Platform. Use the following link to register:
https://tecnico-pt.zoom.us/meeting/register/sNgN0GluTn2Xm_jgQq7Lhg
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
Loading virtual attendance info...
Speakers
Ricardo Bessa of INESC-TEC
Applying Foundation Models to Power Grids and Emerging AI Paradigms
This talk explores the potential and challenges of applying foundation models to power grid operation. It introduces the core concepts behind foundation models, with a particular focus on transformer-based architectures, and discusses how to identify and select promising use cases. Key challenges related to data availability, sharing, and governance are also examined, as these are critical to unlocking the full potential of such models. In addition, the talk highlights emerging AI paradigms that take a complementary approach: rather than relying solely on large-scale general knowledge, they integrate domain expertise with data. In this context, neuro-symbolic learning architectures are presented as a promising direction for combining expert knowledge with data-driven methods in power system applications.
Biography:
Ricardo Bessa (b. 1983, Viseu), IEEE Fellow, earned his 5-years in Electrical and Computer Engineering (2006), M.Sc. in Data Analysis (2008), and Ph.D. in Sustainable Energy Systems (2013) from the University of Porto. He coordinates the Center for Power and Energy Systems at INESC TEC. His research spans energy forecasting, computational intelligence, and smart grids. He led projects like AI4REALNET and contributed to FP6 ANEMOS.plus, H2020 Smart4RES and H2020 InteGrid. Received the ESIG (Energy Systems Integration Group) Excellence Award (2022) for its contributions to renewable energy forecasting.
Agenda