LEVERAGING LARGE LANGUAGE MODELS FOR A RESILIENT ELECTRICITY SUPPLY SYSTEM
| Tipo de publicação: | Artigo |
| Citação: | Binjos, A.; Mermoud, G.; Guo, B.; Vianin, J.; Pouget, J.; Wannier, D. Leveraging Large Language Models for a Resilient Electricity Supply System. CIRED 2025 Conference, Paper 983, 16–19 June 2025, pp. 1–5. https://doi.org/10.1049/icp.2025.2162 |
| Journal: | CIRED 2025 Conference |
| Número: | Paper 983 |
| Ano: | 2025 |
| Mês: | June |
| Páginas: | 1–5 |
| Nota: | Conference dates: 16–19 June 2025. These authors contributed equally: Abdullah Binjos, Grégory Mermoud, Baoling Guo. Corresponding author: David Wannier. |
| URL: | https://doi.org/10.1049/icp.20... |
| DOI: | 10.1049/icp.2025.2162 |
| Resumo: | This work introduces Large Language Model (LLM)-based agent that interacts with a resilient microgrid platform, enabling users to interact via intuitive text-based queries. The system provides both expert knowledge, scenario testing, and decision-making support related to resilient electricity systems while ensuring data privacy through local hosting and GDPR compliance. Users can explore strategies such as leveraging local energy production and storage units, shedding less critical loads to mitigate blackouts' impact. Enhanced with Geographic Information System (GIS) capabilities, the platform, developed during the European project OpenGIS4ET [1], offers actionable insights and visualizations to support microgrid resilience. This framework, tested with the use case and dataset relevant to the wastewater treatment plant (STEP) in Neuchâtel, is designed to be scalable and replicable across different regions in a next step. |
| Palavras-chave: | AI Agents, Large Language Model (LLM), Microgrid Resilience, Retrieval Augmented Generation (RAG), Text-to-SQL |
| Autores | |
| Adicionado por: | [] |
| Total mark: | 0 |
|
Anexos
|
|
|
|
|
Notas
|
|
|
|
|
|
Tópicos
|
|
|
|
|
